Membrane protein complexes can support both the generation and utilization of a transmembrane electrochemical proton potential (Δp), either by supporting transmembrane electron transfer coupled to protolytic reactions on opposite sides of the membrane or by supporting transmembrane proton transfer. Regarding the first mechanism, this has been unequivocally demonstrated to be operational for Δp-dependent catalysis of succinate oxidation by quinone in the case of the dihaem-containing SQR (succinate:menaquinone reductase) from the Gram-positive bacterium Bacillus licheniformis. This is physiologically relevant in that it allows the transmembrane Δp to drive the endergonic oxidation of succinate by menaquinone by the dihaem-containing SQR of Gram-positive bacteria. In the case of a related but different respiratory membrane protein complex, the dihaem-containing QFR (quinol:fumarate reductase) of the ϵ-proteobacterium Wolinella succinogenes, evidence has been obtained indicating that both mechanisms are combined, so as to facilitate transmembrane electron transfer by proton transfer via a both novel and essential compensatory transmembrane proton transfer pathway (‘E-pathway’). This is necessary because, although the reduction of fumarate by menaquinol is exergonic, it is obviously not exergonic enough to support the generation of a Δp. This compensatory E-pathway appears to be required by all dihaem-containing QFR enzymes and the conservation of the essential acidic residue on transmembrane helix V (Glu-C180 in W. succinogenes QFR) is a useful key for the sequence-based discrimination of these QFR enzymes from the dihaem-containing SQR enzymes.
According to Peter Mitchell's chemiosmotic theory , the energy released upon the oxidation of electron donor substrates in both aerobic and anaerobic respiration is transiently stored in the form of an electrochemical proton potential (Δp) across the energy-transducing membranes. Fundamentally, there are two mechanisms by which integral membrane proteins can act as catalysts in this coupling of electron-transfer reactions to the generation or utilization of a transmembrane Δp: the redox loop mechanism and the proton pump mechanism . The former essentially involves transmembrane electron transfer. Reduction reactions on one side of the energy-transducing membrane are associated with proton binding, whereas oxidation reactions on the opposite side of the membrane are associated with proton release (Figure 1A). The proton pump mechanism involves the actual translocation of protons across the membrane. As summarized in the present paper, both of these mechanisms are apparently harnessed together in a specific case of a single respiratory membrane protein complex (Figure 1B). The membrane protein in question is the dihaem-containing QFR (quinol:fumarate reductase) from the anaerobic ϵ-proteobacterium Wolinella succinogenes . QFR is the terminal enzyme of fumarate respiration [4,5], a form of anaerobic respiration which allows anaerobic bacteria to use fumarate instead of dioxygen as the terminal electron acceptor. QFR couples the two-electron reduction of fumarate to succinate (reaction a) to the two-electron oxidation of hydroquinone (quinol) to quinone (reaction b):
Electron and proton transfer in B. licheniformis SQR (A), in W. succinogenes QFR (B, D and E), and in fumarate respiration (C)
This reaction is part of an electron-transfer chain that enables the bacterium to grow with various electron donor substrates such as formate or hydrogen (Figure 1C).
The crystal structure of W. succinogenes QFR has been determined, by X-ray crystallography, in three different crystal forms, initially at a resolution of up to 2.2 Å (1 Å=0.1 nm) . In all three crystal forms, two heterotrimeric complexes of A, B and C subunits form a dimer of 260 kDa. Subunit A contains a covalently bound FAD and the site of fumarate reduction, subunit B, harbours three different iron–sulfur clusters, a [2Fe–2S], a [4Fe–4S] and a [3Fe–4S] centre, and the membrane-embedded subunit C binds two haem b groups (Figure 1D). On the basis of their proximity to the hydrophilic subunits, these are referred to as the proximal haem bP and the distal haem bD. Although it has long been known  that the two haem groups have different redox midpoint potentials, it has only recently been possible to assign the ‘high-potential’ haem to bP and the ‘low-potential’ haem to bD .
