摘要:

Faraday Discuss. Chem. SOC., 1982, 74, 311-329 Electron Transfer in Biology The Function of Cytochrome c BY GEOFFREY R. MOORE, ZHONG-XIAN HUANG,~ CRISPIN G. S. ELEY, HAZEL ROBERT J. P. WILLIAMS A. BARKER, GLYN WILLIAMS, MARTIN N. ROBINSON AND Inorganic Chemistry Laboratory, University of Oxford, South Parks Road, Oxford OX1 3QR Received 2nd June, 1982 A review of the literature concerning the structure and electron-transfer function of cytochrome c is presented. Emphasis is placed upon the conformational heterogeneity of cytochrome c, particularly of the region around the exposed edge of the haem that is the reductase and oxidase interaction region, and upon the variation in sequence of eukaryotic cytochromes c. Five distinct steps can be recognised between the conversion of energy from reduced organic molecules to adenosine triphosphate (ATP). These steps are: (i) the initial injection of electrons in a membrane phase and the simultaneous release of protons from reduced organic molecules; (ii) the flow of electrons to 0,; (iii) the flow of protons to O,, either locally, within or along the membrane, or equilibrated in the bulk aqueous phases; (iv) the generation of a proton gradient across the ATP synthesis site in the membrane; (v) the conversion of the electrochemical potential of the proton gradient into chemical energy by driving the synthesis of ATP.This generally accep- ted sequence of reactions was proposed by Mitchell and Williams.2 The initial charge separation of protons and electrons is maintained by the use of separate pathways for transferring protons and electrons.In this paper one com- ponent of the electron-transfer pathway is analysed. Kell (this Discussion) addresses the problem of proton flow. The conversion of energy from a proton gradient into ATP within an enzyme reaction centre is not yet understood. ELECTRON-TRANSFER CHAINS Biological electron-transfer chains 4 s are generally membrane-bound systems with a preponderance of iron- and copper-containing redox proteins. The organisation of the redox centres depends on the proteins in which they are placed and on interactions between protein components of the chain, both between themselves and with the membrane. First, many of the reactions coupled to electron transfer are reactions that are best carried out in a controlled environment : e.g.dehydration reactions. Second, membrane binding and organisation facilitates compartmentalisation and directional electron transfer. Thus, while connecting primary electron donors to terminal electron acceptors, membrane proteins generate electrochemical gradients and eliminate short- ? Present Address : Department of Chemistry, Fudan University, Shanghai, People’s Republic of China. Electron-transfer chains are membrane-bound for a variety of reasons.3 12 THE FUNCTION OF CYTOCHROME C cytochrome ox id is e MEMBRANE cy t 0 chrome red u c t ase circuits of the chain and accidental by-passing of the energy-conserving reaction sites. Third, the protein centres constituting the redox chain can be held in the correct juxtaposition for reaction.The mitochondrial respiratory chain is shown in fig. 1 . This chain runs along and spans the inner mitochrondrial membrane and consists of 350 redox centres. Electron transfer is coupled to ATP formation at three sites [see ref. (3)]. ElmV 600 400 200 0 - 200 - 400 --- FIG. 1.-The mitochondrial respiratory hai in.^*^ (A) The sequence of redox centres within the different complexes is not known for certain. The arrangement given in the diagram is based on the EM values obtained with de-energized heart mitochondria. Cytochrome c links the cytochrome c reductase and cytochrome c oxidase complexes. (B) The arrangement of cytochrome c with respect to the positions of cytochrome c reductase and cytochrome c oxidase is not well defined It is believed that the reductase and oxidase complexes span the inner mitochondrial membrane with cytochrome c loosely attached to the outside of the membrane.It is not known what the distance (d) is between the reductase and oxidase complexes, and thus whether cytochrome c has to migrate along the membrane to transport electrons.G . R . MOORE et al. 313 ELECTRON-TRANSFER CENTRES Metal-containing redox centres give rise to relatively stable odd-electron centres, and can therefore transfer electrons in one-electron steps. They do not have an absolute requirement for the simultaneous uptake or release of protons. Non-metal redox molecules usually transfer electrons in two-electron steps that do involve protons to compensate the variation in charge. Flavins, quinones and O2 are involved in the mitochondrial respiratory chain and these can all be considered to react in the two- electron step HO .. . O H + O = . . a - -0 + 2H+ + 2e which indicates the change from single bonds in the reduced molecules to double bonds in the oxidised molecules with the consequent release of protons. These two-electron non-metal redox couples are associated with the proton-gradient-producing steps of the respiratory chain. The metal redox centres appear to function mainly as electron- transfer connectors; the terminal oxidase is an exception. This protein has a cataly- tic function, the reduction of 0,. Thus the flow of electrons and the flow of protons are coupled. MITOCHONDRIAL CYTOCHROME C Cytochrome c 6 3 7 is a small (molecular weight M 12 000) monohaem protein that transfers electrons from cytochrome c reductase to cytochrome c oxidase in the mito- chondrial respiratory chain (fig.1). The iron cycles between the ferric and ferrous states with a mid-point redox potential at pH 7 of 260 & 20 mV. There is no evidence to suggest that other groups within the protein undergo a redox transition. Questions associated with cytochrome c that are relevant to this meeting are: (i) How does the protein interact with the metal in the two oxidation states? (ii) To what degree does the protein resemble a solid matrix? (iii) Where is (are) the electron injection and ejection site(s) on cytochrome c ? Where are the binding sites on cyto- chrome c for its redox partners? (iv) How are electrons transferred to and from the iron of cytochrome c ? To answer these questions it will be necessary to know not only what the structures of cytochrome c are, but also what the dynamic properties of the structures are, and the way in which the structures respond to changes in environmental conditions, such as temperature, pH and ionic strength, and to changes in amino-acid composition (mitochondrial cytochromes from different species vary in up to 70% of their amino acids).8 The complications within this study may be realised by observing that cyto- chrome c has been the subject of more than 3000 publications since its rediscovery in 1925, and yet cytochrome c is probably the simplest protein in the cytochrome chain! THE IRON CENTRE OF CYTOCHROME C The redox centre of mitochondrial cytochrome c is a low-spin haem c with histidyl and methionyl axial ligands to the iron (fig.2).9-11 The two neutral ligating groups bind particularly well to the ferrous state and mid-point redox potentials are relatively high compared with those for other ligands. The enhanced affinity of methionine sulphur in water for a ferrous haem rather than a ferric haem has been compared with the fact that cytochrome c is far more stable in the reduced state than it is in the oxidised314 FIG. 2.-The iron the protein via ~ t a t e . ~ . ~ Thus, THE FUNCTION OF CYTOCHROME C c’ I H c q / C H 2 HOOC coordination centre of cytochrome c. The haem group is covalently bound to thioether linkages, and there is a histidine and methionine ligand to the iron. the Fe--S bond of ferricytochrome c is relatively easily broken by increasing pH or increasing temperature, while the Fe-S bond of ferrocytochrome c is particularly stable, even to such extreme conditions as 95 “C at pH 6 and pH 12.5 at 25 “C.An example of the lack of stability of the ferric state is that on freeze-drying a solution of native ferricytochrome c, the resulting powder contains a large proportion of denatured protein in which the Fe-S bond has been broken. On dissolution of the powder in H20 renaturation of the protein rapidly occurs.12 No such transition occurs in the ferrous state. It should not be concluded that the intrinsic weakness of the FelIr-S bond is the only source of instability. The haem is surrounded by a large number of amino-acid residues, some of which will destablise the ferric state relative to the ferrous state.A controversy has developed concerning the importance of the Fe-S bond to the function of cytochrome c. From n.m.r. studies l3 of cobalticytochrome c and native cytochrome c a small change in metal-sulphur interaction was observed with change in oxidation state and interpreted as a bond-length change. However, X-ray crystallo- graphic structure determination l 4 9 l 5 and EXAFS measurements on cytochrome c16 and model complexes17 suggest there is no change in the Fe-S bond length with change in oxidation state. Whether or not there is an oxidation-state dependence to the Fe-S interaction, it is necessary to have information concerning fluctuations of the interaction with time. Chemical and spectroscopic measurements indicate thatG .R . MOORE et al. 315 there are some anomalous dynamic properties associated with the Fe-S bond of ferricytochrome c. The methionine ligand of ferricytochrome c is displaced by reagents such as CN-, N< and imidazole at pH 7 and by an unknown residue at pH 8.8. In all cases reduction of the resulting low-spin complex readily produces native reduced cytochrome c.697 Sutin and Yandell l8 have studied this reaction at pH 7 and proposed the following scheme : ki k2 k - 1 k--2 Cyt"'(Fe-S) CytlI1(Fe - S) -CL Cyt"'(Fe-X) k3 + dithionite k4 I Cyt"(Fe-S) +-- Cyt"(Fe-X) where X = CN-, N3- or imidazole. Rate constants for the reaction of CN- with horse cytochrome c at pH 7 and 25 "C are: kl = 60 & 20 s-l; k2/k-l = 0.5 dm3 mol-l, k--2 = 1.3 x s-l, k3 = 6.9 x lo5 dm3 m o l - ' ~ - ~ ; k4 = 5.0 x lov3 s-l.The nature of the activated state of ferricytochrome c, Cyt"' (Fe - - S), is not known. Sutin and Yandell l8 speculate that it contains five-coordinate iron. Spectroscopic measurements of the charge-transfer interaction between the ferric iron and sulphur ligand of ferricytochrome c also show that Fe-S is a particularly sensitive b ~ n d . ~ . ~ The interaction has been monitored in two ways. First, from observations of a weak charge-transfer absorption band in the near-infrared at 695 nm and, secondly, from observations of the n.m.r. resonances of the methionine ligand of ferricytochrome c.19 The n.m.r. resonances experience large perturbations to their chemical shifts via a mechanism involving delocalisation of the unpaired electron of the ferric ion; the so-called contact shift.The near-infrared band is a charge-transfer band that is present only where there is a Fe"'-S bond. This absorption band is sensitive to a wide range of conditions. Its temperature sensitivity is particularly revealing; with increasing temperature the 695 nm absorbance diminishes although the magnetic susceptibility remains that of a low-spin ferric ion (fig. 3) and the methionine ligand remains bound to the iron.19 It is only at ca. 60 "C that the ferric ion attains some high-spin character, indicating that in an appreciable amount of the ferri- cytocrome c the Fe"'-S bond has been broken. N.m.r. spectra in the range 15-60 "C confirm that the methionine Fe-S bond is intact and confirms that there is a con- tinuous change in the Fe"'-S interaction with increasing temperature. The conclusion from these studies cannot be stated firmly, but it does appear that there are some anomalous physical properties associated with the Fe'II-S bond.These properties involve bond stretching, which in the extreme leads to a rapid bond breaking and remaking of the type suggested by Sutin and Yandell.'