Component C of the Methylreductase System of Methunobucterium*

Component C of the methyl coenzyme M methylre- ductase system of Methanobacterium thermoautotrophicum has been purified to homogeneity with a 17% recovery of initial units. The native protein has a molecular weight of 300,000 and is composed of three different subunits with masses of 68,000, 45,000, and 38,500. They are present in equal proportion, suggest- ing a stoichiometry of a2, &, yz in the native protein. The amino acid composition reveals a preponderance of acidic amino acid residues. The protein is yellow, having an absorption maximum at 425 n m and a shoul- der at 455 nm. Reconstitution of the methyl coenzyme M methylreductase activity was linearly dependent on added component C. Component C has been detected in cell extracts of other methanogens. protein Component C has identified recently as the 2-(methy1thio)ethanesulfonic acid Mg'+, components reduce CH&-CoM1 we report preliminary

Portions of this paper (including "Experimental Procedures," Figs. 1 to 8, and Table I   Peak frdcllonr contdlned pure Connonent C as Judged by nOndissOcIatlng gel eleclmphoresls.

Occasionally a peat f r a c t l m frm t h t l c o l u m was s l i g h t l y Contdnlndted w t h o t h e l proteins and required rechrmdtogrdphy on DLAC-Sephddex A-25 l o dchleve f l n d l purl f l c a t l m as dercrlbed for t h e l h l r d DEAL-Sephader 1-25 column.
A l h r r d DIAL-Stphadex 11-25 column (7.5 Cm x 65 Cm) was equilibrated w ? l h 0 . 2 H NaCl i n

RESULTS
Purification of Component C-Component C of the CHZl-S-CoM methylreductase system of Methanobacterium thermoautotrophicum was purified to homogeneity as judged by i t s behavior upon nondissociating gel electrophoresis and sedimentation velocity centrifugation. As shown in Fig. 1 when component C was subjected to electrophoresis on polyacrylamide gels of varying porosity a t either pH 7 or pH 9, only one major band was detected on gels stained for protein. Sedimentation velocity centrifugation of component C revealed a single, symmetrical sedimenting-boundary with a corrected s value of 15.05.
The purification scheme employed for the purification of component C relied heavily upon the use of gradient sievesorptive chromatography (25). Through the use of this chromatographic method, it was possible to remove extraneous proteins and yet maintain an active component C. Conventional methods of protein purification were either ineffective or fatal to the purification of component C. The progress of one representative purification is summarized in Table I, and an example of gradient sievesorptive chromatography as applied to purification of component C is shown in Fig. 2. The purified protein was enriched only 8.3-fold over cell-free extract with a recovery of 17% of the initial units. The protein has been purified both aerobically and anaerobically without any pronounced differences in activity or physical properties. The purified protein was stable for months on ice.
Physical Properties of Component C-The molecular weight of purified component C has been determined by two methods to be approximately 300,000. When the data obtained from Fig. 1 were plotted with those obtained for suitable standards according to the method of Hedrick and Smith (12), an M , of 330,000 was found for component C (Fig. 3). Similarly, chromatography of component C on a calibrated Sephadex G-200 column gave a M, = 300,000 (Fig. 4). of acidic amino acids to basic amino acids, a fact supporting the observation that the protein complex bound rather tightly to anionic ion exchange resins. The ratio of polar to nonpolar amino acids was 1.36, and V calculated from the amino acid composition was 0.71. The subunit composition of component C after SDS-gel electrophoresis is shown in Fig. 5. The native protein is composed of three different subunits: M , = 68,000, M, = 45,000, and M, = 38,500. The estimated molecular weight of the two smaller subunits vaned slightly, i.e. 53,000. The fact that each subunit was present in equal molar amount in component C is supported by densitometer scans. Graphical integration of each peak area gave the following molar subunit stoichiometries on a total mass basis: 2.1 for the M, = 68,000 subunit; 1.9 for the M , = 45,000 subunit, and 2.15 for the M, = 38,500 subunit in the native protein of M , = 300,000. Thus, the native protein contains two copies of each subunit.
Evidence to support the presence of three different subunits in the native protein was obtained through NH2-terminal analysis. Component C was separated into its constituent subunits by SDS-gel electrophoresis, the band corresponding to each of the subunits was excised, eluted, and subjected to electrophoresis again to assure its purity. These separated subunits were dansylated, hydrolyzed, and then chromatographed on polyamide sheets with the appropriate standards. Methionine was the NHn-terminal amino acid of the M, = 68,000 subunit whereas the two smaller subunits contained Methylreductase Component C 4261 alanine at this position. Treatment of the native protein with dansyl chloride followed by hydrolysis and chromatography gave only two spots on polyamide sheets, corresponding to methionine and alanine. The alanine spot had greater fluorescence intensity when viewed under long wavelength ultraviolet light. Thus, only three subunits were found for component C.
The protein had a characteristic nonfluorescent, yellow color, the spectrum of which is shown in Fig. 6. The protein had an absorbance maximum a t 425 nm and a shoulder a t 455 nm. There were no peaks in the 300 nm region as would be expected for a typical flavin. Over a 2-h incubation period, none of the common reductants such as ascorbate, dithionite, or sodium borohydride or oxidants such as ferricyanide brought about any significant change in the visible spectrum of the protein. Imrnunoreplicate Electrophoresis-Because of the tendency of component C to associate with other proteins in cell extracts, it seemed wise to examine the cross-reactivity of these methanogen cell-free extracts after resolution by non-dissociating gel electrophoresis on a 10% polyacrylamide slab.
The results are shown in Fig. 8. Clearly M. marisnzgri and M . ruminantium have protein bands with the same mobility as purified component C, but as described above, these failed to cross-react with antibodies to component C.

