Bioenergetics of the formyl-methanofuran dehydrogenase and heterodisulfide reductase reactions in Methanothermobacter thermautotrophicus Linda M. I. de Poorter 1 , Wim G. Geerts 1 , Alexander P. R. Theuvenet 2 and Jan T. Keltjens 1 1 Department of Microbiology, Faculty of Science and 2 Department of Cell Biology, Faculty of Science, University of Nijmegen, the Netherlands The synthesis of formyl-methanofuran and the reduction of the heterodisulfide (CoM-S-S-CoB) of coenzyme M (HS-CoM) and coenzyme B (HS-CoB) are two crucial, H 2 -dependent reactions in the energy metabolism of meth- anogenic archaea. The bioenergetics of the reactions in vivo were studied in chemostat cultures and in cell suspensions of Methanothermobacter thermautotrophicus metabolizing at defined dissolved hydrogen partial pressures ( p H 2 ). Formyl- methanofuran synthesis is an endergonic reaction (DG°¢ ¼ +16 kJÆmol )1 ). By analyzing the concentration ratios between formyl-methanofuran and methanofuran in the cells, free energy changes under experimental conditions (DG¢) were found to range between +10 and +35 kJÆmol )1 depending on the p H 2 applied. The comparison with the sodium motive force indicated that the reaction should be driven by the import of a variable number of two to four sodium ions. Heterodisulfide reduction (DG°¢ ¼ )40 kJÆmol )1 )was associated with free energy changes as high as )55 to )80 kJÆmol )1 . The values were determined by analyzing the concentrations of CoM-S-S-CoB, HS-CoM and HS-CoB in methane-forming cells operating under a variety of hydrogen partial pressures. Free energy changes were in equilibrium with the proton motive force to the extent that three to four protons could be translocated out of the cells per reaction. Remarkably, an apparent proton translo- cation stoichiometry of three held for cells that had been grown at p H 2 <0.12 bar, whilst the number was four for cells grown above that concentration. The shift occurred within a narrow p H 2 span around 0.12 bar. The findings suggest that the methanogens regulate the bioenergetic machinery involved in CoM-S-S-CoB reduction and proton pumping in response to the environmental hydrogen concentrations. Keywords: energy conservation; methanogenesis; proton motive force; sodium motive force; Methanothermobacter thermautotrophicus. Methanothermobacter thermautotrophicus is a methano- genic Archaeon that derives the energy for autrophic growth from the reduction of CO 2 with molecular hydrogen as the electron donor. The process of methano- genesis consists of a series of reduction reactions at which the one-carbon unit derived from CO 2 is bound to C 1 carriers of unique nature (for recent reviews see [1,2]). From a bioenergetic point of view, three reactions are of importance, notably the formation of formyl-methanofu- ran, the N 5 -methyl-tetrahydromethanopterin:coenzyme M methyl transfer step and the H 2 -dependent reduction of CoM-S-S-CoB [1,3–5]. Formyl-methanofuran (MFR-NH-CHO; f-MFR) syn- thesis represents the first step in methanogenesis. In this step, CO 2 is bound to methanofuran (MFR-NH 3 + ;MFR) and subsequently reduced to the formyl state with electrons derived from hydrogen (reaction 1). MFR-NH þ 3 þ CO 2 þ H 2 ! MFR-NH-CHO þ H þ þ H 2 O ðDG  1 0 ¼þ16 kJÁmol À1 Þð1Þ The reaction is endergonic under thermodynamic standard conditions [1,6]. Studies with cell suspensions of Methano- sarcina barkeri and Methanothermobacter marburgensis indicated that reaction (1) is driven by a sodium motive Correspondence to J. T. Keltjens, Department of Microbiology, Faculty of Science, University of Nijmegen, Toernooiveld 1, NL-6525 ED Nijmegen, the Netherlands. Tel.: + 31 24 3653437, Fax: + 31 24 3652830; E-mail: jankel@sci.kun.nl Abbreviations: CoM-S-S-CoB, heterodisulfide of HS-CoM and HS-CoB; DiBAC 4 (3), bis-(1,3-dibutylbarbituric acid)trimethine oxonol; DW, dry weight; f-MFR, formyl methanofuran; hdrACB, heterodisulfide reductase; H 4 MPT, 5,6,7,8-tetrahydromethanopterin; HS-CoB, 7-mercaptoheptanoylthreonine phosphate (Coenzyme B); HS-CoM, 2-mercaptoethanesulfonic acid (Coenzyme M); DpH, transmembrane chemical gradient of H + ; DpNa, transmembrane chemical gradient of Na + ; Dw,membranepotential;MCR,methyl- coenzyme M reductase; MFR, 4-[N-(4,5,7-tricarboxy-heptanoyl-c- L - glutamyl-p-(b-aminoethyl)phenoxy-methyl]-2-(aminomethyl)furan (methanofuran); mvhDGAB, methyl viologen-reducing hydrogenase; p H 2 , dissolved hydrogen partial pressure; p CO 2 , dissolved CO 2 partial pressure; pmf, proton motive force; q CH 4 , specific rate of methane formation (molÆh )1 Æg )1 DW); smf, sodium motive force; TCS, 3,3¢,4¢,5-tetrachlorosalicylanilide. (Received 29 August 2002, revised 4 November 2002, accepted 12 November 2002) Eur. J. Biochem. 270, 66–75 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03362.x force (smf) [7,8]. The free energy change derived from sodium import depends on the number (n Na + ) of sodium ions that are translocated per reaction: DG 0 2 ¼ n þ Na Fsmf ðkJÁmol À1 Þð2Þ in which F is the Faraday constant (96.49 kJÆV )1 Æmol )1 ). Kaesler and Scho ¨ nheit [7] estimated a Na + translocation stoichiometry of two to three Na + /CO 2 for M. barkeri.In case of M. marburgensis, the number could be somewhat higher (three to four Na + /CO 2 ). Following the transfer of the formyl group to 5,6,7,8- tetrahydromethanopterin (H 4 MPT), a dehydration step and two subsequent reduction reactions, N 5 -methyl-H 4 MPT is produced. Next, the methyl group is transferred to coenzyme M (HS-CoM) to yield methyl-coenzyme M (CH 3 -S-CoM) (reaction 3). N 5 -methyl-H 4 MPT þ HS-CoM ! H 4 MPT þ CH 3 -S-CoM ðDG  3 0 ¼À30 kJÁmol À1 Þð3Þ This exergonic reaction is catalyzed by the membrane- associated methyltransferase enzyme complex (MtrEDC BAFGH) (for a recent review see [9]). During the reaction, Na + ions are pumped out of the cell, thus creating a sodium motive force. Experiments with everted membrane vesicle preparations of Methanosarcina mazei indicated a Na + translocation stoichiometry of close to two [10]. In the terminal reaction, methane is formed by methyl- coenzyme M reduction with coenzyme B (HS-CoB) as the electron donor [1]. The heterodisulfide of HS-CoM and HS-CoB (CoM-S-S-CoB) is formed as the oxidized product. Theexergonicreaction(DG°¢ ¼ )45 kJÆmol )1 ) is catalyzed by the soluble methylcoenzyme M reductase (MCR). In fact, M. thermautotrophicus contains two different methyl reductases, MCR I and MCR II, encoded by the mcrBDCGA and mrtBDGA operons, respectively. HS-CoM and HS-CoB are recovered by the hydrogen-dependent reduction of CoM-S-S-CoB (reaction 4). CoM-S-S-CoB þ H 2 ! HS-CoM þ HS-CoB ðDG  4 0 ¼À40 kJÁmol À1 Þð4Þ The energy released in the reaction is conserved by the export of protons with the concomitant generation of an electrochemical proton gradient, or proton motive force (pmf). It then holds that DG 0 5 ¼ n þ H Fpmf ðkJÁmol À1 Þð5Þ where DG 0 5 is the free energy change to pump n H + across the cell membrane per reaction. The heterodisulfide reductase reaction has been studied in quite some detail in M. mazei (reviewed in [3–5]). Studies with everted membrane vesicle preparations of the organism initially showed a proton translocation stoichiometry of two H + /CoM-S-S-CoB reduced [11]. Recently, a novel lipophilic low-molecular- weight-electron carrier was identified, called methanophen- azine, which intermediates between hydrogen oxidation and CoM-S-S-CoB reduction. By the participation of methano- phenazine, a total number of four protons can be trans- located per reaction across the cell membrane [12]. M. thermautotrophicus neither contains methanophenazine, nor the cytochrome b-type proteins that are typical for the Methanosarcina electron-transport chain. In an in vitro system from M. marburgensis, reaction (4) is catalyzed by an enzyme complex composed of methyl viologen-reducing hydrogenase (mvhDGAB) and the heterodisulfide reductase (hdrACB) [13]. However, the mechanism by which H + is transported and the proton translocation stoichiometry have as yet not been established in Methanothermobacter. As described, methanogenic archaea use both proton- and sodium motive forces in their energy metabolism. H + and Na + fluxes are linked by the action of a Na + –H + antiporter [14]. H + and Na + movements have to result in the net efflux of protons, which drives ATP synthesis by the H + -translocating A 1 A 0 ATPase complex [5]. Above-given Gibbs free energy changes associated with the formyl-methanofuran dehydrogenase (1) and heterodi- sulfide reductase (4) reactions apply to standard conditions. Actual free energy changes (DG¢) depend on the cellular concentrations of the reactants, including the dissolved hydrogen partial pressure ( p H 2 ). In natural habitats and during growth in the laboratory, hydrogen concentrations may differ by orders of magnitude. Obviously, the differ- ences in p H 2 will affect the free energy changes of the reactions. Moreover, the methanogens have to control pmf and smf over a broad range of hydrogen concentrations, possibly by adapting proton and sodium translocation stoichiometries. In this study, the bioenergetic aspects have been investigated for M. thermautotrophicus grown at defined p H 2 values in a chemostat. Materials and methods Materials Methanofuran was purified from M. thermautotrophicus and converted into formyl-methanofuran as described before [15,16]. HS-CoB and CoM-S-S-CoB were prepared by chemical synthesis [17,18]. Cell extracts of M. thermau- totrophicus were made according to [19]. HS-CoM and benzyl viologen were purchased from Sigma (St. Louis, MO, USA), bis-(1,3-dibutylbarbituric acid)trimethine oxo- nol [DiBAC 4 (3)] was from Molecular Probes (Eugene, OR, USA), r-phtaldialdehyde was from Merck (Darmstadt, Germany), monobromobimane (thiolyte) was from Cal- biochem (Darmstadt, Germany), 3,3¢,4¢,5-tetrachlorosali- cylanilide (TCS) was from Eastman Kodak (Rochester, NY, USA), and p-nitrophenol was from BDH (Poole, UK). All other chemicals were of the highest grade available. Gasses were supplied by Hoek-Loos (Schiedam, the Neth- erlands). To remove traces of oxygen, hydrogen-containing gasses were passed over a BASF RO-20 catalyst at room temperature; nitrogen-containing gasses were passed over a prereduced R3-11 catalyst at 150 °C. The catalysts were a gift of BASF Aktiengesellschaft (Ludwigshafen, Germany). Chemostat culturing of Methanothermobacter thermautotrophicus M. thermautotrophicus (formerly: Methanobacterium ther- moautotrophicum strain DH; DSM 1053) was grown in a 3.0 L fermentor (MBR) operated as a chemostat with a culturing volume of 1.1 L. The fermentor was equipped with probes for the on-line measurement of pH (Ingold, Ó FEBS 2003 Bioenergetics of H 2 -dependent methanogenic reactions (Eur. J. Biochem. 270)67 Maarsenbroek, the Netherlands), p H 2 (see below) and temperature. The medium contained 6.8 gÆL )1 KH 2 PO 4 , 9.0 gÆL )1 NaHCO 3 ,2.1gÆL )1 NH 4 Cl, 0.1% (v/v) trace elements stock solution [20], 0.1 mgÆL )1 sodium resazurin, and 0.6 gÆL )1 cysteine/HCl and 0.5 gÆL )1 Na 2 S 2 O 3 as reducing agents. Growth was performed at 65 °Cand pH 7.0. Cultures were gassed with 80% H 2 : 20% CO 2 (v/v) at a stirring speed of 1500 r.p.m. Gassing rates were varied between 100 and 400 mLÆmin )1 , and dilution rates between 0.06 and 0.3 h )1 were applied to obtain a number of steady states as summarized in Table 1. A steady state was defined as the condition at which the optical density at 600 nm (D 600 ) of the culture, the dissolved hydrogen partial pressure and the rate of methane formation had become constant at a given gassing and dilution rate. Following three to four culture-volume changes after the establishment of a partic- ular steady state, a series of cell samples was rapidly (<10 s) withdrawn into evacuated serum bottles kept in ice-cold water. Cells were subsequently analyzed for the various bioenergetic parameters (intracellular pH, sodium concen- tration, membrane potential), dry weight content, and for the contents of methanofuran, HS-CoM and HS-CoB derivatives. Other portions were used for cell suspension incubations. Chemostat analyses Dissolved hydrogen partial pressures were recorded with an amperometric Ag 2 O/Ag probe [21] prepared from a Clark- type oxygen electrode (Broadly Technologies Corp., Irvine, CAL, USA). Fermentor inflow and outflow gas rates were measured with a soap film meter. To determine the methane content of the outflow gas, a 1 mL gas sample was added to 1 mL of ethane kept in a closed serum bottle. Hereafter, 0.1 mL amounts of the gas mixture were analyzed on a HP 5890 gas chromatograph equipped with a Poropack Q column and a flame ionization detector. Methane production rates (molÆh )1 ) were calculated from the outflow gas rates and the specific methane contents. For dry weight (DW) determination, a known volume (25–50 mL) of cell culture was centrifuged (27 000 g, 2 min, ambient tempera- ture), washed and dried at 60 °C to constant weight. Specific rates of methane formation (q CH 4 ,molÆh )1 ÆgDW )1 )were determined from the methane production rates and cellular dry weight content of the fermentor. Cell suspension incubations Inside an anaerobic glove box, anoxic cell samples from the chemostat were diluted with