* They may also involve rotation about the Fe"'-S bond resulting in poorer overlap between the sulphur lone pair and ferric orbitals. Such changes, which have not been observed in ferrocytochrome c, may also account for the differences between deductions obtained from n.m.r. measurements l3 and those from other methods I4,l6 as to the nature of the metal-sulphur bond.THE STRUCTURE OF MITOCHONDRIAL CYTOCHROME C The structure of the cytochrome c protein in both oxidation states has been investigated with a variety of technique~.~'~ There is general agreement between the316 0 . 4 m Q\ \o td 0 y 0.3- 3 -E 0, s td 0.2 THE FUNCTION OF CYTOCHROME C - - 0.1 4 0 d ' 30 2 I 0 w 20 10 I I I I I i I 2.9 3 - 1 3 . 3 3.5 1 OJK/ T 2.9 3 . 1 3 . 3 3.5 1 03K/ T FIG. 3.-Temperature dependence of the 695 nm band (upper) and the molar magnetic susceptibility, Axmo,, (lower) of horse ferricytochrome c. 'The optical experiments were done with 0.5 mmol dm-3 protein in 10 mmol dmP3 phosphate buffer at pH 5.3 (upper curve) and pH 7.1 (lower curve). The magnetic susceptibility experiments were done with 4-6 mmol dm-3 protein at pH 5.4 (upper curve) and pH 7.0 (lower curve).G .R. MOORE et al. 317 different techniques that the structures deduced from X-ray diffraction studies 14915 are largely conserved in The structure of tuna cytochrome c is represented in fig. 4 by a ribbon-folding diagram. The polypeptide chain folds around the haem group shielding all but one edge of the haem. The amino acids in contact with the haem are overwhelmingly hydrophobic in character, as the haem packing diagram in fig. 5 shows. A conse- 54 FIG. 4.-Ribbon diagram of the structure of tuna cytochrome c. Each segment of the ribbon represents one amino-acid residue. The molecule is viewed from the side bearing the exposed haem edge. [Reproduced with permission from ref. (22).] quence of the hydrophobic character of the haem environment is that the ferrous state is stabilised relative to the ferric state; that is, the redox potential is relatively high.21 The unusual positions22 of the two haem propionate groups within the protein is energetically unfavourable, although there is some compensation from hydrogen bonding to amino-acid side chains.In other haem proteins the position of the haem relative to the polypeptide is different, the edge bearing the propionates is the one exposed at the molecular surface. This structural feature of cytochrome c is designed to prevent the propionates ionising over the physiological pH range, which would result in pH-dependent redox proper tie^.^^ One aspect of the structure of cytochrome c that is of particular relevance to this meeting is: to what extent does the structure of cytochrome c vary with change in oxidation state ? The answer is little, although there is an oxidation-state-linked con- formation change.The nature of this change is not known with certainty. Takano and Dickerson 14*15 have published a detailed model based on X-ray studies that in- volves up to 30% of the protein and includes movement of the haem group and axial ligands, with some groups moving by up to 0.87 nm (Asn52 and Tyr67). Studies in solution with n.m.r.13 have not yet reached the stage where such a detailed model for318 THE FUNCTION OF CYTOCHROME C the conformation change can be presented, although recent work has led to the assign- ment of resonances of 290% of the aromatic and methyl groups. However, the X- ray data and n.m.r.data are in general agreement that the major change in structure is at the back of the molecule close to Ile57; a finding consistent with chemical modific- ation However, the n.m.r. studies indicate that there is also a change in the 0 ” H CH,-C-”/ 25 FIG. 5.-Chain-packing diagram of cytochrome c. Evolutionarily invariant residues (underlined upper case), partially buried side chains (heavy semi-circles), buried side chains (heavy circles), and side chains packed against the haem (black dots). [Adapted from ref. (7).] region around PhelO and Tyr97, although this region is not included in the model based upon X-ray data. Complete definition of the conformation change in solution requires further work, although the main structural conclusions are that there are some minor changes in structure that extend to the surface of the protein.Taken together with the observations concerning the structure of the iron site and the reactivity of the iron site, the conformation change suggests that there are dynamic features of the structure involving small changes over large parts of the molecule.G. R . MOORE et al. 319 DYNAMIC CHARACTERISTICS OF THE STRUCTURE OF CYTOCHROME C The dynamic characteristics of the crystal structure of cytochrome c have been studied by analysis of thermal B-factors 25 and by molecular-dynamics calcula- t i o n ~ . ~ ~ . ~ ~ These studies produced results typical of globular proteins; only surface residues are particularly mobile. Slightly larger motions were observed for the chain bearing Met80 compared with those of the chain bearing Hisl8. In solution different regions of the structure have associated with them different dynamic characteristics.These characteristics have been investigated with n.m.r. spectroscopy. However, complete definition of the dynamic states of cytochrome c remains to be obtained. The motion of aromatic residues about their Cg-Cr bonds (flipping 180 O between equivalent orientations) is one probe of protein mobility.28 At 25 "C the following aromatic residues of mitochondria1 ferrocytochrome c are relatively immobile : 2o PhelO, Phe46, Tyr48, Tyr67 and Tyr97. The phenyl and tyrosyl rings flip about their Cg-C,, bonds with rate constants <lo2 s-l; for Phe46 and Tyr97 of horse ferro- cytochrome c at 25 "C the rate is 9 -+ 3 s-l.Moreover, the activation energy for flipping of the rings is very high: for Phe46 and Tyr97 it is 104 & 8 kJ mo1-l. Other aromatic residues of cytochrome c (Phe36, Tyr74 and Phe82) are relatively mobile, although in most of these cases it is not possible to obtain precise values of the rate constants, For Phe82 the rate is >,lo4 s-l. A similar pattern is observed for ferri- cytochrome c and cobalticytochrome c but there are some qualitative differences between ferricytochrome c and ferrocytochrome c, with residues of the oxidised state being more mobile. Other bulky groups also undergo motion about their Cg-C, bonds although it is not always possible to study it by n.m.r. In the case of tryptophan it is possible to observe vibratory motion, and for cytochrome c whereas the internal residue Trp59 is immobile, the surface residue Trp33 is mobile.20 Another probe of protein mobility is measurement of NH-ND exchange rates.Although there is a controversy surrounding the mechanism of NH-ND exchange there is general agreement that for an NH proton relatively resistant to exchange the rate and activation energy of exchange are related to fluctuations in the protein struc- ture. Ulmer and Kagi 29 have observed that NH protons of ferrocytochrome c are more resistant to exchange than are those of ferricytochrome c, implying that there are more dynamic states available to ferricytochrome c than there are available to ferro- cytochrome c. The indole NH proton of Trp59 is particularly resistant to exchange; 30931 for horse cytochrome c at pH 7 and 45 "C it has a half-life in ferrocytochrome c of 480 min and a half-life in ferricytochrome c of 20 min.This striking oxidation-state dependence is due in part to the oxidation-state-linked conformation change, and in part to different dynamic characteristics of the two oxidation states of cytochrome c. These data lead to a clearer understanding of the structure of cytochrome c. The most mobile residues are those exposed at the surface of the protein, while those residues internal to the protein are subject to packing constraints that restrict their motion (fig. 5). However, differences in the mobility of exposed residues are apparent, showing that factors other than the degree of exposure are important in determining the dynamic properties of cytochrome c.The temperature dependence of the n.m.r. spectrum can also provide information on protein mobility where temperature-dependent changes in structure are the domin- ant effect. Most of the n.m.r. resonances of ferricytochrome c are temperature dependent because of the paramagnetism of the ferric ion; there is a Curie-type dependence to their chemical shifts. However, the chemical shifts of resonances of a diamagnetic protein are temperature independent unless there is a conformation change. A study of the diamagnetic proteins ferrocytochrome c and cobalticyto-320 THE FUNCTION OF CYTOCHROME C chrome c shows that there are only a few sets of resonances that are temperature d e ~ e n d e n t . ' ~ . ~ ~ One set of resonances is from groups at the bottom left of the molecule (as viewed in fig.4) and include Tyr74 and Ile57, while another set of resonances are from groups at the top right of the molecule and include Val20 and haem meso 8. All these residues are close to the surface of the molecule. The temperature dependence of resonances of Ile57 and Tyr74 has been analysed in detail 32 because this region of the protein also experiences a relatively large change in structure upon oxidation-state change. The temperature-dependent effects can be ascribed to a small movement of Tyr74 and to rotation of Ile57 about its C,-Cb bond (fig. 6). Thus, the relatively 75 58 FIG. 6.-FlexibiIity of cytochrome c around Ile57. The two loops of residue, the 50s and 70s loops, shield the base of the haem group (fig.4). Both Tyr74 and He57 rotate about their CB-CY bonds (heavy single arrow), and the peptide backbone may also be flexible (double arrow). Despite the motion, the structural integrity of this part of the moIecule is retained. Although they rotate, the conformation of Tyr74 is little altered and the conformation of He57 is only altered to the extent that the 6CH3 group moves further into the protein as the yCH3 group moves out of the protein (broken arrow). Trp59 is immobile. mobile residues Ile57 and Tyr74 shield the relatively immobile residue Trp59. It is particularly important to discover what are the dynamic aspects of the redox centre of cytochrome c. Some dynamic aspects of the axial sulphur ligation have already been described. Conformational heterogeneity involving the haem of horse ferricytochrome c has been observed by n.m.r.spectroscopy and described 33 as the interconversion between different protein conformations around haem methyl-3 with a rate constant of ca. lo3 s-l. The origin of the conformational heterogeneity is not clear, although it is notable that conformation heterogeneity is reflected by resonances of other groups near to the haem (e.g. Thr28 of horse cytochrome c). It may be rele- vant that the haem is buckled and that the pyrrole ring bearing haem methyl-3 is deformed.l4' l5 Fig. 7 summarises what is known about the dynamic characteristics of the struc-G. R. MOORE et al. 32 1 5 n 7 54 FIG. 7.