* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Adaptations of anaerobic archaea to life under extreme energy
Survey
Document related concepts
Mitochondrion wikipedia , lookup
Photosynthesis wikipedia , lookup
Biochemistry wikipedia , lookup
Magnesium in biology wikipedia , lookup
Metalloprotein wikipedia , lookup
NADH:ubiquinone oxidoreductase (H+-translocating) wikipedia , lookup
Citric acid cycle wikipedia , lookup
Electron transport chain wikipedia , lookup
Microbial metabolism wikipedia , lookup
Photosynthetic reaction centre wikipedia , lookup
Light-dependent reactions wikipedia , lookup
Evolution of metal ions in biological systems wikipedia , lookup
Transcript
REVIEW ARTICLE Adaptations of anaerobic archaea to life under extreme energy limitation Florian Mayer & Volker Müller Department of Molecular Microbiology & Bioenergetics, Institute of Molecular Biosciences, Johann Wolfgang Goethe University Frankfurt/Main, Frankfurt, Germany €ller, Correspondence: Volker Mu Department of Molecular Microbiology & Bioenergetics, Institute of Molecular Biosciences, Johann Wolfgang Goethe University Frankfurt/Main, Max-von-Laue-Str. 9, 60438 Frankfurt, Germany. Tel.: +49 69 79829507; fax: +49 69 79829306; e-mail: [email protected] Received 26 March 2013; revised 30 August 2013; accepted 3 September 2013. Final version published online 5 November 2013. DOI: 10.1111/1574-6976.12043 MICROBIOLOGY REVIEWS Editor: Michael Bott Keywords hydrogenase; Rnf; Ech; phosphorylation potential; electrochemical ion potential; sodium bioenergetics. Abstract Some anaerobic archaea live on substrates that do not allow the synthesis of 1 mol of ATP per mol of substrate. Energy conservation in these cases is only possible by a chemiosmotic mechanism that involves the generation of an electrochemical ion gradient across the cytoplasmatic membrane that then drives ATP synthesis via an A1AO ATP synthase. The minimal amount of energy required is thus depending on the magnitude of the electrochemical ion gradient, the phosphorylation potential, and the ion/ATP ratio of the ATP synthase. Methanogens, Thermococcus, Pyrococcus, and Ignicoccus have evolved different ways to energize their membranes, such as methyltransferases, H+, or NAD+ reducing electron transport systems fueled by reduced ferredoxin or H2-dependent sulfur reduction that all operate at the thermodynamic limit of life. The structure and function of the enzymes involved are discussed. Despite the differences in membrane energization, they have in common an A1AO ATP synthase that shows an extraordinary divergence in rotor composition and structural adaptations to life under these conditions. In sum, adaptation of anaerobic archaea to energylimited substrates involves chemiosmotic energy coupling, often with Na+ as coupling ion and a structurally and functionally highly adapted ATP synthase. Introduction Ever since archaea have been discovered, their unusual lifestyle, their metabolic and structural adaptations to extreme environments, and their use as biocatalyst in industrial processes have attracted much attention (Schiraldi et al., 2002; Stetter, 2006; Sharma et al., 2012). Most of the archaea first discovered grew under extreme conditions such as temperatures from 80 to 121 °C, high salt concentrations, acidic or alkaline pHs, and combinations thereof (Stetter et al., 1983; Fiala & Stetter, 1986; Zillig et al., 1990; Kurr et al., 1991; Bl€ ochl et al., 1997; Andrei et al., 2012; Mesbah & Wiegel, 2012; Sharma et al., 2012). Therefore, the name ‘archaea’ is often synonymous with ‘extremophile’, although this is not correct because a number of archaea are not extremophilic at all. For example, members of the methanogenic archaea inhabit the gastrointestinal tract of humans or the rumen. They also inhabit anoxic zones of freshwater lakes or live in other, nonextreme environments (Miller & Wolin, 1982; Belay et al., 1990; Kulik et al., 2001; Moissl et al., FEMS Microbiol Rev 38 (2014) 449–472 2003; Lepp et al., 2004), not to mention the recently discovered phylum of ammonia-oxidizing Thaumarchaeota (Brochier-Armanet et al., 2008; Schleper & Nicol, 2010). Methanogens can grow on a limited number of substances that are converted to methane (Zinder, 1993). The substrate used by the most is H2 + CO2, and the substrate used by a few species, but with the most impact on production of this greenhouse gas in nature, is acetate. The pathway of methanogenesis has been worked out in the last decades (Thauer, 1998; Deppenmeier & M€ uller, 2008). The question, how this pathway is used to synthesize ATP, has also been answered to a great extent (M€ uller et al., 1999; Sch€afer et al., 1999; Deppenmeier & M€ uller, 2008; Thauer et al., 2008; Schlegel & M€ uller, 2013). The free energy change in methanogenesis from various substrates is low and often not enough for the synthesis of 1 mol of ATP. Methanogens can only use this type of metabolism using sophisticated and unusual biochemical reactions and a chemiosmotic mechanism of energy conservation. The reactions that drive the generation of a transmembrane ion gradient are often close to ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 450 the minimal amount of energy required to translocate an ion against an electrochemical ion gradient across the membrane. Methanogens are the ‘pioneers’ in our studies to understand the principal mechanisms how cells synthesize ATP under extreme energy limitation. They turned out to involve membrane-bound electron transport systems, to use Na+ instead or along with H+, and to use an unusual ATP synthase (M€ uller et al., 1999; Deppenmeier & M€ uller, 2008; Thauer et al., 2008; Schlegel & M€ uller, 2013). Later, it turned out that other anaerobic archaea use similar, principal strategies to energize their membrane for ATP synthesis or substrate transport. It is the purpose of this review to briefly introduce the thermodynamic constraints of (archaeal) life under extreme energy limitation. Next, we will describe the chemiosmotic mechanisms of energy conservation used in selected members of the anaerobic archaea to finally conclude with the structure and function of the most important enzyme in cellular bioenergetics, the ATP synthase. F. Mayer & V. Müller synthase is, thus, the most important and universal enzyme in cellular bioenergetics. As we shall see later, the ATP synthases known to date arose from a common ancestor, have the same principal mode of operation, but evolved differently in terms of structure and function in the three domains of life (M€ uller & Gr€ uber, 2003). The ATP synthase catalyzes the following reaction (Eqn. 1): ATP þ H2 OADP þ Pi 0 DG0 ¼ 31:8 kJ mol1 (1) The free energy change in the reaction under standard conditions is 31.8 kJ mol1 (Thauer et al., 1977). That means that any reaction that is more exergonic than 31.8 kJ mol1 is sufficient to drive ATP synthesis by a direct coupling mechanism. Evolution evolved only a few reactions (Eqns 2–7) in living cells that are more exergonic and used to drive the phosphorylation of ADP (Thauer et al., 1977; Teague & Dobson, 1999): Acetate kinase: Principles of energy conservation or how much energy is required to make an ATP Highly ordered structures such as living cells require energy to maintain an ordered ‘status’. Energy is supplied by different ways, either by light or by chemical reactions (Harold, 1986). Light-driven ATP synthesis is observed not only in plants under aerobic conditions, but also in microorganisms under anaerobic conditions. For example, Rhodospirillaceae, Chromatiaceae, and Chlorobiaceae are well known for their anaerobic photosynthesis. Interestingly, one prominent family of the Archaea, the Halobacteriaceae, is known for their light-driven proton pump, bacteriorhodopsin, which translocates protons from the inside to the outside using a light-driven retinal switch (Oesterhelt & Tittor, 1989; Krebs & Khorana, 1993; Mukohata et al., 1999). In contrast, the archaea discussed here are strictly chemotrophic. However, they differ with respect to their metabolism, and organoheterotrophic as well as lithoautotrophic species are known. Anyway, the primary energy source, be it light or a chemical reaction, is converted to an energy currency that can be used by the living cell. Evolution has brought about two different energy currencies in living cells: ATP and the transmembrane electrochemical ion (H+, Na+) gradient across a membrane. It is essential that both currencies are usually freely convertible. This is done by an enzyme that was in the beginning of bioenergetics called the ‘coupling factor’, nowadays known as ATP synthase (Mitchell, 1966; Senior, 1988; Cross & M€ uller, 2004). This enzyme is present in every living cell, despite the different ways in which ATP is synthesized or the membrane is energized. The ATP ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved Acetyl phosphate þ H2 O ! Acetate þ Pi 0 DG0 ¼ 44:8 kJ mol1 (2) Phosphoglycerate kinase: 1; 3-Bisphosphoglycerate þ H2 O ! Phosphoglycerate þ Pi 0 DG0 ¼ 51:9 kJ mol1 (3) Pyruvate kinase: Phosphoenolpyruvate þ H2 O ! Pyruvate þ Pi 0 DG0 ¼ 51:6 kJ mol1 (4) Creatine kinase: Creatine phosphate þ H2 O ! Creatine þ Pi 0 DG0 ¼ 43:3 kJ mol1 (5) Arginine kinase: Arginine phosphate þ H2 O ! Arginine þ Pi 0 DG0 ¼ 45 kJ mol1 (6) Carbamate kinase: Carbamyl phosphate þ H2 O ! Carbamate þ Pi 0 DG0 ¼ 39:3 kJ mol1 (7) Sugar-fermenting organisms mostly use reactions 2, 3, and 4 for ATP synthesis. Thus, anaerobic chemoorganotrophs that ferment sugars to acetate gain 4 mol of ATP if the sugar is converted to: FEMS Microbiol Rev 38 (2014) 449–472 451 Archaeal life under extreme energy limitation Hexose þ 4ADP þ 4Pi ! 2 acetate þ 4ATP þ 4H2 þ 2CO2 (8) Compared with aerobic bacteria, this is little ATP per mol of sugar, but for an anaerobic lifestyle, this is the most one gets out of a hexose molecule by substrate-level phosphorylation. However, the fermentation according to Eqn. (8) is only possible if the reducing equivalents are ‘blown away’ as hydrogen. Hydrogen production (from ferredoxin) leads us to the second energy currency, the transmembrane electrochemilNaþ ). As we shall see later, cal ion potential (D~ lHþ , D~ proton translocation by proton reduction to hydrogen gas is one way to energize the cytoplasmic membrane. Aerobic and anaerobic archaea have a beautiful diversity of membrane integral enzymes that couple exergonic reactions to the translocation of ions out of the cell. Translocation of the ion, protons in most cases, leads to translocation of a positive charge and the translocation of a substance. Thus, the exergonic reaction drives the establishment of an electrochemical proton gradient (Eqn. 9): D~ lHþ =F ðmVÞ ¼ Dpmf ¼ DW z DpH (9) where pmf = proton-motive force; DW = electrical membrane potential; z = 2.3 9 R 9 T/F; DpH = pHinside pHoutside; R = gas constant; T = temperature; F = Faraday constant. As we shall see later, organisms that live under energy limitation often use Na+ instead of H+. Then, the Eqn. (10) is: D~ lNaþ =F ðmVÞ ¼ Dsmf ¼ DW z DpNa (10) where smf = sodium-motive force. If we reconcile that an exergonic reaction drives the export of an ion against its electrochemical potential, we can now ask how much energy is required to translocate an ion. This is given by the following Eqn. (11): DG ¼ n F D~ lion (11) where DG = free energy change in the reaction; n = number of translocated ions; F = Faraday constant; D~ lion = electrochemical potential of the coupling ion (Na+ or H+). lHþ or D~ lNaþ ) has been measured only in The D~ lion (D~ some bacteria and archaea and seems to be rather constant at around 180 to 220 mV (for simplicity D~ lion rather than D~ lion =F is used from here onwards). With Eqn. (11), this calculates to a free energy change of 17 to 21 kJ mol1 for the exergonic reaction to translocate one proton or sodium ion across the membrane (Schink, 1997; Hoehler et al., 2001). Obviously, the minimal FEMS Microbiol Rev 38 (2014) 449–472 amount of energy required to translocate an ion depends on the magnitude of D~ lion . If DG gets smaller than 17 kJ mol1, the membrane may still be energized if D~ lion is also decreased. Indirect coupling mechanisms for ATP synthesis involve an ATP synthase that is driven by D~ lion according to the following equation: lion DGP ¼ n F D~ (12) where DGP = intracellular phosphorylation potential. The intracellular phosphorylation potential describes the ratio of ATP over ADP and Pi as described in the following equation: 0 DGP ¼ DG0 þ R T lnð½ATP=½ADP ½Pi Þ (13) Synthesis of ATP is a highly endergonic reaction, and the phosphorylation potential is a function of the cellular concentrations of ATP, ADP, and Pi that, again, has been determined only for a few model organisms (Slater et al., 1973; Kashket, 1982). These data revealed a value of around +60 kJ mol1. This value is used throughout in textbooks and is seen nearly as a ‘standard value’ for the phosphorylation potential. But there have been only a few organisms analyzed and only one archaeon. In the latter, the estimates are lower, for example +45 kJ mol1 (Jetten et al., 1991). The second, very important variable is n, the number of ions translocated by the ATP synthase. Let us consider for the moment a constant DGP value. If the D~ lion decreases, ATP synthesis requires a larger number of n, the number of protons translocated. Thus, even at low D~ lion , ATP synthesis is still possible if n increases. How much ‘n’ varies between species or within one species is discussed in detail later. Let us, again for the moment, turn the argument around. If a chemical reaction linked to metabolism translocates two protons, then the reaction has to proceed twice to export the four protons required for ATP synthesis under standard conditions. If only one ion can be pumped by the chemical reaction, only 0.25 mol of ATP can be formed. How could this number be increased? Only if DGP is lower as under standard conditions! For example, a chemical reaction leading to the export of two ions may yield one ATP, if DGP is not +60, but +30 kJ mol1. Thus, the magnitude of the