Reduction by the Methylreductase System in Methanobacterium Bryantii WILLIAM B

Home , Methanofuran, Tetrahydromethanopterin

JOURNAL OF BACTERIOLOGY, Jan. 1987, p. 87-92 Vol. 169, No. 1 0021-9193/87/010087-06$02.00/0 Copyright © 1987, American Society for Microbiology Inhibition by Corrins of the ATP-Dependent Activation and CO2 Reduction by the Methylreductase System in Methanobacterium bryantii WILLIAM B. WHITMAN'* AND RALPH S. WOLFE2 Department of Microbiology, University of Georgia, Athens, Georgia 30602,1 and Department of Microbiology, University ofIllinois, Urbana, Illinois 618012 Received 1 August 1986/Accepted 28 September 1986 Corrins inhibited the ATP-dependent activation of the methylreductase system and the methyl coenzyme M-dependent reduction of CO2 in extracts of Methanobacterium bryantii resolved from low-molecular-weight factors. The concentrations of cobinamides and cobamides required for one-half of maximal inhibition of the ATP-depen4ent activation were between 1 and 5 ,M. Cobinamides were more inhibitory at lower concentra- tiops than cobamides. Deoxyadenosylcobalamin was not inhibitory at concentrations up to 25 ,uM. The inhibition of CO2 reduction was competitive with respect to CO2. The concentration of methylcobalamin required for one-half of maximal inhibition was 5 ,M. Other cobamideg inhibited at similar concentrations, but diaquacobinami4e inhibited at lower concentrations. With respect to their affinities and specificities for corrins, inhibition of both the ATP-dependent activation'and CO2 reduction closely resembled the corrin- dependent activation of the methylreductase described in similar extracts (W. B. Whitman and R. S. Wolfe, J. Bacteriol. 164:165-172, 1985). However, whether the multiple effects of corrins are due to action at a single site is unknown.

The effect of corrins (cobamides and cobinamides) on in CO2 reduction. However, the precise role of CH3-S-CoM is vitro methanogenesis is enigmatic. On one hand, small not known (9, 20). amounts of corrins stimulate the methyl coenzyme M (CH3- We described a corrin-dependent activation of the S-CoM) methylreductase system three- to fivefold (29). On methylreductase system in extracts of the mesophile the other hand, as reportfd in this communication, corrins Methanobacterium bryantii (29). This activation was unlike also inhibit CO2 reduction as well as ATP activation of the previously described effects of corrins in methanogenesis. It methylreductase system. This system (23) catalyzes the required catalytic amounts of corrins. Qther workers have reduction of CH3-S-CoM to form methane: shown that methylcobalamin (MeCbl) is a substrate for CH3-S-COM + H2 + CH4 methanogenesis in extracts of M. bryantii and Methanosar- Mg2+,2+HS-CoMATP, B, FAD cina barkeri (3, 4, 30). In our studies, corrins stimulated the demethylation of CH3-S-CoM. In methylotrophic Most of our information about the methylreductase system methanogens, a cobamide-containing protein is required for comes from studies of Methanobacterium thermoautotro- methylation of HS-CoM (2, 25-27). Furthermore, the phys- phicum. In M. thermoautotrophicum, the enzyme system iological significance of the coffin-dependent activation of contains at least four proteins (11, 17). Component C, which the methylreductase system was not certain. Although acti- contains the nickel tetrapyrrole factor F-430 and tightly vation required low concentrations of corrins, the coenzyme bound CoM, and the colorless A2 protein, which binds to and methyl forms were not required, and the greatest stim- N6-ATP agarose, have been purified to homogeneity (6, 7, ulation was obtained with n9nphysiological cobinamides. 14, 22). The methylreductase system also requires a low- Therefore, we also investigated the effects of cortins on molecular-weight factor of unknown function called compo- other reactions of the methylreductase system, including the nent B, Mg2+, flavin adenine dinucleotide (FAD), and cata- CH3-S-CoM-dependent reduction of C02, which may be the lytic amounts of ATP (10, 17, 18). physiological reaction of this system. The results of these Extracts of M. thermoautotrophicum also catalyze a CH3- studies are described in this report. S-CoM-dependent reduction of CO2 (9, 20): MF, MP, CH3-S-CoM MATERIALS AND METHODS Mg2+, ATP, B, FAD Methylreductase assays. Growth and anaerobic prepara- tion of cell extracts of M. bryantii M.o.H. were described In addition to the components of the methylreductase sys- previously (28, 29). Low-molecular-weight components of tem, this reaction requires two additional coenzymes, the extracts were removed by anaerobic chromatography on methanofuran (MF) and tetrahydromethanopterin (MP) (9, Sephadex G-25 (29). In a total volume of 0.2 ml, the 13, 15, 16, 21, 24). These coenzymes are additional C-1 methylreductase assay contained p.2 ml of Sephadex-treated carriers at the formyl, methine, methylene, and methyl extract (about 3 mg of protein), 2 ,umol of CH3-S-CoM, 2 levels of oxidation (8, 15), and some of the enzymes neces- ,umol of MgCl2, 10 ,ul of partially purified component B, 50 sary for their activity have been identified in extracts (5, 12). nmol of disodium ATP, and an H2 atmosphere (29). Methane Small amounts of CH3-S-CoM are also necessary to initiate was measured as described previously (28), and the velocity of the reaction was calculated from the linear portion of the * Corresponding author. time course of the reaction by linear regression analysis (28). 87 88 WHITMAN AND WOLFE J. BACTERIOL.