DISCUSSION
Since component C is an acidic protein, ion exchange chromatogrzphy appeared to be a logical technique to use for purification. Chromatography of component C on DEAE-cellulose under a variety of conditions gave negligible purification. The use of the stronger ion exchange resin DEAE-Sephadex A-25 resulted in nearly total inactivation of recoverable component C activity. Poor purification was obtained with Sephacryl S300 or Bio-Gel P200. A method of purification was needed that would remove contaminating proteins and still yield an active component C; gradient sievesorptive chromatography fulfiied these requirements. Furthermore component C emerged from these columns in a concentrated band of activity and its behavior was reproducible from column to column. This technique may be ideal for the purification of other large, multisubunit proteins where stability is a problem.
The low increase in specific activity upon purification of component C is puzzling. However, as seen in Fig. 7, component C is one of the major soluble proteins released upon cell breakage. Whether or not it is 12% of the total protein remains to be seen. It also is possible that one or more of the proteins removed during the purification of component C may play a role in efficiently integrating component C into the methylreductase system. Experiments to clarify this possibility are in progress.
When it became evident in 1967 (26) that the major metabolic system used by most methanogens was an anaerobic respiration in which hydrogen was oxidized and carbon dioxide was reduced to methane, we concluded that ATP synthesis in these organisms must occur by electron transport phosphorylaticn; ATP pools and the effect of uncouplers were studied (27,28). Recently, excellent evidence has been presented by Doddema and Vogels (29,30) as well as by Sauer et al. (31,3 ! ! ) that intact vesicles of methanogens oxidize hydrogen, prodwing ATP by electron transport phosphorylation with the reduction of carbon dioxide to methane. The membranous vesicles of Sauer et al. (32) produced only a slight increase in methane formation when ATP or CHa-S-CoM were added, and only a fraction of the methyl moieties of added CH3-S-CoM was converted to methane by these intact vesicles. T o us there appears to be a reasonable explanation for these results; the membranous vesicles may represent a highly integrated system in which ATP and CHz-S-CoM could be generated in nearly saturating amounts either on or inside the membrane environment. Thus, externally added ATP or CH3-S-CoM, both highly charged molecules, may not penetrate readily to the appropriate enzyme sites in the membrane.
To understand the enzymology of this multi-enzyme, multicoenzyme system, we have elected to study specific reactions the components of which can be fractionated in solution.
(Whether or not these proteins are truly soluble is another question.) We have focused on the CH:%-S-CoM methylreductase, and by providing ATP and CHn-S-CoM have simplified the system, component C being the first protein to be purified to homogeneity. The native protein as purified contains six subunits having a stoichiometry of (YZ, j j 2 , and y~. The molec-ular weight of about 300,000 is slightly greater than twice that reported by Gunsalus and Wolfe (1). Although no evidence of an M , = 135,000 component C was found during any stage of purification; it is possible that their protein was a trimer composed of a, p, and y subunits. Component C has a distinct nonfluorescent, yellow color which can be attributed to an acid-or heat-extractable chromophore. Neither the structure nor function of this chromophore is known at the present time. Preliminary observations suggest that the chromophore is the nickel-containing factor FdSO (33).