fresh growth medium to obtain a D 600 % 0.6–0.8. Titanium citrate (1 m M ) was added to remove oxygen traces [22]. Cell suspensions were divided into 10 mL portions kept in 115 mL serum bottles. The bottles were closed with black butyl rubber stoppers and aluminum crimped seals, and pressured to 150 kPa with H 2 /CO 2 (80 : 20, v/v) and N 2 /CO 2 (80 : 20, v/v) gas mix- tures to obtain hydrogen partial pressures between 0.001 and 0.8 bar. Following the addition of 1 mL of ethane, which served as the internal standard for methane measurements, serum bottles were placed in a water bath at 65 °C. At regular time intervals, gas samples were taken to follow methane formation. As soon as methanogenesis had started, the bottles were transferred to a rotary shaking water bath (65 °C, 200 r.p.m.) and incubations were continued for 1 h. Hereafter, reactions were stopped by rapidly cooling the serum bottles in iced water. Cells were subsequently subjected to a number of analyses outlined hereafter. Determination of intracellular pH, Na + concentrations and membrane potential Intracellular pH (pH i ) was measured taking advantage of the pH-dependent fluorescence characteristics of coenzyme F 420 , and the transmembrane electrochemical gradient Table 1. Physiological and bioenergetic properties of M. thermautotrophicus growing in a chemostat. M. thermautotrophicus was cultured at the indicated dilution and 80 H 2 :20CO 2 (v/v) gassing rates. At steady state, dissolved hydrogen partial pressures in the medium ( p H 2 ), optical densities (D 600 ) and specific rates of methanogenesis (q CH 4 ) of the cultures, as well as membrane potentials (Dw ± 10 mV), intracellular pH (pH i ±0.05 units), proton motive (pmf ± 12 mV) and sodium motive (smf ± 12 mV) forces of the cells were measured as described in the text. ND, not determined. Growth conditions Growth properties Bioenergetic parameters Culture (nr) Dilution rate (h )1 ) Gassing rate (mlÆmin )1 ) p H 2 (Bar) D 600 q CH 4 (molÆg )1 Æh )1 ) Dw (mV) pH i pmf (mV) smf (mV) 1 0.06 100 0.015 1.22 0.089 )130 7.55 )175 )90 2 150 0.115 1.41 0.077 )130 7.35 )165 )105 3 200 0.125 2.35 0.066 )125 8.65 )250 )95 4 300 0.140 1.96 0.073 )120 7.45 )160 )95 5 400 0.125 3.35 0.048 )85 8.80 )215 )50 6 0.10 100 0.005 1.65 ND )110 8.20 )180 )95 7 100 0.100 1.10 0.040 )95 8.15 )185 )60 8 200 0.040 1.94 0.074 )115 7.90 )175 )85 9 200 0.160 0.93 0.047 )95 7.60 )140 )90 10 300 0.120 1.68 0.066 )120 7.55 )170 )85 11 400 0.550 0.44 0.137 )110 8.40 )200 )85 12 0.20 400 0.125 1.02 0.168 )125 9.20 )285 )100 13 0.30 400 0.500 0.61 0.171 )130 8.00 )210 )100 68 L. M. I. de Poorter et al.(Eur. J. Biochem. 270) Ó FEBS 2003 (membrane potential, Dw, mV) was measured with the probe DiBAC 4 (3) [23]. Errors in the pH i and Dw measurements were about 0.05 pH units and 10 mV, respectively [23]. To determine the intracellular ([Na + i ]) and extracellular ([Na + o ]) sodium ion concentrations, 10 mL of cells from chemostat cultures or suspension incubations were centri- fuged (10 min, 27 000 g,4°C) immediately after sampling. Supernatants were diluted 500-fold in washing buffer and kept for determination of [Na + o ]. Washing buffer contained 50 m M Tris/HCl buffer (pH 7.0) and 200 m M sucrose. Pellets were washed three times in cold washing buffer with centrifugation each time (10 min, 27 000 g,4°C). Supern- atants and pellets were stored at )20 °C. Before analysis, pellets were suspended in 0.5 mL of 6 M HCl and suspen- sions were placed in a boiling water bath for 1 h to destroy the cells. After cooling and centrifugation (10 min, 27 000 g, 4 °C), the supernatants were diluted in washing buffer to obtain preparations that were suitable for analysis ([Na + ]<500 l M ; [HCl]<1 M ). Na + concentrations were measured by means of flame atomic absorption spectrom- etry. Repeated analyses showed that the Na + contents of the cells could be measured with a standard deviation of less than 10%. For calculation of the intracellular ion concentrations, a cell volume of 1.8 lLÆmg DW )1 was assumed [24]. Analysis of methanofuran and formyl-methanofuran Cell samples were anaerobically divided into two parts. One portion was kept cold, while the other part was incubated for 1 h at 65 °C under an N 2 atmosphere (100%, 200 kPa, 200 r.p.m.) and in the presence of uncoupler (25 l M p-nitrophenol or 25 l M TCS). By the incubation, formyl- methanofuran is quantitatively converted into methanofu- ran. In this way, the total methanofuran content could be determined. The following steps took place under air. Known volumes of incubated and untreated cell samples were centrifuged (10 min, 27 000 g,4°C) and pellets were washed three times in 25 m M KH 2 PO 4 buffer (pH 7.0) containing 5 m M EDTA. Hereafter, cell pellets were taken up in a small volume of washing buffer, such that cells were concentrated about 50-fold (at D 600 ¼ 1). For methanofu- ran extraction, cell suspensions were vigorously suspended in an equal volume of acetone and centrifuged as above. The supernatant, containing the coenzyme, was stored at )20 °C. Next, methanofuran was fluorescently labeled with r-phtaldialdehyde [0.01 g in 10 mL 5% (v/v) 2-mercapto- ethanol] according to reported procedures [25], except that a 0.1 M borate buffer (pH 9.7) was used. Leucine (20 l M )was added as an internal standard. After a 2 min incubation at room temperature, the reaction mixture was separated on a Hewlett-Packard 1090 liquid chromatograph equipped with a HP 1046A programmable fluorescence detector andcontrolledbyHP CHEMSTATION software. Separation took place at 25 °C at a flow rate of 1.0 mLÆmin )1 on a LiChrospher100 RP-18 column (Merck, Darmstadt, Germany) using 20 m M acetate/acetic acid buffer (pH 5.0) and 80% methanol as solvent systems. The eluate was monitored with a diode array UV-visible light detector at 260 nm and a fluorescence detector set at an excitation wavelength of 340 nm and emission wavelength of 455 nm (cut-off filter, 370 nm). Labeled methanofuran, showing a characteristic retention time of 12.5 min, was quantified by the comparison of the fluorescence peak area with a calibration curve prepared from methanofuran standards. By this method, amounts as low as 10 pmol could be readily detected; errors were less than 5–10%. Formyl-methanofu- ran was quantified from the difference between the meth- anofuran contents in incubated and nonincubated cells. Analysis of HS-CoM, HS-CoB and CoM-S-S-CoB Cold, anoxically harvested cells were centrifuged and washed as described for methanofuran quantification. Pellets were taken up in washing buffer such that samples showing a D 600 ¼ 1 were concentrated