greater than 50 puM with 50 nmol of ATP to S to 15 puM with 20 nmol of ATP and 5 ,uM with 10 nmol of ATP (Fig. 1). Because ATP is not required for activation by cobalamins (29), these results implied that CN-Cbl inhibited the activation of the methylreductase by ATP. Methylreductase assays in extracts of M. bryantii contain a brief lag which is dependent on ATP (28). CN-Cbl increased the ATP-dependent lag early in the methylreductase assay (Fig. 2). Phosphoenolpyruvate, which acts like an ATP-generating system in these extracts (28), shortened the lag in the absence of CN-Cbl from 5 to 3 min. In the presence of CN-Cbl, phosphoenolpyruvate shortened the lag from 9 to 6 min, but it did not eliminate the effect of CN-Cbl (Fig. 2). Other cobalamins, cobamides, and cobinamides also length- ened the ATP-dependent lag (data not shown). This effect I 10 was especially dramatic for diaquacobinamnide (Aq2Cbi), E which activated the methylreductase system at very low concentrations (29). Even in the presence of high concentrations of ATP, 25 ,uM Aq2Cbi inhibited the methylreductase (Fig. 2). Phosphoenolpyruvate abolished the inhibition ofthe 20l nmo AT velocity of the methylreductase reaction. However, the ATP-dependent lag was only partially reduced (Fig. 2). The failure of phosphoenolpyruvate to abolish the inhibition by corrins also provided evidence that inhibition was not due to 0 nmol ATP the stimulation of an additional corrin-dependent reaction which competed with the methylreductase for ATP. 10 20 30 40 50 The inhibition by corrins appeared to be a direct effect on the ATP-dependent activation of the methylreductase. The [cobolamin] (EpLM) ATP-dependent activation is complete during the first few minutes of the methylreductase assay, and ATP is rapidly FIG. 1. Activation of the methylreductase by cobalamins at high degraded in these extracts (19, 28). If the activation is the and low concentrations of ATP. The cobalamin used was CN-Cbl. target of inhibition by corrins, corrins added after the No activity was obtained in the absence of ATP in the presence or activation was complete should no longer inhibit. MeCbl (25 absence of cobalamin. ,uM) inhibited the methylreductase only when it was added with low concentrations of ATP (Table 1). When MeCbl was The lag ofthe assay was equal to the intercept in the abscissa added after the activation by low concentrations of ATP, it of the linear portion of the reaction. The rate of activation activated. Like CN-Cbl (Fig. 1), when MeCbl was added was taken to be the reciprocal of the lag (1/lag). One unit of with high concentrations of ATP, activation was also ob- methylreductase activity is 1 nmol of CH4 produced per min. served. Therefore, the reversal of inhibition observed with CH3-S-CoM-dependent CO2 reduction. Assays were iden- ATP and phosphoenolpyruvate was not due to an indirect tical to the methylreductase assays except that the concentration of CH3-S-CoM was reduced to 200 nmol and 1 ,ug of MF (or carbon dioxide reduction factor), 2.5 ,ug of MP, 5% C02, and 50 mM potassium piperazine-N,N'-bis(2-ethane- sulfonate) (PIPES), pH 7.2, were added. Under these con- +PEP ditions, the addition of 5% CO2 to the gas atmosphere of the 800 assay vials caused the pH of the extract to decrease by less than 0.05. Assays were continued until there was no further E increase in the amount of CH4 (at least 60 min). The total 0" 600- CO2 reduced was determined as the total CH4 formed in the presence of CO2 minus the total CH4 formed in the absence of added CO2 (5). 400- Materials. Corrins, CH37S-CoM, and component B were prepared as described previously (29). MF and MP were the gift of J. Leigh. All common reagents were analytical grade or better. Biochemicals were obtained from Sigma Chemical 200- Co. (St. Louis, Mo.).

