A mechanistic principle for proton pumping by cytochrome c oxidase

In aerobic organisms, cellular respiration involves electron transfer to oxygen through a series of membrane-bound protein complexes. The process maintains a transmembrane electrochemical proton gradient that is used, for example, in the synthesis of ATP. In mitochondria and many bacteria, the last enzyme complex in the electron transfer chain is cytochrome c oxidase (CytcO), which catalyses the four-electron reduction of O2 to H2O using electrons delivered by a water-soluble donor, cytochrome c. The electron transfer through CytcO, accompanied by proton uptake to form H2O drives the physical movement (pumping) of four protons across the membrane per reduced O2. So far, the molecular mechanism of such proton pumping driven by electron transfer has not been determined in any biological system. Here we show that proton pumping in CytcO is mechanistically coupled to proton transfer to O2 at the catalytic site, rather than to internal electron transfer. This scenario suggests a principle by which redox-driven proton pumps might operate and puts considerable constraints on possible molecular mechanisms by which CytcO translocates protons.     

Reduction of cytochrome c oxidase by a second electron leads to proton translocation

Cytochrome c oxidase, the terminal enzyme of cellular respiration in mitochondria and many bacteria, reduces O(2) to water. This four-electron reduction process is coupled to translocation (pumping) of four protons across the mitochondrial or bacterial membrane; however, proton pumping is poorly understood. Proton pumping was thought to be linked exclusively to the oxidative phase, that is, to the transfer of the third and fourth electron. Upon re-evaluation of these data, however, this proposal has been questioned, and a transport mechanism including proton pumping in the reductive phase--that is, during the transfer of the first two electrons--was suggested. Subsequently, additional studies reported that proton pumping during the reductive phase can occur, but only when it is immediately preceded by an oxidative phase. To help clarify the issue we have measured the generation of the electric potential across the membrane, starting from a defined one-electron reduced state. Here we show that a second electron transfer into the enzyme leads to charge translocation corresponding to pumping of one proton without necessity for a preceding turnover.     

Identification of the electron transfers in cytochrome oxidase that are coupled to proton-pumping

Mitochondrial cytochrome oxidase is a functionally complex, membrane-bound respiratory enzyme which catalyses both the reduction of O2 to water and proton-pumping. During respiration, an exogenous donor, cytochrome c, donates four electrons to O2 bound at the bimetallic haem alpha 3 Fe-Cu centre within the enzyme. These four electron transfers are mediated by the enzyme's haem alpha and CuA redox centres and result in the translocation of four protons across the inner mitochondrial membrane. The molecular mechanism of proton translocation has not yet been delineated, however, and in the absence of direct experimental evidence all four electron transfers have been assumed to couple equally to proton-pumping. Here, I report the effects of proton-motive force and membrane potential on two equilibria involving intermediates of the bimetallic centre at different levels of O2 reduction. The results show that only two of the electron transfers, to the 'peroxy' and 'oxyferryl' intermediates of the bimetallic centre, are linked to proton translocation, a finding which strongly constrains candidate mechanisms for proton-pumping.     

Proton translocation by cytochrome c oxidase.

Cell respiration in mitochondria and some bacteria is catalysed by cytochrome c oxidase, which reduces O2 to water, coupled with translocation of four protons across the mitochondrial or bacterial membrane. The enzyme's catalytic cycle consists of a reductive phase, in which the oxidized enzyme receives electrons from cytochrome c, and an oxidative phase, in which the reduced enzyme is oxidized by O2. Previous studies indicated that proton translocation is coupled energetically only to the oxidative phase, but this has been challenged. Here, with the purified enzyme inlaid in liposomes, we report time-resolved measurements of membrane potential, which show that half of the electrical charges due to proton-pumping actually cross the membrane during reduction after a preceding oxidative phase. pH measurements confirm that proton translocation also occurs during reduction, but only when immediately preceded by an oxidative phase. We conclude that all the energy for proton translocation is conserved in the enzyme during its oxidation by O2. One half of it is utilized for proton-pumping during oxidation, but the other half is unlatched for this purpose only during re-reduction of the enzyme.     

