Internal charge transfer in cytochrome c oxidase at a limited proton supply: Proton pumping ceases at high pH

Background

In the membrane-bound enzyme cytochrome c oxidase, electron transfer from cytochrome c to O₂ is linked to proton uptake from solution to form H₂O, resulting in a charge separation across the membrane. In addition, the reaction drives pumping of protons across the membrane.

Methods

In this study we have measured voltage changes as a function of pH during reaction of the four-electron reduced cytochrome c oxidase from Rhodobacter sphaeroides with O₂. These electrogenic events were measured across membranes containing purified enzyme reconstituted into lipid vesicles.

Results

The results show that the pH dependence of voltage changes (primarily associated with proton transfer) during O₂ reduction does not match that of the previously studied absorbance changes (primarily associated with electron transfer). Furthermore, the voltage changes decrease with increasing pH.

Conclusions

The data indicate that cytochrome c oxidase does not pump protons at high pH (10.5) (or protons are taken from the “wrong” side of the membrane) and that at this pH the net proton-uptake stoichiometry is ∼ 1/2 of that at pH 8. Furthermore, the results provide a basis for interpretation of results from studies of mutant forms of the enzyme.

General significance

These results provide new insights into the function of cytochrome c oxidase.     

Reaction of wild-type and Glu243Asp variant yeast cytochrome c oxidase with O2

We have studied internal electron transfer during the reaction of Saccharomyces cerevisiae mitochondrial cytochrome c oxidase with dioxygen. Similar absorbance changes were observed with this yeast oxidase as with the previously studied Rhodobacter sphaeroides and bovine mitochondrial oxidases, which suggests that the reaction proceeds along the same trajectory. However, notable differences were observed in rates and electron-transfer equilibrium constants of specific reaction steps, for example the ferryl (F) to oxidized (O) reaction was faster with the yeast (0.4 ms) than with the bovine oxidase (~ 1 ms) and a larger fraction Cu_A was oxidized with the yeast than with the bovine oxidase in the peroxy (P_R) to F reaction. Furthermore, upon replacement of Glu243, located at the end of the so-called D proton pathway, by Asp the P_R → F and F → O reactions were slowed by factors of ~ 3 and ~ 10, respectively, and electron transfer from Cu_A to heme a during the P_R → F reaction was not observed. These data indicate that during reduction of dioxygen protons are transferred through the D pathway, via Glu243, to the catalytic site in the yeast mitochondrial oxidase. This article is part of a Special Issue entitled: 18th European Bioenergetic Conference.     

Transmembrane proton translocation by cytochrome c oxidase

Respiratory heme-copper oxidases are integral membrane proteins that catalyze the reduction of molecular oxygen to water using electrons donated by either quinol (quinol oxidases) or cytochrome c (cytochrome c oxidases, CcOs). Even though the X-ray crystal structures of several heme-copper oxidases and results from functional studies have provided significant insights into the mechanisms of O₂-reduction and, electron and proton transfer, the design of the proton-pumping machinery is not known. Here, we summarize the current knowledge on the identity of the structural elements involved in proton transfer in CcO. Furthermore, we discuss the order and timing of electron-transfer reactions in CcO during O₂ reduction and how these reactions might be energetically coupled to proton pumping across the membrane.     

Electrostatic control of proton pumping in cytochrome c oxidase

As part of the mitochondrial respiratory chain, cytochrome c oxidase utilizes the energy produced by the reduction of O₂ to water to fuel vectorial proton transport. The mechanism coupling proton pumping to redox chemistry is unknown. Recent advances have provided evidence that each of the four observable transitions in the complex catalytic cycle consists of a similar sequence of events. However, the physico-chemical basis underlying this recurring sequence has not been identified. We identify this recurring pattern based on a comprehensive model of the catalytic cycle derived from the analysis of oxygen chemistry and available experimental evidence. The catalytic cycle involves the periodic repetition of a sequence of three states differing in the spatial distribution of charge in the active site: [0|1], [1|0], and [1|1], where the total charge of heme a and the binuclear center appears on the left and on the right, respectively. This sequence recurs four times per turnover despite differences in the redox chemistry. This model leads to a simple, robust, and reproducible sequence of electron and proton transfer steps and rationalizes the pumping mechanism in terms of electrostatic coupling of proton translocation to redox chemistry. Continuum electrostatic calculations support the proposed mechanism and suggest an electrostatic origin for the decoupled and inactive phenotypes of ionic mutants in the principal proton-uptake pathway.     