RESULTS 10 20 30 10 20 30 Inhibition of the ATP-dependent activation. When the TIME (min) TIME (min) methylreductase was assayed at suboptimal concentrations FIG. 2. Effect of phosphoenolpyruvate on the corrin activation of ATP, the activation by cyanocobalamin (CN-Cbl) was of the methylreductase. ATP (50 nmol) was present in all assays. greatly reduced (Fig. 1). Furthermore, high concentrations Phosphoenolpyruvate (PEP; 2 ,umol) was added where indicated. of CN-Cbl were actually inhibitory, so that the concentration Symbols: 0, no corrins added; 0, 25 pLM CN-Cbl added; A, 25 ,uM of CN-Cbl required for maximal activity was reduced from Aq2Cbi added. VOL. 169, 1987 INHIBITION BY CORRINS 89

TABLE 1. Corrin inhibition of the ATP-dependent activation of activation declined by 65% in the presence of high concen- the methylreductase system trations of Aq2Cbi. This result confirmed that the apparent Additiona at time (min): inhibition of the ATP-dependent activation was independent * ~~~~Methylreductase of the final velocity of the methylreductase reaction. 0 14 activityb Inhibition of CH3-S-CoM-dependent CO2 reduction. CO2 50 nmol of ATP 1.1 reduction was measured in a coupled assay with the 50 nmol of ATP + MeCbl 9.2 methylreductase. Corrins either activated or inhibited the 50 nmol of ATP MeCbl 8.0 methylreductase system, depending on the concentration of 20 nmol of ATP 1.0 ATP. Therefore, it was important to determine if the 20 nmol of ATP + MeCbl 0.4 methylreductase was inhibited under the conditions of CO2 20 nmol of ATP MeCbl 4.4 reduction. The addition of MP or MF had little effect on the a The methylreductase assay also contained 3.1 mg of protein, 5 ,umol of methylreductase or the activation by corrins (Table 2). In the MgCl2, 2 ,umol of CH3-S-CoM, component B, and H2 gas. Where indicated, 5 presence of 20% CO2, the methylreductase was stimulated, nmol of MeCbl (25 ,uM) was added. and the activation by corrins decreased somewhat. These b Units of methylreductase per milligram of protein. effects were attributed largely to the decrease in pH caused by CO2. In subsequent assays which included PIPES buffer effect on the methylreductase or corrins in these complex and 5% CO2, these effects disappeared, and the pH remained extracts. constant. In any case, no inhibition by corrins was observed Corrins also activated the methylreductase system (29). in the presence of CO2, MF, and MP (Table 2). In addition, The concentrations required for one-half of maximal activa- no inhibition by corrins was observed when the concentration varied from 1 ,uM for cobinamides (Aq2Cbi and tion of CH3-S-CoM was reduced to the level used in the CO2 diaquacobinic acid pentaamide) to about 5 ,uM for corrins reduction assay (Table 2). At this concentration, 200 nmol, with one nucleoside ligand to the cobalt (MeCbl, CN-Cbl, the level of CH3-S-CoM appeared to be rate limiting, and Coa-[5-hydroxybenzimidazoyl]-Co,-cyanocobamide, and little activation by corrins was observed. 