about 200-fold. Hereafter, suspensions were anaerobically boiled for 30 min under H 2 atmosphere (100%, 120 kPa). Cell debris were removed by centrifugation and supernatants were stored under 100% H 2 at )20 °C. For CoM-S-S-CoB determin- ation, part of the supernatant was adjusted to pH 8.0 with 1 M Tris buffer (pH 8), and incubated under 100% H 2 (120 kPa) in the presence of 5 lL cell extract and 20 l M benzyl viologen at 60 °C for 30 min. By the incubation, CoM-S-S-CoB is quantitatively reduced to HS-CoM and HS-CoB. Benzyl viologen was included, because it strongly stimulates heterodisulfide reduction catalyzed by the cell free extract, while the compound completely inhibits the methyl transferase and methylcoenzyme M reduction reactions [18]. Subsequently, HS-CoM and HS-CoB present in the boiled cell extracts were fluorescently labeled with monobromobimane reagent [26]. For this purpose, a 1 mL assay mixture was prepared containing 25 lL boiled cell extract, 13 m M Tris-methanesulfonic acid (pH 8.0) and 5m M monobromobimane (stock solution, 100 m M in acetonitril). 2-Thiouracil (0.1 m M ) was added as an internal standard. After a 15 min incubation in the dark, 5 lLofa 500 m M methanesulfonic acid solution was added to stop the derivatization [26]. Immediately hereafter, reaction mixtures were separated on a Hewlett-Packard 1090 liquid chromatograph as described above, using acetic acid buffer (0.25%, pH 3.5) and 100% methanol as solvent systems. The eluate was monitored by simultaneously recording the absorbance at 260 nm and the fluorescence intensity at 231 nm excitation and 460 nm emission wavelength. Labe- led HS-CoM and HS-CoB, that were eluted from the column at 8.5 and 24 min, respectively, were quantified by comparing the fluorescence peak areas with calibration curves made from HS-CoM and HS-CoB standards. Detection limits of both compounds were approximately 10 pmol and errors in the analyses were less than 5–10%. CoM-S-S-CoB was determined from the difference between the HS-CoM and HS-CoB contents in reduced vs. non- reduced boiled cell extracts. Data analysis The formyl-methanofuran synthesis (1) is associated with a DG 1 °¢ ¼ +16 kJÆmol )1 [1,6]. Using artificial electron acceptors, Bertram and Thauer [27] measured a midpoint potential for the CO 2 + methanofuran/formyl-methanofu- ran couple of approximately )530 mV at 60 °C and pH 7.0. From this value a somewhat lower DG 1 ° ¼ +13.0 kJÆmol )1 is derived at 60 °C, which was used in our calculations. In reaction (1) one proton is formed. Considering that the Ó FEBS 2003 Bioenergetics of H 2 -dependent methanogenic reactions (Eur. J. Biochem. 270)69 reaction takes place in the cytoplasm, DG 1 ° varies with the intracellular pH (pH i ): DG  1 ¼ 13:0 À 2:303RTðpH i À 7ÞðkJÁmol À1 Þð6Þ where R is the gas constant (8.314.10 )3 kJÆmol )1 ÆK )1 )and T is the absolute temperature (K). Under experimental conditions, the free energy change (DG 1 ¢) depends on the concentrations of the dissolved (nonenzyme-bound) reac- tants and product according to the Nernst equation: DG 0 1 ¼ DG  1 þ RT ln ½f-MFR p H 2 Áp CO 2 ½MFR ðkJÁmol À1 Þð7Þ Similarly, the Gibbs free energy change of the heterodisul- fide reaction (4) (DG 4 °¢ ¼ )40 kJÆmol )1 [1]), is related with the dissolved reactant and product concentrations accord- ing to: DG 0 4 ¼DG  4 0 þRTln ½HS-CoM½HS-CoB p H 2 ½CoM-S-S-CoB ðkJÁmol À1 Þð8Þ In our calculations, it was assumed that the experimentally determined MFR, f-MFR, HS-CoM, HS-CoB and CoM- S-S-CoB levels represented the free (nonenzyme-bound) species. CO 2 is the reactive species in formyl-methanofuran formation [28] and a dissolved partial CO 2 pressure p CO 2 ¼ 0.2 was taken for Eqn (7). In addition, it was assumed that the intracellular p H 2 equals the dissolved hydrogen partial pressure measured with the hydrogen probe (chemostat cultures) and that p H 2 in cell suspensions equals the partial hydrogen pressures applied in the headspace. Introductory studies substantiated the latter assumptions to be valid [29]. Finally, it was anticipated that hydrogen oxidation takes place inside the cells. Data were also evaluated assuming oxidation to occur at the outer space of the cell membrane. This gave, however, highly inconsistent results. Methanogens utilize both transmembrane electrochemi- cal potentials of protons (pmf expressed in mV) and of sodium ions (smf in mV) in their energy metabolism (see introduction). According to the Mitchell hypothesis, pmf is composed of the membrane potential (Dw,mV)andthe chemical gradient of H + (DpH): pmf ¼ Dw À Z:DpH ðmVÞð9Þ where Z ¼ 2.303(RT/F)andDpH ¼ pH i –pH o ;pH i and pH o refer to the intra- and extracellular pH, respectively. At the experimental temperature (65 °C) Z ¼ 67 mV. The sodium motive force is described analogously smf ¼ Dw À Z:DpNa ðmVÞð10Þ where DpNa ¼ )log([Na + i ]/([Na + o ]). By using Eqns (9) and (10), pmf and smf were quantified from the experiment- ally measured Dw,pH i and pH o ,aswellas[Na + i ]and [Na + o ]. Results Growth of M. thermautotrophicus in the chemostat M. thermautotophicus wasculturedinachemostatatvaried dilution rates and gassing rates with 80% H 2 /20% CO 2 (Table 1). This gave a number of steady state cultures in which dissolved hydrogen partial pressures differed more than 100-fold (0.005–0.55 bar). For each steady state culture, the specific rate of methane formation (q CH 4 , molÆh )1 Æg )1 DW) was determined. In addition, cells were analyzed for a number of bioenergetic parameters (Dw,pmf and smf). Cells were also analyzed for their contents of methanofuran, HS-CoM and HS-CoB derivatives. Results are summarized in Tables 1 and 2 and will be discussed later. Proton- and sodium motive forces during growth in the chemostat Despite the over 100-fold difference in p H 2 values, cells maintained an approximately constant membrane potential (Dw ¼ )115 ± 15 mV) (Fig. 1). Likewise, pmf values did not vary much over the broad p H 2 range and were )180 to )200 mV. The values readily compared with data ()160 to )200 mV) measured by other authors [30–33]. Large deviations, however, were seen in a narrow region around p H 2 ¼ 0.125 bar (Fig. 1). During growth in this region, cells were highly alkaline, resulting in aberrant pmfs (Table 1, Fig. 1). Also smf was approximately constant (c. )90 mV), except for the p H 2 ¼ 0.125 bar region. In our study, cells were grown in a medium containing 100 m M Na + .Since intracellular sodium concentrations were generally twofold to threefold higher, smf was less than Dw. From Fig. 1 it may be noted that an increase or decrease in pmf may be accompanied by an opposite change in