-Summary of the dynamic characteristics of cytochrome c. The positions are indicated of groups known to possess anomalous dynamic properties (solid areas), and of groups known to be relatively immobile (cross-hatched areas).Side-chains of tuna cytochrome c are illustrated in the diagram. The dynamic characteristics of other mitochondria1 cytochromes c are similar to those of tuna cytochrome c. ture of cytochrome c. Residues internal to the protein are rclatively immobile, while residues on the surface are relatively mobile. It is particularly notable that the region of the protein around the exposed edge of the haem is relatively mobile, since this is probably the electron injection and ejection site (see later). STRUCTURAL INTEGRITY OF CYTOCHROME C Many chemically modified derivatives of cytochrome c have been obtained in attempts to locate its essential functional groups.34 Some of the modifications deac- tivate the protein by causing the Fe-S bond to rupture. This is another manifesta- tion of the weakness of the Fe-S bond (see earlier) and it has provided an impetus to the determination of the structure of cytochrome c in which there is no Fe-S bond.The structure has been determined from n.m.r.35 data of the stable cyanoferricyto- chrome c and the unstable cyanoferrocytochrome c (fig. 8). Displacement of the Met80 sulphur from the iron by CN- only perturbs the left side of the molecule, specificially the loop of residues from Phe82 to Tyr74. Phe82 is displaced from its position close to thioether bridge-2 but it remains reasonably close to the haem; the para proton is 50.5 nm away. Also Met80 is not completely expelled from the haem pocket.The conformation of the box, Tyr74, Trp59 and Ile57, is only slightly perturbed. The residues Phe82 and Tyr74 mark the extent of the322 THE FUNCTION OF CYTOCHROME C perturbations along the polypeptide chain. Although it is possible that other changes are transmitted through the protein, any such change must be small. A striking feature of the structure is that not only is the right side unperturbed by the ligand displacement but also the dynamic characteristics of the structure are largely unchanged. Thus, the same aromatic residues that are relatively immobile in 87 54 FIG. 8.-The structure of cyanocytochrome c. Only the loop of residues from positions 74-82 is seriously affected, but the S-ligand is replaced by CN-.native cytochrome c are also immobile in cyanocytochrome c. (We note that the extent of the similarities between native cytochrome c and cyanocytochrome c lends support to the idea ’ that one of the last steps in the folding of the polypeptide chain to form native cytochrome c is the formation of the Fe-S bond.) THE ELECTRON-TRANSFER PROCESS The structure of cytochrome c has now been described in some detail. Further definition of the electron-transfer process requires knowledge of the lifetime of con- tact between redox partners (rates of binding and association constants), and of the geometric and chemical relationships between the redox centres (the interaction sites) and of the driving force of the reaction (relative redox potentials). Only when these data are available will the rates of electron transfer be comprehensible. The reactions of cytochrome c with its physiological partners are exceedingly complex (fig.l).3-6 Both the reductase and oxidase complexes are coupled to ATP synthesis, and the oxidase complex has the additional function of reducing O2 to H20. In addition to these complexities, pure preparations of the oxidase and reductase complexes are not yet routinely available. Thus many workers have turned to simpler systems to investigate the electron-transfer function of cytochrome c.G. R . MOORE et al. 323 The reaction of iron hexacyanides with cytochrome c is one such system. First studied 36 by Sutin and Christman in 1961, this system has now been extensively studied (see later) and holds the prospect of allowing various theories of electron trans- fer to be experimentally tested.The reactions of cytochrome c with cytochrome b, and cytochrome c peroxidase have also been extensively studied, and those investi- gations aimed at defining the electron-transfer complexes are described next. INTERACTION SITES ON CYTOCHROME C FOR ITS PHYSIOLOGICAL REACTANTS The interaction sites on cytochrome c for the reductase and oxidase complexes have been determined by a number of te~hniques.~~-~O The different studies agree that the interaction region (fig. 9) is largely the same for both complexes and includes \ 1 I FIG. 9.--The reductase-oxidase interaction domain of cytochrome c. The protein is viewed from the same orientation as in fig. 4. The haem is viewed edge-on (solid bar) and the molecule is cut by an imaginary plane just behind the haem and perpendicular to the haem plane.The approximate location of &NHz groups of the lysyl residues is in front of and behind the imaginary plane and is indicated by closed and dashed circles, respectively. Cross-hatched circles indicate lysyl residues of the interaction domain. Residues peripheral to the interaction domain are indicated by hatched circles and residues not part of the domain are indicated by open circles. [Reproduced with permis- sion from ref. (39).] the upper part of the exposed haem edge. Computer modelling of the interactions between cytochrome c with cytochrome b5 41 and cytochrome c peroxidase 42 have also implicated this region of cytochrome c.These studies go further and suggest a struc- ture for the complexes formed (fig. lo). The computer-modelling and chemical- modification studies show that lysine residues 13, 27, 72, 86 and 87 play a key role in holding the complex together and that Phe82 is in the centre of the interaction site. The iron to iron distance in the complex is 2.46 nm, and the closest haem-edge to haem-edge distance is 1.65 nm. This is the closest edge to edge distance that can be attained without involving considerable changes in protein structure. The postulated 41 complex between cytochrome c and cytochrome b, resembles the complex in fig. 10 in one important detail. In both complexes the haem planes are parallel, although in the cb5 complex the haem-edge to haem-edge distance is only 0.84 nm..324 THE FUNCTION OF CYTOCHROME C FIG. 10.-The inter-haem region of the complex between cytochrome c and cytochrome c peroxida~e.~~ The haems of cytochrome c (left) and cytochrome c peroxidase (right) are separated by two loops of amino-acid residues. One from cytochrome c contains Met80, Ile81, Phe82, Ala83 and Gly84, and the other from peroxidase contains Thr179, Hisl80, Leu181 and Ala182. THE REACTION OF CYTOCHROME C WITH IRON HEXACYANIDES Chemical-modification studies 43,44 and n.m.r. spectroscopic investigations 4s*46 have shown that the interaction site for the negatively charged reagents [Fe(CN),I3-/ [Fe(CN),I4- is at the exposed haem edge, in the same region as for cytochrome c oxidase and cytochrome c reductase (fig. 9). Thus the first criterion for modelling the physiological electron-transfer function of cytochrome c is met.However, it is not clear how many reaction sites there are for the iron hexacyanides. At present the most plausible interpretation of the data is that there are two sites for horse cyto- chrome c, one below and one above the haem plane (fig. 11). The Fe-Fe distances in these two sites are ca. 1.20 0.20 nm, respectively, and the haem- NC distances are 1 .OO & 0.1 5 and 1 .OO & 0.20 nm, respectively. 0.15 and ca. 1.30 The following mechanism has been proposed by Ohno and Cusanovich:47 ki, kf k23 kE k34 kzt kr k32 k , k43 Cyt"' + Fe4- - C, TL: C; - Ci. C, T- Cyt" + Fe3- (1) where CI and C, represent collision complexes, and C'I and C', intermediates between which electron transfer occurs.Fe4- and Fe3- are ferrocyanide and ferricyanide, respectively, and Cyt"' and Cyt" are ferricytochrome c and ferrocytochrome c, respectively. The parameters given by Ohno and Cusanovich for horse cytochrome c are: k,, = 5 x lo8 dm3 mol s-I K , = - = 10 dm3 mol k,, = 5 x 107 s-1 k21G . R. MOORE et al. 325 kf = 1 x lo4 s-l kr = 710 s-' k23 = 2600 S-' k32 = 2500 s k; = 240 s-' k: = 2 x lo6 S-' k' K$ = 4 = 1.2 x kr k34 = 5 x lo7 S-' K3 = = 1 x 10-1 mol dm-3 k43 = 5 x lo8 dm3 mol -' s-l k43 kik, = 140 dm3 mo1-' k3kj = 1.2 x loy5 mol dmy3. If there are two hexacyanide interaction sites from which electron transfer can occur mechanistic analysis of the kinetic data will be difficult. For the present it has to be assumed that the probability of electron transfer is the same for both sites.Cusanovich 48 has made a preliminary analysis of the kinetic data using Marcus t h e ~ r y . ~ ~ * ~ * When the collision complex is short-lived the correlation equation for relating cross-reaction rates to equilibrium constants is kl2 = (k11k22keq.f >+ (1) FIG. 11.-The haem crevice binding sites for [Fe(CN)J3- on horse cytochrome c. The orientation of the diagram is that of fig. 4 oriented ca. 90 "C to the left. The cross-hatching represents residues close to the bound [Fe(CN)J-, The arrows indicate the positions where [Fe(CN)6]3- binds. (Note that because of the type of spectroscopy used not all of the residues close to the bound [Fe(CN,)13- have been positively identified.}326 THE FUNCTION OF CYTOCHROME C where k12 is the cross-reaction rate constant, k,, and k22 are the self-exchange rate constants for the reactants, Kes is the overall equilibrium constant andfis a correction factor to take into account the collision frequency.It is usually unity for small molecules. This equation does not include work terms for complex formation which may be important for proteins. The self-exchange rate constant for horse and Candida krusei cytochromes c have been shown 51 to be lo4 dm3 mo1-I s-' and lo3 dm3 mol-1 s-l, respectively, at high ionic strength. Preliminary results with cow and chicken cytochromes c show 35 that the self-exchange is similar to horse cytochrome c. The cross-reaction rate constants k23 and k32 for reaction (I) are 2.6 x lo3 s-' and 2.5 x lo3 s-l, respectively, for horse, cow and chicken cytochromes c but are 4.6 x lo3 s-l and 5.6 x lo3 s-l, respectively, for Candida krusei cytochrome c, and the overall rate constant for [Fe(CN),14- reduction is 3.2 x lo5 dm3 mol-1 s-l for horse cytochrome c and 6 x lo5 dm3 mol-1 s-' for Candida krusei cytochrome c.The order of variation in the cross-reaction rate constants is contrary to that expected from the cytochrome c self-exchange rate constants and eqn (I) when f = 1. However, if the complexes formed between iron hexacyanides and Candida krusei cytochrome c are different from those involving the other mitochondria1 cytochromes, the apparent discrepancy might be clarified. Preliminary n.m.r. results 35745 indicate that there is indeed a difference between the interactions of iron hexacyanides with different cytochromes c, but de- tailed structural models are not yet available.All the studies on the proposed electron-transfer complexes of biological origin and on the model complexes indicate that for an electron-transfer rate of lo4 s-l to be achieved the distance between centres, including conjugated ligands, is <2 nm. Detailed analysis of the reactions of different cytochromes c should reveal the effect of changes in disposition of groups at the binding site and should throw light on the chemical groups which can be tolerated between the electron-transfer centres. ELECTRON-TRANSFER PATHWAY Although the electron-transfer pathway need not be part of the reductase and oxidase binding site, there are no reasons for believing it is not.Thus the electron- transfer pathway is assumed to include the exposed edge of the haem. In the postul- ated cytochromes cb, complex 41 the haem-edge to haem-edge distance is probably short enough to permit direct electron transfer from one haem to the other. Also in the complex between cytochrome c and iron hexacyanide (fig. 11) the close approach of the hexacyanide to the haem probably permits direct electron transfer. However, in the postulated complex between cytochrome c and cytochrome c peroxidase (fig. 10) the inter-haem distance is probably too great for direct electron transfer to occur from haem-edge to haem-edge at the observed rate. This may also be the case with cyto- chrome c and its respiratory oxidase and reductase, although there may be vibrational states within