5'-deoxyadenosylcobinamide) (29). Deoxyadenosylcobala- CO2 reduction by extracts of M. bryantii was inhibited mn, with two nucleoside ligands, activated only slightly. 58% by 25 ,uM CN-Cbl (Fig. 5). However, the initial rate of Similarly, low concentrations of corrins inhibited the ATP- methanogenesis was not affected (Fig. 5). Other corrins also dependent activation (Fig. 3). The concentrations of corrins inhibited CO2 reduction; 25 ,uM MeCbl inhibited 57%, 25 ,uM required for one-half of maximal inhibition of the rate of activation (as determined from 1/lag) were obtained from Fig. 3 and additional data (not shown). They were Aq2Cbi, 1 ,uM; diaquacobinic acid, 1 ,uM; CN-Cbl, 3 puM; MeCbl, 2 F.M; Coa-(5-hydroxybenzimidazoyl)-Co3-cyanocobamide, 50' 1.5 jiM; and 5'-deoxyadenosylcobinamide, 2.5 puM. Deoxy- adenosylcobalamin did not inhibit the ATP-dependent activation at concentrations up to 25 ,uM (data not shown). However, because of variation in the measurements of the rate of activation, it was difficult to determine with confi- dence whether the concentrations of cobinamides and 0 F-30( cobamides required for one-half of maximal inhibition were truly different. In addition, the 1/lag measured only the time for complete activation. A systematic error could result when the 10' methylreductase activity varied greatly. For example, in the concentration range of 0 to 2 [tM, cobinamides activated the methylreductase sixfold in the presence of phosphoenolpyruvate while the apparent rate of activation declined by 35%, or about 75% of the inhibition observed at 5 ,uM (Fig. 3). Part or all of this decline could be due to the fact that simply more enzyme was being activated. Alternatively, the rate of the I ATP-dependent activation could be independent of the final level of methylreductase activity in the presence of corrins. E In this case, the apparent decline in the rate of activation would accurately measure inhibition even though the measurements were not precise. High concentrations of Aq2Cbi appeared to inhibit the rate of activation more strongly than CN-Cbl (Fig. 2). Therefore, it was possible to test the accuracy of the assay for ATP- dependent activation under conditions where the velocity of 2.5 5.0 10 20 the methykeductase was constant. The methylreductase was ECbi (,M) LcblU (MM) titrated with in the of a Aq2Cbi presence high concentration FIG. 3. Inhibition of the ATP-dependent activation and activa- of CN-Cbl (Fig. 4). Under these conditions, the corrin- tion of the methylreductase by corrins. The cobinamides (Cbi) used binding site(s) should be fully saturated with one corrin or were Aq2Cbi and diaquacobinic acid pentaamide. The cobalamins the other, and the methylreductase should be fully activated. (Cbl) used were MeCbl and CN-Cbl. All assays included 2 ,umol of Consequently, the velocity of the methylreductase reaction phosphoenolpyruvate.- Symbols: 0, Aq2Cbi; 0, diaquacobinic acid varied no more than +20% (Fig. 4). However, the rate of pentaamide; M, MeCbl; El, CN-Cbl. 90 WHITMAN AND WOLFE J. BACTERIOL.