Cytochrome c1 exhibits two binding sites for cytochrome c in plants

In plants, channeling of cytochrome c molecules between complexes III and IV has been purported to shuttle electrons within the supercomplexes instead of carrying electrons by random diffusion across the intermembrane bulk phase. However, the mode plant cytochrome c behaves inside a supercomplex such as the respirasome, formed by complexes I, III and IV, remains obscure from a structural point of view. Here, we report ab-initio Brownian dynamics calculations and nuclear magnetic resonance-driven docking computations showing two binding sites for plant cytochrome c at the head soluble domain of plant cytochrome c₁, namely a non-productive (or distal) site with a long heme-to-heme distance and a functional (or proximal) site with the two heme groups close enough as to allow electron transfer. As inferred from isothermal titration calorimetry experiments, the two binding sites exhibit different equilibrium dissociation constants, for both reduced and oxidized species, that are all within the micromolar range, thus revealing the transient nature of such a respiratory complex. Although the docking of cytochrome c at the distal site occurs at the interface between cytochrome c₁ and the Rieske subunit, it is fully compatible with the complex III structure. In our model, the extra distal site in complex III could indeed facilitate the functional cytochrome c channeling towards complex IV by building a “floating boat bridge” of cytochrome c molecules (between complexes III and IV) in plant respirasome.     

Mechanism and energetics by which glutamic acid 242 prevents leaks in cytochrome c oxidase

Cytochrome c oxidase (CcO) is the terminal enzyme of aerobic respiration. The energy released from the reduction of molecular oxygen to water is used to pump protons across the mitochondrial or bacterial membrane. The pump function introduces a mechanistic requirement of a valve that prevents protons from flowing backwards during the process. It was recently found that Glu-242, a key amino acid in transferring protons to be pumped across the membrane and to the site of oxygen reduction, fulfils the function of such a valve by preventing simultaneous contact to the pump site and to the proton-conducting D-channel (Kaila V.R.I. et al. Proc. Natl. Acad. Sci. USA 105, 2008). Here we have incorporated the valve model into the framework of the reaction mechanism. The function of the Glu valve is studied by exploring how the redox state of the surrounding metal centers, dielectric effects, and membrane potential, affects the energetics and leaks of this valve. Parallels are drawn between the dynamics of Glu-242 and the long-standing obscure difference between the metastable O_H and stable O states of the binuclear center. Our model provides a suggestion for why reduction of the former state is coupled to proton translocation while reduction of the latter is not.     

Cyanobacterial cytochrome cM: Probing its role as electron donor for CuA of cytochrome c oxidase

It is well known that efficient functioning of photosynthetic (PET) and respiratory electron transport (RET) in cyanobacteria requires the presence of either cytochrome c₆ (Cytc₆) or plastocyanin (PC). By contrast, the interaction of an additional redox carrier, cytochrome c_M (Cytc_M), with either PET or RET is still under discussion. Here, we focus on the (putative) role of Cytc_M in cyanobacterial respiration. It is demonstrated that genes encoding the main terminal oxidase (cytochrome c oxidase, COX) and cytochrome c_M are found in all 44 totally or partially sequenced cyanobacteria (except one strain). In order to check whether Cytc_M can act as electron donor to COX, we investigated the intermolecular electron transfer kinetics between Cytc_M and the soluble Cu_A domain (i.e. the donor binding and electron entry site) of subunit II of COX. Both proteins from Synechocystis PCC6803 were expressed heterologously in E. coli. The forward and the reverse electron transfer reactions were studied yielding apparent bimolecular rate constants of (2.4 ± 0.1) × 10⁵ M^− 1 s^− 1 and (9.6 ± 0.4) × 10³ M^− 1 s^− 1 (5 mM phosphate buffer, pH 7, 50 mM KCl). A comparative analysis with Cytc₆ and PC demonstrates that Cytc_M functions as electron donor to Cu_A as efficiently as Cytc₆ but more efficient than PC. Furthermore, we demonstrate the association of Cytc_M with the cytoplasmic and thylakoid membrane fractions by immunobloting and discuss the potential role of Cytc_M as electron donor for COX under stress conditions.     