phosphorylation potential is another important variable that determines how much energy is required to sustain life. In sum, direct and indirect coupling mechanisms differ with respect to the amount of energy required to phosphorylate ADP. In the former, the complete amount is required by one chemical reaction; for the latter, the energy can be sequestered to the number of ions ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 452 F. Mayer & V. Müller transported across the membrane. There, the minimum amount is that required to translocate one ion across the membrane. After having described the thermodynamic principles, we will now move on to discuss the different ways that energy-limited archaea use to drive ATP synthesis. Because energy limitation always includes a chemiosmotic way of ATP synthesis, we will describe the beauty of diversity of membrane energization in selected anaerobic archaea. Energy conservation in methanogenic archaea Methanogens are a nutritionally unique, but phylogenetically diverse group of organisms that all belong to the phylum Euryarchaeota (Wolfe, 1992). Methanogens have attracted much attention because of their unique metabolism. Methanogenic substrates are H2 + CO2, methyl groups as in methanol or trimethylamine, or acetate. The pathway of methanogenesis, the biochemistry of the enzymes involved with their unique coenzymes, and (a) the way ATP is synthesized have been the subject of recent, excellent reviews (Blaut et al., 1992; Gottschalk & Thauer, 2001; Deppenmeier, 2002a, b; Deppenmeier & M€ uller, 2008; Thauer et al., 2008). Therefore, we will keep this paragraph short and focus on the coupling principles. The conversion of H2 + CO2 to CH4 requires unique coenzymes with unprecedented reaction mechanisms. This linear pathway is functional even at high temperatures up to 113 °C as shown for Methanopyrus kandleri (Kurr et al., 1991). Methanogenesis from H2 + CO2 could allow for the synthesis of about 2 mol ATP under standard conditions (Deppenmeier & M€ uller, 2008; Thauer et al., 2008) according to Eqn. (14): 4H2 þ CO2 ! CH4 þ H2 O 0 DG0 ¼ 131 kJ mol1 (14) Considering the low hydrogen partial pressure in natural environments, the energetics allows for the synthesis of only about one-third of an ATP (Deppenmeier & M€ uller, 2008; Thauer et al., 2008). Methanogens use a chemiosmotic way of energy conservation and couple (b) Fig. 1. Energy conservation in methanogens without (a) and with cytochromes (b). Methanogens couple their metabolism to the generation of a primary Na+ gradient that is established by the methyltetrahydromethanopterin–coenzyme M methyltransferase (Mtr), which transfers the methyl group from methyltetrahydromethanopterin to coenzyme M. In cytochrome-free methanogens, this Na+ gradient is used by the A1AO ATP synthase to synthesize ATP. The reduction of the electron acceptor, a heterodisulfide of coenzyme M and coenzyme B, is catalyzed by a soluble complex of heterodisulfide reductase and the methyl viologen-reducing hydrogenase (Mvh–Hdr complex). Exergonic electron transfer from hydrogen to the heterodisulfide drives endergonic electron transfer from hydrogen to ferredoxin by electron bifurcation (Kaster et al., 2011). The reduction of the heterodisulfide is in cytochrome-containing methanogens achieved by a proton-motive electron transport chain (Vho–Hdr complex), which contains cytochromes and methanophenazine. These H+ and Na+ gradients are then also used by a promiscuous A1AO ATP synthase that concurrently translocates Na+ and H+. In cytochrome-containing methanogens, the endergonic electron transfer from hydrogen to ferredoxin is driven by the D~ lHþ catalyzed by the Ech hydrogenase. The ion stoichiometries depicted are mostly based on thermodynamic calculations and have some uncertainties. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved FEMS Microbiol Rev 38 (2014) 449–472 453 Archaeal life under extreme energy limitation methanogenesis to the generation of two primary ion gradients (Deppenmeier, 2002a). One of these ion gradients is a primary, electrochemical sodium ion gradient established by the methyltetrahydromethanopterin–coenzyme M methyltransferase (Mtr; M€ uller et al., 1988; Gottschalk & Thauer, 2001), a unique enzyme in the bioenergetics of methanogens. This membrane-bound, multisubunit enzyme complex transfers the methyl group from methyltetrahydromethanopterin to coenzyme M; this exergonic reaction is used to drive sodium ion translocation across the cytoplasmatic membrane (Gottschalk & Thauer, 2001; Fig. 1a). The DG0′ value of the reaction is 30 kJ mol1 that is sufficient to translocate 2 Na+ (Sch€ onheit & Beimborn, 1985; Thauer et al., 2008). Methyl-CoM is then reduced with coenzyme B (CoB, 7-mercaptoheptanoylthreoninephosphate) to CH4, and a heterodisulfide of CoM and CoB, CoM-S-S-CoB, is formed. The reduction of this heterodisulfide is the ultimate step in the pathway. The process, by which the heterodisulfide is reduced, is different in the two groups of methanogens (Deppenmeier, 2002b; Thauer et al., 2008). The cytochrome-free methanogens such as Methanothermobacter thermautotrophicus and Methanocaldococcus jannaschii have a soluble complex of heterodisulfide reductase and the methyl viologen-reducing hydrogenase (Mvh–Hdr complex) that use the recently established mechanism of electron bifurcation (Buckel & Thauer, 2013) to use the energy liberated during exergonic heterodisulfide reduction to drive endergonic electron transfer from hydrogen (E0′ = 420 mV) to ferredoxin (E0′ = 500 mV) [the redox potential of ferredoxin is set to 500 mV, which is the redox potential of the CO2/formylmethanofuran couple (Thauer et al., 2008)] (Fig. 1a). The reduced ferredoxin is then used as a ‘high-potential’ electron donor to drive the first reaction in methanogenesis, the endergonic formylmethanofuran dehydrogenase reaction (Thauer et al., 2008). These organisms have only one coupling site that is Na+ motive, and thus, the ATP synthase should be Na+ selective. This is confirmed by sequence analysis (M€ uller & Gr€ uber, 2003; Mayer et al., 2012a) as well as, in one case, supported by experimental analysis (McMillan et al., 2011). Cytochrome-containing methanogens such as Methanosarcina acetivorans, Methanosarcina mazei, and Methanosarcina barkeri are evolutionarily more advanced and have evolved in addition to Mtr a proton-motive electron transport chain that includes cytochromes and the electron carrier methanophenazine (Deppenmeier, 2002b). Therefore, cytochrome-containing methanogenic archaea couple methanogenesis to the generation of a proton and a sodium ion gradient at the same time (Fig. 1b; Schlegel & M€ uller, 2013). The proton-motive electron transport chain involves a membrane-bound heterodisulfide FEMS Microbiol Rev 38 (2014) 449–472 reductase (Hdr) anchored to the membrane by a cytochrome b, the soluble electron carrier methanophenazine, and electron input modules that change with the growth substrate. During methanogenesis from H2 + CO2, hydrogen is activated by a membrane-bound F420 nonreducing hydrogenase (Vho) that also acts as an input module for the electron transport chain (Vho–Hdr complex in Fig. 1b). Altogether, four protons, two pumped and two scalar ones, are finally on the outside of the membrane (Deppenmeier, 2002b). During growth on methyl groupcontaining substrates, the methyl groups are oxidized to CO2 with the deazaflavin cofactor F420 as an electron acceptor. Reduced F420 is reoxidized by a membranebound F420H2 dehydrogenase that is, based on biochemical data and sequence analyses, very similar to complex I of bacterial electron transport chains. Altogether four protons are translocated (Deppenmeier, 2002b). The thermodynamically unfavorable, H2-dependent reduction of ferredoxin is not catalyzed by a direct coupling mechanism via electron bifurcation, but by a secondary coupling mechanism using D~ lHþ as driving force, catalyzed by the energy-converting hydrogenase (Ech; Welte et al., 2010b), as we will see later. Acetate is the worst substrate for methanogenesis with a standard free energy change of only 36 kJ mol1 and only used by cytochrome-containing methanogens. It is converted according to: CH3 COOH ! CO2 þ CH4 0 DG0 ¼ 36 kJ mol1 (15) To conserve as much as possible of the little energy available, acetotrophic methanogens have evolved sophisticated mechanisms. Acetate is cleaved to a methyl group and CO. The methyl group is transferred onto tetrahydromethanopterin, and the subsequent reduction again involves the sodium-motive Mtr. CO is oxidized to CO2 with subsequent reduction of ferredoxin. Reduced ferredoxin is the electron donor for heterodisulfide reduction, but the electron input modules differ in methanogens. M. mazei has an Ech-type hydrogenase, a remarkable and unique coupling site in methanogens (Welte et al., 2010a, b). This membrane integral enzyme contains iron– sulfur centers as electron carriers and transfers electrons from reduced ferredoxin to protons, thereby producing H2. It is a small electron transport chain that uses protons as terminal electron acceptor and couples the exergonic electron transfer from reduced ferredoxin (E0′ 500 mV) to protons (E0′ = 414 mV) to the export of protons from the cytoplasm to the medium, thus establishing a D~ lHþ across the membrane (Hedderich, 2004; Fig. 2a). The reaction is reversible, and the enzyme uses D~ lHþ to drive the endergonic reduction ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 454 (a) F. Mayer & V. Müller (b) Fig. 2. Energy conservation in Methanosarcina mazei and Methanosarcina acetivorans by aceticlastic methanogenesis. In M. mazei (a), the Ech complex is used to reduce protons to H2 gas with electrons derived from reduced ferredoxin in a process, which is coupled to the generation of a transmembrane electrochemical proton gradient. M. acetivorans (b) does not have hydrogenases, but a Na+-translocating Rnf complex. The electron transfer from ferredoxin to methanophenazine is coupled to vectorial Na+ transport. The subsequent heterodisulfide reductase reaction is proton motive. The ion stoichiometries depicted are mostly based on thermodynamic calculations and have some uncertainties. of ferredoxin with H2, required in the first step of o1 methanogenesis from H2 + CO2 in M. mazei G€ (Fig. 1a). This remarkable, energy-converting hydrogenase may be regarded as the prototype of energy-converting hydrogenases found in other energy-limited archaea discussed below. Hydrogen produced by Ech is then fed into the proton-motive electron transport chain (Vho–Hdr complex) by the membrane-bound F420 nonreducing hydrogenase (Vho) described above (Fig. 2a; Deppenmeier et al., 1991; Welte et al., 2010a, b). Methanosarcina acetivorans does not have a hydrogenase, and thus, hydrogen is not an intermediate in electron transfer from reduced ferredoxin to the heterodisulfide reductase (Hdr). Again, the diversity in this small group of organisms is amazing. M. acetivorans employs a ferredoxin:methanophenazine oxidoreductase encoded by the rnf genes (Li et al., 2006; Rohlin & Gunsalus, 2010). This electron transfer protein complex contains several subunits with flavins, iron–sulfur cluster, and a cytochrome c as an electron carrier. The bacterial homologue (without cytochrome c) has been shown to be a vectorial Na+ pump (Biegel & M€ uller, 2010; Biegel et al., 2011). In M. acetivorans, exergonic electron transfer from reduced ferredoxin (E0′ 500 mV) to methanophenazine (E0′ = 165 mV) is coupled to vectorial Na+ transport out of the cell (Fig. 2b; Schlegel et al., 2012a). Reduced methanophenazine then donates the electrons to the heterodisulfide reductase, and this reaction drives proton translocation out of the cell (Schlegel et al., 2012a). Altogether, this rather simple reaction sequence leading from acetate to methane and carbon dioxide involves three coupling sites. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved Most interestingly, not only Mtr, but also Rnf translocates Na+. That raises the question about the ion specificity of the ATP synthase that is discussed below. In sum, methanogens have evolved sophisticated coupling sites early in the history of life (Schlegel & M€ uller, 2013). First was the sodium-motive Mtr that, so far, has only been found in methanogenic archaea. This reaction lNaþ generation ( 2 Na+ uses 30 kJ mol1 to drive D~ translocated). Second, with the advent of the cytochromes, was the evolution of electron transfer chains that are coupled to proton translocation. Remarkable are also Rnf and Ech. Rnf is widespread in bacteria. In anaerobes, it uses NAD+ as the electron acceptor, whereas aerobes may use it to drive the endergonic reaction of ferredoxin with NADH at the expense of D~ lNaþ . Rnf is the ancestor of the sodium-motive Nqr complex found in some bacteria (Tokuda & Unemoto, 1984; Juarez & Barquera, 2012). Ech is found in anaerobic archaea and some bacteria as well, but also was the ancestor of complex I of the respiratory chain (Biegel et al., 2011). Energy conservation by formate oxidation in Thermococcus onnurineus Thermococcus onnurineus is an anaerobic hyperthermophilic archaeon isolated from a deep-sea hydrothermal vent within the PACMANUS field near the Manus Basin in the Pacific Ocean. It grows optimally at pH 8.5 and 80 °C and requires complex substrates such as yeast extract, peptone, casein, or starch (Bae et al., 2006). The FEMS Microbiol Rev 38 (2014) 449–472 455 Archaeal life under extreme energy limitation outstanding feature of T. onnurineus is its growth on formate as carbon and energy source (Kim et al., 2010). Due to the small energy change in the reaction, HCOO þ H2 O ! HCO 3 þ H2 DG0 ¼ þ1:3 kJ mol1 (16) it had been theoretically excluded that microorganisms can grow by conversion of formate to carbon dioxide and hydrogen. So far, growth under anaerobic conditions by formate oxidation according to Eqn. (16) has only been observed by syntrophic bacteria in mixed cultures. In these consortia, a partner organism oxidizes hydrogen and, thus, lowers the hydrogen partial pressure in the environment and makes growth