I 30 -* + 25/ttM CN-Cbl 0.3

C ._ 20 0.2

o S. ._ 3

_. 0

0 10o 0.1 0

I1 5 10 15 20 TIME (minutes) Cbi] (PM) [Aq;2 FIG. 5. Inhibition of CO2 reduction by comns. The complete FIG. 4. Inhibition of the ATP-dependent activation by reaction contained 200 nmol of CH3-S-CoM, the other components cobinamide in the presence ofcobalamin. Assays included 2 ,umol of of the methylreductase assay, MP, MF, and an H2-CO2 (80:20 phosphoenolpyruvate. The methylreductase activity in the absence [vol/vol]) atmosphere. Where indicated, 25 ,M CN-Cbl was added. of corrins was 3.9 U/mg of protein, and 1/lag was 0.42 min-'. Symbols: 0, units per milligram of protein; 0, 1/lag. Therefore, the concentration of MeCbl required for 50% inhibition was determined from replots of the data in Fig. 7 and two additional experiments not shown. It was 4.8, 5.5, Coa-(5-hydroxybenzimidazoyl)-Co,3-cyanocobamide inhib- and 5.0 F.M, which was very close to the apparent Km for ited 74%, and 5 ,uM Aq2Cbi inhibited 95%. None of these activation of the methylreductase and the concentration corrins inhibited the initial rate of methanogenesis under required for one-half of maximal inhibition of the ATP- these conditions. dependent activation by MeCbl under similar conditions (see The specificity and low concentrations of corrins required below). In the absence of MeCbl, concentration of CO2 for inhibition of CO2 reduction suggested that inhibition may reqtiired for 50% of maximum C02 reduction was 0.64% or have resulted from action at the same site or sites involved in 0.21 ,uM (Fig. 7). activation of the methylreductase or inhibition of the ATP- However, CO2 had little effect on the corrin-dependent dependent activation. To test this hypothesis, all three effects of corrins were measured in parallel assays (Fig. 6). A striking coincidence of the inhibition of CO2 reduction and ATP-dependent activation with activation of the methylreductase was observed. Furthermore, the concentrations of CN-Cbl required for 50% inhibition of CO2 reduction and the 200_ ATP-dependent activations were also very similar (Fig. 6). CO2 relieved the inhibition of CO2 reduction in an appar- ently competitive manner with respect to corrins (Fig. 7).

TABLE 2. Activation of the nlethylreductase by corrins under conditions of the CO2 reduction assay 100 Methylreductase activityb Assay coilditionsa +CblV-Cbl -Cbl +Cbl ratio Standard 3.0 14.9 5.0 +MF 2.8 15.0 5.4 +MP 3.7 16.7 4.5 10 20 +CO2 5.4 10.3 1.9 +MF + MP 4.1 15.0 3.7 [CN-Cb3 (t4M) +MF + MP + CO2 7.5 11.5 1.5 FIG. 6. Effect of corrins on the methylreductase system under +200 nmol of CH3-S-CoM 3.2 3.5 1.1 the conditions of CO2 reduction. The conditions were identical to those described in the legend to Fig. 5 except that the a The assay contained the standard components of the methylreductase assay (3.1 mg of protein, 5 ,Lmol of MgCl2, component B, 50 nmol of ATP nmethylreductase assays contained 2 ,umol of CH3-S-CoM. 100% under an atmosphere of H2, and 2 ,umol of CH3-S-CoM except where noted) activity was 6.4 U/mg of protein for the nmethylreductase, 133 nmol and MF (1.0 p.g), MP (2.5 Fg), and CO2 (20%) where noted. of CH4 for CO2 reduction, and 0.270 min-' for the ATP-dependent bMethylreductase activity (units per milligram of protein) in the presence activation. Symbols: A, methylreductase activity; 0, CO2 reduc- (+Cbl) or absence (-Cbl) of 25 ,uM CN-Cbl. tion; 0, ATP-dependent activation. VOL. 169, 1987 INHIBITION BY CORRINS 91

appears to be a direct effect on the ATP-dependent activation. Corrins also inhibit the CH3-S-CoM-dependent reduction of CO2. Evidence that the inhibition is specific to this 0.02 reaction includes the following. (i) Corrins do not inhibit the initial rate of the reaction but only the final amount of CH4 + cobolomin produced. Likewise, the methylreductase is not inhibited under the conditions of CO2 reduction. So, the apparent inhibition is not due to an indirect effect on the methylreductase system. (ii) Moreover, inhibition of CO2 reduction is competitive with respect to C02, while activa- (.) tion of the methylreductase and inhibition of the ATP-