Structural insights into electron transfer in caa3-type cytochrome oxidase

Cytochrome c oxidase is a member of the haem copper oxidase superfamily (HCO). HCOs function as the terminal enzymes in the respiratory chain of mitochondria and aerobic prokaryotes, coupling molecular oxygen reduction to transmembrane proton pumping. Integral to the enzyme's function is the transfer of electrons from cytochrome c to the oxidase via a transient association of the two proteins. Electron entry and exit are proposed to occur from the same site on cytochrome c. Here we report the crystal structure of the caa3-type cytochrome oxidase from Thermus thermophilus, which has a covalently tethered cytochrome c domain. Crystals were grown in a bicontinuous mesophase using a synthetic short-chain monoacylglycerol as the hosting lipid. From the electron density map, at 2.36?? resolution, a novel integral membrane subunit and a native glycoglycerophospholipid embedded in the complex were identified. Contrary to previous electron transfer mechanisms observed for soluble cytochrome c, the structure reveals the architecture of the electron transfer complex for the fused cupredoxin/cytochrome c domain, which implicates different sites on cytochrome c for electron entry and exit. Support for an alternative to the classical proton gate characteristic of this HCO class is presented.     

Pumping of protons from the mitochondrial matrix by cytochrome oxidase

The stoichiometry and mechanism of redox-linked proton translocation by the mitochondrial respiratory chain is a major issue of debate in membrane bioenergetics. The function of cytochrome oxidase is a focal point of disagreement. In 1977 it was suggested that the terminal component of the respiratory chain, cytochrome oxidase, functions as a redox-linked proton pump. That and subsequent studies were based mainly on measurements of proton ejection from mitochondria or from vesicles reconstituted with isolated cytochrome oxidase, or on measurements of translocation of electrical charge equivalents across mitochondrial and vesicle membranes. This proton-translocating function of cytochrome oxidase is confirmed here by a quantitative determination of proton uptake from the inside (matrix) of intact mitochondria.     

Oxygen activation and the conservation of energy in cell respiration.

Many of the membrane-associated oxidases that catalyse respiratory reduction of O2 to water simultaneously couple this exergonic reaction to the translocation of protons across the inner mitochondrial membrane, or the cell membrane in prokaryotes, a process by which metabolic energy is conserved for subsequent synthesis of ATP. The molecular mechanism of O2 reduction and its linkage to H+ translocation are now emerging. The bimetallic haem iron-copper reaction centre in this family of enzymes is the critical structure for catalysis of both these processes.     

An atypical haem in the cytochrome b(6)f complex

Photosystems I and II (PSI and II) are reaction centres that capture light energy in order to drive oxygenic photosynthesis; however, they can only do so by interacting with the multisubunit cytochrome b(6)f complex. This complex receives electrons from PSII and passes them to PSI, pumping protons across the membrane and powering the Q-cycle. Unlike the mitochondrial and bacterial homologue cytochrome bc(1), cytochrome b(6)f can switch to a cyclic mode of electron transfer around PSI using an unknown pathway. Here we present the X-ray structure at 3.1 A of cytochrome b(6)f from the alga Chlamydomonas reinhardtii. The structure bears similarities to cytochrome bc(1) but also exhibits some unique features, such as binding chlorophyll, beta-carotene and an unexpected haem sharing a quinone site. This haem is atypical as it is covalently bound by one thioether linkage and has no axial amino acid ligand. This haem may be the missing link in oxygenic photosynthesis.     

Molecular mechanism of vectorial proton translocation by bacteriorhodopsin

Bacteriorhodopsin, a membrane protein with a relative molecular mass of 27,000, is a light driven pump which transports protons across the cell membrane of the halophilic organism Halobacterium salinarum. The chromophore retinal is covalently attached to the protein via a protonated Schiff base. Upon illumination, retinal is isomerized. The Schiff base then releases a proton to the extracellular medium, and is subsequently reprotonated from the cytoplasm. An atomic model for bacteriorhodopsin was first determined by Henderson et al, and has been confirmed and extended by work in a number of laboratories in the last few years. Here we present an atomic model for structural changes involved in the vectorial, light-driven transport of protons by bacteriorhodopsin. A 'switch' mechanism ensures the vectorial nature of pumping. First, retinal unbends, triggered by loss of the Schiff base proton, and second, a protein conformational change occurs. This conformational change, which we have determined by electron crystallography at atomic (3.2 A in-plane and 3.6 A vertical) resolution, is largely localized to helices F and G, and provides an 'opening' of the protein to protons on the cytoplasmic side of the membrane.     

