Methanogenic archaea live at the thermodynamic limit of life and use sophisticated mechanisms for ATP synthesis and energy coupling. The group of methanogens without cytochromes use an Na+ current across the membrane for ATP synthesis, whereas the cytochrome-containing methanogens have additional coupling sites that also translocate protons. The ATP synthase in this group is promiscuous and uses Na+ and H+ simultaneously.
- ATP synthase
- ferredoxin: heterodisulfide oxidoreductase
- proton potential
- sodium potential
Methanogens are strictly anaerobic archaea that live on a limited number of substrates . The most common substrate used for methanogenesis is CO2 (carbon dioxide) that is reduced in a linear pathway to methane with electrons derived from H2 (molecular hydrogen). In short, CO2 is first converted into formylmethanofuran from which the formyl group is transferred to another coenzyme, H4MPT (tetrahydromethanopterin). Water is split off and the resulting methenyl-H4MPT is reduced via methylene- to methyl-H4MPT. From there, the methyl group is transferred to CoM-SH (coenzyme M) yielding methyl-CoM, which is the ultimate precursor of methane  (Figure 1). Methanogenesis from H2 and CO2 according to (1)has a free energy change (∆G0′) of −131 kJ/mol under standard conditions which is sufficient for the synthesis of approximately 2 mol of ATP . However, at the low hydrogen pressure found in Nature, this value decreases to approximately one-third of an ATP per mol of methane formed .
Members of the order Methanosarcinales can also grow on acetate according to (2)
Aceticlastic methanogenesis according to eqn (2) has the lowest ∆G0′ (−36 kJ/mol of methane) of any methanogenic reaction. Methanogenesis from acetate starts with an activation of acetate to acetyl-CoA and a subsequent cleavage of the C-C bond by the CODH/ACS (carbon monoxide dehydrogenase/acetyl-CoA synthase) yielding methyl-H4SPT [Methanosarcina strains have a derivative of H4MPT, H4SPT (tetrahydrosarcinapterin)], CO and CoA. The methyl group of methyl-H4SPT is transferred to CoM-SH yielding methyl-CoM (Figure 2B).
Overall energetics of methanogenesis: all species have an Na+ ion-motive methyltransferase that generates an electrochemical ion gradient across the membrane
The overall enzymology of methanogenesis involves only a few reactions and the question is how the carbon and/or electron flow is coupled to the generation of ATP. ATP is not synthesized by substrate level phosphorylation, but by ion-gradient driven phosphorylation [3,5]. On the basis of the thermodynamics, only two reactions of the pathway are exergonic enough to translocate ions out of the cell to establish a transmembrane electrochemical ion gradient to be used by the ATP synthase for ATP synthesis. One is the methyl transfer by the methyl-H4M/SPT-CoM methyltransferase Mtr that is a membrane-integral six-subunit-containing enzyme complex that couples the methyl transfer reaction with the translocation of Na+ across the cytoplasmic membrane, thus establishing an Na+ ion-motive force across the membrane (eqn 3) [6–8]. (3)
This reaction (ΔG0′=−29 kJ/mol is common to all methanogenic pathways and thus every methanogen that grows on H2+CO2 or acetate produces a primary Na+ gradient (Figures 1 and 2). The Na+/methyl group stoichiometry of this reaction has been determined to be 1.7, which is in good agreement with the thermodynamic data of this reaction .
Overall energetics of methanogenesis: some species have in addition a membrane-bound ion-motive electron transfer system
The second exergonic reaction is the reduction of methyl-CoM that is a more complex reaction then assumed at first sight. Another coenzyme, CoB (coenzyme B), is involved in the methyl-CoM reductase reaction that produces methane, but also a mixed disulfide composed of CoB and CoM, the so-called heterodisulfide. In the last step of the pathway, the heterodisulfide is reduced. This reduction is exergonic, but the way energy is conserved is different in methanogens [4,10].
Methanogens differ with respect to the presence of cytochromes and thus the bioenergetics are different in these two groups. Most methanogens do not have cytochromes and use this exergonic reaction to drive the endergonic reduction of the electron carrier ferredoxin by electron bifurcation [11,12] (Figure 1). Fdred (reduced ferredoxin) then drives the first reaction in the pathway, the endergonic reduction of CO2 to formylmethanofuran. Thus these methanogens indeed only have the electrochemical Na+ gradient established by methyl-H4M/SPT-CoM methyltransferase to drive the synthesis of ATP.