the complex which considerably shorten the electron-transfer distance.Thus the electron-transfer pathway may involve the medium formed by amino-acid side chains of either or both cytochrome c and cytochrome c peroxidase. Margo- liash 37 has drawn attention to the fact that Phe82 is in the centre of the interaction site close to the exposed edge of the haem, and Poulos and Kraut have emphasised 42 this by suggesting that Phe82 is part of the electron-transfer pathway. In particular they suggest that e!ectron transfer from cytochrome c to cytochrome c peroxidase involves a " super-molecular " conduction orbital formed by overlap of 7t orbitals from the haem and a variety of conjugated residues of cytochrome c peroxidase, and from Phe82 and the haem of cytochrome c.The previously described observations of con-G . R. MOORE et al. 327 formational flexibility in this region of cytochrome c, particularly of Phe82, are con- sistent with the requirements for structural relaxation about the electron-transfer site, although whether such flexibility persists within the electron-transfer complex is not known. The fact that Phe82 is one of the invariant residues of cytochrome c is also important. Further studies of the proposed electron-transfer pathway should aim to replace Phe82 with a non-aromatic group (e.g. alanine or leucine) without grossly dis- rupting the protein structure. SEQUENCE VARIATION OF MITOCHONDRIAL CYTOCHROME C The description of cytochrome c given in the previous sections illustrates how the polypeptide chain folds around the haem to create a hydrophobic haem environment with only one edge exposed at the molecular surface, to create interaction sites for cytochrome c reductase and cytochrome c oxidase, and to create electron-transfer pathways, all of which require formation of the Fe"'-S bond. These structural and functional requirements place severe constraints upon the sequence variation of cyto- chrome c, constraints which are reflected in the fact that 24 out of 103 amino-acid residues are invariant in 82 mitochondrial cytochromes c and 25 residues are con- servatively replaced.* Of the conserved residues (fig.5), 4 are haem binding residues, 9 are glycine and proline (which are required for bends and turns in the polypeptide), and the remaining 11 residues are either haem contact residues or residues involved in the redox inter- action site.Interestingly, only one of the charged residues involved in the interaction is totally invariant (Lys73), although another residue is only conservatively replaced (Lysl3/Argl3). The interaction site encompasses a region of the surface of cyto- chrome c which includes 6-8 positively charged residues, and although some of these residues are variable each particular cytochrome contains many of them. Thus a key property of cytochrome c is retained even though some of the amino-acid residues controlling the property are not invariant. This must also be true of other properties. Although there is a degree of constancy of amino-acid sequences, 21 amino-acid residues are variable and 33 are hypervariable.In other words, a third of the amino- acid sequence appears to have little functional significance. However, this view must be wrong. Despite the sequence variations redox potentials are maintained at 260 -)= 20 mV.6* 7*34 By way of contrast, structurally related non-mitochondria1 cytochromes c which operate in different systems from mitochondrial cytochromes c, and which have the same coordination sphere and similar protein fold to the mitochondrial proteins, have redox potentials varying from 100 to 500 mV.6952 It seems that some amino-acid substitutions of mitochondrial cytochrome c compensate each other so as to maintain the redox potential of cytochrome c near 260 mV. That it is important to maintain the potential near this value can be recognised from the redox potentials of the cytochrome c reductase and cytochrome c oxidase donor and acceptor sites.These are ca. 220 and ca. 300 mV, re~pectively.~.~ Thus, given a plethora of mechan- sisrn to control the redox potential, and an interacting protein structure designed to maintain the redox potential, it is necessary to use chemically modified proteins to greatly perturb the redox potential. For example, replacement of the conservatively substituted Tyr67 of horse cytochrome c by leucine produces a protein with a redox potential of 200 mV.53 This decrease is not due to a gross structural change,54 and indeed the protein is active with isolated cytochrome c o x i d a ~ e . ~ ~ The idea of compensating amino-acid substitutions is supported by detailed n.m.r.55 and Raman spectroscopic 56 studies of mitochondrial cytochromes c.These have revealed that even for small changes in amino-acid type there are minor changes328 THE FUNCTION OF CYTOCHROME C in protein conformation that can be transmitted over large distances. Thus a change at position 92, which is a hypervariable residue, can be transmitted over ca. 1.6 nm to the centre of the p ~ r p h y r i n , ~ ~ and a change at position 65, another hypervariable residue, can be transmitted over 1.8 nm to the region of the protein about PhelO and Tyr97 .57 IS CYTOCHROME C A N IDEAL ELECTRON-TRANSFER REAGENT ? The suggestion has frequently been made that biological systems have found by random searching almost perfect solutions to chemical problems.The observation of constant redox potential of many different mitochondria1 cytochromes c from very different species, and the similarities in electron self-exchange rate constants, support this argument. It appears as if mitochondria have " homed in '' on these properties despite sequence variations. Many of the requirements for fast electron transfer are known. The redox poten- tial is one such control. A second is the ability of the coordination sphere to relax. Phe82 is mobile and the Fe"'-S bond is adjustable, perhaps all the surrounding protein at the haem edge is adjustable. Good relaxation conditions are therefore met. The distance between redox centres within complexes of ca. 1-2 nm is long, but it may well be the optimum since two recognition surfaces are required between the centres. Thus given the requirements of biological systems, especially of self-assembly, cyto- chrome c may well be close to being an ideal electron-transfer reagent. The experimental work discussed in this paper is supported by the S.E.R.C. and the Royal Society.G. R. M. thanks the S.E.R.C. for an Advanced Fellowship. We thank our colleagues, Drs J. Angstrom, P. Boon, H. A. 0. Hill, J. J. G. Moura, G. W. Pettigrew, C . J. A. Wallace, M. T. Wilson and A. V. Xavier, for many helpful discussions and advice concerning the function of cytochrome c. P. Mitchell, Nature (London), 1961, 191, 144. R. J. P. Williams, J. Theor. Viol., 1961, 1, 1. D. B. Kell and G. D. Hitchens, Faraday Discuss. Chem. Soc., 1982, 74, 377. Tunnelling in Biological Systems, ed.B. Chance, D. DeVault, M. Fraunfelder, R. A. Marens, J. R. Scheffer and N. Sutin (Academic Press, New York, 1979). D. G. Nicholls, Bioenergetics (Academic Press, New York, 1982). R, Lemberg and J. Barrett, Cytochromes (Academic Press, New York, 1973). R. E. Dickerson and R. Timkovich, in The Enzymes, ed. P. D. Boyer (Academic Press, New York, 1975), vol. XI. * M. 0. Dayhoff, Atlas ofprotein Sequence and Structure (National Biomedical Research Founda- tion, Silver Spring, Maryland, 1968-78). H. A. Harbury, J. R. Cronin, M. W. Fanger, J. P. Hettinger, A. J. Murphy, Y. P. Myer and S. N. Vinogradov, Proc. Natl Acad. Sci. USA, 1965, 52, 1462. lo C . C. McDonald, W. D. Phillips and S . N. Vinogradov, Biochem. Biophys. Res. Commun., 1969, 36, 442.l1 K. Wuthrich, Proc. Natl Acad. Sci. USA, 1969, 63, 1071. l2 I. Aviram and A. Schejter, Biopolymers, 1972, 11, 2141. l3 G R. Moore, R. J. P. Williams, J. C. W. Chien and L. C. Dickinson, J. Inorg. Biochem., l4 T. Takano and R. E. Dickerson, J. Mol. Biol., 1981,153, 79. l5 T. Takano and R. E. Dickerson, J. Mol. Biol, 1981, 153, 95. l6 A. Labhardt and C. Yuen, Nature (London), 1979, 277, 150. l7 T. Mashiko, C. A. Reed, K. J. Haller, M. E. Kastner and W. R. Scheidt, J. Am. Chem. SOC.. 1980, 12, 1. 1981, 103, 5758. N. Sutin and J. K. Yandell, J Biol. Chem., 1972, 247, 6932. l9 J. Angstrom, G. R. Moore and R. J. P. Williams, Biochim. Biophys. Acta, 1982, 703, 87. 2o G. R. Moore and R. J. P. Williams, Eur. J. Biochem., 1980, 103, 533.G . R. MOORE et al.329 21 R. J. Kassner, Proc. Natl Acad. Sci. USA, 1972,69, 2263. 22 R. E. Dickerson, Sci. Am., 1980,242, 136. 23 G. R. Moore, G. W. Pettigrew, R. C. Pitt and R. J. P. Williams, Biochim. Biophys. Acta, 1980, 24 H. R. Bosshard and M. Zurrer, J . Biol. Chem., 1980,255, 6694. 25 S. H. Northrup, M. R. Pear, J. A. McCammon, M. Karplus and T. Takano, Nature (London), 26 S . H. Northrup, M. R. Pear, J. A. McCammon and M. Karpluc, Nature (London), 1980, 286, 27 S. H. Northrup, M. R. Pear, J. D. Morgan, J. A. McCammon and M. Karplus, J . Mol. Biol., 28 I. D. Campbell, C. M. Dobson, G. R. Moore, S. J. Perkins and R. J. P. Williams, FEBS Lett., 29 D. D. Ulmer and J. H. R. Kagi, Biochemistry, 1968, 7, 2710. 30 D. J. Pate1 and L. L Canuel, Proc. Natl Acad. Sci. USA, 1976, 73, 1398. 31 G.R. Moore, A. B. V. P. De Aguiar, N. D. Pluck and R. J. P. Williams, in Biochemical and Medical Aspects of Tryptophan Metabolism, ed. 0, Hayaishi, Y . Ishimura and R. Kido (Elsevier/ North Holland, Amsterdam, 1980), pp. 83-94. 32 M. N. Robinson, C. G . S. Eley, A. P. Boswell and G . R. Moore, Biochem. J., submitted for publication. 33 P. D. Burns and G. LaMar, J . Biol, Chem., 1981,256, 4934. j4 S. Ferguson-Miller, D. L. Brautigan and E. Margoliash, in The Porphyrins, ed. D. Dolphin (Academic Press, New York, 1979), vol. VII, pp. 149-240. j5 G. R. Moore et al., unpublished data. 36 N. Sutin and D. R. Christman, J. Am. Chem. SOC., 1961,83, 1773. 37 S. Ferguson-Miller, D. L. Brautigan and E. Margoliash, J . Biol Chem.: 1978, 253, 149. j8 G. W. Pettigrew, FEBSLett., 1978, 86, 14. 39 R. Rieder and H. R. Bosshard, J. Biol. Chem., 1980,255, 4732. 40 H. T. Smith, N. Staudenmayer and F. Millett, Biochemistry, 1977, 16, 4971. 41 F. R. Salemme, J. Mol. Biol., 1976, 102, 563. 42 T. L. Poulos and J. Kraut, J. Biol. Chem., 1980,255, 10 322. 43 J. Butler, D. M. Davies, A. G. Sykes, W. H. Koppenol, N. Osheroff and E. Margoliash, J. Am. 44 A. J. Ahmed and F. Millett, J. Biol. Chern., 1981, 256, 1611. " C. G. S. Eley, G. R. Moore, G. Williams and R. J. P. Williams, Eur. J . Biochem., 1982, 124, 46 J. J. Hopfield and K. Ugurbil, in Interaction Between Iron and Proteins in Oxygen and Electron 47 N. Ohno and M. A. Cusanovich, Biophys. J . , 1981,36, 589. 48 M. A. Cusanovich, Bio-org. Chem., 1978, IV, 11 7. 49 R. A. Marcus, J. Phys. Chem., 1963, 67, 853. 'O S. Wherland and H. B. Gray, in Biological Aspects of Inorganic Chemistry, ed. A. W. Addison, 51 R. K. Gupta, Biochim. Biophys. Acta, 1973, 292, 291. 52 G. W. Pettigrew, R. G. Bartsch, T. E. Meyer and M. D. Kamen, Biochim. Biophys. Acta, 53 P. J. Boon, Doctoral Thesis (University of Nijmegen, 1981). 54 C. G. S. Eley, G. R. Moore, R. J. P. Williams, P. J. Boon, H. K. Brinckhof, R. J. F. Nivard, G. I. Tesser and W. Neupert, Biochem. J., 1982, 205, 153. 55 G. R. Moore and R. J. P. Williams, Eur. J . Biochem., 1980, 103, 543. 56 J. A. Shelnutt, D. L. Rousseau, J. K. Dethmers and E. Margoliash, Biochemistry, 1981, 20, 57 A. P. Boswell, G. R. Moore, R. J. P. Williams, C . J. A. Wallace, P. J. Boon, R. J. F. Nivard, 590, 261. 1980,287, 659. 304. 1981, 153, 1087. 1976, 70, 96. Chem. SOC., 1981, 103,469. 295. Transport, C. Ho (Elsevier/North Holland, Amsterdam), in press. W. R Cullen, D. Dolphin and B. R. James (Wiley, New York. 1977), pp. 289-368. 1978, 503, 509. 6485. and G. I. Tesser, Biochem. J., 1981, 193,493.