Site-directed mutagenesis of cytochrome c shows that an invariant Phe is not essential for function

Phenylalanine 87 of yeast iso-1-cytochrome c (Phe 82 in horse heart and bonito) is phylogenetically conserved and occurs near the surface of the protein. It has been suggested that this residue is directly involved in electron transfer between cytochrome c and cytochrome c peroxidase (CCP) and may also control the polarity of the haem environment. Because Phe residues are not susceptible to chemical modification, no direct means of studying the functional role of Phe 87 has been available, so we have chosen Phe 87 as our initial target here to test the feasibility of using site-directed mutagenesis as a means of studying structure-function relationships in cytochrome c. We have changed the codon for Phe 87 to that of either a Ser, a Tyr or a Gly. The mutated genes have been introduced into a yeast strain lacking both isozymes of cytochrome c. Unlike the recipient strain, transformants grow on a non-fermentable carbon source, indicating that the mutant proteins can reduce cytochrome oxidase. The purified mutant proteins are similar to wild type with respect to their visible spectra, 20-70% as active as wild-type protein in the CCP assay, and their reduction potentials are lowered by as much as 50 mV. Thus Phe 87 is not essential for cytochrome c to transfer electrons but is involved in determining the reduction potential.     

Ferryl and hydroxy intermediates in the reaction of oxygen with reduced cytochrome c oxidase

Cytochrome c oxidase catalyses the 4-electron reduction of dioxygen to water and translocates protons vectorially across the inner mitochondrial membrane. Proposed reaction pathways for the catalytic cycle of the O2 reduction are difficult to verify without knowing the structures of the intermediates, but we now have such information for the catalytic intermediates in the first steps of the reaction of O2 with cytochrome c oxidase from resonance Raman spectroscopy, a technique that enables iron-ligand stretching modes to be identified. Here we report on two more key intermediates: a ferryl-oxo (Fe4 = O2-) and a ferric-hydroxy (Fe3+--OH-) intermediate at the level of 3- and 4-electron reduction, respectively. We identified these intermediates by their characteristic iron-oxygen stretching frequencies (786 cm-1 for Fe4+ = O2-, and 450 cm-1 for Fe3+ -- OH-) and oxygen and deuterium isotope shifts. The oxo atom in the ferryl intermediate is hydrogen-bonded and the iron-oxygen bond in the hydroxy intermediate is anomalously weak. With the identification of the primary, ferryl and hydroxy intermediates, the predominant structures at almost all stages of O2 reduction are now known and the catalytic pathway can be described with more certainty.     

Structure of the membrane domain of respiratory complex I

Complex I is the first and largest enzyme of the respiratory chain, coupling electron transfer between NADH and ubiquinone to the translocation of four protons across the membrane. It has a central role in cellular energy production and has been implicated in many human neurodegenerative diseases. The L-shaped enzyme consists of hydrophilic and membrane domains. Previously, we determined the structure of the hydrophilic domain. Here we report the crystal structure of the Esherichia coli complex I membrane domain at 3.0?? resolution. It includes six subunits, NuoL, NuoM, NuoN, NuoA, NuoJ and NuoK, with 55 transmembrane helices. The fold of the homologous antiporter-like subunits L, M and N is novel, with two inverted structural repeats of five transmembrane helices arranged, unusually, face-to-back. Each repeat includes a discontinuous transmembrane helix and forms half of a channel across the membrane. A network of conserved polar residues connects the two half-channels, completing the proton translocation pathway. Unexpectedly, lysines rather than carboxylate residues act as the main elements of the proton pump in these subunits. The fourth probable proton-translocation channel is at the interface of subunits N, K, J and A. The structure indicates that proton translocation in complex I, uniquely, involves coordinated conformational changes in six symmetrical structural elements.     