Subunit C also contains the active site of MKH2 (menaquinol) oxidation, close to the haem bD, and it has been shown by site-directed mutagenesis and by structurally and functionally characterizing the variant enzyme E66Q, in which the nearby Glu-C66 had been replaced by a glycine residue, that this residue is specifically essential for quinol oxidation . The orientation of the catalytic sites of fumarate reduction , associated with proton binding, and MKH2 oxidation , associated with proton release, towards opposite sites of the membrane indicated that quinol oxidation by fumarate should be an electrogenic process in W. succinogenes (Figure 1D), i.e. associated directly with the establishment of an electrochemical proton potential across the membrane.
Electrogenic catalysis by dihaemcontaining succinate:menaquinone oxidoreductases from Gram-positive bacteria
This electrogenic catalysis indeed appeared to be the case for some dihaem-containing representatives of the superfamily of QFRs and SQRs (succinate:menaquinone reductases). Succinate oxidation by MK (menaquinone), an endergonic reaction under standard conditions, had been proposed to be driven by the electrochemical proton potential in the Gram-positive bacterium Bacillus subtilis and other prokaryotes containing SQRs [11–13]. This is the analogous reaction to that suggested in Figure 1(D), but in the opposite direction (Figure 1A). However, the experiments on SQR had previously been performed only with whole cells and isolated membranes, and it has been questioned whether the observed effects are associated specifically with SQR . Recently, it has been shown that the dihaem-containing SQR, isolated from the Gram-positive bacterium Bacillus licheniformis and reconstituted into proteoliposomes, is unable to support the reduction of the soluble MK analogue EMN (2-ethyl-3-methyl-1,4-naphthoquinone) unless the protonophore CCCP (carbonyl cyanide m-chlorophenylhydrazone) is added . Apparently, the driving force of the reaction, even with a large excess of educt over product, was insufficient to support the establishment of a Δp across the proteoliposomal membrane. In order to increase this driving force, the quinone substrate was modified to increase its oxidation/reduction midpoint potential, as is the case for the soluble ubiquinone analogue EQ-0 (2,3-dimethoxy-5-ethyl-6-methyl-1,4-benzoquinone) (Figure 2). For an increased driving force of the reverse reaction, a lower redox midpoint potential than that of EMN and DMN (2,3-dimethyl-1,4-naphthoquinone) was achieved by designing the substrate MMANH2 (2-methyl-3-methylamino-1,4-naphthoquinol) .
Chemical structures of the quinones used for the summarized conclusive characterization experiments [15,16]
Two components contribute to Δp: the membrane potential, ΔΨ, and the proton gradient, ΔpH. The latter was measured by monitoring pH changes in the lumen of the proteoliposomes via the absorption properties of the proton-sensitive dye pyranine. Oxidation of the low-potential quinol MMANH2 by fumarate as catalysed by the proteoliposomal SQR from B. licheniformis was found to be associated with the lowering of the luminal pH . Conversely, reduction of the high-potential quinone EQ-0 was found to be associated with the increase of the luminal pH . The ΔΨ was measured with the help of electrodes selective for the lipophilic ions TPP+ (tetraphenylphosphonium) and TPB− (tetraphenylborate) respectively. In the case of quinol oxidation as well as in the case of quinone reduction, a ΔΨ could be measured, thus unequivocally demonstrating electrogenic catalysis in the case of this enzyme .
Electroneutral catalysis by dihaem-containing menaquinol:fumarate oxidoreductases from ϵ-proteobacteria
However, as discussed earlier , analogous experiments for isolated W. succinogenes QFR reconstituted into liposomes had shown that the oxidation of quinol by fumarate as catalysed by this enzyme is electroneutral [16–19].
To reconcile these apparently conflicting experimental observations, the so-called ‘E-pathway hypothesis’ (Figure 1E) was proposed . According to this working hypothesis, the transmembrane transfer of two electrons in W. succinogenes QFR is coupled to the compensatory parallel translocation of one proton per electron from the periplasm to the cytoplasm. The proton-transfer pathway used is transiently established during reduction of the haem groups and is closed in the oxidized enzyme. The two most prominent constituents of the proposed novel pathway were suggested to be the ring C propionate of the distal haem bD and, in particular, the amino acid residue Glu-C180, after which the ‘E-pathway’ was named (Figure 1E). Since the first proposal of this hypothesis, a number of theoretical  and experimental results [21–23] have been obtained that support it (reviewed in ).