0 dependent activation are independent of CO2. This result supports the conclusion that the observed inhibition is a separate effect of corrins and not a consequence of the indirect assay. CO2 reduction and the methylreductase are interlinked processes in the cell, being the first and last steps of 0~~~~~.2.0i methanogenesis, respectively. Furthermore, the methylreductase system is required for activation of CO2 reduction, the so-called RPG effect (9). Therefore, the multiple effects of corrins could result from action at a single site. Inhibition of both the ATP-dependent activation and CO2 reduction are similar to the corrin-dependent activation in their affinities (% C02) and specificities for corrins. Although these results may FIG. 7. Competitive inhibition of C02 reduction by corrns. The suggest that all three effects are due to action at a single corrn used was 12 ,uM MeCbl. Assays included 50 mM K-PIPES corrin-binding site, neither the affinity nor the specificity for buffer. Symbols: *, without MeCbl; O, with MeCbl. corrins is sufficiently remarkable to eliminate the possibility that more than one site is involved. Recently, reduced cobalamin was shown to be required for a simplified methylreductase system containing only activation of the methylreductase or the inhibition of the CH3-S-CoM, component B, dithiols (or SnCl2), and partially ATP-dependent activation (data not shown). For the corrin- purified component C from M. thermoautotrophicum (1). dependent activation, the average apparent Kms for two This activity differs from the effects described in M. bryantii experiments were 3.6 and 4.7 ,uM in the presence and because ATP is not required, and much higher concentra- absence of C02, respectively. Similarly, the apparent Vma,,Xs tions of cobalamins are needed. Moreover, activation of the were 10.7 and 12.5 U/mg of protein in the presence and methylreductase from M. bryantii was independent of thiols absence of C02, respectively. In the presence or absence of (29). In the M. thermoautotrophicum system, reduced C02, the concentration of MeCbl required for one-half the cobalamin may function as an electron carrier to the maximal inhibition ofthe ATP-dependent activation was less methylreductase system (1). This conclusion is not inconsis- than 2 ,uM in these experiments. tent with the results presented here. In both the M. bryantii and the M. thermoautotrophicum systems, the effects of corrins could best be explained by one or more direct DISCUSSION interactions with the methylreductase system. It is not surprising that these interactions are complex and depend Corrins activate methanogenesis by the methylreductase greatly on the details of the assay conditions. system from M. bryantii three- to fivefold (29). Therefore, it In conclusion, corrins inhibit the ATP-dependent activa- was of interest to determine the effect of corrins on other tion of the methylreductase system and CH3-S-CoM- reactions involving this system. Two such effects are de- dependent CO2 reduction in extracts of M. bryantii. In both scribed here. First, corrins inhibit the activation of the cases, the affinity and specificity for corrins closely resemble methylreductase by ATP. Secondly, corrins inhibit the CHY- the corrin-dependent activation of the methylreductase. S-CoM-dependent reduction of C02. However, this evidence alone is not sufficient to conclude a Evidence that corrins inhibit the ATP-dependent activa- common mechanism, although several are possible. Because tion includes the following. (i) ATP and phosphoenolpy- of the complexity of corrin interactions in these extracts, it is ruvate reverse the inhibition of the methylreductase by high not possible to determine whether any or all of these effects concentrations of corrins. Therefore, inhibition appears to are physiologically significant at this time. Nevertheless, be due to a failure to fully activate the enzyme system. (ii) corrins may prove to be valuable tools in understanding the Corrins are not inhibitory when added after the activation by complex interactions which occur during methanogenesis. ATP is complete. (iii) Corrins lengthen the ATP-dependent lag in the methylreductase assay. Because high concentra- ACKNOWLEDGMENTS tions of phosphoenolpyruvate only partially relieve this This work was supported by Public Health Service grant A112277 effect, the increased lag is not due to an inadequate supply of from the National Institutes of Health and grants PCM81-18178 and ATP, as might occur if corrins stimulate another ATP- DMB83-51355 from the National Science Foundation. consuming reaction in the same extract. (iv) The extent of We are indebted to D. P. Nagle, J. E. Rogers, and L. J. Shimkets the lag is independent of the final velocity of the methyl- for valuable discussions; J. Leigh and J. S. Shieh for providing some reductase assay. For these reasons, the inhibition by corrins of the materials used; and V. Gabriel for technical assistance. 92 WHITMAN AND WOLFE J. BACTERIOL.