ISSN:0301-7249    

DOI:10.1039/DC9827400311

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Faraday DiJcuss. Chem. SOC., 1982, 74, 331-341 Factors Influencing the Electron-transfer Rates of Redox Proteins BY MARK J. EDDOWES AND H. ALLEN 0. HILL Inorganic Chemistry Laboratory, South Parks Road, Oxford OX1 3QR Received 10th May, 1982 The electron-transfer reaction of cytochrome c at a surface-modified gold electrode has been investigated using a variety of electrochemical techniques. The reaction is rapid and the mechanism, deduced from the observed electrode kinetics, involves adsorption of the protein at the electrode surface prior to electron transfer. This adsorption step is a crucial factor in the enhancement of the observed, overall rate of the electron-transfer reaction, for which there is a large free energy of activation for the electron-transfer step itself.The relationship between the heterogeneous electron- transfer reaction of cytochrome c and its reaction with its physiological redox partners in homo- geneous solution is discussed. The rates of electron transfer between horse-heart cytochrome-c and its physio- logical redox partners are fast.l-s There has been 6-10 much discussion of these facile electron-transfer reactions. Most emphasis has been placed on mechanisms which invoke a " facilitated pathway " for electron transfer through the fabric of the protein via strategically placed amino-acid residues. Attention has also been focused lo on the possible role of the ligand environment of the redox centre in minimizing the reorganisation energy consequent upon a change in redox state and thus maximizing the rate of electron transfer.However, in addition to accounting for the rapid rates of electron transfer, any proposed mechanism should also be consistent with the degree of specificity exhibited l1 by cytochrome c towards reaction with its physiological redox partners. Comparative kinetic and X-ray crystallographic studies l2 and many detailed and selective chemical modification studies 12-23 suggest that this specificity is associated with interactions due to charged amino-acid residues on the protein surface. These charge interactions appear 23*24 to be responsible for the form- ation of the precursor complexes, formed prior to electron transfer, consistent with their influence upon the observed overall rate of electron transfer. Redox reactions between metalloproteins and small-molecule redox reagents have been studied 25 in attempts to elucidate the mechanism of electron transfer.The formation -of tightly bound, protein-redox-reagent precursor complexes prior to electron transfer has been observed 26-29 in the reactions of cytochrome c, and other redox p r ~ t e i n s . ~ ~ - ~ ~ As in the case of protein-protein redox reactions, such binding between reactants would be expected to have a significant influence on the observed overall electron-transfer rates. In contrast to the rapid rates often encountered in the study of protein redox reactions in homogeneous solution, there have been few reports of rapid, direct elec- tron transfer between electrodes and redox proteins in solution. Indeed, for cyto- chrome-c the rate of heterogeneous electron transfer at gold and platinum electrodes has been found36*37 to be so slow that electron transfer is essentially undetectable using conventional electrochemical techniques involving current measurement.Thus332 PROTEIN ELECTRON-TRANSFER RATES the electrochemist, rather than asking why protein electron transfer is so fast, might often need to ask why it is so slow! However, rapid, direct heterogeneous electron transfer has been reported 38 -44 for several redox proteins using certain electrode materials. In these cases adsorption of the protein at the electrode surface has often been observed. In this respect the heterogeneous reactions are analogous to the homogeneous protein-protein reactions -5 in that rapid electron transfer coincides with a significant binding interaction between the reactants.The study of the heterogeneous electron-transfer reactions of proteins should offer certain advantages in the investigation of the factors influencing their redox kinetics. Electrochemical methods, allowing experimental control over the mass-transport rate and, through the electrode potential, over the electron-transfer rate, facilitate the separation of the contribution to the overall rate made by the kinetics of the component elementary steps, thus enabling a detailed kinetic analysis. Here we present the results of the study of the kinetics of the electron-transfer reaction of cytochrome c at a surface-modified gold electrode. These results are compared with those obtained from the study of the homogeneous electron-transfer reactions of cytochrome c, and their indirect implications for biological electron transfer in general are discussed.ELECTROCHEMISTRY OF CYTOCHROME c AT A SURFACE-MODIFIED GOLD ELECTRODE STATIONARY-ELECTRODE STUDIES Though the electrochemical reduction of cytochrome c at an unmodified gold electrode is slow 37*40 at potentials close to its standard electrode potential, it has been found to undergo a rapid and reversible electron-transfer reaction at a gold electrode, whose surface has been modified by an adsorbed layer of 4,4’-bipyridyl, as demon- strated by d.c. [fig. l(a)] and a.c. cyclic voltammetry. A.c. impedance studies con- firm that the electrode reaction is indeed fast, and the standard electrochemical rate constant, kz z 1.5 x m s-l, has been d e t e ~ m i n e d .~ ~ ~ ~ ~ The 4,4’-bipyridyl, which is not electroactive in the potential region of the observed electron-transfer reaction and therefore cannot function as a conventional mediator, acts by adsorbing on the electrode surface, thereby modifying it to produce a suitable interface at which rapid electron transfer can take place. Cyclic-voltammetry studies have demonstrated 45 some significant analogies be- tween the heterogeneous electron-transfer reaction of cytochrome c and its reaction with its physiological redox partners, indicative of the role of the modified electrode surface in promoting electron transfer. The effect of chemical modification of the cytochrome c lysine residues on the electrode reaction has_ been investigated.The enzymatically inactive trifluoroacetyl and maleyl derivatives, in which the positively charged, protonated &-amino groups of the lysine residues are replaced by neutral and negatively charged functional groups, respectively, were both found to be electro- inactive at the modified gold electrode. On the other hand, the enzymatically active acetimidyl and guanidinyl derivatives, in which the modified forms retain protonated, basic nitrogeneous functional groups, give rise to d.c. cyclic voltammograms in- distinguishable from those of the native protein. The importance of these lysine residues in determining the protein-protein interactions responsible for binding in the precursor complexes formed by physiological redox partners has been demon- ~ t r a t e d .l ~ - ~ ~ It has therefore been proposed 45 that the heterogeneous electron- transfer reaction takes place by an analogous mechanism, in which the protein binds to the modified electrode surface prior to electron transfer. The interaction may involve hydrogen bonding between those lysines adjacent to the partially protrudingM. J . EDDOWES AND H . A . 0. HILL 333 0.4 I 0.2 4 0.0 2 -0.2 - 0.4 - 0.2 0.0 0.2 E/V us. SCE FIG. 1.-D.c. cyclic voltammograms of horse-heart ferricytochrome c ( 5 mg ~ m - ~ ) in NaC10, (0.1 mol dmP3), phosphate buffer (0.02 mol dm-3) at pH 7 in the presence of 1,2-bis(.l-pyridyl)- ethylene in the potential range +0.2 to -0.2 V us. SCE with poly-r-lysine (a) 0, (6) 1, ( c ) 1.5 mg ~ r n - ~ . The increasing irreversiblity with increasing poly-L- lysine concentration illustrates its inhibitory effect due to blockage of adsorption sites on the electrode haem edge and the 4,4’-bipyridyl which may be “ perpendicularly ” adsorbed on the surface.The importance of the lysine residues in the electrode reaction is further indicated by the effect 45 of polylysine, a competitive inhibitor of the physiological redox reactions of cytochrome c , ~ ’ which acts by preferentially binding to its physio- logical redox partners. The presence of polylysine inhibits the electrode reaction, as shown by its effect on the cyclic voltammetry [fig. 1 (b) and ( c ) ] , consistent with adsorp- tion of polylysine on to the electrode surface, decreasing the effective free electrode area for reaction with cytochrome c.D.c. potential scan rate 100 mV s - ’ . surface [from ref. (45); by permission]. ROTATING-DISC ELECTRODE STUDIES The reaction mechanism, outlined in scheme 1, involving adsorption of cyto- chrome c at the electrode prior to electron transfer, was further investigated 48 by using the rotating-disc method. For the steady state we can write the following equations for the flux of ferricytochrome c (Ox) to ferrocytochrome c (Red) at the rotating-disc electrode : j = kD(OX, - 0x0) = kads(l - == rLk-e,ox - rLke,red = rLkdesored - kads(l - 00, - 6red)Red0 - ered)OXO - rLkdesoox = kD(Redo - Red,)334 PROTEIN ELECTRON-TRANSFER RATES kD kada kdea Oxm 0% transport in solution 0x0 Ti oxads adsorption and desorption of reactant ke k-c Oxads @ Red,,, electron transfer kdes Redads + Redo desorption and adsorption of product kads k, Redo Redm transport in solution SCHEME 1 .-Reaction mechanism.where kD is the mass-transport rate constant for a rotating-disc electrode, given 49 by the Levich equation, kads and kdes are the rate constants for adsorption and desorp- tion of cytochrome c and k, and k-, are the potential-dependent first-order rate constants for the oxidation and reduction of adsorbed ferro- and ferri-cytochrome c, respectively. O, and Ored are the fractional surface coverages of Ox and Red and is the total number of adsorption sites per unit area of the modified electrode. The unknowns, Oox, Ore& Ox, and Red,, can be eliminated by equating these flux equations to give the general expression for the flux in terms of the four rate constants and the bulk concentrations of the reacting species : Although this general expression is complex, since the electrode potential can be experimentally selected such that either terms in k,(oxidation)(k,,,,) or k-,(reduction)- (ke,red) dominate, the expression for the limiting flu simplifies to a form of the Koutecky-Levich equation,50 and for reduction is given by (3) 1 1 1 1 Red +- - _ - - j , - k,Oxm + kadsOXm rLkdes + rLkdesOXm In eqn (3) the first three terms on the right-hand side describe the possible rate-limit- ing processes (mass transport, adsorption and desorption, respectively), whilst the fourth term describes the competition between reactant and product for adsorption sites on the electrode.This latter feature is analogous to the product inhibition observed 5*51 in the physiological redox reactions of cytochrome c.The rotation-speed dependence of the limiting current, iL, for both oxidation and reduction of cytochrome c was investigated 48 as a function of concentration. Fig. 2 shows the corresponding Koutecky-Levich plots, illustrating the dependence on the concentration of both reactant and product. From eqn (3) the intercept, I, of the Koutecky-Levich plot gives the rate of the surface step : where w is the number of electrons, F is the Faraday constant and A is the area of the electrode.M . J . EDDOWES AND H . A . 0. HILL 335 0.4 4 I 3 . 5 0.2 .2 W 1.5 1.c 0.5 0.2 0.4 0.6 (W/Hz)-* FIG. 2.