by formate oxidation thermodynamically possible (Dolfing et al., 2008; Stams & Plugge, 2009). Amazingly, T. onnurineus can do the same type of metabolism in pure culture! At 80 °C, the DG0 becomes slightly negative (2.6 kJ mol1). This is a good example for a type of metabolism that is not possible under ambient, but only at higher temperatures. By analyzing the concentrations of educts and products, the DG of the reaction was calculated during growth on formate. These elegant studies revealed that T. onnurineus does grow from 20 to 8 kJ mol1 (as calculated from the measured concentrations of formate, HCO 3 , and H2), the lowest value ever reported for an organism (Kim et al., 2010). This finding challenges the current concept of the minimal energy value that sustains life. How is formate oxidation coupled to energy conservation? According to molecular and genetic studies, it is postulated that formate is taken up by a formate transporter and oxidized to carbon dioxide and hydrogen by a soluble formate dehydrogenase (Fdh) coupled to a proton-reducing hydrogenase (Kim et al., 2010; Fig. 3). The genetic organization of this hydrogenase is similar to the membrane-bound hydrogenase (Mbh) from Pyrococcus furiosus (Schut et al., 2012). Both are more complex than the Ech hydrogenase, but all are variations of one theme: proton reduction to hydrogen gas coupled to proton export from the cell, thus establishing a D~ lHþ . In analogy to the formate hydrogen lyase system of Escherichia coli, it is suggested that the electron is directly transferred from the Fdh to the hydrogenase via an electron-transferring subunit. Although neither ion transport nor ion dependence of these multisubunit hydrogenases have been proven experimentally, it is argued that they are proton pumps. Because DG values below 17 kJ mol1 are well below the minimal amount of energy required to energize the membrane at a D~ lHþ of 180 mV, it may be possible that the D~ lHþ established is smaller than 180 mV. Separated by two genes that encode a protein of unknown function and a potential formate transporter are genes that encode for subunits with similarities to multisubunit Na+/H+ antiporter (Mrp). Such an Mrp-like system was first discovered in the alkaliphilic Bacillus halodurans and is necessary for pH homeostasis under alkaline conditions in the habitat (Hamamoto et al., 1994). Thus, it is postulated that the primary proton potential is converted to a secondary sodium ion gradient by these antiporters (Kim et al., 2010). This argument is strengthened by the finding of a conserved Na+-binding motif in subunit c of the ATP synthase. However, these postulates have to be seen with caution because methanoarchaeal ATP synthases also have the Na+-binding motif (M€ uller et al., 1999), but, so far, clear evidence for Na+ transport has not been obtained, and in one case (M. acetivorans), the enzyme was shown to translocate Na+ and H+ concurrently (Schlegel et al., 2012b). Second, Na+/H+ antiporter-like modules are also present in complex I of the respiratory chain of bacteria, and there is a clear evidence that complex I translocates protons (Efremov et al., Fig. 3. Model of energy conservation by formate oxidation in Thermococcus onnurineus. Formate is oxidized by the Fdh. Electrons are shuttled directly to the membrane-bound hydrogenase (Mfh, similar to Mbh of the same organism and Pyrococcus furiosus), which drives electrogenic H+ translocation. The H+ gradient may be changed to a Na+ gradient by a Na+/H+ antiporter (Mrp), and the A1AO ATP synthase then uses this Na+ gradient for ATP synthesis (Kim et al., 2010). The figure shows the nature of the ions pumped, but not the stoichiometries of the reactions, which so far are not known. FEMS Microbiol Rev 38 (2014) 449–472 ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 456 2010; Hunte et al., 2010; Brandt, 2011; Dr€ ose et al., 2011). Clearly, the fascinating bioenergetics of T. onnurineus deserves much more attention in future studies. Energy conservation in Pyrococcus furiosus Pyrococcus furiosus, a hyperthermophilic strictly anaerobic archaeon that belongs to the phylum Euryarchaeota, conserves energy for growth by fermentation of carbohydrates and peptides to CO2, H2, and organic acids (e.g. acetate) in the absence of elemental sulfur. In the presence of sulfur, P. furiosus rather reduces elemental sulfur to H2S than producing H2 (Fiala & Stetter, 1986). Therefore, it does not live under extreme energy limitations, but nevertheless is included in this review because it is the paradigm for an anaerobe using a novel protontranslocating, proton-reducing electron transport chain. Sugars are oxidized by a modified Embden–Meyerhof– Parnas pathway. The important modification is a ferredoxin-dependent glyceraldehyde-3-phosphate dehydrogenase reaction coupled to ferredoxin reduction (GAP: ferredoxin oxidoreductase, GAPOR). The organism also does not employ a pyruvate dehydrogenase that would reduce NAD+, but a pyruvate:ferredoxin oxidoreductase F. Mayer & V. Müller instead (Sapra et al., 2003). Why is this important? The redox potential of ferredoxin (E0′ 500 mV) is low enough to transfer electrons directly to protons (E0′ = 414 mV), thereby producing hydrogen gas. Thus, the reducing equivalents are literally ‘blown away’ as hydrogen. This allows oxidation of glucose to two acetates with the production of 2 mol of ATP by substratelevel phosphorylation (Fig. 4). At a first glance, the invention of the GAPOR looks like a waste of energy because one ‘substrate-level phosphorylation’ ATP is lost. But ‘blowing away’ electrons as hydrogen eliminates the need to produce reduced end products such as ethanol and allows acetate production by the ADP-forming acetylCoA synthetase (ACD) that also produces ATP (Br€asen et al., 2008). Thus, there is no loss of ATP synthesized by substrate-level phosphorylation. Even better, if hydrogen production from reduced ferredoxin is coupled to D~ lHþ generation, more ATP can be synthesized. This is where the multisubunit, Mbh comes into play. Hydrogen is evolved by hydrogenases, and the genome of P. furiosus encodes three different hydrogenases. Two of them (SH-I and SH-II) are soluble, consisting of four subunits each, and are extremely thermostable (Bryant & Adams, 1989; Ma et al., 2000; Robb et al., 2001). For example, purified SH-I only loses 50% of activity after Fig. 4. Model of energy conservation in Pyrococcus furiosus. Enzymes involved in energy conservation are a Mbh, an Mrp-like Na+/H+ antiporter module, and a Na+-dependent A1AO ATP synthase. P. furiosus ferments different carbohydrates and peptides and produces organic acids, for example, acetate, CO2, and H2. All oxidative steps of this modified Embden-Meyerhof-Parnas pathway, the reactions catalyzed by the glyceraldehyde-3-phosphate:ferredoxin oxidoreductase (GAPOR) and the pyruvate:ferredoxin oxidoreductase (POR), use ferredoxin as an electron acceptor. Hydrogen gas is produced by the Mbh, coupled to the oxidation of reduced ferredoxin and generation of a proton gradient. This gradient may be changed to a Na+ gradient by the Mrp-like Na+/H+ antiporter module of the Mbh. This Na+ gradient is then used by the Na+dependent A1AO ATP synthase to synthesize ATP. The figure shows the nature of the ions pumped, but the stoichiometries of the reactions are unknown. GAP, glyceraldehyde-3-phosphate; 3-PG, 3-phosphoglycerate; ACD, ADP-forming acetyl-CoA synthetase. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved FEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation 12 h of incubation at 80 °C (Ma & Adams, 2001; Jenney & Adams, 2008). The third hydrogenase, encoded in the genome of P. furiosus, is membrane-bound (Mbh). The Mbh operon contains 14 genes (mbhA–mbhN), but a purified Mbh complex had only two subunits (MbhK and MbhL), in which subunit MbhL has the catalytic activity (Jenney & Adams, 2008; Schut et al., 2012). It is thought that this Mbh uses reduced ferredoxin, generated during glycolysis at the GAPOR (Mukund & Adams, 1995; van der Oost et al., 1998) and the POR reactions (Blamey & Adams, 1993), as an electron donor and transfers electrons to protons, thereby producing hydrogen gas. This exergonic reaction is used to pump ions out of the cell, and the electrochemical ion gradient established is then used to drive ATP synthesis (Fig. 4). In principle, the reaction is similar to the Ech-catalyzed reaction. Measurements of ATP synthesis and hydrogen production in inverted membrane vesicles of P. furiosus clearly showed that hydrogen production is coupled to ATP synthesis (Sapra et al., 2003). Although the coupling ion was not addressed in this study, it was hypothesized that protons were translocated. However, with the finding that the ATP synthase of P. furiosus uses Na+ (Pisa et al., 2007), the question about the ion translocated by the Mbh arose. The Mbh operon (14 genes in total) encodes eight proteins (MbhA–MbhH) that are similar to subunits MrpB–MrpG of the Mrp-type Na+/ H+ antiporter (Schut et al., 2012). Therefore, it might be possible that the Mbh of P. furiosus, like the one of T. onnurineus, consists of two modules: one is for hydrogenase activity to produce a proton gradient, and the other module, the Mrp-type Na+/H+ antiporter module, changes the proton gradient to a Na+ gradient (Schut et al., 2012). The Na+ gradient can then be utilized by the Na+-dependent A1AO ATP synthase of P. furiosus (Pisa et al., 2007) to synthesize ATP at high temperatures. There is also a striking sequence identity of Mbh to complex I of respiratory chains. Its soluble domain, consisting of modules N and Q, catalyzes NADH oxidation and transfers electrons via a chain of iron–sulfur center to the membrane. The soluble domain is indeed similar to soluble hydrogenases. The membrane-embedded domain, which is called P-module or pump module, contains two main modules, PP and PD, with a row of four Na+/H+-antiporter-similar subunits, which are connected by a long helical ‘transmission element’ (Efremov et al., 2010; Hunte et al., 2010; Brandt, 2011; Dr€ ose et al., 2011). These Na+/H+ antiporter-like modules are the predicted sites of H+ translocation. In complex I of E. coli and Rhodothermus marinus, not only an export of H+ was observed, but also a Na+ transport in the other direction was observed. One may work as a H+ pump and the other as Na+/H+ antiporter. But this FEMS Microbiol Rev 38 (2014) 449–472 457 Na+/H+ antiport activity does not seem to be a general property of complex I because for Paracoccus denitrificans, only H+ translocation was observed. No Na+ transport in the opposite direction could be shown (Batista et al., 2010; Batista & Pereira, 2011). Is this system of energy conservation, composed of Mbh, Mrp-type Na+/H+ antiporter module, and A1AO ATP synthase also present in other hyperthermophiles? Yes, such an Mrp–Mbh system is also genetically encoded in the genera Thermococcus, Desulfurococcus, Thermosphaera, Ignisphaera, and Staphylothermus (Schut et al., 2012), all hyperthermophilic archaea. It might be possible that such a system of energy conservation is an adaptation of hyperthermophilic archaea to environments with high temperatures. Energy conservation in Ignicoccus hospitalis and Nanoarchaeum equitans Ignicoccus hospitalis is a hyperthermophilic, strictly anaerobic crenarchaeon isolated from a submarine hydrothermal system at the Kolbeinsey Ridge. It grows optimally at 90 °C and has a chemolithoautotrophic metabolism (Paper et al., 2007). Together with Nanoarchaeum equitans cells, which are small cocci with only 400 nm in diameter, I. hospitalis forms an ‘intimate association’, the only known stable association between two archaea (Huber et al., 2002; Jahn et al., 2008). Species of the genus Ignicoccus are so far the only described members of archaea who have an unusual cell architecture as well as an unusual energy conservation mechanism (Fig. 5). They are the only archaea that have an inner membrane and an outer cellular membrane. These two membranes enclose an intermembrane compartment with a width up to 500 nm (Huber et al., 2000, 2012; Rachel et al., 2002). First insights into the unusual energy conservation mechanisms in Ignicoccus cells came from immuno electron microscopy of ultrathin sections of I. hospitalis cells. With the use of antibodies against the A1AO ATP synthase of I. hospitalis, the enzyme was detected exclusively in the outer cellular membrane (K€ uper et al., 2010). This was a major breakthrough into a new era of bioenergetics, because till then, only cytoplasmic membranes have been described as harboring ATP synthases (Lewalter & M€ uller, 2006), and it was believed that outer membranes are generally ‘non-energy-conserving’ (Nicholls & Ferguson, 1992). However, not only an A1AO ATP synthase is essential to drive ATP synthesis, but also an ion gradient across the membrane is necessary. This ion gradient is produced by the H2:sulfur oxidoreductase reaction, in which sulfur is reduced with H2 to H2S (Dirmeier et al., 1998). The reduction of sulfur with hydrogen is not an exclusive feature of Ignicoccus, but has been observed ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 458 F. Mayer & V. Müller before in archaea and bacteria (Hedderich et al., 1999). The free energy change in the reaction H2 þ S ! HS þ Hþ 0 DG0 ¼ 28 kJ mol1 (17) is large enough to allow for proton translocation (Thauer et al., 1977; Amend & Shock, 2001). However, the mechanism of membrane energization is obscure. If the hydrogenase and sulfur reductase face to the outside, it must involve an active transport of protons from the inside to the outside because there is no net charge generation in the redox reaction. The c subunit of the A1AO ATP synthase of I. hospitalis has no conserved Na+-binding motif. This is consistent with proton-driven ATP synthesis and the presence of a primary proton gradient across the membrane. Consistent with the finding that the outermost membrane is the place of energy conservation (K€ uper et al., 2010), an ATP-consuming key enzyme of autotrophic lifestyle was found in the intermembrane