LITERATURE CITED factor and methanopterin, two coenzymes required for CO2 formation reduction to methane by extracts ofMethanobacterium. J. Biol. 1. Ankel-Fuchs, D., and R. K. Thauer. 1986. Methane Chem. 258:7536-7540. in a system containing methyl- from methyl-coenzyme M 17. Nagle, D. P., and R. S. Wolfe. 1983. Component A of the methyl coenzyme M reductase, component B and reduced cobalamin. Eur. J. Biochem. 156:171-177. coenzyme M reductase system of Methanobacterium: resolu- 2. Blaylock, B. A. 1968. Cobamide-dependent methanol-cyano- tion into four components. Proc. Natl. Acad. Sci. USA cob(l)alamin methyltransferase of Methanosarcina barkeri. 80:2121-2155. Arch. Biochem. Biophys. 124:314-324. 18. Noll, K. M., K. L. Rinehart, Jr., R. S. Tanner, and R. S. Wolfe. 3. Blaylock, B. A., and T. C. Stadtman. 1963. Biosynthesis of 1986. Structure of component B (7-mercaptoheptanoylthreonine methane from the methyl moiety of methylcobalamin. Biochem. phosphate) of the methylcoenzyme M methylreductase system Biophys. Res. Commun. 11:34-38. of Methanobacterium thermoautotrophicum. Proc. Natl. Acad. 4. Blaylock, B. A., and T. C. Stadtman. 1966. Methane biosynthe- Sci. USA 83:4238-4243. of the soluble en- 19. Roberton, A. M., and R. S. Wolfe. 1969. ATP requirement for sis by Methanosarcina barkeri. Properties strain zyme system. Arch. Biochem. Biophys. 116:138-152. methanogenesis in cell extracts of Methanobacterium 5. Donnelly, M. I., J. C. Escalante-Semerena, K. L. Rinehart, Jr., M.o.H. Biochim. Biophys. Acta 192:420-429. and R. S. Wolfe. 1985. Methenyl-tetrahydromethanopterin 20. Romesser, J. A., and R. S. Wolfe. 1982. Coupling of methyl cyclohydrolase in cell extracts of Methanobacterium. Arch. coenzyme M reduction with carbon dioxide activation in ex- Biochem. Biophys. 242:430-439. tracts of Methanobacterium thermoautotrophicum. J. Bacte- 6. Eliefson, W. L., W. B. Whitman, and R. S. Wolfe. 1982. riol. 152:840-847. Nickel-containing factor 430: chromophore of the methylreduc- 21. Romesser, J. A., and R. S. Wolfe. 1982. CDR factor, a new tase of Methanobacterium. Ptoc. Natl. Acad. Sci. USA coenzyme required for carbon dioxide reduction to methane by 79:3707-3710. extracts of Methanobacterium thermoautotrophicum. Zentralb. 7. Eliefson, W. L., and R. S. Wolfe. 1981. Component C of the Bakteriol. Mikrobiol. Hyg. 1 Abt. Orig. Reihe C 3:271-276. methylreductase system of Methanobacterium. J. Biol. Chem. 22. Rouviere, P. E., J. C. Escalante-Semerena, and R. S. Wolfe. 256:4259-4262. 1985. Component A2 of the methylcoenzyme M methyl- 8. Escalante-Semerena, J. C., K. L. Rinehart, and R. S. Wolfe. reductase system from Methanobacterium thermoauto- 1984. Tetrahydromethanopterin, a carbon carrier in methano- trophicum. J. Bacteriol. 162:61-66. genesis. J. Biol. Chem. 259:9447-9455. 23. Taylor, C. D., and R. S. Wolfe. 1974. Structure and methylation 9. Gunsalus, R. P., and R. S. Wolfe. 1977. Stimulation of CO2 of coenzyme M (HSCH2CH2SO3). J. Biol. Chem. 249:4879- reduction to methane by methylcoenzyme M in extracts of 4885. Methanobacterium. Biochem. Biophys. Res. Commun. 