Cytochrome c reactivity in its complexes with mammalian cytochrome c oxidase and yeast peroxidase

1.

1. The ascorbate reducibility of cytochrome c (beef or horse heart) in its complexes with cytochrome c oxidase (beef heart) and cytochrome c peroxidase (yeast) has been studied.     

The inside pH determines rates of electron and proton transfer in vesicle-reconstituted cytochrome c oxidase

Cytochrome c oxidase is the terminal enzyme in the respiratory chains of mitochondria and many bacteria where it translocates protons across a membrane thereby maintaining an electrochemical proton gradient. Results from earlier studies on detergent-solubilized cytochrome c oxidase have shown that individual reaction steps associated with proton pumping display pH-dependent kinetics. Here, we investigated the effect of pH on the kinetics of these reaction steps with membrane-reconstituted cytochrome c oxidase such that the pH was adjusted to different values on the inside and outside of the membrane. The results show that the pH on the inside of the membrane fully determines the kinetics of internal electron transfers that are linked to proton pumping. Thus, even though proton release is rate limiting for these reaction steps (Salomonsson et al., Proc. Natl. Acad. Sci. USA, 2005, 102, 17624), the transition kinetics is insensitive to the outside pH (in the range 6-9.5).     

Combined DFT and electrostatics study of the proton pumping mechanism in cytochrome c oxidase

Cytochrome c oxidase is a redox-driven proton pump which converts atmospheric oxygen to water and couples the oxygen reduction reaction to the creation of a membrane proton gradient. The structure of the enzyme has been solved; however, the mechanism of proton pumping is still poorly understood. Recent calculations from this group indicate that one of the histidine ligands of enzyme's Cu_B center, His291, may play the role of the pumping element. In this paper, we report on the results of calculations that combined first principles DFT and continuum electrostatics to evaluate the energetics of the key energy generating step of the model—the transfer of the chemical proton to the binuclear center of the enzyme, where the hydroxyl group is converted to water, and the concerted expulsion of the proton from δ-nitrogen of His291 ligand of Cu_B center. We show that the energy generated in this step is sufficient to push a proton against an electrochemical membrane gradient of about 200 mV. We have also re-calculated the pK_a of His291 for an extended model in which the whole Fe_a3-Cu_B center with their ligands is treated by DFT. Two different DFT functionals (B3LYP and PBE0), and various dielectric models of the protein have been used in an attempt to estimate potential errors of the calculations. Although current methods of calculations do not allow unambiguous predictions of energetics in proteins within few pK_a units, as required in this case, the present calculation provides further support for the proposed His291 model of CcO pump and makes a specific prediction that could be targeted in the experimental test.     

Redox-coupled proton pumping in cytochrome c oxidase: Further insights from computer simulation

The membrane-bound enzyme cytochrome c oxidase, the terminal member in the respiratory chain, converts oxygen into water and generates an electrochemical gradient by coupling the electron transfer to proton pumping across the membrane. Here we have investigated the dynamics of an excess proton and the surrounding protein environment near the active sites. The multi-state empirical valence bond (MS-EVB) molecular dynamics method was used to simulate the explicit dynamics of proton transfer through the critically important hydrophobic channel between Glu242 (bovine notation) and the D-propionate of heme a₃ (PRDa₃) for the first time. The results from these molecular dynamics simulations indicate that the PRDa₃ can indeed re-orientate and dissociate from Arg438, despite the high stability of such an ion pair, and has the ability to accept protons via bound water molecules. Any large conformational change of the adjacent heme a D-propionate group is, however, sterically blocked directly by the protein. Free energy calculations of the PRDa₃ side chain isomerization and the proton translocation between Glu242 and the PRDa₃ site have also been performed. The results exhibit a redox state-dependent dynamical behavior and indicate that reduction of the low-spin heme a may initiate internal transfer of the pumped proton from Glu242 to the PRDa₃ site.     