smf. Methanogenesis and proton motive force As outlined earlier, methane formation is connected to the net extrusion of protons, thus generating a proton motive Table 2. Cellular contents of methanofuran, coenzyme B and coenzyme M derivatives of M. thermautotrophicus growing in a chemostat. The organism was cultured under the conditions specified in Table 1. Methanofuran (MFR), coenzyme M (HS-CoM), coenzyme B (HS- CoB) and the heterodisulfide (CoM-S-S-CoB) of HS-CoM and HS- CoB were quantified as described in the Materials and methods section. For all growth conditions applied, total methanofuran (MFR + formyl-MFR) contents were 2.00 ± 0.10 nmolÆmg DW )1 of cells. ND, not determined. Culture (nr) Coenzyme content (nmolÆmg DW )1 ) MFR HS-CoM HS-CoB CoM-S-S-CoB 1 0.44 0.06 0.18 1.80 2 0.77 0.03 0.18 1.70 3 0.53 0.04 0.24 1.30 4 0.92 0.15 0.33 1.10 5 0.60 0.19 0.42 1.00 6 ND 0.10 0.25 1.30 7 ND 0.11 0.45 1.50 8 ND 0.05 0.21 1.40 9 ND 0.45 0.85 0.70 10 0.27 0.06 0.33 1.20 11 0.21 0.05 0.05 1.50 12 0.22 0.14 0.38 1.20 13 0.18 0.01 0.01 2.00 70 L. M. I. de Poorter et al.(Eur. J. Biochem. 270) Ó FEBS 2003 force. It appeared that pmf increased with the specific rate of methane formation by the cells to approach some maximum value (Fig. 2). Remarkably, two distinct curves were obtained showing apparent maxima of )215 mV and )290 mV. The latter applied to cells that had grown at p H 2 around 0.125 bar, whereas cells growing at the other dissolved hydrogen partial pressures took the lower curve. Bioenergetics of formyl-methanofuran synthesis in chemostat cultures Cells collected from the different steady state cultures were analyzed for their total (MFR + f-MFR) and specific (MFR) methanofuran contents (Table 2). Under the growth conditions applied, total methanofuran contents were quite constant (2.00 ± 0.10 nmolÆmg DW )1 ). Previ- ously, Jones et al. [34] measured a comparable content of 1.8 nmolÆmg DW )1 for M. thermautotrophicus.Formyl- methanofuran levels were calculated from the difference between the total and specific methanofuran contents. Using Eqns (6) and (7), free energy changes (DG 1 ¢)were calculated from the experimental formyl-methanofuran and methanofuran concentrations, intracellular pH values, and the p H 2 and p CO 2 at which growth had taken place (Fig. 3). As expected, reactions were endergonic and DG¢ values depen- ded on the dissolved hydrogen partial pressures. At p H 2 0.005–0.01 bar, DG 0 1 was about +30 kJÆmol )1 , while a DG 0 1 % +20 kJÆmol )1 held at p H 2 0.5–0.55 bar. Notable variations occurred around p H 2 ¼ 0.12 bar. In the analyses, formyl- methanofuran was always the major species, even at the low hydrogen concentrations (Table 2). This implies that formyl- methanofuran synthesis should be driven. We then com- pared the free energy changes with those generated by the sodium motive force using Eqn (2) (Figs 1 and 3). The comparison showed that at p H 2 <0.12bartheimportof approximately three Na + ions per reaction would be required to drive the reaction, whereas an import of two Na + would suffice at p H 2 > 0.12 bar. In the small p H 2 region around 0.12 bar, the stoichiometry was either two or three. Fig. 2. Generation of proton motive forces and related specific methane- forming activities of M. therm autotr ophicu s growinginthechemostat. The organism was cultured under the conditions summarized in Table 1. Proton motive force (pmf, mV) and specific methane-forming activity (q CH 4 ,molÆh )1 ÆgDW )1 ) were determined as described in Materials and methods for cells growing at p H 2 ¼ 0.12 bar (j)andat the other dissolved hydrogen partial pressures (r). Fig. 3. Bioenergetics of formyl-methanofuran synthesis in chemostat cultures of M. thermautotrophicus. M. thermautotrophicus was grown at the indicated dissolved hydrogen partial pressures (p H 2 ,bar).Gibbs free energy changes of formyl-methanofuran synthesis at the experi- mental conditions (DG¢,kJ.mol )1 )(m) were calculated as described in the Text. The values were compared with the smf-related energy changes DG¢ ¼ n Na + Fsmf (kJÆmol )1 ) assuming the reaction to be dri- ven by the import of n Na + ¼ 2(e)orn Na + ¼ 3Na + (s). Fig. 1. Membrane potentials, proton and sodium motive forces during growth of M. thermautotrophicus in a chemostat. M. thermautotrophi- cus was cultured at the indicated dissolved hydrogen partial pressures (p H 2 , bar) as summarized in Table 1. Membrane potential (Dw,mV) (m), proton motive force (pmf, mV) (s) and sodium motive force (smf, mV) (h) were determined as described in the Materials and methods section. Ó FEBS 2003 Bioenergetics of H 2 -dependent methanogenic reactions (Eur. J. Biochem. 270)71 Bioenergetics of CoM-S-S-CoB reduction in chemostat cultures Cells from the chemostat cultures were also analyzed for the contents of HS-CoM, HS-CoB and CoM-S-S-CoB (Table 2). Contents of HS-CoM derivatives readily com- pared to those described in literature [35]. From the experimental coenzyme concentrations and the in situ p H 2 values, the free energy changes associated with heterodisul- fide reduction (DG 0 4 ) were calculated using Eqn (8) (Fig. 4). The reaction was, indeed, very exergonic, notably at high p H 2 ,whereDG 0 4 amounted to )80 kJÆmol )1 . Although reaction thermodynamics would have favored the quantita- tive reduction of CoM-S-S-CoB, also at a p H 2 % 0.01 bar, the compound was the major species. Since heterodisulfide reduction is most likely linked with the generation of a proton motive force, we related the free energy changes to pmf using Eqn (5) (Figs 1 and 4). The comparison suggested that DG 0 4 at p H 2 < 0.12 bar permitted the export of three to four protons per reaction. At p H 2 > 0.12 bar the value was close to four. As mentioned above, pmf varied considerably in the p H 2 ¼ 0.12 bar region. Here, the putative proton translocation stoichiometry could be either three or four. Free energy changes and sodium motive forces associated with the formyl-methanofuran dehydrogenase reaction catalyzed by cell suspensions Direct measurements on growing cells from the chemostat suggested formyl-methanofuran synthesis to be driven by the import of a distinct, yet integral number of two (p H 2 >0.12 bar) or three (p H 2 <0.12 bar) sodium ions. To study the stoichiometry in more detail, cells were anoxically collected from the different steady state cultures listed in Table 1. Hereafter, series of cell suspensions from a particular culture were incubated under 20% CO 2 and in the presence of 0.001–0.80 bar hydrogen. In the course of the incubations, methane formation was followed. Meth- anogenesis always proceeded linearly in time