Some methanogens such as Methanosarcina use an evolutionarily more advanced system to catalyse the reduction of the heterodisulfide. They evolved membrane integral ion carriers such as cytochromes and methanophenazine [13,14]. Electrons are fed into this electron transport chain from different electron input modules . In the case of H2 as electron donor, a membrane-bound hydrogenase, Vho (cofactor F420 non-reducing hydrogenase), is the electron input module and electrons are channelled to the terminal electron acceptor, the heterodisulfide that is reduced to the corresponding thiols by action of a membrane-bound heterodisulfide reductase, Hdr . This exergonic electron transport is used to translocate two protons out of the cell, thus establishing an electrochemical proton gradient across the membrane. The ferredoxin required for the first step in CO2 reduction is generated by a membrane-bound energy-converting hydrogenase, Ech, that uses the to drive the endergonic electron transfer from H2 (E0′≈−414 mV) to ferredoxin (E0′≈−500 mV) (Figure 2A).
If the electron donor is acetate, the carbonyl group is oxidized to carbon monoxide with concomitant reduction of ferredoxin . The redox potential difference of the donor (ΔE0′ CO/CO2=−520 mV) and the acceptor methanophenazine (ΔE0′=−165 mV) allows for additional ion translocation , but the fate of the Fdred is different in different species. Most Methanosarcina species use the aforementioned Ech hydrogenase to reduce protons to H2 gas with electrons derived from Fdred in a process that is coupled to the generation of a transmembrane electrochemical proton gradient [19,20] (Figure 2B). Methanosarcina acetivorans does not have an Ech hydrogenase. Instead, its genome encodes a potential ferredoxin:NAD+ oxidoreductase of the Rnf type . The Rnf complex has been shown before in the bacterium Acetobacterium woodii to be an Na+-translocating ferredoxin:NAD+oxidoreductase . Recently, we described that IMVs (inside-out membrane vesicles) of M. acetivorans catalysed Na+ transport coupled to an electron transport catalysed by the ferredoxin:heterodisulfide oxidoreductase activity  (Figure 2C). Ionophore studies revealed that Na+ transport was primary and electrogenic. An ∆rnf mutant was unable to grow on acetate and the ferredoxin:heterodisulfide oxidoreductase-coupled Na+ transport was abolished. These data are consistent with the hypothesis that the Rnf complex of M. acetivorans is an Na+-translocating coupling site and the entry point of electrons derived from Fdred into the electron transport chain leading to the heterodisulfide.
In summary, methanogenesis in cytochrome-free methanogens generates only an electrochemical Na+ potential as driving force for ATP synthesis, whereas, with the advent of cytochromes, additional proton bioenergetics were evolved . Since the overall process of methanogenesis involves only one to three coupling sites and the number of ions translocated is less than required for ATP synthesis and since some species have mixed (Na+ and H+ gradients), the question is how these gradients are used by the ATP synthase(s) to drive the synthesis of ATP in the different species of methanogens.
Structure and function of archaeal ATP synthases
The ATP synthases from archaea are different from the well-known bacterial F1Fo ATP synthase and the eukaryal V1Vo ATPase and thus they are classified as a separate class, the A1Ao ATP synthase . These enzymes are complex membrane-bound transport machineries with at least eight non-identical subunits in various stochiometries in its simplest, i.e. bacterial, case [25,26]. A common structural feature of the ATP synthases/ATPases is that they comprise two motors that are connected by a central stalk and peripheral stalks [27,28]. In the hydrolysis mode, ATP is hydrolysed in the A1/F1/V1 motor . This is coupled to a rotation of the central stalk that is connected to the membrane-bound motor and thus rotation of the membrane-embedded rotor is the consequence. Since rotation of the membrane-embedded rotor is obligatorily coupled to the translocations of ions, ions are transported from the cytoplasm to the outside of the cell, thus establishing an electrochemical ion gradient across the cytoplasmic membrane. F1Fo ATP synthases and the A1Ao ATP synthases are reversible in vivo and their cellular function is to synthesize ATP at the expense of the electrochemical ion gradient across the membrane [30–32].