Electron transfer kinetics between soluble modules of Paracoccus denitrificans cytochrome c1 and its physiological redox partners

The transient electron transfer (ET) interactions between cytochrome c₁ of the bc₁-complex from Paracoccus denitrificans and its physiological redox partners cytochrome c₅₅₂ and cytochrome c₅₅₀ have been characterized functionally by stopped-flow spectroscopy. Two different soluble fragments of cytochrome c₁ were generated and used together with a soluble cytochrome c₅₅₂ module as a model system for interprotein ET reactions. Both c₁ fragments lack the membrane anchor; the c₁ core fragment (c_1CF) consists of only the hydrophilic heme-carrying domain, whereas the c₁ acidic fragment (c_1AF) additionally contains the acidic domain unique to P. denitrificans. In order to determine the ionic strength dependencies of the ET rate constants, an optimized stopped-flow protocol was developed to overcome problems of spectral overlap, heme autoxidation and the prevalent non-pseudo first order conditions. Cytochrome c₁ reveals fast bimolecular rate constants (10⁷ to 10⁸ M^− 1 s^− 1) for the ET reaction with its physiological substrates c₅₅₂ and c₅₅₀, thus approaching the limit of a diffusion-controlled process, with 2 to 3 effective charges of opposite sign contributing to these interactions. No direct involvement of the N-terminal acidic c₁-domain in electrostatically attracting its substrates could be detected. However, a slight preference for cytochrome c₅₅₀ over c₅₅₂ reacting with cyochrome c₁ was found and attributed to the different functions of both cytochromes in the respiratory chain of P. denitrificans.     

Importance of cytochrome c redox state for ceramide-induced apoptosis of human mammary adenocarcinoma cells

Background

Ceramides are intracellular lipid mediator implicated in various cellular responses, including oxidative stress and programmed cell death. Studies demonstrated strong links between ceramide and the mitochondria in the regulation of apoptosis. However, the mechanism of apoptosis induced by ceramides is not fully understood. The present study delineates importance of the redox state of cytochrome c for release of cytochrome c and apoptosis of human mammary adenocarcinoma MCF-7 and MDA-MB-231 cells induced by ceramides.

Methods

The study uses MCF-7 and MDA-MB-231 cells, isolated mitochondria, submitochondrial particles, and oxidized and reduced cytochrome c. Methods used include flow cytometry, immunoblotting, spectroscopy, and respirometry.

Results

We show that ceramides induce mitochondrial oxidative stress and release of cytochrome c from the mitochondria of these cells. Our findings show that ceramides react with oxidized cytochrome c whereas reduced cytochrome c does not react with ceramides. We also show that oxidized cytochrome c reacted with ceramides exerts lower reducibility and function to support mitochondrial respiration. Furthermore, our data show that glutathione protects cytochrome c of reacting with ceramides by increasing the reduced state of cytochrome c.

Conclusions

Ceramides induce oxidative stress and apoptosis in human mammary adenocarcinoma cells by interacting with oxidized cytochrome c leading to the release of cytochrome c from the mitochondria. Our findings suggest a novel mechanism for protective role of glutathione.

General significance

Our study suggests that the redox state of cytochrome c is important in oxidative stress and apoptosis induced by ceramides.     

The chemistry of the CuB site in cytochrome c oxidase and the importance of its unique His-Tyr bond

The Cu_B metal center is at the core of the active site of the heme-copper oxidases, comprising a copper atom ligating three histidine residues one of which is covalently bonded to a tyrosine residue. Using quantum chemical methodology, we have studied the Cu_B site in several redox and ligand states proposed to be intermediates of the catalytic cycle. The importance of the His-Tyr crosslink was investigated by comparing energetics, charge, and spin distributions between systems with and without the crosslink. The His-Tyr bond was shown to decrease the proton affinity and increase the electron affinity of both Tyr-244 and the copper. A previously unnoticed internal electronic equilibrium between the copper atom and the tyrosine was observed, which seems to be coupled to the unique structure of the system. In certain states the copper and Tyr-244 compete for the unpaired electron, the localization of which is determined by the oxygenous ligand of the copper. This electronic equilibrium was found to be sensitive to the presence of a positive charge 10 Å away from the center, simulating the effect of Lys-319 in the K-pathway of proton transfer. The combined results provide an explanation for why the heme-copper oxidases need two pathways of proton uptake, and why the K-pathway is active only in the second half of the reaction cycle.     