Directional electron transfer in ruthenium-modified horse heart cytochrome c

Cytochrome c can be modified by [(NH3)5RuII/III-] specifically at the imidazole moiety of histidine 33, and we have recently discussed the thermodynamics and kinetics of electron transfer within this modified protein. X-ray crystal structures of the oxidized and reduced forms of tuna cytochrome c indicate that the separation between the haem group of cytochrome c and the ruthenium label is 12-16 A. Internal electron transfer from the [(NH3)5RuII-] centre to the Fe(III) haem centre occurs with a rate constant k congruent to 53 s-1 (25 degrees C) (delta H = 3.5 kcal mol-1, delta S = -39 EU), as measured by pulse radiolysis. The measured unimolecular rate constant, k congruent to 53 s-1, is on the same timescale as a number of conformational changes that occur within the cytochrome c molecule. These results raise the question of whether electron transfer or protein conformational change is the rate limiting step in this process. We describe here an experiment that probes this intramolecular electron transfer step further. It involves reversing the direction of electron transfer by changing the redox potential of the ruthenium label. Electron transfer in the new ruthenium-cytochrome c derivative described here is from haem(II) to the Ru(III) label, whereas in (NH3)5Ru-cytochrome c the electron transfer is from Ru(II) to haem(III). Intramolecular electron transfer from haem(II) to Ru(III) in the new ruthenium-cytochrome c described here proceeds much slower (greater than 10(5) times) than the electron transfer from Ru(II) to haem(III) in the (NH3)5Ru-cytochrome c. We therefore conclude that electron transfer in cytochrome c is directional, with the protein envelope presumably involved in this directionality.     

Crystal structure of a membrane-embedded H+-translocating pyrophosphatase

H(+)-translocating pyrophosphatases (H(+)-PPases) are active proton transporters that establish a proton gradient across the endomembrane by means of pyrophosphate (PP(i)) hydrolysis. H(+)-PPases are found primarily as homodimers in the vacuolar membrane of plants and the plasma membrane of several protozoa and prokaryotes. The three-dimensional structure and detailed mechanisms underlying the enzymatic and proton translocation reactions of H(+)-PPases are unclear. Here we report the crystal structure of a Vigna radiata H(+)-PPase (VrH(+)-PPase) in complex with a non-hydrolysable substrate analogue, imidodiphosphate (IDP), at 2.35?? resolution. Each VrH(+)-PPase subunit consists of an integral membrane domain formed by 16 transmembrane helices. IDP is bound in the cytosolic region of each subunit and trapped by numerous charged residues and five Mg(2+) ions. A previously undescribed proton translocation pathway is formed by six core transmembrane helices. Proton pumping can be initialized by PP(i) hydrolysis, and H(+) is then transported into the vacuolar lumen through a pathway consisting of Arg?242, Asp?294, Lys?742 and Glu?301. We propose a working model of the mechanism for the coupling between proton pumping and PP(i) hydrolysis by H(+)-PPases.     

Helix deformation is coupled to vectorial proton transport in the photocycle of bacteriorhodopsin

A wide variety of mechanisms are used to generate a proton-motive potential across cell membranes, a function lying at the heart of bioenergetics. Bacteriorhodopsin, the simplest known proton pump, provides a paradigm for understanding this process. Here we report, at 2.1 A resolution, the structural changes in bacteriorhodopsin immediately preceding the primary proton transfer event in its photocycle. The early structural rearrangements propagate from the protein's core towards the extracellular surface, disrupting the network of hydrogen-bonded water molecules that stabilizes helix C in the ground state. Concomitantly, a bend of this helix enables the negatively charged primary proton acceptor, Asp 85, to approach closer to the positively charged primary proton donor, the Schiff base. The primary proton transfer event would then neutralize these two groups, cancelling their electrostatic attraction and facilitating a relaxation of helix C to a less strained geometry. Reprotonation of the Schiff base by Asp 85 would thereby be impeded, ensuring vectorial proton transport. Structural rearrangements also occur near the protein's surface, aiding proton release to the extracellular medium.     

High-resolution X-ray structure of an early intermediate in the bacteriorhodopsin photocycle.