A1Ao ATP synthases have unusual structural features such as two peripheral stalks that connect the two motors and a collar-like structure that is perpendicular to the membrane domain located in the cytoplasm [27,28]. Another notable unusual feature is the primary structure of the subunit that makes the membrane-embedded rotor, subunit c. Generally, it is a ring made by multiple copies of subunit c, an extremely hydrophobic protein with, in its simplest form, two transmembrane helices and a short connecting cytoplasmic loop . The structure of the c rings has been solved for several bacterial species and, interestingly, the c ring stoichiometry was shown to be species-specific with values of 10–15 .
The ion-binding site of the ATP synthases/ATPases resides in subunit c. Most ATP synthases/ATPases use protons as coupling ions and the proton-binding site is a conserved carboxy group in transmembrane helix 2 . Some can also use Na+, and complexation of Na+ requires five amino acids, two of which (Glu65, Ser66 for Ilyobacter tartaricus) are well conserved and constitute the Na+-binding motif easily detectable in the primary sequences [36,37].
The ion specificity of methanoarchaeal ATP synthases: protons, Na+ ions and promiscuity
As early as 1978, it was shown that an artificial ∆pH created by the addition of HCl to cell suspensions of Methanosarcina barkeri led to instantaneous ATP synthesis, demonstrating the use of protons as coupling ions in M. barkeri . With the discovery of the Na+-dependence of methanogenesis  and the Na+ ion-motive methyl-H4M/SPT-CoM methyltransferase , the question arose whether or not methanoarchaeal ATP synthases can use Na+ as coupling ion for ATP synthesis. In fact, the conserved Na+-binding motif is present in every methanoarchaeal c subunit, but, to date, the question of the ion specificity has been answered for only a few species (for a review, see ).
In the cytochrome-free methanogens such as Methanothermobacter thermoautotrophicum  and Methanobrevibacter ruminantium , Na+-dependence of ATP hydrolysis has been demonstrated and the residues involved in Na+ binding as shown for I. tartaricus are well conserved (Figures 3A and 3B). Recently, a model of the c ring of M. ruminantium was built  (Figures 3A and 3B). The c subunit of M. ruminantium arose by gene duplication and subsequent fusion of the two copies and thus has four transmembrane helices with an Na+-binding site each. The model predicts an Na+-binding site within one c subunit and one that bridges two c subunits. The ion specificity was addressed by molecular dynamics simulations that clearly revealed a high selectivity for Na+ over H+. This finding is in line with the notion that these methanogens only have an Na+ gradient across the cytoplasmic membrane and thus the ion translocated under physiological conditions (10−7 M H+, 10−1–10−2 M Na+) is Na+ . Of course, under non-physiological conditions such as addition of HCl to cell suspensions, which increases the H+ concentration to 100–10−1 M H+, ATP synthesis can be driven by H+ transport.
The situation is different in cytochrome-containing methanogens. They generate a primary electrochemical H+ and a primary electrochemical Na+ gradient across the membrane at the same time . Therefore the long-standing question was whether or not both ions are used for ATP synthesis or whether one gradient is converted into the other by means of Na+/H+ antiporter. This question was solved recently for the marine organism M. acetivorans . ATP hydrolysis measured on IMVs under physiological conditions (pH 7) was stimulated by Na+. Under these conditions, an ATP-dependent primary and electrogenic Na+ transport could be observed. The transport was effectively inhibited by DCCD (dicyclohexylcarbodi-imide) or DES (diethylstilbestrol), two known ATP synthase inhibitors, clearly demonstrating that the A1Ao ATP synthase is the enzyme responsible for Na+ translocation. Interestingly, the Na+-dependence of ATP-hydrolysis was abolished at lower pH (H+ concentration 10−5 M). Under these conditions, no Na+ transport by the ATP synthase was observed. To analyse further the ion usage of this exceptional enzyme, H+ translocation in IMVs was analysed. Indeed an ATP-dependent primary H+ transport by the ATP synthase was observed, even under physiological conditions, clearly demonstrating a dual-ion usage by this ATP synthase. The basis for this ion promiscuity was revealed by computational analyses of the H+/Na+ selectivity of the ion binding in the subunit c  (Figures 3C and 3D). The H+ selectivity for the M. acetivorans ATP synthase was greater than the one calculated for the Na+-dependent ATP synthase of I. tartaricus, but far less than the H+ selectivity of the ion-binding site of the protonmotive ATP synthase of Bacillus pseudofirmus OF4. In summary, these data demonstrated that the ATP synthase from M. acetivorans is H+-selective, but is still able to bind excess Na+ for ATP synthesis. It is assumed that the presence of the hydrophilic charge of Thr67 in the binding site of subunit c is responsible for this ion promiscuity, as strictly H+-coupled ATP synthases harbour hydrophobic residues (e.g. alanine in Spirulina platensis) at this position . Ion promiscuity has been observed before for other Na+/H+-coupled transporters, such as the flagellar motor  or solute transporter [46,47], but this was the first demonstration of a promiscuous ATP synthase.