Production and preliminary characterization of a recombinant triheme cytochrome c7 from Geobacter sulfurreducens in Escherichia coli

Multiheme cytochromes c have been found in a number of sulfate- and metal ion-reducing bacteria. Geobacter sulfurreducens is one of a family of microorganisms that oxidize organic compounds, with Fe(III) oxide as the terminal electron acceptor. A triheme 9.6 kDa cytochrome c₇ from G. sulfurreducens is a part of the metal ion reduction pathway. We cloned the gene for cytochrome c₇ and expressed it in Escherichiacoli together with the cytochrome c maturation gene cluster, ccmABCDEFGH, on a separate plasmid. We designed two constructs, with and without an N-terminal His-tag. The untagged version provided a good yield (up to 6 mg/l of aerobic culture) of the fully matured protein, with all three hemes attached, while the N-terminal His-tag appeared to be detrimental for proper heme incorporation. The recombinant protein (untagged) is properly folded, it has the same molecular weight and displays the same absorption spectra, both in reduced and in oxidized forms, as the protein isolated from G. sulfurreducens and it is capable of reducing metal ions in vitro. The shape parameters for the recombinant cytochrome c₇ determined by small angle X-ray scattering are in good agreement with the ones calculated from a homologous cytochrome c₇ of known structure.     

Elevated Proton Leak of the Intermediate OH in Cytochrome c Oxidase

The kinetics of the formation and relaxation of transmembrane electric potential (Δψ) during the complete single turnover of CcO was studied in the bovine heart mitochondrial and the aa₃-type Paracoccus denitrificans enzymes incorporated into proteoliposome membrane. The real-time Δψ kinetics was followed by the direct electrometry technique. The prompt oxidation of CcO and formation of the activated, oxidized (O_H) state of the enzyme leaves the enzyme trapped in the open state that provides an internal leak for protons and thus facilitates dissipation of Δψ (τ^app ≤ 0.5-0.8 s). By contrast, when the enzyme in the O_H state is rapidly re-reduced by sequential electron delivery, Δψ dissipates much slower (τ^app > 3 s). In P. denitrificans CcO proteoliposomes the accelerated Δψ dissipation is slowed down by a mutational block of the proton conductance through the D-, but not K-channel. We concluded that in contrast to the other intermediates the O_H state of CcO is vulnerable to the elevated internal proton leak that proceeds via the D-channel.     

Yeast acetic acid-induced programmed cell death can occur without cytochrome c release which requires metacaspase YCA1

To investigate the role of cytochrome c (cyt c) release in yeast acetic acid-induced programmed cell death (AA-PCD), wild type (wt) and cells lacking metacaspase (Δyca1), cytochrome c (Δcyc1,7) and both (Δcyc1,7Δyca1) were compared for AA-PCD occurrence, hydrogen peroxide (H₂O₂) production and caspase activity. AA-PCD occurs in Δcyc1,7 and Δcyc1,7Δyca1 cells slower than in wt, but similar to that in Δyca1 cells, in which no cytochrome c release occurs. Both H₂O₂ production and caspase activation occur in these cells with early and extra-activation in Δcyc1,7 cells. We conclude that alternative death pathways can be activated in yeast AA-PCD, one dependent on cyt c release, which requires YCA1, and the other(s) independent on it.     