and the rates depended on the p H 2 applied. After incubation, cells were analyzed for the contents of methanofuran, formyl-metha- nofuran, Dw, and for intra- and extracellular pH and sodium concentrations. Results of a typical experiment are shown in Fig. 5. Despite the 800-fold variation in p H 2 , concentration ratios between formyl-methanofuran and methanofuran varied only threefold (Fig. 5A). Quite remarkably, formyl-methanofuran was the predominant Fig. 5. Bioenergetics of formyl-methanofuran synthesis in cell suspen- sions of M. thermautotrophicus. Cell suspensions of M. thermautotro- phicus grown in the chemostat at p H 2 ¼ 0.005 bar (dilution rate, 0.1 h )1 ; gassing rate with 80% H 2 :20%CO 2 , 100 mLÆmin )1 )were incubated under 20% CO 2 and at the indicated hydrogen partial pressures (p H 2 , bar). Methane-forming cells were subsequently ana- lyzed for (A) the concentration ratios between formyl-methanofuran and methanofuran ([f-MFR] : [MFR]) and (B) membrane potential (Dw,mV)(h) and sodium motive force (smf, mV) (r). In (C) the Gibbs free energy changes of formyl-methanofuran synthesis at the experimental conditions (DG¢,kJÆmol )1 )(h)arecomparedwith the energy changes generated by smf, assuming the reactions to be coupled by the import of n Na + ¼ 2(m), 3 (r)or4(d)Na + . Fig. 4. Bioenergetics of CoM-S-S-CoB reduction in chemostat cultures of M. thermautotrophicus. M. thermautotrophicus was grown at the indicated dissolved hydrogen partial pressures (p H 2 , bar). Gibbs free energy changes of the heterodisulfide reduction at the experimental conditions (DG¢,kJÆmol )1 )(m) were calculated as described in the text. The values were compared with the proton motive force-related (pmf) energy changes DG¢ ¼ n H + Fpmf (kJÆmol )1 ) assuming CoM-S-S-CoB reduction to be coupled to the export of n H + ¼ 3(e)orn H + ¼ 4H + (s). 72 L. M. I. de Poorter et al.(Eur. J. Biochem. 270) Ó FEBS 2003 derivative, especially at p H 2 <0.1bar.Dw and smf tended to change in parallel, becoming more negative with increasing p H 2 (Fig. 5B). From the concentration ratios between formyl-methanofuran and methanofuran, p H 2 and p CO 2 , DG 0 1 values were calculated and compared to the energy generated by the sodium motive force using Equa- tion 2 and assuming the translocation of integral numbers of two, three or four sodium ions per reaction (Fig. 5C). As above (Fig. 3), DG 0 1 varied between +30 to +10 kJÆmol )1 in the p H 2 range between 0.001 and 0.8 bar. In addition, the comparison with the sodium motive force indicated that the importoftwoNa + was sufficient to drive the reaction at p H 2 > 0.1 bar, whereas an import of three Na + would be required in the p H 2 range 0.01–0.1 bar. At p H 2 <0.01bar, however, formyl-methanofuran synthesis required the translocation of even four Na + .Moreover,thedata presented in Fig. 5C rather point to a variable, and also nonintegral, number of two to four sodium ions to be involved in the coupling. Suspension incubations were performed with cells from the different steady states. Irrespective of the chemostat conditions and p H 2 at which growth had occurred, similar results were obtained as showninFig.5. Free energy changes and proton motive forces associated with CoM-S-S-CoB reduction catalyzed by cell suspensions Chemostat analyses suggested the energy gain from hetero- disulfide reduction to be in equilibrium with a proton motive force, permitting the translocation of three to four protons. Using the experimental conditions described in the previous section, the reaction was studied with cells collected from the chemostat. After incubation of the cell suspensions under 20% (v/v) CO 2 and varied hydrogen concentrations (0.001–0.8 bar), cells were analyzed for the HS-CoM, HS-CoB and CoM-S-S-CoB concentrations, Dw,andfor the intra- and extracellular pH values. From these data, DG 0 4 and pmf were determined. In cells that had been cultured at p H 2 ¼ 0.005 bar, DG 0 4 changed from )50 to )57 kJÆmol )1 in the p H 2 range 0.001–0.8 bar (Fig. 6A). Pmf changed in parallel with DG 0 4 . The comparison between both param- eters showed that heterodisulfide reduction enabled the export of exactly three protons. The same results, including the fixed translocation stoichiometry n H + ¼ 3, were obtained for all cells suspensions grown at p H 2 <0.12bar. A different result was obtained with cells that had been cultured at p H 2 > 0.12 bar (Fig. 6B). Again, DG 0 4 and pmf increased in parallel with the hydrogen concentrations at which incubations had taken place. Free energy changes ()55 to )70 kJÆmol )1 ) were more negative than above (Fig. 6) and permitted the export of exactly four protons. Whereas apparent proton translocation stoichiometries n H + ¼ 3andn H + ¼ 4 were observed for suspensions grown at p H 2 < 0.12 bar and p H 2 > 0.12 bar, respectively, n H + was either three or four for cells grown around p H 2 ¼ 0.12 bar. Discussion Hydrogen-dependent formyl-methanofuran synthesis and heterodisulfide reduction are two central reactions in the energy metabolism of methanogenic archaea. The thermo- dynamics of the reactions were studied in M. thermautot- rophicus growing in a chemostat under a variety of dissolved hydrogen partial pressures and in experiments with cell suspensions of the organism collected from steady state cultures. Formyl-methanofuran synthesis, the first step in methane formation from CO 2 , is an endergonic reaction for which a DG°¢ ¼ +16 kJÆmol )1 was calculated [1,6]. Data presented here show the free energy changes under experimental conditions (DG¢) to vary between +10 and +35 kJÆmol )1 (Figs 3 and 5). As one might expect, values depended on the in situ hydrogen concentrations. Previous studies demon- strated that the reaction is driven by the import of sodium ions [7]. This was concluded from experiments in which reactions were followed from the opposite direction, notably CO 2 formation from formaldehyde. By measuring the rates of CO 2 formation and sodium ion extrusion, Kaesler and Scho ¨ nheit [7] concluded that formyl-methanofuran synthe- sis is connected to the translocation of two to three Na + per reaction in case of M. barkeri; the number could be three to four for Methanothermobacter. In this study, we measured Fig. 6. Bioenergetics of CoM-S-S-CoB reduction in cell suspensions of M. th erma utotroph icus. Cells of M. thermautotrophicus collected from the chemostat growing (A) at p H 2 ¼ 0.005 bar (dilution rate, 0.1 h )1 ; gassing rate with 100 mLÆmin )1 80% H 2 :20%CO 2 ,v/v)and(B) p H 2 ¼ 0.16 bar (dilution rate, 0.1 h )1 ; gassing rate, 200 mLÆmin )1 ) were incubated under 20% CO 2 and at the indicated hydrogen partial pressures (p H 2 , bar). Gibbs free energy changes of heterodisulfide reduction at the experimental conditions (DG¢,kJÆmol )1 )(h)were calculated as described in the text and compared with the energy changes n H + Fpmf (kJÆmol )1 )(m) required to pump (A) n H + ¼ 3and (B) n H + ¼ 4H + across the cell membrane. Ó FEBS 2003 Bioenergetics of H 2 -dependent methanogenic reactions (Eur. J. Biochem. 