Bioenergetics of methanogenesis: a synopsis
Methanogens are a phylogenetically diverse group, which have in common the production of methane as the main metabolic product. The development of cytochromes in early life history marked a watershed in the methanogenic lifestyle. As described above, methanogens without cytochromes build up an Na+ gradient over the membrane using the methyl-H4MPT-CoM methyltransferase Mcr (Figure 1). This reaction (∆G0′=−29 kJ/mol) allows for the translocation of approximately two Na+ per methyl group transferred, clearly demonstrating the extreme energy limitation of this metabolism. The low-potential electrons of ferredoxin needed for the reduction of CO2 are gained by a bifurcating enzyme complex consisting of the Hdr and the Methyl Viologen-reducing hydrogenase (MvhADG–HdrABC). Other hydrogenases such as Eha or Ehb found in this organism are hypothesized to only provide ferredoxin required for anaplerotic or anabolic reactions . The ATP synthase in these organisms is highly Na+-specific. Assuming an Na+/ATP stoichiometry of 4, only 0.5 mol of ATP can be synthesized during methanogenesis from H2 and CO2.
Methanosarcinales are evolutionarily more advanced since they developed cytochromes and thus adopted an additional way to energize the cytoplasmic membrane. Methanosarcina mazei uses a protonmotive electron transport chain containing a hydrogenase as input module, cytochromes as electron carriers and a membrane-bound Hdr as an electron output module. Electron flow from H2 to the heterodisulfide is coupled to the translocation of proton. Thus methanogenesis is coupled to the generation of a mixed gradient of Na+ and H+. The ATP synthase of M. mazei is similar to the one from M. acetivorans and is thus also expected to use Na+ and H+ simultaneously. The overall number of ions translocated is higher than in cytochrome-free methanogens, but one H+ has to be used by the Ech hydrogenase to reduce ferredoxin that is required for the first step of methanogenesis. Thus three ions are left that can be used by the promiscuous ATP synthase to synthesize ATP.
Another variation is found in M. acetivorans that cannot grow on H2 and CO2, but can on CO, methylated substrates or acetate. Fdred generated during acetate oxidation feeds electrons into an electron transport chain including the Rnf complex as well as the Hdr. This electron flow is coupled to translocation of Na+ and H+. This apparently drove the evolution of an ATP synthase that uses both ions, i.e. Na+ and H+, simultaneously.
If one considers the development of cytochromes to be a later step in evolution, the bioenergetics of methanogens started with Na+ as coupling ions, resulting in the need for an Na+-specific ATP synthase. Na+ ions are preferred over H+ in early bioenergetics since membranes are tighter for Na+ than for H+ . Since these organisms live at the thermodynamic limit of life, any loss of energy by diffusion would be disadvantageous to the cell.
This work was supported by the Deutsche Forschungsgemeinschaft [grant number SFB807] and through the LOEWE funding programme of Hesse's Ministry of Higher Education, Research and Arts.
Molecular Biology of Archaea 3: An Independent Meeting held at the Max Planck Institute for Terrestrial Microbiology, Marburg, Germany, 2–4 July 2012. Organized and Edited by Sonja-Verena Albers (Max Planck Institute for Terrestrial Microbiology, Germany), Bettina Siebers (University of Duisberg-Essen, Germany) and Finn Werner (University College London, U.K.).
Abbreviations: CoB(-SH), coenzyme B; CoM(-SH), coenzyme M; Fdred, reduced ferredoxin; H4SPT, tetrahydrosarcinapterin; H4MPT, tetrahydromethanopterin; IMV, inside-out membrane vesicle
- © The Authors Journal compilation © 2013 Biochemical Society