-(a) Koutecky-Levich plots, according to eqn (4), of the limiting current, iL, for the reduction of ferricytochrorpe c against square root of rotation speed, W*; ferricytochrome c concentration: A, 26; 0, 48; 0, 101 pmol dm-3.The slope of these plots represents the reciprocal transport rate and the intercept represents the reciprocal overall rate of the surface step. (b) Koutecky- Levich plots for the oxidation of ferrocytochrome c (86 pmol dm-3), in the presence of different concentrations of ferricytochrome c: 0, 0; 0, 105; A, 225 pmol dm-3. The increase in the intercept (representing the reciprocal rate of the surface step) with increasing product concentration illustrates its inhibitory effect as predicted by eqn (4). Thus the contribution of the mass-transport rate to the observed rate can be accounted for, allowing determination of the rate of the reactions at the surface.The dependence of the surface kinetics, as determined from the Koutecky-Levich intercept, on both reactant concentration [fig. 2(a)] and product concentration [fig. 2(b)] is that predicted by eqn (4). Thus it is concluded that the heterogeneous electron-transfer reaction of cytochrome c follows the mechanism outlined in scheme 1. Analysis of the con- centration dependence of the Koutecky-Levich intercept allows determination of kads and r&des, which are given in table 1. TABLE 1 .-VALUES OF ADSORPTION AND DESORPTION RATE CONSTANTS DETERMINED4' FROM THE KOUTECKY-LEVICH PLOTS (FIG. 2) oxidation of reduction of ferrocytochrome c ferricytochrome c kadslms - (3.3 f 0.1) x 1 0 - 4 rLkdes/mol s-l (6.3 0.4) x (5.9 5 0.4) x 1 0 - 5 (2.3 f 0.3) x With the mass-transport, adsorption and desorption rates known from the analysis of the limiting currents, their contributions to the overall electron-transfer rate, as described by' the general expression eqn (2), can be determined. Thus, by analysis of the current-voltage curves at the rotating-disc electrode the electron-transfer rate constant may also be determined.We define kz as the standard first-order electron-336 PROTEIN ELECTRON-TRANSFER RATES transfer rate constant between adsorbed species and the electrode at the standard electrode potential, E", of the system. The potential-dependent, first-order electron- transfer rate constants are given by for the for the k, = k:exp[a(nF/RT)(E - E")] k - , = k:exp[-/?(nF/RT)(E - E")] anodic process and cathodic process. From eqn (2) we have where iLev is the diffusion-controlled current predicted 49 by the Levich equation, k,/Ox,.Writing k, and k-, in terms of kz, as expressed by eqn (5a) and (5b), gives an expression for the potential dependence of the current function, loglo(i/iL - i). Differentiation shows that the slope, aOb,F/2.3RT, of the modified Tafel plot, loglo(i/ iL - i ) against E, at E = E+ is given by = for the anodic reaction and OLobs = for the cathodic reaction. Thus - $1 + - a / [ l + with iLev, iL and kdes known from the rotation-speed dependence of the limiting current, i E may be determined from the slope of the modified Tafel plot, and it is found 48 that k: x kdes. Alternating-current ring-disc experiments have also been carried out 48 and con- firm the adsorption of cytochrome c at the modified gold electrode.These experi- ments also provide a measure of the limiting surface coverage, rL = 1.2 x mol m-2, which is consistent with the limiting coverage corresponding to monolayer form- ation. This, and the rate constant data, are given in table 2, and summarized in fig. 3. TABLE 2.-sUMMARY OF RATE CONSTANTS FOR THE ELECTRON-TRANSFER REACTION OF CYTO- CHROME C AT THE SURFACE-MODIFIED GOLD ELECTRODE diffusion coefficient, D = m2 s-' adsorption rate constant, kads = 3 x desorption rate constant, kdes = 50 s-l electron-transfer rate constant, k," = 50 s-' limiting surface coverage, rL = 1.2 x m s-' mol mF2 DISCUSSION Applying the transition-state theory expression, k = Zexp( -AG+/RT), the free- energy profile for the heterogeneous electron-transfer reaction of cytochrome c at E = E" at the surface-modified gold electrode has been c o n s t r ~ c t e d .~ ~ Using a frequency factor for the adsorption step of 2 = lo2 m s-l (the heterogeneous collision frequency from the Marcus treatment) and for the electron-transfer and desorption steps of 2 = kT/h (the usual first-order frequency factor), the following values for the free energies of activation are obtained: adsorption, AG# z 30 kJ mol-l, electron transfer and desorption, AG+ z 60 kJ mol-l. The free-energy profile [fig. 4 ( ~ ) ]M . J . EDDOWES AND H . A. 0. HILL 337 diffusion D=10-'Om2s-' Q diffusion bulk solution Q deswption < kd,, = 50 s-l electrode surface FIG. 3.-Reaction scheme for the electron-transfer reaction of cytochrome c at the 4,4'-bipyridyl surface-modified gold electrode.contains the following important features: the free energy of activation for the elec- tron-transfer step is large and the free energy for binding of the protein at the electrode surface is also large and negative. Indeed, it is the large binding free energy which overcomes the substantial free energy of activation for electron transfer and thus allows the reaction to proceed at a measurable rate at the modified gold electrode. This is illustrated by comparison of the free-energy profiles for electron transfer with and without binding of the protein at the electrode surface, shown in fig. 4. For the re- 40t - 40 L Oxads G' FIG. 4.-(a) Free-energy profile for the electron-transfer reaction of cytochrome c at the 4,4'-bipyridy surface-modified electrode at a potential E = Ee, illustrating the importance of adsorption in determining the overall rate, kobs = ( Z / 3) exp (- AGZ ,,,/RT).(b) Comparable free-energy profile for the reaction where adsorption of cytochrome c at the electrode surface does not occur, kobs = 2 exp (AG#,,/RT).338 PROTEIN ELECTRON-TRANSFER RATES action with binding the expression for the apparent rate constant, kobs, for reduction, where the product, Red, is absent from solution, is obtained by equating the relevant fluxes in eqn (1): At low concentrations, where the kadsOxO term in the denominator is negligible, kobs attains its maximum value, and since from table 2 we have that k,"x kdes eqn (8) reduces to (9) where Zhet is the heterogeneous collision frequency.binding the rate constant would be given simply by On the other hand, without kobs = Zhet exp (%). Using the values of the activation energies, AGZS and AGZt, obtained from the rate data we find that the rate constant for the reaction with binding, given by eqn (9), represents a rate enhancement by a factor of roughly lo5 above that for the reaction without binding, given by eqn (10). Thus we conclude that binding of the protein to the surface is a crucial factor in the enhancement of the overall rate of electron transfer at the surface-modified electrode and that the substantial free energy of activation for electron transfer may explain why the reaction is so slow at the unmodified electrode, where there may be no binding interaction.As mentioned briefly above, in general, rapid heterogeneous electron transfer between redox proteins in solution and electrodes is not observed. However, several c-type cytochromes 38-43 and ferridoxins 44s2 have been shown to undergo rapid electron-transfer reactions at certain electrode materials. Adsorption of the protein at the electrode surface appears to be a general phenomenon encountered in these examples of rapid heterogeneous electron-transfer reactions of redox proteins. We therefore suggest that the principles outlined for the reaction of cytochrome c at the surface-modified gold electrode may be generally applicable to the heterogeneous electron-transfer reactions of redox proteins. That is to say, a large free energy of activation for the electron-transfer step may be a general feature of the redox reactions of proteins and therefore rapid electron transfer of electrodes might be expected only in those cases where rate enhancement due to adsorption of the protein at the electrode surface occurs.We may now consider the relationship between the electrode kinetics of cyto- chrome c and its physiological electron-transfer reactions, in particular, the importance of binding in the catalysis of both the homogeneous and heterogeneous reactions and the importance of the large activation free energy for electron transfer in providing selectivity. Kinetic studies have shown that cytochrome c forms complexes with its physiological redox partners prior to electron transfer. The importance of the lysine residues of cytochrome c in determining the strength of binding in these pre- cursor complexes and in controlling the observed electron-transfer kinetics has been demonstrated l3 by chemical modification studies.Specific chemical modification in conjunction with the known X-ray crystal structure, have defined the binding surface on cytochrome c which interacts with its physiological redox partners. This surface contains the " exposed haem edge " of the protein, essentially surrounded by a ring of lysine residues. It has therefore been suggested9 that binding in thisM. J . EDDOWES AND H. A . 0. HILL 339 orientation might minimize the distance between redox centres in the protein-protein complexes, allowing “ facile ” electron transfer. A model of one such precursor complex has been p r o p o ~ e d .~ ~ ~ ~ Since the crystal structures of both cytochrome c and one of its physiological partners, cytochrome c peroxidase, are known, it has been possible to examine the surfaces of the two proteins seeking likely interaction sites. That preferred incorporates hydrogen bonding between lysine residues on the cyto- chrome c and acidic residues on the peroxidase and has the two haem groups parallel. Some of these features of the physiological redox reactions may also be present in the electrode reaction of cytochrome c, as described above. We have proposed 4s that these similarities may arise because of a common requirement: binding, in a preferred orientation caused by interaction with the lysine residues around the exposed haem edge, is necessary for rapid electron transfer in both the heterogeneous and homo- geneous redox reactions of cytochrome c.The chemical principles which may provide the basis for efficient enzyme catalysis have been con~idered.’~-~’ First, binding of the substrate, which essentially effects catalysis by holding the reactants together, has been recognized as an important and perhaps the simplest way in which the rate of the catalysed reaction might be maxim- ized. Second, the elementary step might be specificially catalysed within the enzyme- substrate complex. Possible factors contributing to catalysis of the latter type have been considered for protein electron-transfer reactions : providing a pathway through the protein fabric which facilitates electron ensuring binding between reactants such that the distance over which the electron is transferred is short,9 or providing a ligand environment for the redox centre such that the reorganisation energy consequent upon a change in redox state is minimized.’