compartment. The AMP-forming acetyl-CoA synthetase uses the ATP by the A1AO ATP synthase for activation of acetate and conversion of acetate to acetyl-CoA, the primary acceptor molecule of CO2 fixation in I. hospitalis (Mayer et al., 2012b). Nanoarchaeum equitans, the small riding companion of I. hospitalis has apparently, as deduced from its genome sequence, no genetic capability for energy conservation (Giannone et al., 2011). Thus, the working hypothesis is that it gains ATP from its host. Even if this is correct, every living cell requires an energized membrane. In principle, this can be achieved by a proton-translocating ATP synthase/ATPase at the expense of ATP. In fact, this is not uncommon and observed in many strictly fermenting bacteria. However, the genome of N. equitans harbors only a subset of genes that are usually required to encode a functional ATP synthase/ATPase (Lewalter & M€ uller, 2006; Giannone et al., 2011). Whether N. equitans has the most rudimentary and simplest ATPase known to date is discussed below. ATP synthases: key enzymes in cellular bioenergetics Despite all the differences in the ways the electrochemical ion gradient is generated, the common feature of archaea (and other life forms) is an enzyme that synthesizes ATP, according to Eqn. (1), at the expense of the transmembrane electrical ion gradient: the ATP synthase (M€ uller et al., 1999; M€ uller & Gr€ uber, 2003). All present rotary ATP synthases/ATPases arose from a common ancestor and evolved into the three classes known to date: the F1FO ATP synthase present in bacteria, mitochondria, and chloroplasts; the V1VO ATPase present in eukaryotes; and the A1AO ATP synthase present in archaea (Fig. 6; Cross & Taiz, 1990; Nelson, 1992; M€ uller & Gr€ uber, 2003; Cross & M€ uller, 2004). Their common feature is a structure Fig. 5. Model of energy conservation in Ignicoccus hospitalis. This archaeon has two membranes, an outer cellular membrane and an inner membrane. Protein complexes for energy conservation are localized in the outer cellular membrane, so far exclusively. For energy conservation, the H2:sulfur oxidoreductase reduces sulfur with H2 to H2S. This reaction is coupled to the generation of an electrochemical ion (most likely proton) potential by an unknown mechanism. It may involve a proton-motive Q-cycle or equivalent. Then, the proton gradient is used by the A1AO ATP synthase to synthesize ATP. This ATP is used, for example, by the AMP-forming acetyl-CoA synthetase (ACS) for the activation of acetate and the conversion of acetate to acetyl-CoA, the primary acceptor molecule of CO2 fixation in I. hospitalis. The figure shows the nature of the ions pumped, but not the stoichiometries of the reaction. Adapted from Huber et al. (2012). ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved FEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation 459 Fig. 6. Comparison of the three classes of ATP synthases/ATPases. The archaeal A1AO ATP synthase consists of nine subunits, whereas the wellstudied bacterial F1FO ATP synthase has only eight different subunits. The yeast V1VO ATPase has 14 different subunits. Same or similar subunits of A1AO ATP synthases, F1FO ATP synthases, and V1VO ATPases are illustrated with the same color. The yeast V1VO ATPase has three different c subunits (c, c′, and c″), encoded by different genes, that build the rotor. Subunit a of the yeast V1VO ATPase has two isoforms Vph1p (complex targeting to the vacuole) and Stv1p (complex targeting to the Golgi apparatus), which are not mentioned in the figure. X1 = A1 or V1 or F1 and XO = AO or FO or VO. made of two rotary motors, the A1/F1/V1 and the AO/FO/VO motors, connected by a central stalk and one to three peripheral stalks. A1AO ATP synthases, F1FO ATP synthases, and V1VO ATPases catalyze a reaction that in vitro is fully reversible. In the ATP hydrolysis mode, ATP hydrolyzed by the soluble motor, localized in the cytoplasm of bacteria and archaea, drives rotation of the central stalk that is physically connected to the membrane-embedded, ion-translocating motor. Rotation of this motor is obligatorily coupled to the translocation of ions, in the hydrolysis mode from the inside to the outside. In the synthesis mode, the electrochemical ion potential drives the influx of ions into the motor, forcing it to rotate. Ions are translocated from the outside to the inside, and the central stalk starts to rotate. Rotational energy is then transmitted to the soluble motor where it is used to synthesize ATP (M€ uller & Gr€ uber, 2003). A major functional difference is that the predominant physiological function of A1AO and F1FO ATP synthases is to synthesize ATP at the expense of an electrochemical ion gradient, whereas the V1VO ATPases are designed by nature to drive ion transport at the expense of ATP hydrolysis in organelles of eukaryotes. ATP synthesis in eukaryotes is by F1FO ATP synthases that reside in mitochondria and chloroplasts. Interestingly, despite their major difference in function, the A1AO ATP synthase is evolutionarily more related to V1VO ATPases than to F1FO ATP synthases. This is consistent with the phylogenetic tree of life that shows a rather late divergence of archaea and eukarya from a common ancestor. It should be noted in this connection that the ‘tree of life’ has actually become a ‘bush of life’. FEMS Microbiol Rev 38 (2014) 449–472 Genes are transferred between the domains of life by horizontal gene transfer. This also holds true for archaea and bacteria (Averhoff, 2009). Actually, a number of bacteria have ‘stolen’ the A1AO ATP synthase genes from archaea. This is true for Thermus thermophilus and Enterococcus hirae and may be also true for other bacteria, as evident from genome sequences. ATPase genes from bacteria annotated as V1VO ATPase genes are actually A1AO ATP synthase genes! As long as these wrong annotations are not taken out of data bases, they will continue due to the use of annotation programs. How is the rotational energy used for ATP synthesis in the A1/F1 domain? The A1 domain consists of three alternating A and B subunits and the F1 domain of three alternating a and b subunits (Table 1). All subunits contain the Walker motifs A and B, which are located at the boundary surface of each subunit. These Walker motifs have conserved amino acids for the binding of nucleotides (Boyer, 1997). The three catalytic centers for ATP synthesis and ATP hydrolysis are built predominantly by amino acids of subunit A in A1AO ATP synthases and by amino acids from subunit b in F1FO ATP synthases (Weber et al., 1995; Weber & Senior, 1996; Kumar et al., 2010). The Walker A motif, also named P-loop (Walker et al., 1982), has a hairpin structure with the consensus sequence (GXXXXGKT), which is directly involved in coordination of the c-phosphoryl group of the nucleotide (Hanson & Whiteheart, 2005). Similar to B subunits of V1VO ATPases, the B subunits of A1AO ATP synthases (functionally equivalent to the a subunits of F1FO ATP synthases) do not contain a conserved P-loop motif ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 460 F. Mayer & V. Müller Table 1. Comparison of subunit composition of A1AO ATP synthases, V1VO ATPases, and F1FO ATP synthases* A1AO ATP synthase (nine subunits)† V1VO ATPase (14 subunits)‡ F1FO ATP synthase (eight subunits)§ H a c E C F A B D – – – – – G a c, c′, c″ E d F A B D C H – – e – a c – – e b a c – – d b – *Homologous subunits are in the same row. † Subunit composition of the A1AO ATP synthase from Pyrococcus furiosus. ‡ Subunit composition of the V1VO ATPase from yeast. § Subunit composition of the F1FO ATP synthase from Escherichia coli. (Olendzenski et al., 1998), but they are able to selectively bind ADP and ATP (Weber et al., 1995; Weber & Senior, 1996; Kumar et al., 2008). The Walker B motif is involved in coordination of the nucleotides and contains a glutamate, which activates a water molecule, which is essential for the cleavage of the phosphoanhydride bond during ATP hydrolysis (Abrahams et al., 1994). ATP is synthesized at the boundary surface of subunits A and B (a and b in F1FO ATP synthases) in a ‘binding change’mechanism using the rotational energy from the membrane-embedded motor and the central stalk. In the ‘open state’, the boundary surface between subunits A and B (a and b in F1FO ATP synthases) is open for the binding of ADP and Pi. After the binding of ADP and Pi, the central stalk is rotating by 120°, and ATP is synthesized from ADP and Pi, which is called the ‘tight state’. A further rotation of the central stalk by 120° leads to the ‘loose state’ in which the ATP is released from the boundary surface. After a rotation of the central stalk of another 120°, the boundary surface is again in the ‘open state’. With this mechanism, three ATPs can be synthesized by one rotation of the membrane-embedded motor (Boyer, 1989; Yoshida et al., 2001; Nakanishi-Matsui et al., 2010). ATP synthesis is reversible in vitro, but it is important to note that this is different in vivo. For example, a thiol modification as observed in plants turns off the ‘hydrolysis mode’ at night (Hisabori et al., 2002). Moreover, energy conservation in eukaryotes is in mitochondria or chloroplasts and catalyzed by F1FO ATP synthases. In contrast, energization of membranes of eukaryal ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved compartments such as the vacuole or the endoplasmic reticulum is by the generation of an ion gradient across the organelle membranes by the action of rotary ATP hydrolases (the V1VO ATPases). Nature evolved those enzymes to rest in the ‘hydrolysis mode’ by changing the number of ion-binding sites in the membrane-embedded rotor. The number of ion-binding sites in the rotor is determined by the primary structure of the c subunit and the number of c subunits in the ring (Fig. 7). In its simplest form, the c subunit contains two transmembrane helices connected by a small, cytoplasmic loop. Each c subunit has one ion-binding site, and a carboxylate as in a glutamate or aspartate residue is essential to bind and translocate H+. The H+ is bound between the carboxyl oxygen of a conserved carboxylate and the backbone carbonyl of another amino acid, for example phenylalanine (Pogoryelov et al., 2009). Very few c subunits, notably from anaerobic bacteria and archaea, bind Na+ instead uller et al., 2001). The experiof H+ (Dimroth, 1992; M€ mentally determined Na+-binding site from Ilyobacter tartaricus contains the before-mentioned carboxylate (Glu65) and four more amino acid residues that coordinate the Na+ (Murata et al., 2005b; Meier et al., 2009), notably Gln32, Val63, Ser66, and Thr67 (over a buried structural water molecule). Tyr70 is forming a hydrogen bond with Glu65 and stabilizes the geometry of the ion coordination shell, but is not directly involved in Na+ coordination. Noteworthy, only Gln32, Glu65, and Ser66 are sufficiently conserved in evolution and found in all methanogens, Pyrococci, and Thermococci. (However, it should be noted that Gln32 can be changed to Glu, Glu65 to Asp, and Ser66 often to Thr.) Therefore, the switch from a sodium ion (P. furiosus) to a proton-binding site (Spirulina platensis) is just by a change in three residues within the large complex of altogether 6657 residues (as calculated for the A1AO ATP synthase from P. furiosus)! The individual c subunits assemble to a ring that rotates against the stator subunit a. Not only the ionbinding site with its conserved carboxylate, but also a conserved arginine of the membrane-embedded part of subunit a is essential for ion translocation (Fig. 8). It is suggested that the deprotonated carboxylate of the c subunit is forming a salt bridge with the positively charged arginine of subunit a and keeps the carboxylate in the so-called ‘unlocked conformation’ (state 2). An ion (H+ or Na+), transported by the first ion channel of subunit a from the periplasm to the hydrophilic a/c interface will compete with this ion pair, which will allow the protonation of (or binding of Na+ to) the carboxylate (state 3). A c subunit with a protonated (Na+-bound) carboxylate can enter the lipid interface in this ‘ion-locked conformaFEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation 461 Fig. 7. Rotor subunits and rotor composition in archaeal A1AO ATP synthases. A1AO ATP synthases have a higher diversity in c subunits than F1FO ATP synthases or V1VO ATPases. Most archaea, like Methanosarcina mazei, have a c subunit with one hairpin and one ion-binding site (E, glutamate residue). Methanothermobacter thermautotrophicus has a c subunit consisting of two hairpins with two ion-binding sites. The c subunit of Methanocaldococcus jannaschii has a triplicated hairpin with two ion-binding sites, and the c subunit of Methanopyrus kandleri consists of 13 hairpins with 13 ion-binding sites. Based on a hypothetical c ring composed of 24 transmembrane helices, the number of monomers decreases with the temperature from 12 to 1 in the different species. The ion/ATP stoichiometry was calculated using the hypothetical c ring composition. tion’ (state 0). After rotation of the c ring and re-entering the hydrophilic a/c interface from the other side, the carboxylate in the ‘ion-locked conformation’ rotates outwards and releases the ion, with the help of the electrostatic influence of the arginine, over the second ion channel of subunit a into the cytoplasm (state 1). Afterward, the carboxylate is in the ‘unlocked conformation’ (state 2) again and opened for re-entering of the next ion (Pogoryelov et al., 2009). The rotational movement of the c ring is coupled to the rotation of the central stalk, which transmits this rotational energy into the soluble domain (Fig. 8). As mentioned above, the number of ion-binding sites in the rotor is determined by the primary structure of the c subunit and the number of c subunits in the ring (M€ uller, 2004). For example, if the c ring has 12 monomers, with two transmembrane helices each, the c ring has 12 ionbinding sites. Because the soluble motor has three ATPhydrolyzing/ATP-synthesizing centers, the ion-to-ATP ratio is four. One way to change this ratio is to change the number of ion-binding sites in the c ring either by changing the stoichiometry of monomers in the ring or by changing the primary structure of the monomers. Actually, the c subunit of V1VO ATPases arose by gene duplication of an ancestral 8-kDa c subunit gene followed by fusion of the products. A c subunit twice as big has no consequences for the efficiency of the enzyme (Jones & Fillingame, 1998). FEMS Microbiol Rev 38 (2014) 449–472 However, one ion-binding site was lost during the duplication event. Thus, if the same number of transmembrane helices is compared, the rotor of V1VO ATPases has only half the number of ion-binding sites. This is nature’s very clever way to lock the enzyme in the ‘hydrolysis mode’ in vivo. ATP synthesis is thermodynamically impossible, but the enzyme is a very good ion pump: because it translocates fewer ions, it is able to generate steeper gradients (Nelson, 1992; c.f. Eqn. 11). Up to the discovery of the A1AO ATP synthases, there was a clear difference in structure and function of ATP synthases/ATPases. The evolution of a duplicated c subunit with only half the number of ion-binding sites was seen as the starting point for the evolution of ATPases locked in the hydrolysis mode (V1VO ATPases). Later, additional subunits were evolved to reflect the need for the interaction of the enzyme with a complex regulatory network in the eukaryotic cell (Wieczorek et al., 2000; Kane & Smardon, 2003; Forgac, 2007). With the improvement in DNA sequencing technology, genome sequences from archaea became available. These data revealed that the A1AO ATP synthases form a phylogenetically distinct group of enzymes. Moreover, these data showed that A1AO ATP synthases have an extraordinary variety in the size of c subunits as well as the number of ion-binding sites per monomer (M€ uller, 2004), questioning the suggested evolution of ATP synthases/ATPases. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 462 Structure of the A1AO ATP synthase The first insights into the structure of the A1AO ATP synthases came from analyses of electron micrographs of enzymes isolated from hyperthermophiles. The global structures of the A1AO ATP synthases from M. jannaschii Fig. 8. Proposed mechanism of ion translocation in the membraneembedded domain of the ATP synthase. Individual c subunits assemble to a ring that rotates against the stator subunit a. It is suggested that the deprotonated carboxylate of the c subunit is forming a salt bridge with the positively charged arginine of subunit a and keeps the carboxylate in the ‘unlocked conformation’ (state 2). An ion (H+ or Na+) transported by the first ion channel of subunit a will compete with this ion pair and allows the protonation of (or binding of Na+ to) the carboxylate (state 3). A c subunit with a protonated (Na+ bound) carboxylate can enter the lipid interface in the ‘ion-locked conformation’ (state 0). After rotation of the c ring and re-entering the hydrophilic a/c interface from the other side, the carboxylate in the ‘ion-locked conformation’ rotates outwards and releases the ion over the second ion channel of subunit a into the cytoplasm (state 1). Afterward, the carboxylate is in the ‘unlocked conformation’ (state 2) again and opened for re-entering of the next ion (Pogoryelov et al., 2009). F. Mayer & V. Müller and P. furiosus at a resolution of 1.8 and 2.3 nm, respectively, showed features similar to F1FO ATP synthases, but also some unexpected characteristics (Coskun et al., 2004; Vonck et al., 2009), which might be an adaptation to the high temperatures of their environments. In general, the A1AO ATP synthase has a bipartite structure, consisting of an A1 domain and AO domain, which form a pair of coupled rotary motors connected with one central and two peripheral stalks (Fig. 9). Analyses of the molecular mass of the A1AO ATP synthase of P. furiosus by laser-induced liquid bead ion desorption mass spectrometry (LILBID-MS) revealed a molecular mass of 730 10 kDa and a stoichiometry of the enzyme complex of A3B3CDE2FH2ac10. The soluble A1 domain has the catalytic activity, and the hydrophobic membrane-embedded AO domain is responsible for ion translocation across the membrane. The A1 domain, often also described as catalytic head domain, shows pseudo-3-fold symmetry and clear asymmetry because of the presence of the central stalk (Vonck et al., 2009) and has a strong resemo1 blance to the isolated A1 domain from M. mazei G€ (Coskun et al., 2004). The head domain has a central cavity, which is surrounded by three alternating A and B subunits, where A is the catalytic subunit comparable to the b subunit in F1FO ATP synthases (Fig. 10). Subunit A can be clearly identified in the 3D reconstruction by its spike-like structures, and the gen atpA contains a 270bp-long intein, which makes atpA from archaea longer than its homologues (Bowman et al., 1988; Zimniak et al., 1988; Hirata et al., 1990; Puopolo et al., 1991; Vonck et al., 2009). The asymmetric head domain is connected to the central stalk, which is composed of subunits C, D, and F. Subunit C has a funnel-shape structure, composed of three domains, and is directly placed on top of the membrane-embedded c ring (Bernal & Stock, 2004; Iwata et al., 2004; Murata et al., 2005a; Vonck et al., 2009). Furthermore, a knob-like structure over subunit C Fig. 9. 3D reconstruction of the A1AO ATP synthase from Pyrococcus furiosus. Surface representation of the ATP synthase in top view, bottom view, and side view orientation. The global structure was obtained by single particle analyses and 3D reconstruction of the A1AO ATP synthase from P. furiosus. Main features of this new class of A1AO ATP synthases are a collar-like structure (subunit a) and two peripheral stalks (each composed of an EH complex). Scale bar represents 100 A (Vonck et al., 2009). ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved FEMS Microbiol Rev 38 (2014) 449–472 463 Archaeal life under extreme energy limitation (a) (e) (b) (f) (c) (g) (d) (h) Fig. 10. Global structure of the A1AO ATP synthase from Pyrococcus furiosus. Fitting of known X-ray structures into the electron density map of the A1AO ATP synthase from P. furiosus. (a–f) Cut open views of the map. (a–d) Vertical views of the map, cut at the levels indicated in (e). (a) Fitting of A3B3, subunit A (blue) from Pyrococcus horikoshii (PDB: 1VDZ) and subunit B (green) from Methanosarcina mazei (PDB: 2C61). (b) Fitting of the F subunit (purple) from P. furiosus (PDB: 2QAI). (c) Fitting of the C subunit (turquoise) from Thermus thermophilus (PDB: 1R5Z). (d) Fitting of the 20-hairpin A-type c ring (gold) from Enterococcus hirae (PDB: 2BL2). (e, f) Two perpendicular views along the membrane plane. The EC domain (pink) from P. horikoshii is fitted to the knobs (PDB: 2DM9). (g) All fitted subunits in the same orientation as (f). (h) Model for the stator with the a subunit in yellow and the EH dimers in pink (Vonck et al., 2009). is visible in the 3D reconstruction. This is subunit F, which has a globular N-terminal domain and an elongated C-terminal part, which extends deep into the A1 head domain and interacts there with subunit B (Gayen et al., 2007; Vonck et al., 2009). The third subunit of the central stalk is D, which has a high a-helical content and can be cross-linked to the inside of subunit B (Xu et al., 1999; Arata et al., 2002). Unfortunately, subunit D could not be detected in the 3D reconstruction of the A1AO ATP synthase of P. furiosus, because of the low resolution of 2.3 nm of the density map (Vonck et al., 2009). New structural features are a collar-like structure located perpendicular to the c ring and two peripheral stalks (Fig. 10). In contrast, the F1FO ATP synthases of bacteria, mitochondria, and chloroplasts have only one peripheral stalk and no collar. Each of the two peripheral stalks of A1AO ATP synthases consists of a dimer of subunits E and H (Vonck et al., 2009). As shown for the A1AO ATP synthase of T. thermophilus, the EH heterodimer is formed by two distinct domains, a globular head domain and a 140- A-long right-handed coiled coil, a protein fold, which is quite unusual in nature (Lee et al., 2010). This EH dimer is connected to the A1 domain. Fluorescence correlation spectroscopy showed that subunit E of the EH dimer is binding stronger to subunit B FEMS Microbiol Rev 38 (2014) 449–472 than to the catalytic subunit A (Hunke et al., 2011). The stalks are mounted to the collar, located perpendicular to the c ring that has two long crossed a-helices with four bab-motifs. It might be possible that these four motifs are the connecting sites for the peripheral stalks to subunit a and therefore to the AO domain. Subunit a has also a membrane-embedded part, which is composed of 7–8 transmembrane a-helices (Vonck et al., 2009). These a-helices are in close contact with the c ring, and they presumably form two ‘ion channels’ to load the c ring with coupling ions over the first channel and to remove the ion again from the c ring over the second ion channel (Lau & Rubinstein, 2012). Rotor composition and coupling efficiencies in A1AO ATP synthases An outstanding feature of A1AO ATP synthases is their difference in size of c subunits and the number of iontranslocating residues (M€ uller, 2004). Most archaea have the typical 8-kDa c subunit, like F-type ATP synthases (Inatomi et al., 1989; Wilms et al., 1996; Ihara et al., 1997; Steinert et al., 1997), but some archaea have very unusual c subunits. Species of the genera Pyrococcus (experimentally proved) and Thermococcus (deduced from ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 464 DNA data) have a 16-kDa c subunit with four transmembrane helices, but only one ion-binding site (Mayer et al., 2012a). In contrast, the 16-kDa c subunits from the methanogens M. thermautotrophicus (experimentally; Ruppert et al., 2001) and Methanosphaera stadtmanae (deduced) have two ion-binding sites per monomer (M€ uller, 2004; Fricke et al., 2006). M. jannaschii has a triplicated c subunit (experimentally) with a molecular mass of around 21 kDa and two ion-binding sites; the third one was lost during evolution (Ruppert et al., 1999). Also M. kandleri is very special, because it is the only archaeon that has a c ring consisting of only one c subunit with 13 hairpins and therefore 13 ion-binding sites (Slesarev et al., 2002; M€ uller, 2004), as deduced from the DNA sequence. What are the physiological consequences of the different numbers of ion-binding sites per c ring? Obviously, the ion-to-ATP ratio changes and this is an important, if not the most important, parameter to adapt ATP synthesis to low-energy environments. According to Eqn. (12), a phosphorylation potential (DGp) of ~ +50 to +70 kJ mol1 would be sustained by the use of n = 2.9–4.0 ions/ATP at a physiological electrochemical ion potential of 180 mV (D~ lion ; Sch€afer et al., 1999). Three ATP-synthesizing subunits in A1 and 12 ion-translocating subunits in the c ring calculate actually to four ions per ATP. In this example, the loss of one ion-binding site in every second hairpin of the rotor would lower the number of ions by half. The resulting value for ‘n’ of two is not enough to synthesize ATP at a phosphorylalHþ tion potential of ~ +50 to +70 kJ mol1 at a given D~ of 180 to 200 mV, and thus, the enzyme is no longer able to pump ions against this potential and thus unable to synthesize ATP. Apparently, the ATP synthase of P. furiosus is an ATP synthase that operates in the synthesis mode, although it has a c subunit with only one ion-binding site in four transmembrane helices (Mayer et al., 2012a) that would prevent the enzyme from working as ATP synthase. Moreover, the function as ATP synthase is of physiological importance because the ferredoxin-driven H+ translocation by the Mbh enables additional membrane energization (see Fig. 4) that should be and is used for ATP synthesis (Sapra et al., 2003). But how can this apparent contradiction be solved? The answer is rather simple: by adding more c subunits and thus more ion-binding sites to the c ring. Indeed, the c ring of P. furiosus contains 10 c subunits, each having one ion-binding site in four transmembrane helices (Vonck et al., 2009). This gives an ion/ATP ratio of 3.3. The ratio is not unusual and, for example, also predicted for yeast ATP synthase (Petersen et al., 2012) or the Na+ F1FO ATP synthase from Acetobacterium woodii (Fritz et al., 2008). ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved F. Mayer & V. Müller The deduction of the number of ion-binding sites is not unambiguous for the other species. For example, M. jannaschii has a c subunit containing three hairpins, but only two ion-binding sites. If we take four subunits to make the c ring, we have 12 transmembrane helices with eight ion-binding sites; if we take 5, we will have 15 transmembrane helices and 10 ion-binding sites; six gives 18; and seven gives 21 with 12 or 14 ion-binding sites. These values are all within the physiological range. Clearly, the variation in c subunits of archaeal A1AO ATP synthases has to be of evolutionary advantage and may reflect an adaptation to the environment these archaea live in. Adaptations of A1AO ATP synthases to life under extreme conditions How proteins adapt to high temperature has been the subject of many excellent reviews (Somero, 1995; Macario et al., 1999; Robb & Clark, 1999; Stetter, 1999; Sterner & Liebl, 2001). However, one should keep in mind that most of the studies have been carried out with soluble enzymes and not with membrane proteins, which are more exposed to the ‘heat’ than soluble proteins that may be protected by, for example, solutes (Martins & Santos, 1995; Martins et al., 1997). Among the membrane protein complexes, the ATP synthases are special because they work by a rotational mechanism. Because a number of archaea live at the thermodynamic limit of life, the coupling mechanism has to be very tight, and any leakage of the coupling ion through the motor has to be avoided. To date, different features may classify as adaptations of A1AO ATP synthases to high temperatures. Some of these features were apparently evolved very early in the evolution, but then kept also by mesophiles. The stator of A1AO ATP synthases is built by subunit a on which two peripheral stalks are mounted that keep the A3B3 subassembly in place. Ion transport drives rotation of the central stalk within the static A3B3 subassembly. X-ray structures of subunit E from Pyrococcus horikoshii showed that the 112- A-long N-terminal a-helix is S-shaped like the backbone of a human being and would facilitate the storage of elastic energy (Balakrishna et al., 2012). The A1 domain is placed asymmetrically on the central stalk, which causes a wobbling of the head domain during rotation of the central stalk. Both peripheral stalks, consisting of EH dimers, are stabilizing the ‘wobbling head’ and are therefore essential to counteract the torque generated by the central stalk during ATP hydrolysis and ATP synthesis (Stewart et al., 2012). The most important difference to other ATP synthases is the length of the c subunits. Most mesophilic archaea have a 8-kDa c subunit with two transmembrane helices, FEMS Microbiol Rev 38 (2014) 449–472 465 Archaeal life under extreme energy limitation and the copy number in the ring is not known, but in F1FO ATP synthases, this varies from 8 to 15 (Mitome et al., 2004; Meier et al., 2005; Pogoryelov et al., 2009; Vollmar et al., 2009; Preiss et al., 2010; Watt et al., 2010). Let us assume that the rotor of the ATP synthases consists of 8–15 noncovalently bound c subunits that form a ring that rotates against the a subunit that is coupled to ion translocation, one ion per two transmembrane helices. It is easily conceivable that thermal fluctuations will result in less-stable interactions of c subunits in the ring. One way to circumvent the problem is to covalently link individual c subunits in the ring, and the more the covalent bonds, the higher is its stability (Fig. 7). Indeed, while most mesophiles have a c subunit with two transmembrane helices, Methanothermobacter and Methanosphaera as well as Pyrococcus and Thermococcus species have a c subunit with four transmembrane helices, and Methanocaldococcus species have a c subunit with six transmembrane helices. In M. kandleri, the c ring is most likely made by just one c subunit that has 26 predicted transmembrane helices (M€ uller, 2004). Not only the size of the c subunits, but also the coupling ion used may be seen as an adaptation to high temperatures and for low-energy environments. Methanogens, for example, live at the thermodynamic limit of life (Deppenmeier & M€ uller, 2008). In the cytochrome-free methanogens, Mtr is the only coupling site, and the entire pathway yields only two ions (Na+) translocated per mol methane formed, equivalent to 0.5 mol of ATP per mol CH4 (Thauer et al., 2008). Any loss of ions back into the cell by leakage would be detrimental. Biological membranes are leakier for protons than for sodium ions, and this effect is even bigger at high temperatures (van de Vossenberg et al., 1995). Thus, a sodium bioenergetics is advantageous and is seen as the primary event in the evolution of bioenergetics (Mulkidjanian et al., 2008; Poehlein et al., 2012). Indeed, cytochrome-free methanogens rely on sodium bioenergetics (Thauer et al., 2008). Mtr is the only coupling site in the metabolism of methanogens, and it is a Na+ pump. Thus, the ATP synthase should be highly Na+ specific. This is obviously the case (McMillan et al., 2011; Mayer et al., 2012a). The cytochrome-containing methanogens have a mixed H+ and Na+ potential, and as an adaptation to these two ion gradients, the ATP synthase is concurrently driven by Na+ and H+ gradients (Schlegel et al., 2012b). The ATP synthase of P. furiosus requires Na+ for ATP hydrolysis (Pisa et al., 2007), Na+ interacts with the c subunit (Mayer et al., 2012a), and a sodium ion-binding motif is conserved in subunit c. This would argue that the Mbh antiporter module generates a Na+ gradient. The same bioenergetic scenario may be present in the close relative T. onnurineus. FEMS Microbiol Rev 38 (2014) 449–472 A1AO ATP synthases have undergone multiple variations as adaptation to their environment, but the biggest change may be encountered in N. equitans. A search of the genome sequence revealed only genes encoding subunits A, B, D, a, and c. Apparently, subunits C, F, E, and H are not encoded (Waters et al., 2003). Although this has to be interpreted with caution, this may also be seen as an adaptation to the energy metabolism of N. equitans. It lacks respiratory enzymes that energize its membranes, and apparently, the A1AO ATPase/ATP synthase is the only ion pump present. The lack of subunits may be a result of a transition from an ATP synthase to an ATP-driven ion pump. Unfortunately, the cells are hard to grow, and purification and biochemical analysis of the enzyme have not been possible yet due to a lack of biomass. Concluding remarks Apparently, anaerobic archaeal life under extreme energy limitations has evolved sophisticated mechanisms to conserve even very little energy increments. Thermodynamics does not allow a direct coupling via substrate-level phosphorylation, but allows for membrane energization reactions. These involve enzymes such as methyltransferases that use an exergonic (DG0′ = 30 kJ mol1) methyl transfer reaction to generate a D~ lNaþ . However, this type of reaction is restricted to methanogenic archaea. A more widespread reaction is the reversible conversion of hydrogen according to Eqn. (18): Ferredoxinred þ 2Hþ þ n ionin $ Ferredoxinox þ H2 þ n ionout (18) Exergonic electron flow from reduced ferredoxin to H+ (E = 414 mV) is used to generate a D~ lion (Buckel & Thauer, 2013). This type of reaction is catalyzed by the Ech hydrogenase and the much more complex Mbhs of Pyrococcus or Thermococcus. The Ech hydrogenase is a proton pump, and the complexity of the Mbh may arise from the fusion of the hydrogenase module to a Na+/H+ antiporter module. A similar type of reaction is catalyzed by the Rnf complex that is found in some archaea, but very widespread in bacteria (Biegel et al., 2011). There, the electron acceptor of the ferredoxin-fueled electron transportation is uller, 2010) or methanophenNAD+ (bacteria; Biegel & M€ azine (Methanosarcina sp.; Schlegel et al., 2012a). Thus, the common principle is the use of a ferredoxin-fueled electron transport chain inherent to one or more protein complexes. Electron acceptors are NAD+ (bacteria), methanophenazine or protons. A second principle observed in archaea that live under extreme energy limitations is the use of Na+ instead of 0′ ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 466 H+ as coupling ion. A sodium ion-based bioenergetics was expected to be present in organisms that live at high NaCl concentrations, but astonishingly, none of the known halophilic or moderately halophilic archaea and bacteria has a sodium ion-based primary bioenergetics. They all have a secondary Na+ potential built by Na+/H+ antiporter used for various purposes such as pH regulation or flagellar rotation. The only exception are some marine bacteria that have evolved a Na+-pumping NADH:ubiquinone oxidoreductase (Nqr; Tokuda & Unemoto, 1984), but that also have proton-motive complexes III and IV of their respiratory chains. Flagellar rotation (Imae & Atsumi, 1989), but not ATP synthesis is driven by D~ lNaþ . This was always discussed as an adaptation to the marine environment, but in addition these environments are somewhat energy-limited because the pH in these ecosystems is well above 8. These bacteria and other alkaliphiles have a low D~ lHþ due to an inverse DpH, and the secondary sodium ion potential helps to move and take up nutrients (Krulwich et al., 1996). How ATP is synthesized by this low D~ lHþ is still a mystery (Meier et al., 2007; Matthies et al., 2009; Hicks et al., 2010). Apparently, the only primary sodium bioenergetics is found in anaerobes, bacteria, and archaea, the latter have been described here. Why is this? To answer this question, we have to have a look at the ecosystem: anaerobic environments often have high concentrations of organic acids such as acetate, propionate, or butyrate. These act as protonophores that enter the cell in the protonated form by diffusion and dissociate inside the cells, due to a higher pH in the cytoplasm. Thus, organic acids act as ‘proton ferries’. The ion pumped out sneaks back in through the back door and cannot be used for ATP synthesis. If you live on a diet, you better watch out that your food does not diffuse away. Organic acids cannot act as ‘Na+ ferries’, and this is why, some anaerobes have a sodium ion-based bioenergetics. But, it can get worse. Membranes are, in general, leakier to protons than to sodium ions (Lane & Martin, 2012). This difference increases at high temperature. Thus, Thermococcus that lives at 8 to 20 kJ mol1 (Kim et al., 2010) has to avoid back leakage of ions into the cytoplasm, and sodium ions are clearly superior over H+ in this regard. The third principle is that the ATP synthases are adapted to life under extreme energy limitation. If the primary gradient is a sodium ion gradient, then the ATP synthase should be driven by D~ lNaþ . That is indeed observed. Only a few (three) subtile changes in amino acids are required to change the ion specificity. Apart from the use of Na+, the A1AO ATP synthase shows remarkable structural adaptations: an exciting variety of c ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved F. Mayer & V. Müller subunits that most likely lead to rotors with different coupling efficiencies. The general trend is that the ion/ ATP stoichiometry is decreased. Why and how can this be of physiological importance? Considering life at only 17 kJ mol1, only one ion can be translocated across the membrane, and several rounds of metabolism are required to pump out the number of ions required to make one ATP. The lower the ion/ATP stoichiometry, the less is the number of substrates converted to make an ATP. Of course, this comparison is odious because ATP synthesis only works if the phosphorylation potential is much below +60 kJ mol1. This will bring us to the fourth principle. Maybe the phosphorylation potential in these archaea is indeed considerably lower than +60 kJ mol1. Data are not available, but in one case (Jetten et al., 1991), one can calculate, based on the measurement of the nucleotides, a DGP of around +40 kJ mol1. Describing the basis for life at thermodynamic equilibrium is a challenge for further studies. Be it Thermococcus or anaerobic methane oxidizers that both live at around equilibrium (Kim et al., 2010; Holler et al., 2011), the thermodynamics and, more important, mechanistic basis of energy conservation in these creatures are not understood. In the last decades, the basis of energy conservation in bacterial and mitochondrial electron transport chains has been deciphered down to the atomistic level. The principles are well understood. Now, it is the time to discover the beauty of diversity in microbial bioenergetics in unusual organisms such as the archaea. Much can be learned, also about the origin of bioenergetics. Acknowledgements This work was supported by a grant from the Deutsche Forschungsgemeinschaft (through SFB 807). References Abrahams JP, Leslie AGW, Lutter R & Walker JE (1994) Structure at 2.8 A resolution of F1-ATPase from bovine heart mitochondria. Nature 370: 621–628. Amend JP & Shock EL (2001) Energetics of overall metabolic reactions of thermophilic and hyperthermophilic archaea and bacteria. FEMS Microbiol Rev 25: 175–243. Andrei AS, Banciu HL & Oren A (2012) Living with salt: metabolic and phylogenetic diversity of archaea inhabiting saline ecosystems. FEMS Microbiol Lett 330: 1–9. Arata Y, Baleja JD & Forgac M (2002) Localization of subunits D, E, and G in the yeast V-ATPase complex using cysteine-mediated cross-linking to subunit B. Biochemistry 41: 11301–11307. Averhoff B (2009) Shuffling genes around in hot environments: the unique DNA transporter of Thermus thermophilus. FEMS Microbiol Rev 33: 611–626. FEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation Bae SS, Kim YJ, Yang SH, Lim JK, Jeon JH, Lee HS, Kang SG, Kim SJ & Lee JH (2006) Thermococcus onnurineus sp nov., a hyperthermophilic Archaeon isolated from a deep-sea hydrothermal vent area at the PACMANUS field. J Microbiol Biotechnol 16: 1826–1831. Balakrishna AM, Hunke C & Gr€ uber G (2012) The structure of subunit E of the Pyrococcus horikoshii OT3 A-ATP synthase gives insight into the elasticity of the peripheral stalk. J Mol Biol 420: 155–163. Batista AP & Pereira MM (2011) Sodium influence on energy transduction by complexes I from Escherichia coli and Paracoccus denitrificans. Biochim Biophys Acta 1807: 286–292. Batista AP, Fernandes AS, Louro RO, Steuber J & Pereira MM (2010) Energy conservation by Rhodothermus marinus respiratory complex I. Biochim Biophys Acta 1797: 509–515. Belay N, Mukhopadhyay B, Conway de Macario E, Galask R & Daniels L (1990) Methanogenic bacteria in human vaginal samples. J Clin Microbiol 28: 1666–1668. Bernal RA & Stock D (2004) Three-dimensional structure of the intact Thermus thermophilus H+-ATPase/synthase by electron microscopy. Structure 12: 1789–1798. Biegel E & M€ uller V (2010) Bacterial Na+-translocating ferredoxin:NAD+ oxidoreductase. P Natl Acad Sci USA 107: 18138–18142. Biegel E, Schmidt S, Gonzalez JM & M€ uller V (2011) Biochemistry, evolution and physiological function of the Rnf complex, a novel ion-motive electron transport complex in prokaryotes. Cell Mol Life Sci 68: 613–634. Blamey JM & Adams MW (1993) Purification and characterization of pyruvate ferredoxin oxidoreductase from the hyperthermophilic archaeon Pyrococcus furiosus. Biochim Biophys Acta 1161: 19–27. Blaut M, M€ uller V & Gottschalk G (1992) Energetics of methanogenesis studied in vesicular systems. J Bioenerg Biomembr 24: 529–546. Bl€ ochl E, Rachel R, Burggraf S, Hafenbradl D, Jannasch HW & Stetter KO (1997) Pyrolobus fumarii, gen. and sp. nov., represents a novel group of