Proof of the E-pathway hypothesis
A possible haem propionate involvement as a participant in the E-pathway has been investigated by combining 13C-labelling of the haem propionates with redox-induced FTIR spectroscopy . The redox transition of the distal haem led to protonation and/or environmental changes of (at least) one of the two distal haem propionates. Since it was established that the ring D propionate of the low-potential haem is involved in an extensive salt-bridge interaction with a nearby arginine residue [6,8], the obvious candidate for the observed effects is the ring C propionate, which is fully consistent with the proposed role of this residue in the E-pathway hypothesis.
The role of Glu-C180 in this context was first supported by multi-conformation continuum electrostatics calculations , predicting that this residue undergoes the combination of a change in protonation and conformation upon reduction of the haem groups, a result that was also obtained experimentally by the combination of FTIR difference spectroscopy  and site-directed mutagenesis, involving the replacement of Glu-C180 with a glycine residue . The mutant E180Q was unable to grow with fumarate as the terminal electron acceptor, an observation that we can now understand as a demonstration of the essential nature of this pathway for life under the conditions of fumarate respiration. The mutant did grow when fumarate was replaced by nitrate, and the variant QFR was produced. After refining the structure of the variant QFR at 2.2 Å resolution, any major structural changes compared to the structure of the wild-type enzyme could be ruled out .
The variant enzyme, after reconstitution into proteoliposomes, is unable to support the oxidation of DMNH2 (2,3-dimethyl-1,4-naphthoquinol) unless CCCP is added . Oxidation of the low-potential quinol is supported even in the absence of CCCP. In contrast with the results obtained with the wild-type enzyme, quinol oxidation by the E180Q variant was clearly associated with an acidification of the interior of the proteoliposomes (indicative of ΔpH generation)  as well as TPB− entry into the proteoliposomes (indicative of ΔΨ generation) . Taken together with the results obtained for the proteoliposomal wild-type enzyme, these results clearly demonstrate the presence and absence of the E-pathway in the wild-type and E180Q variant enzymes respectively .
Although the E-pathway hypothesis as depicted in Figure 1(E) is the simplest model compatible with the experimental data obtained so far, it is by no means unique. In particular, the simplifying assumption that both protons released upon quinol oxidation are released via the same pathway may turn out not to be true, as may the assumption that both E-pathway protons have the same entry point. In particular, the scenario that one proton is transferred directly from the quinol oxidation site to the E-pathway (Figure 3) also explains the available data. However, there is no such proton-transfer connectivity apparent from the structure of the oxidized enzyme, so this scenario would require an appropriate conformational change during catalysis.
Alternative implementation of the E-pathway hypothesis
Integration of Structures, Spectroscopies and Mechanisms: Second Joint German/British Bioenergetics Conference, a Biochemical Society Focused Meeting held at University of Edinburgh, U.K., 2–4 April 2008. Organized by Ulrich Brandt (Frankfurt, Germany), Steve Chapman (Edinburgh, U.K.), Peter Heathcoate (Queen Mary, University of London, U.K.), John Ingledew (St Andrews, U.K.), Mike Jones (Bristol, U.K.), Bernd Ludwig (Frankfurt, Germany), Fraser MacMillan (University of East Anglia, Norwich, U.K.), Hartmut Michel (Max-Planck-Institute for Biophysics, Frankfurt am Main, Germany), Peter Rich (University College London, U.K.) and John Walker (MRC Dunn Human Nutrition Unit, Cambridge, U.K.). Edited by Ulrich Brandt and Peter Rich.
carbonyl cyanide m-chlorophenylhydrazone
Present address: Department of Physiology, University of California, Los Angeles, CA 90095, U.S.A.