* The ability to maximize the rate of reaction is not the only criterion by which to assess enzyme efficiency.For efficiency in electron-transport chains it is essential that electrons follow a certain route ; i.e. any component of the redox chain should react only with its own electron acceptor and donor, and not with other redox species. Specificity might be achieved where the free energy of activation for the electron-transport step is large, ensuring that unwanted reactions are inhibited by the kinetic barrier, but where binding between physiological redox partners effects considerable rate enhancement such that the required reactions take place at an appreciable rate.In addition, this might allow further control of reaction rate by feedback inhibition which influences the strength of binding between redox partners. The relatively large free energy of activation for electron transfer determined from the heterogeneous electron-transfer reaction of cytochrome c and the binding between redox partners observed in its physiological redox reactions are consistent with this mechanism for the achievement of specificity. It appears 58 that a prior binding step is a common feature of both the homogeneous and hetero- geneous electron-transfer reactions of a variety of electron transfer proteins.Con- sequently, it has been proposed ’’ that binding may be a prerequisite for rapid electron transfer reactions of redox proteins in general, essential for achievement of the ob- served selectivity. CONCLUSIONS The free-energy profile determined for the electron-transfer reaction of cytochrome c at a surface-modified gold electrode shows that the free energy of activation for the electron-transfer step is large and that considerable rate enhancement due to binding of the protein at the electrode surface is essential for the achievement of a rapid rate. This binding is analogous to that observed between cytochrome c and its physiological redox partners, involving the charged lysine residues of the protein, and it is proposed that formation of a tightly bound precursor complex may also be essential for achieve-340 PROTEIN ELECTRON-TRANSFER RATES ment of a rapid rate in the physiological redox reactions. The large free energy of activation for the electron-transfer step may be responsible for the specificity of cyto- chrome c towards reaction with its physiological redox partners, ensuring that elec- tron transfer may proceed at an appreciable rate only where significant rate enhance- ment due to binding occurs. P.Nicholls, Biochim. Biophys. Acta, 1974, 346, 261. L. Smith and H. Conrad, Arch. Biochem Biophys., 1956, 63, 403. K. Minnaert, Biochim. Biophys. Acra, 1961, 50, 23. T. Yonetani and G. S Hay, J . Biol Chem., 1966, 241, 700. R. E. Dickerson, T.Takano, D. Eisenberg, 0. B. Kallai, L. Samson, A. Cooper and E. Margoliash, J. Biol. Chem., 1971, 246, 151 1. J. Kraut, Biochem. SOC. Trans., 1981,9, 197. Tunnelling in Biological Systems, ed. B. Chance, D. C. DeVault, M. Frauenfelder, R. A. Marcus, J. R Schrieffer and N. Sutin (Academic Press, New York, 1979). F. R. Salemme, Annu. Rev. Biochem., 1977,46, 299. M, D. Kamen and B. Errede, Biochemistry, 1978, 17, 1015. ’ P. Nicholls and E Mochan, Biochem. J., 1971, 121, 55. lo B. L. Vallee and R. J. P. Williams, Proc. Natl Acad. Sci. USA, 1968, 59, 498. l2 F. R. Salemme, J. Kraut and M. D Kamen, J. Biol. Chem., 1973, 248, 7701. l 3 S . Takemori, K. Wada, K. Ando, M. Hosokawa, Z . Sekuzu and K. Okunuki, J. Biochem. l4 K. Wada and K. Okunuki, J . Biochem (Tokyo), 1969, 66, 249.l6 H. T. Smith, N. Staudenmayer and F. S Millet, Biochemistry, 1978, 17, 2479. l7 A. J. Ahmed, H. T. Smith, M. B. Smith and F. S Millet, Biochemistry, 1978, 17, 2479. l8 G. Pettigrew, FEBS Lett., 1978, 86, 14. l9 R. Rieder and H. R. Bosshard, FEBS Lett., 1978, 92, 223. *O S. Ferguson-Miller, D. L. Brautigan and E Margoliash, J. Biol. Chem., 1978,253, 149. 22 C . H. Kang, D. L Brautigan, N. Osheroff and E Margoliash, J. Biol. Chem., 1978,253, 6502. 23 N. Osheroff, D. L. Brautigan, and E. Margoliash, J. Biol. Chem., 1980,255, 8245. 24 S. Ferguson-Miller, D. L. Brautigan and E. Margoliash, J. Biol. Chem., 1976, 251, 1104. 25 S. Wherland and H. B. Gray, in Biological Aspects of Inorganic Chemistry, ed. A. W. Addison, W. R. Cullen, D. Dolphin and B. R. James (Wiley-Interscience, New York, 1977), pp.289-368. 26 E. Stellwagen and R. G . Shulman, J. Mol. Biol., 1973, 80, 559. 27 E. Stellwagen and R. D. Cass, J . Biol. Chem., 1975, 250, 2095. 28 W, G. Miller and M. A. Cusanovich, Biophys. Struct. Mech., 1975, 1, 97. 29 J. A, McCray and T. Kihara, Biochim. Biophys. Acta, 1979,548, 417. 30 M. G. Segal and A. G. Sykes, J . Am. Chem. Soc., 1978,100, 4585. 31 A. G. Lappin, M. G. Segal, D. C. Weatherburn and A. G. Sykes, J. Am. Chem. SOC., 1979, 101, 2297. 32 A. G. Lappin, M. G. Segal, D C. Weatherburn, R. A. Henderson and A. G. Sykes, J. Am. Chem. Soc., 1979, 101, 2302. 33 F. A. Armstrong and A. G. Sykes, J. Am. Chern. Soc , 1978, 101, 7710. 34 F. A. Armstrong, R. A. Henderson and A. G. Sykes, J . Am. Chem. Soc., 1979, 101, 6912. 35 F.A. Armstrong, R. A. Henderson and A. G. Sykes, J. Am. Chem. SOC., 1980,102, 6545. 36 T. Kono and S . Nakamura, Bull. Agric. Chem. Soc. Jpn, 1958,22, 399. 37 W, R. Heineman, B. J. Norris and J. P. Goelz, Anal. Chem., 1975, 47, 79. 38 S. R. Betso, M. H. Klapper and L. B. Anderson, J. Am. Chem. SOC., 1972, 94, 8197. 39 F. Scheller, M. Janchen, J. Lampe. H-J. Prumke, J. Blanck and E. Palecek, Biochim. Biophys. 40 M. J. Eddowes and H. A. 0. Hill, J . Am. Chem. Soc., 1979, 101, 4461. 41 P. Yeh and T. Kuwana, Chem. Lett., 1977, 1145. 42 M. J. Eddowes, H. Elzanowska and H. A. 0. Hill, Biochem. Soc. Trans. (London), 1979, 7, 43 J. Niki, T. Yagi, H. Inokuchi and J. Kimura, J. Am. Chem. Soc., 1979, 101, 3335. 44 T. Kakutani, K. Toriyama, T. Ikeda and M. Sanda, Bull. Chem. SOC. Jpn, 1980,53, 947. 45 M. J. Eddowes, H. A. 0. Hill and K. Uosaki, J . Am. Chem. SOC., 1979, 101, 7113. 46 M. J . Eddowes, H. A. 0. Hill and K, Uosaki, Bioelectrochem. Bioenerg., 1980, 7 , 527. (Tokyo), 1962, 52, 28. N. Staudenmayer, S. Ng, M. B. Smith and F. S . Millet, Biochemistry, 1977, 16, 600. R. Rieder and H. R. Bosshard, J. Biol. Chem., 1978, 253, 6045. Acta, 1976, 412, 157. 735.M . J . EDDOWES AND H . A . 0. HILL 34 1 47 C. H. A. Seiter, R. Margalit and R. A. Perreault, Biochem. Biophys. Res. Commun., 1976, 68, 48 W. J. Albery, M. J. Eddowes, H. A. 0. Hill and A. R. Hillman, J. Am. Chem. Soc., 1981,103, 49 V. G. Levich, Physicochemical Hydrodynamics (Prentice Hall, Englewood Cliffs, N.J., 1962). 50 J. Koutecky and V. G. Levich, Zh. Fiz. Khim., 1958, 32, 1565. 51 W. W. Wainis and E. T. McGuinness, J. Biol. Chem., 1962, 237, 3273. 52 F. A. Armstrong, H. A. 0. Hill and N. J. Walton, FEBS Lett., 1983, 145, 241. 53 T. L. Poulos and J. Kraut, J. Biol. Chem., 1980,255, 10 322. 54 M. I. Page and W. F. Jencks, Proc. Natl Acad. Sci. USA, 1971,68, 1867. 55 W. P. Jencks, Mol. Biol. Biochem. Biophys., 1980, 32, 3. 56 W. J. Albery and J. R. Knowles, Biochemistry, 1976, 15, 5631. 57 A. R. Fersht, in Enzymic and Non-Enzymic Catalysis, ed. P. Dunnill, A. Wiseman and N. Blakebrough (Ellis Horwood, Chichester, 1980), pp. 13-24. 58 M. J. Eddowes and H. A. 0. Hill, in Electrochemicai and Spectrochemicai Studies of Biological Redox Components, A.C. S. Ado. Chern. Ser. (Am. Chem. SOC., Washington, 1982), in press. 807. 3904.

ISSN:0301-7249    

DOI:10.1039/DC9827400331

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Faraday Discuss. Chem. SOC., 1982,74, 343-348 Activation and Reaction Volumes for Redox Reactions of Horse-heart Cytochrome c with Inorganic Reagents BY KAREL HEREMANS, MARC BORMANS, JOHAN SNAUWAERT AND HERMAN VANDERSYPEN Department of Chemistry, Katholieke Universiteit Leuven, Celest ijnenlaan 200 D, B-3030 Heverlee, Belgium Received 20th May, 1982 Negative activation volumes have generally been considered as a criterion for outer-sphere redox reactions. Marcus and Sutin have, however, emphasized the effect of the overall enthalpy and entropy changes of the reaction on the activation enthalpy and entropy. The effects are expected to be even more pronounced for the activation volumes in view of the large electrostriction accompany- ing cross-redox reactions. This is reflected in the large overall volume changes.We show that the activation volumes for outer-sphere redox reactions of cytochrome c with positive and negative inorganic complexes can be either positive or negative, depending on the overall volume change of the reaction. Studies on the effect of pressure on reactions in solution have been shown to be useful with regard to mechanisms in both organic and inorganic chemistry.l The activation volume is obtained from the following equation : A VobsS = --RT (alnklap),. (1) The observed activation volume (A Yob,S) is considered to be made up from two contri- butions, chemical bond breaking or formation (A VbondS) and solvent effects (A Vsolv*) : For reactions in which no charges appear or disappear in the transition state the first term is predominant.When charged species are involved, the second term takes over owing to the strong electrostriction effect of ions on the solvent. Outer-sphere electron-exchange reactions are an important class of reactions where no bond formation or rupture occurs. The observed activation volumes should therefore largely reflect solvent effects from the redistribution of charges. This has been observed for a number of slow exchange reactions such as Fe,,2+*3+.2 The agreement with theory is excellent. Fast exchange reactions have been measured for Fe(CN),3- v 4 - with electrochemical The results were, however, obtained under different medium conditions. Activation and reaction volumes for cross- reactions have not been reported due to the lack of methods such as temperature jump and stopped flow to study fast reactions under pressure.In this paper we present the results of our pressure studies on the cross-reaction between horse-heart cytochrome c with inorganic and organic redox couples : Cyt"' + Red =+ Cyt" + Ox (1) where Cyt"' and Cyt" stand for ferri- and ferro-cytochrome c, Red and Ox for the344 PRESSURE EFFECTS ON CYTOCHROME C REDOX REACTIONS reduced and oxidized species of Fe(CN)63-p4-, Cobhen)?+ p 3 + and ascorbic acid This study was possible by the design and construction in our laboratory of a Joule-heating temperature-jump relaxation instrument 6* and a stopped-flow instru- ment * for the study of fast reactions under pressure. W 2 N . EXPERIMENTAL Horse-heart cytochrome c (type VI) was purchased from Sigma Chemicals. Concentra- tions were determined with E = 27 600 dm3 mol-' cm-' at 550 nm for Cyt".Cyt" was obtained by treating Cyt"' with a 20 mol dm-3 excess of ascorbate. The solution was then passed over a Sephadex G25 column. Iron hexacyanides were obtained from Merck, Darmstadt, and sodium ascorbate from Fluka, Buchs. [C0(phen)~](Cl0,)~.2H~O was prepared by a described pro~edure.~ All experiments were performed at 25 "C in Tris buffer (0.05 mol dm-j) pH 7, total ionic strength 0.2 mol dm-3 (with Na2S04). Equilibrium studies were done with methods described by Brandt et aZ.'* The effect of pressure on the equilibria was studied with a Zeiss spectrophotometer equipped with a high- pressure optical Temperature-jump relaxation studies were performed in a Messanlagen instrument adapted for high-pressure studies.6*7 The experimental conditions and the treatment of the data was essentially as described by Brandt et The stopped-flow instrument has been described previously. Basically it consists of a motor-driven stopped-flow unit which is immersed in a high-pressure vessel. * Kinetic studies with stopped flow followed closely the procedures described t y McArdle et al." for the reaction with C ~ ( p h e n ) ~ ~ + . For the reaction with ascorbate we foliowed the studies of AI-Ayash and Wilson.I2 THEORY Stranks has shown that all current electron-transfer theories lend themselves easily to a theoretical calculation of activation volumes. For exchange reactions the theory predicts negative activation volumes when the reaction proceeds via an outer- sphere mechanism.A positive activation volume is expected for an inner-sphere mechanism. This is due to the fact that the rate-limiting step is the expulsion of a solvent or ligand molecule from the metal coordination sphere prior to the formation of a bridged intermediate. Volumes of activation have therefore been advocated as a criterion for distinguishing between inner- and outer-sphere mechani~rns.