Physical and catalytic properties of a peroxidase derived from cytochrome c

Except for its redox properties, cytochrome c is an inert protein. However, dissociation of the bond between methionine-80 and the heme iron converts the cytochrome into a peroxidase. Dissociation is accomplished by subjecting the cytochrome to various conditions, including proteolysis and hydrogen peroxide (H₂O₂)-mediated oxidation. In affected cells of various neurological diseases, including Parkinson's disease, cytochrome c is released from the mitochondrial membrane and enters the cytosol. In the cytosol cytochrome c is exposed to cellular proteases and to H₂O₂ produced by dysfunctional mitochondria and activated microglial cells. These could promote the formation of the peroxidase form of cytochrome c. In this study we investigated the catalytic and cytolytic properties of the peroxidase form of cytochrome c. These properties are qualitatively similar to those of other heme-containing peroxidases. Dopamine as well as sulfhydryl group-containing metabolites, including reduced glutathione and coenzyme A, are readily oxidized in the presence of H₂O₂. This peroxidase also has cytolytic properties similar to myeloperoxidase, lactoperoxidase, and horseradish peroxidase. Cytolysis is inhibited by various reducing agents, including dopamine. Our data show that the peroxidase form of cytochrome c has catalytic and cytolytic properties that could account for at least some of the damage that leads to neuronal death in the parkinsonian brain.     

Modelling O2 and O2 unidirectional fluxes in plants. III: Fitting of experimental data by a simple model

Photosynthetic assimilation of CO₂ in plants results in the balance between the photochemical energy developed by light in chloroplasts, and the consumption of that energy by the oxygenation processes, mainly the photorespiration in C₃ plants. The analysis of classical biological models shows the difficulties to bring to fore the oxygenation rate due to the photorespiration pathway. As for other parameters, the most important key point is the estimation of the electron transport rate (ETR or J), i.e. the flux of biochemical energy, which is shared between the reductive and oxidative cycles of carbon. The only reliable method to quantify the linear electron flux responsible for the production of reductive energy is to directly measure the O₂ evolution by ¹⁸O₂ labelling and mass spectrometry. The hypothesis that the respective rates of reductive and oxidative cycles of carbon are only determined by the kinetic parameters of Rubisco, the respective concentrations of CO₂ and O₂ at the Rubisco site and the available electron transport rate, ultimately leads to propose new expressions of biochemical model equations. The modelling of ¹⁸O₂ and ¹⁶O₂ unidirectional fluxes in plants shows that a simple model can fit the photosynthetic and photorespiration exchanges for a wide range of environmental conditions. Its originality is to express the carboxylation and the oxygenation as a function of external gas concentrations, by the definition of a plant specificity factor Sp that mimics the internal reactions of Rubisco in plants. The difference between the specificity factors of plant (Sp) and of Rubisco (Sr) is directly related to the conductance values to CO₂ transfer between the atmosphere and the Rubisco site. This clearly illustrates that the values and the variation of conductance are much more important, in higher C₃ plants, than the small variations of the Rubisco specificity factor. The simple model systematically expresses the reciprocal variations of carboxylation and oxygenation exchanges illustrated by a “mirror effect”. It explains the protective sink effect of photorespiration, e.g. during water stress. The importance of the CO₂ compensation point, in classical models, is reduced at the benefit of the crossing points Cx and Ox, concentration values where carboxylation and oxygenation are equal or where the gross O₂ uptake is half of the gross O₂ evolution. This concept is useful to illustrate the feedback effects of photorespiration in the atmosphere regulation. The constancy of Sp and of Cx for a great variation of P under several irradiance levels shows that the regulation of the conductance maintains constant the internal CO₂ and the ratio of photorespiration to photosynthesis (PR/P). The maintenance of the ratio PR/P, in conditions of which PR could be reduced and the carboxylation increased, reinforces the hypothesis of a positive role of photorespiration and its involvement in the plant-atmosphere co-evolution.     

Cytochrome c oxidase loses catalytic activity and structural integrity during the aging process in Drosophila melanogaster

The hypothesis, that structural deterioration of cytochrome c oxidase (CcO) is a causal factor in the age-related decline in mitochondrial respiratory activity and an increase in H₂O₂ generation, was tested in Drosophila melanogaster. CcO activity and the levels of seven different nuclear DNA-encoded CcO subunits were determined at three different stages of adult life, namely, young-, middle-, and old-age. CcO activity declined progressively with age by 33%. Western blot analysis, using antibodies specific to Drosophila CcO subunits IV, Va, Vb, VIb, VIc, VIIc, and VIII, indicated that the abundance these polypeptides decreased, ranging from 11% to 40%, during aging. These and previous results suggest that CcO is a specific intra-mitochondrial site of age-related deterioration, which may have a broad impact on mitochondrial physiology.