Bacteriorhodopsin is the simplest known photon-driven proton pump and as such provides a model for the study of a basic function in bioenergetics. Its seven transmembrane helices encompass a proton translocation pathway containing the chromophore, a retinal molecule covalently bound to lysine 216 through a protonated Schiff base, and a series of proton donors and acceptors. Photoisomerization of the all-trans retinal to the 13-cis configuration initiates the vectorial translocation of a proton from the Schiff base, the primary proton donor, to the extracellular side, followed by reprotonation of the Schiff base from the cytoplasm. Here we describe the high-resolution X-ray structure of an early intermediate in the photocycle of bacteriorhodopsin, which is formed directly after photoexcitation. A key water molecule is dislocated, allowing the primary proton acceptor, Asp 85, to move. Movement of the main-chain Lys 216 locally disrupts the hydrogen-bonding network of helix G, facilitating structural changes later in the photocycle.     

Bis-methionine axial ligation of haem in bacterioferritin from Pseudomonas aeruginosa

The iron-containing bacterioferritins contain the protoporphyrin IX haem group. It has been established that Escherichia coli cytochrome b1, cytochrome b557 and bacterioferritin are identical. The optical spectra at room temperature of the haem group show it to be predominantly low-spin in both the ferrous and ferric states. The nature of the axial ligands binding the haem group to the polypeptide has, however, remained unknown. Low-spin, bis-coordinate haem centres in proteins typically have a role in rapid electron transfer as redox changes at the metal ion lead to little structural rearrangement. There are only four amino acids with side-chains that have ligand field strengths sufficient to generate the low-spin state of haem, namely, histidine, lysine, methionine and cysteine. Hence there are, potentially, ten different pairs of these four ligands which could be discovered in electron transfer haemoproteins. To date only three have been established with certainty. They are bis-histidine, as in mammalian cytochrome b5, methionine-histidine, typified by cytochrome c and lysine-histidine, recently recognized by spectroscopic methods in cytochrome f. Here we report the electron paramagnetic resonance and near infrared magnetic circular dichroism spectra of the oxidized state of Ps. aeruginosa bacterioferritin which enable the axial ligands to be identified as the thioether side chains of two methionine residues, a ligation scheme not previously reported for haem in any protein.     

Identification of the BAL-labile factor

One of us has previously reported that treatment of the Keilin and Hartree heart-muscle preparation with 2,3-dimercaptopropanol (BAL), in the presence of air, leads to the complete inactivation of the succinate oxidase system with little if any effect on the activities of succinate dehydrogenase (until more than half the BAL was oxidized) or cytochrome c oxidase. The inactivation of the complete succinate oxidase system requires the oxidation of BAL by air in the presence of the enzyme. It is not caused by H2O2 or BAL disulphides produced during the oxidation of BAL. Spectroscopic studies identified the block as lying between cytochromes b and c. It was suggested that a BAL-labile factor is present which transfers electrons from cytochrome b to cytochrome c and which is destroyed by coupled oxidation with BAL. The factor is also required for NADH oxidation. Subsequent work showed it is not identical with cytochrome c1 (ref. 4), myoglobin present in the preparation or the antimycin-binding site. We report here that this factor is identical to the iron-sulphur protein in the central portion of the respiratory chain first identified by Rieske.     

Reversible redox energy coupling in electron transfer chains

Reversibility is a common theme in respiratory and photosynthetic systems that couple electron transfer with a transmembrane proton gradient driving ATP production. This includes the intensely studied cytochrome bc1, which catalyses electron transfer between quinone and cytochrome c. To understand how efficient reversible energy coupling works, here we have progressively inactivated individual cofactors comprising cytochrome bc1. We have resolved millisecond reversibility in all electron-tunnelling steps and coupled proton exchanges, including charge-separating hydroquinone-quinone catalysis at the Q(o) site, which shows that redox equilibria are relevant on a catalytic timescale. Such rapid reversibility renders popular models based on a semiquinone in Q(o) site catalysis prone to short-circuit failure. Two mechanisms allow reversible function and safely relegate short-circuits to long-distance electron tunnelling on a timescale of seconds: conformational gating of semiquinone for both forward and reverse electron transfer, or concerted two-electron quinone redox chemistry that avoids the semiquinone intermediate altogether.