archaea, extending the upper temperature limit for life to 113°C. Extremophiles 1: 14–21. Bowman EJ, Tenney K & Bowman BJ (1988) Isolation of genes encoding the Neurospora vacuolar ATPase. Analysis of vma-1 encoding the 67 kDa subunit reveals homology to other ATPases. J Biol Chem 263: 13994–14001. Boyer PD (1989) A perspective of the binding change mechanism for ATP synthesis. FASEB J 3: 2164–2178. Boyer PD (1997) The ATP synthase – a splendid molecular machine. Annu Rev Biochem 66: 717–749. Brandt U (2011) A two-state stabilization-change mechanism for proton-pumping complex I. Biochim Biophys Acta 1807: 1364–1369. Br€asen C, Schmidt M, Grotzinger J & Sch€ onheit P (2008) Reaction mechanism and structural model of ADP-forming Acetyl-CoA synthetase from the hyperthermophilic archaeon Pyrococcus furiosus: evidence for a second active site histidine residue. J Biol Chem 283: 15409–15418. FEMS Microbiol Rev 38 (2014) 449–472 467 Brochier-Armanet C, Boussau B, Gribaldo S & Forterre P (2008) Mesophilic Crenarchaeota: proposal for a third archaeal phylum, the Thaumarchaeota. Nat Rev Microbiol 6: 245–252. Bryant FO & Adams MW (1989) Characterization of hydrogenase from the hyperthermophilic archaebacterium, Pyrococcus furiosus. J Biol Chem 264: 5070–5079. Buckel W & Thauer RK (2013) Energy conservation via electron bifurcating ferredoxin reduction and proton/Na+ translocating ferredoxin oxidation. Biochim Biophys Acta 1827: 94–113. € Radermacher M, M€ Coskun U, uller V, Ruiz T & Gr€ uber G (2004) Three-dimensional organization of the archaeal A1-ATPase from Methanosarcina mazei G€ o1. J Biol Chem 279: 22759–22764. Cross RL & M€ uller V (2004) The evolution of A-, F-, and V-type ATP synthases and ATPases: reversals in function and changes in the H+/ATP stoichiometry. FEBS Lett 576: 1–4. Cross RL & Taiz L (1990) Gene duplication as a means for altering H+/ATP ratios during the evolution of F0F1 ATPases and synthases. FEBS Lett 259: 22227–22229. Deppenmeier U (2002a) The unique biochemistry of methanogenesis. Prog Nucleic Acid Res Mol Biol 71: 223–283. Deppenmeier U (2002b) Redox-driven proton translocation in methanogenic archaea. Cell Mol Life Sci 59: 1–21. Deppenmeier U & M€ uller V (2008) Life close to the thermodynamic limit: how methanogenic archaea conserve energy. Results Probl Cell Differ 45: 123–152. Deppenmeier U, Blaut M & Gottschalk G (1991) H2: heterodisulfide oxidoreductase, a second energy- conserving system in the methanogenic strain G€ o1. Arch Microbiol 155: 272–277. Dimroth P (1992) Structure and function of the Na+-translocating ATPase of Propionigenium modestum. Acta Physiol Scand 146: 97–102. Dirmeier R, Keller M, Frey G, Huber H & Stetter KO (1998) Purification and properties of an extremely thermostable membrane-bound sulfur-reducing complex from the hyperthermophilic Pyrodictium abyssi. Eur J Biochem 252: 486–491. Dolfing J, Jiang B, Henstra AM, Stams AJ & Plugge CM (2008) Syntrophic growth on formate: a new microbial niche in anoxic environments. Appl Environ Microbiol 74: 6126–6131. Dr€ ose S, Krack S, Sokolova L et al. (2011) Functional dissection of the proton pumping modules of mitochondrial complex I. PLoS Biol 9: e1001128. Efremov RG, Baradaran R & Sazanov LA (2010) The architecture of respiratory complex I. Nature 465: 441–445. Fiala G & Stetter KO (1986) Pyrococcus furiosus sp. nov. represents a novel genus of marine heterotrophic archaebacteria growing optimally at 100°C. Arch Microbiol 145: 56–61. Forgac M (2007) Vacuolar ATPases: rotary proton pumps in physiology and pathophysiology. Nat Rev Mol Cell Biol 8: 917–929. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 468 Fricke WF, Seedorf H, Henne A, Kruer M, Liesegang H, Hedderich R, Gottschalk G & Thauer RK (2006) The genome sequence of Methanosphaera stadtmanae reveals why this human intestinal archaeon is restricted to methanol and H2 for methane formation and ATP synthesis. J Bacteriol 188: 642–658. Fritz M, Klyszejko AL, Morgner N, Vonck J, Brutschy B, M€ uller DJ, Meier T & M€ uller V (2008) An intermediate step in the evolution of ATPases: a hybrid F1FO rotor in a bacterial Na+ F1FO ATP synthase. FEBS J 275: 1999–2007. Gayen S, Vivekanandan S, Biukovic G, Gr€ uber G & Yoon HS (2007) NMR solution structure of subunit F of the methanogenic A1AO adenosine triphosphate synthase and its interaction with the nucleotide-binding subunit B. Biochemistry 46: 11684–11694. Giannone RJ, Huber H, Karpinets T, Heimerl T, Kuper U, Rachel R, Keller M, Hettich RL & Podar M (2011) Proteomic characterization of cellular and molecular processes that enable the Nanoarchaeum equitans–Ignicoccus hospitalis relationship. PLoS ONE 6: e22942. Gottschalk G & Thauer RK (2001) The Na+-translocating methyltransferase complex from methanogenic archaea. Biochim Biophys Acta 1505: 28–36. Hamamoto T, Hashimoto M, Hino M, Kitada M, Seto Y, Kudo T & Horikoshi K (1994) Characterization of a gene responsible for the Na+/H+ antiporter system of alkalophilic Bacillus species strain C-125. Mol Microbiol 14: 939–946. Hanson PI & Whiteheart SW (2005) AAA+ proteins: have engine, will work. Nat Rev Mol Cell Biol 6: 519–529. Harold FM (1986) The Vital Force: A Study of Bioenergetics. W.H. Freeman, New York. Hedderich R (2004) Energy-converting [NiFe] hydrogenases from archaea and extremophiles: ancestors of complex I. J Bioenerg Biomembr 36: 65–75. Hedderich R, Klimmek O, Kr€ oger A, Dirmeier R, Keller M & Stetter KO (1999) Anaerobic respiration with elemental sulfur and with disulfides. FEMS Microbiol Rev 22: 353–381. Hicks DB, Liu J, Fujisawa M & Krulwich TA (2010) F1FO-ATP synthases of alkaliphilic bacteria: lessons from their adaptations. Biochim Biophys Acta 1797: 1362–1377. Hirata R, Ohsumk Y, Nakano A, Kawasaki H, Suzuki K & Anraku Y (1990) Molecular structure of a gene, VMA1, encoding the catalytic subunit of H+-translocating adenosine triphosphatase from vacuolar membranes of Saccharomyces cerevisiae. J Biol Chem 265: 6726–6733. Hisabori T, Konno H, Ichimura H, Strotmann H & Bald D (2002) Molecular devices of chloroplast F1-ATP synthase for the regulation. Biochim Biophys Acta 1555: 140–146. Hoehler TM, Alperin MJ, Albert DB & Martens CS (2001) Apparent minimum free energy requirements for methanogenic archaea and sulfate-reducing bacteria in an anoxic marine sediment. FEMS Microbiol Ecol 38: 33–41. Holler T, Wegener G, Niemann H, Deusner C, Ferdelman TG, Boetius A, Brunner B & Widdel F (2011) Carbon and sulfur back flux during anaerobic microbial oxidation of methane ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved F. Mayer & V. Müller and coupled sulfate reduction. P Natl Acad Sci USA 108: 1484–1490. Huber H, Burggraf S, Mayer T, Wyschkony I, Rachel R & Stetter KO (2000) Ignicoccus gen. nov., a novel genus of hyperthermophilic, chemolithoautotrophic archaea, represented by two new species, Ignicoccus islandicus sp nov and Ignicoccus pacificus sp nov. Int J Syst Evol Microbiol 50: 2093–2100. Huber H, Hohn MJ, Rachel R, Fuchs T, Wimmer VC & Stetter KO (2002) A new phylum of archaea represented by a nanosized hyperthermophilic symbiont. Nature 417: 63–67. Huber H, K€ uper U, Daxer S & Rachel R (2012) The unusual cell biology of the hyperthermophilic Crenarchaeon Ignicoccus hospitalis. Antonie Van Leeuwenhoek 102: 203–219. Hunke C, Antosch M, M€ uller V & Gr€ uber G (2011) Binding of subunit E into the A-B interface of the A1AO ATP synthase. Biochim Biophys Acta 1808: 2111–2118. Hunte C, Zickermann V & Brandt U (2010) Functional modules and structural basis of conformational coupling in mitochondrial complex I. Science 329: 448–451. Ihara K, Watanabe S, Sugimura K & Mukohata Y (1997) Identification of proteolipid from an extremely halophilic archaeon Halobacterium salinarum as an N′,N′-dicyclohexyl-carbodiimide binding subunit of ATP synthase. Arch Biochem Biophys 341: 267–272. Imae Y & Atsumi T (1989) Na+-driven flagellar motors. J Bioenerg Biomembr 21: 705–716. Inatomi KI, Maeda M & Futai M (1989) Dicyclohexylcarbodiimide-binding protein is a subunit of the Methanosarcina barkeri ATPase complex. Biochem Biophys Res Commun 162: 1585–1590. Iwata M, Imamura H, Stambouli E et al. (2004) Crystal structure of a central stalk subunit C and reversible association/dissociation of vacuole-type ATPase. P Natl Acad Sci USA 101: 59–64. Jahn U, Gallenberger M, Paper W, Junglas B, Eisenreich W, Stetter KO, Rachel R & Huber H (2008) Nanoarchaeum equitans and Ignicoccus hospitalis: new insights into a unique, intimate association of two archaea. J Bacteriol 190: 1743–1750. Jenney FE Jr & Adams MW (2008) Hydrogenases of the model hyperthermophiles. Ann NY Acad Sci 1125: 252–266. Jetten MSM, Stams AJM & Zehnder AJB (1991) Adenine nucleotide content and energy charge of Methanothrix soehngenii during acetate degradation. FEMS Microbiol Lett 84: 313–317. Jones PC & Fillingame RH (1998) Genetic fusions of subunit c in the F0 sector of H+-transporting ATP synthase. Functional dimers and trimers and determination of stoichiometry by cross-linking analysis. J Biol Chem 273: 29701–29705. Juarez O & Barquera B (2012) Insights into the mechanism of electron transfer and sodium translocation of the Na+-pumping NADH:quinone oxidoreductase. Biochim Biophys Acta 1817: 1823–1832. FEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation Kane PM & Smardon AM (2003) Assembly and regulation of the yeast vacuolar H+-ATPase. J Bioenerg Biomembr 35: 313–321. Kashket ER (1982) Stoichiometry of the H+-ATPase of growing and resting, aerobic Escherichia coli. Biochemistry 21: 5534–5538. Kaster AK, Moll J, Parey K & Thauer RK (2011) Coupling of ferredoxin and heterodisulfide reduction via electron bifurcation in hydrogenotrophic methanogenic archaea. P Natl Acad Sci USA 108: 2981–2986. Kim YJ, Lee HS, Kim ES et al. (2010) Formate-driven growth coupled with H2 production. Nature 467: 352–355. Krebs MP & Khorana HG (1993) Mechanism of light-dependent proton translocation by bacteriorhodopsin. J Bacteriol 175: 1555–1560. Krulwich TA, Ito M, Gilmour R, Sturr MG, Guffanti AA & Hicks DB (1996) Energetic problems of extremely alkaliphilic aerobes. Biochim Biophys Acta 1275: 21–26. Kulik EM, Sandmeier H, Hinni K & Meyer J (2001) Identification of archaeal rDNA from subgingival dental plaque by PCR amplification and sequence analysis. FEMS Microbiol Lett 196: 129–133. Kumar A, Manimekalai MS & Gr€ uber G (2008) Structure of the nucleotide-binding subunit B of the energy producer A1AO ATP synthase in complex with adenosine diphosphate. Acta Crystallogr D Biol Crystallogr 64: 1110–1115. Kumar A, Manimekalai MS, Balakrishna AM, Jeyakanthan J & Gr€ uber G (2010) Nucleotide binding states of subunit A of the A-ATP synthase and the implication of P-loop switch in evolution. J Mol Biol 396: 301–320. K€ uper U, Meyer C, M€ uller V, Rachel R & Huber H (2010) Energized outer membrane and spatial separation of metabolic processes in the hyperthermophilic Archaeon Ignicoccus hospitalis. P Natl Acad Sci USA 107: 3152–3156. Kurr M, Huber R, Konig H, Jannasch HW, Fricke H, Trincone A, Kristjansson JK & Stetter KO (1991) Methanopyrus kandleri, gen. and sp. nov. represents a novel group of hyperthermophilic methanogens, growing at 110°C. Arch Microbiol 156: 239–247. Lane N & Martin WF (2012) The origin of membrane bioenergetics. Cell 151: 1406–1416. Lau WC & Rubinstein JL (2012) Subnanometre-resolution structure of the intact Thermus thermophilus H+-driven ATP synthase. Nature 481: 214–218. Lee LK, Stewart AG, Donohoe M, Bernal RA & Stock D (2010) The structure of the peripheral stalk of Thermus thermophilus H+-ATPase/synthase. Nat Struct Mol Biol 17: 373–378. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC & Relman DA (2004) Methanogenic archaea and human periodontal disease. P Natl Acad Sci USA 101: 6176–6181. Lewalter K & M€ uller V (2006) Bioenergetics of archaea: ancient energy conserving mechanisms developed in the early history of life. Biochim Biophys Acta 1757: 437–445. Li Q, Li L, Rejtar T, Lessner DJ, Karger BL & Ferry JG (2006) Electron transport in the pathway of acetate conversion to FEMS Microbiol Rev 38 (2014) 449–472 469 methane in the marine archaeon Methanosarcina acetivorans. J Bacteriol 188: 702–710. Ma K & Adams MW (2001) Hydrogenases I and II from Pyrococcus furiosus. Methods Enzymol 331: 208–216. Ma K, Weiss R & Adams MW (2000) Characterization of hydrogenase II from the hyperthermophilic archaeon Pyrococcus furiosus and assessment of its role in sulfur reduction. J Bacteriol 182: 1864–1871. Macario AJ, Lange M, Ahring BK & Conway de Macario E (1999) Stress genes and proteins in the archaea. Microbiol Mol Biol Rev 63: 923–967. Martins LO & Santos H (1995) Accumulation of mannosylglycerate and di-myo-inositol-phosphate by Pyrococcus furiosus in response to salinity and temperature. Appl Environ Microbiol 61: 3299–3303. Martins LO, Huber R, Huber H, Stetter KO, DaCosta MS & Santos H (1997) Organic solutes in hyperthermophilic archaea. Appl Environ Microbiol 63: 896–902. Matthies D, Preiss L, Klyszejko AL, Muller DJ, Cook GM, Vonck J & Meier T (2009) The c13 ring from a thermoalkaliphilic ATP synthase reveals an extended diameter due to a special structural region. J Mol Biol 388: 611–618. Mayer F, Leone V, Langer JD, Faraldo-G omez JD & M€ uller V (2012a) A c subunit with four transmembrane helices and one ion (Na+) binding site in an archaeal ATP synthase: implications for c ring function and structure. J Biol Chem 287: 39327–39337. Mayer F, K€ uper U, Meyer C, Daxer S, M€ uller V, Rachel R & Huber H (2012b) AMP-forming acetyl coenzyme A synthetase in the outermost membrane of the hyperthermophilic crenarchaeon Ignicoccus hospitalis. J Bacteriol 194: 1572–1581. McMillan DG, Ferguson SA, Dey D et al. (2011) A1AO-ATP synthase of Methanobrevibacter ruminantium couples sodium ions for ATP synthesis under physiological conditions. J Biol Chem 286: 39882–39892. Meier T, Polzer P, Diederichs K, Welte W & Dimroth P (2005) Structure of the rotor ring of F-Type Na+-ATPase from Ilyobacter tartaricus. Science 308: 659–662. Meier T, Morgner N, Matthies D, Pogoryelov D, Keis S, Cook GM, Dimroth P & Brutschy B (2007) A tridecameric c ring of the adenosine triphosphate (ATP) synthase from the thermoalkaliphilic Bacillus sp. strain TA2.A1 facilitates ATP synthesis at low electrochemical proton potential. Mol Microbiol 65: 1181–1192. Meier T, Krah A, Bond PJ, Pogoryelov D, Diederichs K & Faraldo-G omez JD (2009) Complete ion-coordination structure in the rotor ring of Na+-dependent F-ATP synthases. J Mol Biol 391: 498–507. Mesbah NM & Wiegel J (2012) Life under multiple extreme conditions: diversity and physiology of the halophilic alkalithermophiles. Appl Environ Microbiol 78: 4074–4082. Miller TL & Wolin MJ (1982) Enumeration of Methanobrevibacter smithii in human feces. Arch Microbiol 131: 14–18. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 470 Mitchell P (1966) Chemiosmotic coupling in oxidative and photosynthetic phosphorylation. Biol Rev Camb Philos Soc 41: 445–502. Mitome N, Suzuki T, Hayashi S & Yoshida M (2004) Thermophilic ATP synthase has a decamer c-ring: indication of noninteger 10:3 H+/ATP ratio and permissive elastic coupling. P Natl Acad Sci USA 101: 12159–12164. Moissl C, Rudolph C, Rachel R, Koch M & Huber R (2003) In situ growth of the novel SM1 euryarchaeon from a string-of-pearls-like microbial community in its cold biotope, its physical separation and insights into its structure and physiology. Arch Microbiol 180: 211–217. Mukohata Y, Ihara K, Tamura T & Sugiyama Y (1999) Halobacterial rhodopsins. J Biochem 125: 649–657. Mukund S & Adams MW (1995) Glyceraldehyde-3-phosphate ferredoxin oxidoreductase, a novel tungsten-containing enzyme with a potential glycolytic role in the hyperthermophilic archaeon Pyrococcus furiosus. J Biol Chem 270: 8389–8392. Mulkidjanian AY, Dibrov P & Galperin MY (2008) The past and present of sodium energetics: may the sodium-motive force be with you. Biochim Biophys Acta 1777: 985–992. M€ uller V (2004) An exceptional variability in the motor of archaeal A1AO ATPases: from multimeric to monomeric rotors comprising 6–13 ion binding sites. J Bioenerg Biomembr 36: 115–125. M€ uller V & Gr€ uber G (2003) ATP synthases: structure, function and evolution of unique energy converters. Cell Mol Life Sci 60: 474–494. M€ uller V, Winner C & Gottschalk G (1988) Electron transport-driven sodium extrusion during methanogenesis from formaldehyde + H2 by Methanosarcina barkeri. Eur J Biochem 178: 519–525. M€ uller V, Ruppert C & Lemker T (1999) Structure and function of the A1AO ATPases from methanogenic archaea. J Bioenerg Biomembr 31: 15–28. M€ uller V, Aufurth S & Rahlfs S (2001) The Na+ cycle in Acetobacterium woodii: identification and characterization of a Na+-translocating F1FO-ATPase with a mixed oligomer of 8 and 16 kDa proteolipids. Biochim Biophys Acta 1505: 108–120. Murata T, Yamato I & Kakinuma Y (2005a) Structure and mechanism of vacuolar Na+-translocating ATPase from Enterococcus hirae. J Bioenerg Biomembr 37: 411–413. Murata T, Yamato I, Kakinuma Y, Leslie AG & Walker JE (2005b) Structure of the rotor of the V-Type Na+-ATPase from Enterococcus hirae. Science 308: 654–659. Nakanishi-Matsui M, Sekiya M, Nakamoto RK & Futai M (2010) The mechanism of rotating proton pumping ATPases. Biochim Biophys Acta 1797: 1343–1352. Nelson N (1992) The vacuolar H+-ATPase – one of the most fundamental ion pumps in nature. J Exp Biol 172: 19–27. Nicholls DG & Ferguson SJ (1992) Bioenergetics, 2nd edn. Academic, London. Oesterhelt D & Tittor J (1989) Two pumps, one principle: light-driven ion transport in halobacteria. Trends Biochem Sci 14: 57–61. ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved F. Mayer & V. Müller Olendzenski L, Hilario E & Gogarten JP (1998) Horizontal gene transfer and fusing lines of descent: the archaeabacteria – a chimera? Horizontal Gene Transfer (Syvanen M & Kado C, eds), pp. 349–362. Chapman & Hall, London. Paper W, Jahn U, Hohn MJ, Kronner M, N€ather DJ, Burghardt T, Rachel R, Stetter KO & Huber H (2007) Ignicoccus hospitalis sp. nov., the host of ‘Nanoarchaeum equitans’. Int J Syst Evol Microbiol 57: 803–808. Petersen J, Forster K, Turina P & Graber P (2012) Comparison of the H+/ATP ratios of the H+-ATP synthases from yeast and from chloroplast. P Natl Acad Sci USA 109: 11150–11155. Pisa KY, Huber H, Thomm M & M€ uller V (2007) A sodium ion-dependent A1AO ATP synthase from the hyperthermophilic archaeon Pyrococcus furiosus. FEBS J 274: 3928–3938. Poehlein A, Schmidt S, Kaster A-K et al. (2012) An ancient pathway combining carbon dioxide fixation with the generation and utilization of a sodium ion gradient for ATP synthesis. PLoS ONE 7: e33439. Pogoryelov D, Yildiz O, Faraldo-G omez JD & Meier T (2009) High-resolution structure of the rotor ring of a proton-dependent ATP synthase. Nat Struct Mol Biol 16: 1068–1073. Preiss L, Yildiz O, Hicks DB, Krulwich TA & Meier T (2010) A new type of proton coordination in an F1FO-ATP synthase rotor ring. PLoS Biol 8: e1000443. Puopolo K, Kumamoto C, Adachi I & Forgac M (1991) A single gene encodes the catalytic A subunit of the bovine vacuolar H+-ATPase. J Biol Chem 266: 24564–24572. Rachel R, Wyschkony I, Riehl S & Huber H (2002) The ultrastructure of Ignicoccus: evidence for a novel outer membrane and for intracellular vesicle budding in an archaeon. Archaea 1: 9–18. Robb FT & Clark DS (1999) Adaptation of proteins from hyperthermophiles to high pressure and high temperature. J Mol Microbiol Biotechnol 1: 101–105. Robb FT, Maeder DL, Brown JR, DiRuggiero J, Stump MD, Yeh RK, Weiss RB & Dunn DM (2001) Genomic sequence of hyperthermophile Pyrococcus furiosus: implications for physiology and enzymology. Methods Enzymol 330: 134–157. Rohlin L & Gunsalus RP (2010) Carbon-dependent control of electron transfer and central carbon pathway genes for methane biosynthesis in the Archaean, Methanosarcina acetivorans strain C2A. BMC Microbiol 10: 62. Ruppert C, Kavermann H, Wimmers S, Schmid R, Kellermann J, Lottspeich F, Huber H, Stetter KO & M€ uller V (1999) The proteolipid of the A1AO ATP synthase from Methanococcus jannaschii has six predicted transmembrane helices but only two proton-translocating carboxyl groups. J Biol Chem 274: 25281–25284. Ruppert C, Schmid R, Hedderich R & M€ uller V (2001) Selective extraction of subunit D of the Na+-translocating methyltransferase and subunit c of the A1AO ATPase from the cytoplasmic membrane of methanogenic archaea by chloroform/methanol and characterization of subunit c of FEMS Microbiol Rev 38 (2014) 449–472 Archaeal life under extreme energy limitation Methanothermobacter thermoautotrophicus as a 16-kDa proteolipid. FEMS Microbiol Lett 195: 47–51. Sapra R, Bagramyan K & Adams MWW (2003) A simple energy-conserving system: proton reduction coupled to proton translocation. P Natl Acad Sci USA 100: 7545–7550. Sch€afer G, Engelhard M & M€ uller V (1999) Bioenergetics of the Archaea. Microbiol Mol Biol Rev 63: 570–620. Schink B (1997) Energetics of syntrophic cooperation in methanogenic degradation. Microbiol Mol Biol Rev 61: 262–280. Schiraldi C, Giuliano M & De Rosa M (2002) Perspectives on biotechnological applications of archaea. Archaea 1: 75–86. Schlegel K & M€ uller V (2013) Evolution of Na+ and H+ bioenergetics in methanogenic archaea. Biochem Soc Trans 41: 421–426. Schlegel K, Welte C, Deppenmeier U & M€ uller V (2012a) Electron transport during aceticlastic methanogenesis by Methanosarcina acetivorans involves a sodium-translocating Rnf complex. FEBS J 279: 4444–4452. Schlegel K, Leone V, Faraldo-Gomez JD & M€ uller V (2012b) Promiscuous archaeal ATP synthase concurrently coupled to Na+ and H+ translocation. P Natl Acad Sci USA 109: 947–952. Schleper C & Nicol GW (2010) Ammonia-oxidising archaea – physiology, ecology and evolution. Adv Microb Physiol 57: 1–41. Sch€ onheit P & Beimborn DB (1985) ATP synthesis in Methanobacterium thermoautotrophicum coupled to CH4 formation from H2-CO2 in the apparent absence of an electrochemical proton potential across the cytoplasmic membrane. Eur J Biochem 148: 545–550. Schut GJ, Boyd ES, Peters JW & Adams MW (2012) The modular respiratory complexes involved in hydrogen and sulfur metabolism by heterotrophic hyperthermophilic archaea and their evolutionary implications. FEMS Microbiol Rev 37: 182–203. Senior AE (1988) ATP synthesis by oxidative phosphorylation. Physiol Rev 68: 177–231. Sharma A, Kawarabayasi Y & Satyanarayana T (2012) Acidophilic bacteria and archaea: acid stable biocatalysts and their potential applications. Extremophiles 16: 1–19. Slater EC, Rosing J & Mol A (1973) The phosphorylation potential generated by respiring mitochondria. Biochim Biophys Acta 292: 534–553. Slesarev AI, Mezhevaya KV, Makarova KS et al. (2002) The complete genome of hyperthermophile Methanopyrus kandleri AV19 and monophyly of archaeal methanogens. P Natl Acad Sci USA 99: 4644–4649. Somero GN (1995) Proteins and temperature. Annu Rev Physiol 57: 43–68. Stams AJ & Plugge CM (2009) Electron transfer in syntrophic communities of anaerobic bacteria and archaea. Nat Rev Microbiol 7: 568–577. Steinert K, Wagner V, Kroth-Pancic PG & Bickel-Sandk€ otter S (1997) Characterization and subunit structure of the ATP FEMS Microbiol Rev 38 (2014) 449–472 471 synthase of the halophilic archaeon Haloferax volcanii and organization of the ATP synthase genes. J Biol Chem 272: 6261–6269. Sterner R & Liebl W (2001) Thermophilic adaptation of proteins. Crit Rev Biochem Mol Biol 36: 39–106. Stetter KO (1999) Extremophiles and their adaptation to hot environments. FEBS Lett 452: 22–25. Stetter KO (2006) History of discovery of the first hyperthermophiles. Extremophiles 10: 357–362. Stetter KO, K€ onig H & Stackebrandt E (1983) Pyrodictium gen. nov., a new genus of submarine disc-shaped sulphur reducing archaebacteria growing optimally at 105°C. Syst Appl Microbiol 4: 535–551. Stewart AG, Lee LK, Donohoe M, Chaston JJ & Stock D (2012) The dynamic stator stalk of rotary ATPases. Nat Commun 3: 687. Teague WE Jr & Dobson GP (1999) Thermodynamics of the arginine kinase reaction. J Biol Chem 274: 22459–22463. Thauer RK (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144: 2377–2406. Thauer RK, Jungermann K & Decker K (1977) Energy conservation in chemotrophic anaerobic bacteria. Bacteriol Rev 41: 100–180. Thauer RK, Kaster AK, Seedorf H, Buckel W & Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6: 579–591. Tokuda H & Unemoto T (1984) Na+ is translocated at NADH: quinone oxidoreductase segment in the respiratory chain of Vibrio alginolyticus. J Biol Chem 259: 7785–7790. van de Vossenberg JLCM, Ubbink-Kok T, Elferink MGL, Driessen AJM & Konings WN (1995) Ion permeability of the cytoplasmic membrane limits the maximum growth temperature of bacteria and archaea. Mol Microbiol 18: 925–932. van der Oost J, Schut G, Kengen SW, Hagen WR, Thomm M & de Vos WM (1998) The ferredoxin-dependent conversion of glyceraldehyde-3-phosphate in the hyperthermophilic archaeon Pyrococcus furiosus represents a novel site of glycolytic regulation. J Biol Chem 273: 28149–28154. Vollmar M, Schlieper D, Winn M, Buchner C & Groth G (2009) Structure of the c14 rotor ring of the proton translocating chloroplast ATP synthase. J Biol Chem 284: 18228–18235. Vonck J, Pisa KY, Morgner N, Brutschy B & M€ uller V (2009) Three-dimensional structure of A1AO ATP synthase from the hyperthermophilic archaeon Pyrococcus furiosus by electron microscopy. J Biol Chem 284: 10110–10119. Walker JE, Saraste M, Runswick MJ & Gay NJ (1982) Distantly related sequences in the alpha- and beta- subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J 1: 945–951. Waters E, Hohn MJ, Ahel I et al. (2003) The genome of Nanoarchaeum equitans: insights into early archaeal ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved 472 evolution and derived parasitism. P Natl Acad Sci USA 100: 12984–12988. Watt IN, Montgomery MG, Runswick MJ, Leslie AG & Walker JE (2010) Bioenergetic cost of making an adenosine triphosphate molecule in animal mitochondria. P Natl Acad Sci USA 107: 16823–16827. Weber J & Senior AE (1996) Binding and hydrolysis of TNP-ATP by Escherichia coli F1-ATPase. J Biol Chem 271: 3474–3477. Weber J, Bowman C, Wilkemounts S & Senior AE (1995) Alpha-aspartate 261 is a key residue in noncatalytic sites of Escherichia coli F-1-ATPase. J Biol Chem 270: 21045–21049. Welte C, Kr€atzer C & Deppenmeier U (2010a) Involvement of Ech hydrogenase in energy conservation of Methanosarcina mazei. FEBS J 277: 3396–3403. Welte C, Kallnik V, Grapp M, Bender G, Ragsdale S & Deppenmeier U (2010b) Function of Ech hydrogenase in ferredoxin-dependent, membrane-bound electron transport in Methanosarcina mazei. J Bacteriol 192: 674–678. Wieczorek H, Gr€ uber G, Harvey WR, Huss M, Merzendorfer H & Zeiske W (2000) Structure and regulation of insect plasma membrane H+ V-ATPase. J Exp Biol 203(Pt 1): 127–135. Wilms R, Freiberg C, Wegerle E, Meier I, Mayer F & M€ uller V (1996) Subunit structure and organization of the genes of ª 2013 Federation of European Microbiological Societies. Published by John Wiley & Sons Ltd. All rights reserved F. Mayer & V. Müller the A1AO ATPase from the archaeon Methanosarcina mazei G€ o1. J Biol Chem 271: 18843–18852. Wolfe RS (1992) Biochemistry of methanogenesis. Archaebacteria: Biochemistry and Biotechnology, Vol. 58, (Danson MJ, Hough DW & Lunt GG, eds), pp. 41–49. Biochemical Soc, London. Xu T, Vasilyeva E & Forgac M (1999) Subunit interactions in the clathrin-coated vesicle vacuolar H+- ATPase complex. J Biol Chem 274: 28909–28915. Yoshida M, Muneyuki E & Hisabori T (2001) ATP synthase – a marvellous rotary engine of the cell. Nat Rev Mol Cell Biol 2: 669–677. Zillig W, Holz I, Janekovic D, Klenk HP, Imsel E, Trent J, Wunderl S, Forjaz VH, Coutinho R & Ferreira T (1990) Hyperthermus butylicus, a hyperthermophilic sulfur-reducing archaebacterium that ferments peptides. J Bacteriol 172: 3959–3965. Zimniak L, Dittrich P, Gogarten JP, Kibak H & Taiz L (1988) The cDNA sequence of the 69-kDa subunit of the carrot vacuolar H+-ATPase. Homology to the beta-chain of F0F1-ATPases. J Biol Chem 263: 9102–9112. Zinder SH (1993) Physiological Ecology of Methanogens. Chapman & Hall, New York, NY. FEMS Microbiol Rev 38 (2014) 449–472