270)73 the free energy changes related with formyl-methanofuran synthesis and compared those with the corresponding sodium motive force values that were maintained in methane-forming cells. The results, indeed, support a Na + translocation stoichiometry of two to four (Figs 3 and 5). Our analyses indicate variable, also nonintegral numbers of sodium ions to be involved in thermodynamic coupling (Fig. 5). The findings, however, do not rule out that formyl- methanofuran synthesis is kinetically coupled with the import of a fixed, integral number of (maximally four) sodium ions. Experiments with cell suspensions showed that the numbers were independent of the hydrogen concentra- tion at which growth was performed. They were controlled instead by the in situ p H 2 during methanogenesis. The reduction of CoM-S-S-CoB with hydrogen is an exergonic reaction showing a DG°¢ ¼ )40 kJÆmol )1 [1,6]. Results presented here demonstrate that the free energy changes under physiological conditions are considerably more negative (DG¢ ¼ )55 to )80 kJÆmol )1 ). DG¢ changed with the hydrogen partial pressures being more negative in cells that had grown at higher p H 2 (Figs 4 and 6). Detailed studies with M. mazei established that the energy released in heterodisulfide reduction is utilized to pump protons out of the cell, thus creating the proton motive force [3–5,11,12]. Although the mechanism in M. thermautotrophicus is as yet not understood, CoM-S-S-CoB reduction must also be the crucial reaction in pmf generation in this organism. In agreement with this, free energy changes associated with heterodisulfide reduction were always in equilibrium with pmf to the degree that three to four protons could be translocated per reaction. Quite remarkably, cells that had been grown at p H 2 < 0.12 bar coupled heterodisulfide reduction free energy changes to proton motive force sizes in a way that permitted the export of three H + , whilst an apparent proton translocation stoichiometry of four held for cells that had been cultured above 0.12 bar. It should be stressed that the proton translocation numbers that are deduced from our approach represent theoretical maximal values. Actual numbers can be lower as the result of (heat- producing) proton-slipping processes. Results described here demonstrate a shift in proton translocation stoichiometry around p H 2 ¼ 0.12 bar. This observation is supported by recent growth studies in our lab [29]. Experiments in fed-batch and continuous culture systems showed that M. thermautotrophicus displays two distinct theoretical maximal growth yields (Y CH 4MAX ), notably 3.1 and 6.7 g DW per mole of methane formed. The former value applies to cells growing below p H 2 % 0.12 bar and the higher value is observed, when growth proceeds above that concentration. Assuming 10 g of dry cells to be produced from one mole of ATP [36] and assuming ATP synthesis to be coupled to the translocation of three H + ions per reaction, a Y CH 4MAX ¼ 3.3 g DWÆmol CH À1 4 is realized by the net export of one proton per methane formed. The about two-fold higher Y CH 4MAX ¼ 6.7 g DWÆmol CH À1 4 would require the net translocation of one additional H + . The change in proton translocation stoichiometry around p H 2 ¼ 0.12 bar is con- sistent with this change in Y CH 4MAX values. The shift in proton translocation stoichiometry occurs in a narrow p H 2 span around 0.12 bar. Cells that had been grown within the zone showed dramatic, almost hyperbolic, deviations in pmf values (Fig. 1). The deviations are, in fact, the direct consequence of the stoichiometry shift. Hetero- disulfide reduction at p H 2 % 0.12 bar was associated with a DG¢ of about )70 kJ per reaction (Fig. 4). The translocation of four H + would require a pmf % )180 mV, whereas the proton motive force had to be increased to )250 mV in the case of three H + ions. Data shown in Fig. 1 are in agreement with the pmf differences. Moreover, the maximal pmf ¼ )290 mV of cells growing at p H 2 % 0.12 bar was higher than in cells growing at other hydrogen partial pressures ()215 mV) (Fig. 2), the ratio (4 : 3) reflecting the H + translocation stoichiometries. Methanogenic archaea growing on hydrogen and CO 2 have to cope with vast changes in their energy source, H 2 . Here it is shown that p H 2 has a direct effect on the bioenergetics of the formyl-methanofuran dehydrogenase and heterodisulfide reductase reactions, forcing the organ- isms to control the Na + and H + translocation numbers in the respective reactions. Control can be exerted in two different ways, instantaneously by the regulation of enzyme activity or genetically at the level enzyme expression. The former mechanism seems to apply to Na + -dependent formyl-methanofuran synthesis. The finding that, notably, H + translocation stoichiometries were associated with the hydrogen concentration at which growth was performed indicates genetic control of the bioenergetic machinery involved. Acknowledgements The work of Linda de Poorter was supported by the Life Sciences Foundation (ALW), which is subsidized by the Netherlands Organ- ization for Scientific Research (NWO). We would like to thank Mr John Hermans from our Department for the assistance with HPLC analyses. Mrs Henk de Haas and Peter Albers from the Technical Department at the Faculty are greatly acknowledged for the develop- ment and testing of the p H 2 probe. References 1. Thauer, R.K. (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144, 2377–2406. 2. Ferry, J.G. (1999) Enzymology of one-carbon metabolism in methanogenic pathways. FEMS Microbiol. Rev. 23, 13–38. 3. Deppenmeier, U., Mu ¨ ller, V. & Gottschalk, G. (1996) Pathways of energy conservation in methanogenic archaea. Arch. Microbiol. 165, 149–163. 4. Deppenmeier, U., Lienard, T. & Gottschalk, G. (1999) Novel reactions involved in energy conservation by methanogenic archaea. FEBS Lett. 457, 291–297. 5. Scha ¨ fer,G.,Engelhard,M.&Mu ¨ ller, V. (1999) Bioenergetics of the archaea. Microbiol. Molec. Biol. Rev. 63, 570–620. 6. Keltjens, J.T. & van der Drift, C. (1986) Electron transfer reac- tions in methanogens. FEMS Microbiol. Rev. 39, 259–303. 