24. van Beelen, P., A. P. M. Stassen, J. W. G. Bosch, G. D. Vogels, 76:790-795. W. Guijt, and C. A. G. Haasnoot. 1984. Elucidation of the 10. Gunsalus, R. P., and R. S. Wolfe. 1978. ATP activation and structure of methanopterin, a coenzyme from Methano- properties of the methyl coenzyme M reductase system in bacterium thermoautotrophicum, using two-dimensional nucle- Methanobacterium thermoautotrophicum. J. Bacteriol. 135: ar-magnetic-resonance techniques. Eur. J. Biochem. 138:563- 851-857. 571. 11. Gunsalus, R. P., and R. S. Wolfe. 1980. Methyl coenzyme M 25. van der Meijden, P., H. J. Heythuysen, A. Pouwels, F. Houwen, reductase from Methanobacterium thermoautotrophicum. J. C. van der Drift, and G. D. Vogels. 1983. Methyltransferases Biol. Chem. 255:1891-1895. involved in methanol conversion by Methanosarcina barkeri. 12. Hartzell, P. L., G. Zvilius, J. C. Escalante-Semerena, and M. I. Arch. Microbiol. 134:238-242. Donnelly. 1985. Coenzyme F420 dependence of the methylene- 26. van der MeUden, P., H. J. Heythuysen, H. T. Sliepenbeek, F. P. tetrahydromethanopterin dehydrogenase of Methanobacterium Houwen, C. van der Drift, and G. D. Vogels. 1983. Activation thermoautotrophicum. Biochem. Biophys. Res. Commun. and inactivation of methanol: 2-mercaptoethanesulfonic acid 133:884-890. methyltransferase from Methanosarcina barkeri. J. Bacteriol. 13. Keltjens, J. T., L. Daniels, H. G. Jannsen, P. J. Borm, and G. D. 153:6-11. Vogels. 1983. A novel one-carbon carrier (carboxy-5,6,7,8- 27. van der Meijden, P., L. P. J. M. Jansen, C. van der Drift, and tetrahydromethanopterin) isolated from Methanobacterium G. D. Vogels. 1983. Involvement of corrinoids in the methyla- thermoautotrophicum and derived from methanopterin. Eur. J. tion of coenzyme M (2-mercaptoethanesulfonic acid) by meth- Biochem. 130:545-552. anol and enzymes from Methanosarcina barkeri. FEMS Micro- 14. Keltjens, J. T., W. B. Whitman, C. G. Caerteling, A. M. van biol. Lett. 19:247-251. Kooten, R. S. Wolfe, and G. D. Vogels. 1982. Presence of 28. Whitman, W. B., and R. S. Wolfe. 1983. Activation of the coenzyme M derivatives in the prosthetic group (coenzyme methylreductase system from Methanobacterium bryantii by MF430) of methylcoenzyme M reductase from Methanobacter- ATP. J. Bacteriol. 154:640-649. ium thermoautotrophicum. Biochem. Biophys. Res. Commun. 29. Whitman, W. B., and R. S. Wolfe. 1985. Activation of the 108:495-503. methylreductase system from Methanobacterium bryantii by 15. Leigh, J. A., K. L. Rinehart, and R. S. Wolfe. 1984. Structure of corrins. J. Bacteriol. 164:165-172. methanofuran, the carbon dioxide reduction factor of 30. Wood, J. M., I. Moura, J. J. G. Moura, M. H. Santos, A. V. Methanobacterium thermoautotrophicum. J. Am. Chem. Soc. Xavier, J. LeGall, and M. Scandellari. 1982. Role of vitamin B12 106:3636-3640. in methyl transfer for methane biosynthesis by Methanosarcina 16. Leigh, J. A., and R. S. Wolfe. 1983. Carbon dioxide reduction barkeri. Science 216:303-305.