7. Kaesler, B. & Scho ¨ nheit, P. (1989) The role of sodium ions in methanogenesis. Formaldehyde oxidation to CO 2 and 2 H 2 in methanogenic bacteria is coupled with primary electrogenic Na + translocation at a stoichiometry of 2–3 Na + /CO 2 . Eur. J. Bio- chem. 184, 223–232. 8. Kaesler, B. & Scho ¨ nheit, P. (1989) The sodium cycle in metha- nogenesis. CO 2 reduction to the formaldehyde level in methano- genic bacteria is driven by a primary electrochemical potential of Na + generated by formaldehyde reduction to CH 4 . Eur. J. Bio- chem. 186, 309–316. 74 L. M. I. de Poorter et al.(Eur. J. Biochem. 270) Ó FEBS 2003 9. Gottschalk, G. & Thauer, R.K. (2001) The Na + -translocating methyltransferase complex from methanogenic archaea. Biochim. Biophys. Acta 1505, 28–36. 10. Lienard, T., Becher, B., Marschall, M., Bowien, S. & Gottschalk, G. (1996) Sodium ion translocation by N 5 -methyltetrahy- dromethanopterin: coenzyme M methyltransferase from Metha- nosarcina mazei Go ¨ 1 reconstituted in ether lipid liposomes. Eur. J. Biochem. 239, 857–864. 11. Deppenmeier, U., Blaut, M. & Gottschalk, G. (1991) H 2 :het- erodisulfide oxidoreductase, a second energy-conserving system in the methanogenic strain Go ¨ 1. Arch. Microbiol. 155, 272–277. 12. Ide, T., Bau ¨ mer, S. & Deppenmeier, U. (1999) Energy conserva- tion by the H 2 : heterodisulfide oxidoreductase from Methano- sarcina mazei Go ¨ 1: Identification of two proton-translocating segments. J. Bacteriol. 181, 4076–4080. 13. Setzke,E.,Hedderich,R.,Heiden,S.&Thauer,R.K.(1994)H 2 : heterodisulfide oxidoreductase complex from Methanobacterium thermoautotrophicum: composition and properties. Eur. J. Bio- chem. 220, 139–148. 14. Mu ¨ ller, V., Blaut, M., Heise, R., Winner, C. & Gottschalk, G. (1990) Sodium bioenergetics in methanogens and acetogens. FEMS Microbiol. Rev. 87, 373–376. 15. Breitung, J. & Thauer, R.K. (1990) Formylmethanofuran: tetra- hydromethanopterin formyltransferase from Methanosarcina barkeri: identification of N 5 -formyltetrahydromethanopterin as the product. FEBS Lett. 275, 226–230. 16. Keltjens, J.T., Brugman, A.J.A.M., Kesseleer, J.M.A., te Bro ¨ m- melstroet, B.W.J., van der Drift, C. & Vogels, G.D. (1992) 5-Formyl-5,6,7,8-tetrahydromethanopterin is the intermediate in the proces of methanogenesis in Methanosarcina barkeri. Biofactors 3, 249–255. 17. Ellermann, J., Hedderich, R., Bu ¨ cher, R. & Thauer, R.K. (1988) The final step in methane formation. Eur. J. Biochem. 172,669– 677. 18. Kengen, S.W.M., Daas, P.J.H., Keltjens, J.T., van der Drift, C. & Vogels, G.D. (1990) Stimulation of the methyl-tetra- hydromethanopterin: coenzyme M methyltransferase reaction in cell-free extracts of Methanobacterium thermoautotrophicum by the heterodisulfide of coenzyme M and 7-mercaptoheptanoyl- threonine phosphate. Arch. Microbiol. 154, 156–161. 19. te Bro ¨ mmelstroet, B.W., Hensgens, C.M.H., Keltjens, J.T., van der Drift, C. & Vogels, G.D. (1990) Purification and properties of 5,10-methylenetetrahydromethanopterin reductase, a coenzyme F 420 -dependent enzyme, from Methanobacterium thermo- autotrophicum strain DH. J. Biol. Chem. 265, 1852–1857. 20. Scho ¨ nheit, P., Moll, J. & Thauer, R.K. (1979) Nickel, cobalt and molybdenum requirement for growth of Methanobacterium ther- moautotrophicum. Arch. Microbiol. 123, 105–107. 21. Schill, N., van Gulik, W.M., Voisard, D. & von Stockar, U. (1996) Continuous cultures limited by a gaseous substrate: Development of a simple, unstructured mathematical model and experimental verification with Methanobacterium thermoautotrophicum. Bio- technol. Bioeng. 51, 645–658. 22. Zehnder, A.J.B. & Wuhrmann, K. (1976) Titanium (III) citrate as a nontoxic oxidation-reduction buffering system for the culture of obligate anaerobes. Science 194, 1165–1166. 23. De Poorter, L.M.I. & Keltjens, J.T. (2001) Convenient fluores- cence-based methods to measure membrane potential and intracellular pH in the Archaeon Methanobacterium thermo- autotrophicum. J. Microbiol. Methods 47, 233–241. 24. Scho ¨ nheit, P. & Perski, H.J. (1983) ATP synthesis driven by a potassium diffusion potential in Methanobacterium thermo- autotrophicum is stimulated by sodium. FEMS Microbiol. Lett. 20, 263–267. 25. Lai, C.Y. (1977) Detection of peptides by fluorescence methods. Methods Enzymol. 47, 236–243. 26. Fahey, R.C. & Newton, G.L. (1987) Determination of low-mo- lecular-weight thiols using monobromobimane fluorescent label- ing and high-performance liquid chromatography. Methods. Enzymol. 143, 85–96. 27. Bertram, P.A. & Thauer, R.K. (1994) Thermodynamics of the formylmethanofuran dehydrogenase reaction in Methanobacter- ium thermoautotrophicum. Eur. J. Biochem. 226, 811–818. 28. Vorholt, J.A. & Thauer, R.K. (1997) The active species of ÔCO2Õ utilized by formylmethanofuran dehydrogenase from methano- genic Archaea. Eur. J. Biochem. 248, 919–924. 29. De Poorter, L.M.I. (2002) Regulation of the hydrogen metabolism in Methanothermobacter thermautotrophicus. PhD Thesis, Uni- versity of Nijmegen, Nijmegen, the Netherlands. 30. Butsch, B.M. & Bachofen, R. (1984) The membrane potential in whole cells of Methanobacterium thermoautotrophicum. Arch. Microbiol. 138, 293–298. 31. Kaesler, B. & Scho ¨ nheit, P. (1988) Methanogenesis and ATP synthesis in methanogenic bacteria at low electrochemical proton potentials. An explanation for the apparent uncoupler insensitivity of ATP synthesis. Eur. J. Biochem. 174, 189–197. 32.Sauer,F.D.,Blackwell,B.A.&Kramer,J.K.G.(1994)Ion transport and methane production in Methanobacterium thermo- autotrophicum. Proc. Natl Acad. Sci. USA 91, 4466–4470. 33. Scho ¨ nheit, P., Beimborn, D.B. & Perski, H.J. (1984) Potassium accumulation in growing Methanobacterium thermoautotrophicum and its relation to the electrochemical proton gradient. Arch. Microbiol. 140, 247–251. 34. Jones, W.J., Donnelly, M.I. & Wolfe, R.S. (1985) Evidence for a common pathway of carbon dioxide reduction to methane in methanogens. J. Bacteriol. 163, 126–131. 35. Balch, W.E. & Wolfe, R.S. (1979) Specificity and biological distribution of coenzyme M (2-mercaptoethanesulfonic acid). J. Bacteriol. 137, 256–263. 36. Stouthamer, A.H. & Bettenhausen, C. (1973) Utilization of energy for growth and maintenance in continuous and batch cultures of microorganisms. Biochim. Biophys. Acta 301, 53–70. Ó FEBS 2003 Bioenergetics of H 2 -dependent methanogenic reactions (Eur. J. Biochem. 270)75 . Bioenergetics of the formyl-methanofuran dehydrogenase and heterodisulfide reductase reactions in Methanothermobacter thermautotrophicus Linda M Science, University of Nijmegen, the Netherlands The synthesis of formyl-methanofuran and the reduction of the heterodisulfide (CoM-S-S-CoB) of coenzyme M (HS-CoM) and