~. l3-I5 Marcus and Sutin l6 have, however, indicated the role of overall enthalpy and entropy changes on the activation enthalpy and entropy for cross-reactions. Here we explore the effect of the overall volume changes of the reaction on the observed activation volume. We start from the ion-pair pre-equilibrium or the diffusional m ~ d e l .' ~ - ' ~ The rate constant for an exchange or cross reaction is given by k = Kok,, (2) (3) (4) where KO is the association constant for the ion-pair equilibrium: KO = (4nNr3p/3000) exp (-w,/RT). AG,,t = --RTln(4nNr3p/3000) + w, + AGi* + AGO*. The free energy of activation for an exchange reaction (AGO = 0) is w, is the electrostatic work required to bring the reactants together, AGi* and AGO* are the free energies of activation for ligand and solvent rearrangement, respectively,K . HEREMANS, M . BORMANS, J . SNAUWAERT AND H . VANDERSYPEN 345 r is the sum of the radii of the reactants (a, + a,). Differentiating eqn (4) with respect to pressure, assuming that r is independent of pressure, gives Ps is the compressibility of the solvent. For water the last term is -1.12 cm3.,0 There is no difference between the experimentally obtained activation volume (A V*) and that obtained from theory (AV*) in contrast to the other activation parameters.16 Since we do not have the pressure dependence of the self-exchange reactions, we have to follow an a priori way for the calculation of the activation volumes of the cross-reactions.It can be shown that eqn (5) becomes AV11S = aw,/ap + AVi* + AVO* - RTP,. (5) AV12S = 6wr/6p + AVi12* + AVOl2* + (0.5 + a)AV,," - 1.12 cm3 (6) with a = AGlz0/[4(AGil2* + AGo12*]. The term a is small and will be neglected subsequently. A similar expression can be derived from the free energies of activation in terms of the self-exchange reactions (assuming a = O).16 AG12* = 0.5(AGll* + AG22* + AG1Z0) AV12* = 0.5(AVl1* + AV22* + AV12") (7) We see that AVi12* and AVOl2* in eqn (6) are the average of the activation volumes of the inner- and outer-sphere reorganisation of ligands and solvent of the exchange reactions.If we also include the electrostatic work term, we have AV12$ = SwJSp + O.5(AVl1* + AV22* + AV120) - 1.12 cm3. (8) We now consider 6wr/6p, AVi* and AVO*. We take for w, the expression for the Debye-Huckel interaction potential l7 wr = (z1z2e2)/[Ds41 + x r ) l (9) where Z is the ionic-strength parameter, which also includes D,, the dielectric con- stant of the solvent. The limits of application of this expression have been dis- Differentiating with respect to pressure gives 6wr/6p = -wr/2(1 + x r ) [ x r P s + (2 + 2r)(8lnDs/+)]- AVO = Sw,/dp - RTP,.(10) (1 1) For AVO we obtain At infinite dilution this expression reduces to an equation already obtained by Hernmes.,, If the reacting partners are assumed to be incompressible, it can be shown that AV,* = 0.2 This is a very good approximation. AVO* can be calculated from AGO* = e2/4[(1/2al + 1/2a, - l/r)(l/n2 - l/D,)]. n is the refractive index of the solvent at optical frequencies. Assuming that a,, a, and r are independent of pressure we have A V,* = e2/4{(a12 + a2'/2ala2r)[6( l/n2)/6p - S( l/D,)/dp]}. (12) Values for the change of n and D , with pressure can be found in ref. (20). from three factors: Concluding this section we see that the observed activation volume is made up AV12Z = 6w,/dp + AVO* + 0.5AV1," - 1.12 cm3. (1 3)346 PRESSURE EFFECTS ON CYTOCHROME C REDOX REACTIONS Sw,/Sp is the volume change on bringing the reactants together, AVO* is the activation volume for the solvent reorganisation and A V I 2 O is the total reaction volume.RESULTS AND DISCUSSION EQUILIBRIUM STUDIES The thermodynamic data are collected in table 1. KO and AVO were calculated with eqn (3) and (11). The following radii were used: Fe(CN),3-v4-, 0.45 nm; C0(phen),2+*~+, 0.7 nm; ascorbate2-*'-, 0.35 nm; ~ y t + ~ ' ~ * + ~ . ' , 1.66 nm. For the reaction of cytochrome with the iron hexacyanides AVI2O is large. This is primarily TABLE 1.-THERMODYNAMIC DATA FOR CYTOCHROME c REDOX REACTIONS AT pH 7, Z = 0.2 mol dm-3, 25 "C. reaction KO" A Vob K12 A V12O /dm3 mo1-I /cm3 /cm3 mol-' mol-I ~~~~~~ Cyt"' -f Fe(CN)64- 278 2.8 2.9 x 1 0 - 3 37 Cyt" + Fe(CN)b3- 118 1.8 350 - 37 Cyt"' + C~(phen)~*+ 12 - 1 .1 7.7 x 10-3 - 20 Cyt" + C~(phen),~+ 9 -1.5 130 20 Cyt"' + A*- 80 1.5 - Cyt" + A*- 37 0.6 - 6.5" - 6.5" ' Calculated using eqn (3); calculated using eqn (11); ' calculated (see text). due to the difference in partial molar volume between Fe(CN)64- (95.6 cm3) and Fe(CN),3- (137 cm3).22 The reaction volume for the reaction with ascorbate is cal- culated from the partial molar volumes of A2- and A*- using an expression proposed by H e ~ l e r . ~ ~ KINETIC RESULTS The experimental A V $ collected in table 2 are the sum of the volume change for the precursor complex formation and the activation volume of the electron transfer. From eqn (2) we have AVZ = AVO +- AVet$. (14) AVO can be calculated using eqn (1 1).We consider first the reaction of cytochrome with the ironhexacyanides. The volume changes for the ion-pair pre-equilibrium are small. They are in the range of available experimental values for ion-pair formation as collected by Hamann.20 The activation volumes calculated using eqn (1 3) are in satisfactory agreement with the experimental results considering the assumptions we have made. For the reaction of Cyt" with Co(phen):+ we use the same equation, eqn (1 l), to calculate AVO, although the charges on the reactants are now both positive. This is acceptable since eqn (3) can be obtained from a consideration of diffusion of the reacting partners. The agreement between calculated and experimental activation volumes is satisfactory. The binding of Co(phen)i+ to Cyt" was considered to be electrostatic.Including the hydrophobic stacking between the phenanthrolineK . HEREMANS, M . BORMANS, J . SNAUWAERT A N D H . VANDERSYPEN 347 ligands and the porphyrin system makes the calculated A V, value more negative. For stacking processes A V" = - 5 cm3.24 The agreement between experiment and theory is then worse. The observed activation volume ( A V = -23 cm3) for the reduction of Cyt"' with ascorbate is too negative to be explained by a negative reaction volume. Re- TABLE 2.-KINETIC DATA FOR CYTOCHROME C REDOX REACTIONS AT pH 7, I = 0.2 mOl dm-3, 25 "C reaction kI2/dm3 AVXobs AVO*' AV12tb AV12" mo1-' s-' /cm3 mol-' /cm3 /cm3 /cm3 mol-' mo1-' mol-l Cyt"' + Fe(CN)64- 1.25 x 104 13 -3.5 16.7 37 Cyt" + C~(phen),~ + 1.87 x 103 8.5 -2.25 5.1 20 Cyt" + A*- - Cyt" f Fe(CN)b3- 2.97 x lo6 -24 -3.5 -21.3 -37 Cy t"' + C ~ ( p h e n ) ~ ~ + - -11.5' -2.25 -14.5 -20 Cyt"' + A*- 3.6 x 105d -lid -4.7 - 1 6.5" - 17.5' -4.7 - 8.5 - 6.5' ' Calculated using eqn (12); * calculated using eqn (13); ' calculated; corrected for substrate ionization; calculated (see text).cently l2 it was shown that the reaction proceeds with the doubly charged ascorbate anion species. At pH 7 AVobs$ then includes the ionization of the substrate: AH- +A2- + H+ (11) (111) (1 5) Cyt"' + A2- -L Cyt" + A*-. A*- is the ascorbate radical anion. For this mechanism: AVobs$ = AVI$ + AViH-[H+/(KAH- + H+)]. An estimate of the volume change for reaction (11) can be made from the entropy change of the reaction.25 From the pK(AH-) = 11.34 (25 "C) and AHo = 38.5 kJ l2 we obtain ASo = -88 J K-l for reaction (11). AVl $ still includes the precursor complex.The agreement between the calculated and ex- perimental activation volumes is not good. This is largely due to the uncertainty in the ionization volume of the substrate. The discussion in the preceding paragraphs shows that the agreement between experiment and theory is generally good. We would certainly not argue against the theory in view of the assumptions it involves.'6 Although it would be desirable to have experimental data for the precursor complex formation, the AVO values available in the literature for ion-pair formation 2o provide a good basis for the calculation using eqn (1 1). The most important result of our studies is that both theory and experiment show that the activation volume can be either negative or positive depending on the total volume change and the direction of the reaction. This makes the activation volume in itself a doubtful criterion for an outer-sphere electron-transfer mechanism.The influence of the ionic strength on the activation parameters for redox reactions is of some interest for testing theories.'' The results presented here were all obtained at I = 0.2 mol dm-3. Prefiminary studies of the effect of ionic strength on the reduction of Cyt"' with ascorbate show a trend in AVobs$ which is consistent with theory.26 We conclude that the volume of activation of a redox reaction can only be regarded This gives AVO = -12 cm3.348 PRESSURE EFFECTS ON CYTOCHROME C REDOX REACTIONS as a good criterion for outer-sphere redox reactions when the total volume change of the reaction is taken into account.Reversing the argument we can say that given an outer-sphere mechanism the overall reaction volume can be estimated from observed activation volumes. W. J. le Noble and H. Kelm, Angew. Chem., 1980,19, 841. G. J. Hills, in The Efects of Pressure on Organisms, ed. M. A. Sleigh and A. G. Macdonald (Cambridge University Press, Cambridge, 1972), chap. 1, pp. 1-26. B. E. Conway and J. C. Curie, J. Electrochem. SOC., 1978, 125, 257. M. Sato and T. Yamada, in High Pressure Science and Technology, ed. B. Vodar and Ph. Marteau (Pergamon Press, Oxford, 1980), vol. 2, pp. 812-814. R. Vanhorebeek, Doctoral Thesis (University of Leuven, 1974).K. Heremans, J. Snauwaert and J. Rijkenberg, Reu. Sci. Instrum., 1980, 51, 806. R. Berkoff, K. Krist and H. D. Gafney, Znorg. Chem., 1980,19, 1. ’ D. R. Stranks, Pure Appl. Chem., 1974,38, 303. ’ K. Heremans, in High Pressure Chemistry, ed. H. Kelm (Reidel, Dordrecht, 1978), pp. 31 1-324. lo K. G. Brandt, P. C. Parks, G. H. Czerlinski and G. P. Hess, J. Biol. Chem., 1966,241, 4180. l1 J. V. McArdle, H. B. Gray, C. Creutz and N. Sutin, J. Am. Chem. SOC., 1974, 96, 5737. l2 A. I. Al-Ayash and M. T. Wilson, Biochem. J., 1979, 177, 641. l3 J. P. Candlin and J. Halpern, Znorg. Chem., 1965, 4, 1086. l4 R. Van Eldik, D. A. Palmer and H. Kelm, Znorg. Chim. Actu, 1978,29, 253, Is G. A. Lawrance and D. R. Stranks, Acc. Chem. Res., 1979, 12,403. l6 R. A. Marcus and N. Sutin, Inorg. Chem., 1975, 14, 213. l7 N. Sutin, in Inorganic Biochemistry, ed. G. L. Eichhorn (Elsevier, Amsterdam, 1973), vol. 2, chap. 19, pp. 611-653. N. Sutin, in Tunnelling in Biological Systems, ed. B. Chance, D. C. Devault, H. Frauenfelder, R. A. Marcus, J. R. Schieffer and N. Sutin (Academic Press, New York, 1979), pp. 201-224. l9 G. M. Brown and N. Sutin, J. Am. Chem. SOC., 1979,101, 883. ’O S. D. Hamann, in Modern Aspects of Electrochemistry, ed. B. E. Conway and J. O’M. Bockris ’l P. Hemmes, J. Phys. Chem., 1972, 76, 895. 22 L. G. Hepler, J. M. Stokes and R. H. Stokes, Trans. Furuduy Soc., 1965, 61, 20. 23 L. G. Hepler, J. Phys. Chem., 1957, 61, 1426. 24 K. Heremans, Annu. Rev. Biophys. Bioeng., 1982, 11, 1. 25 L. G. Hepler, J. Phys. Chem., 1965, 69, 565. 26 M. Bormans and K. Heremans, unpublished data. (Plenum, New York, 1974), vol. 9, chap. 2, pp. 47-158.

ISSN:0301-7249    

DOI:10.1039/DC9827400343

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Faraday Discuss. Chem. SOC., 1982, 74, 349-364 Electron and Proton Transfers in Chemical and Biological Quinone Systems BY PETER R. RICH Department of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW Receiued 2nd June, 1982 Some redox chemistry and biochemistry of p-benzoquinones are summarised. In particular, the behaviour of quinols and quinones at electrode surfaces and in solution, together with the mechanism of reduction of soluble cytochrome c by quinols, produces a model of electron transfer relevant to biological systems. In such a model the biological transfers occur by collisional reactions of the quinone and quinol with the protein donors and acceptors. After collision, the detailed reactions occur via bound enzyme intermediates but electronic mobility between the donors and acceptors is provided.by diffusional mobility of the quinone molecules. Such considerations have allowed a detailed investigation of the redox reactions of quinols with a complex biological multiprotein system, the bc complex. A probable mechanism of electron transfer into and through this complex is presented, together with some discussion of the mechanism of energy transduction which would operate in the intact biological system. p-Quinones are integral components of many types of biological electron-transfer The use of synthetic quinones offers a versatile system for the study of these quinone-related biological electron-transfer reactions since a wide range of chemically substituted p-benzoquinones of known physical properties are available.This has allowed detailed study of the relation between structural, thermodynamic and kinetic'factors in a number of chemical and biological electron-transfer processes. A well defined system with these molecules can be developed for studying even complex biological systems such as the bc complex, a multiprotein electron-transfer device which is found in many respiratory and photosynthetic electron-transfer chains and which is capable in intact systems of transducing electrical into osmotic energy. THE CHEMISTRY OF QUINONES The redox reactions of quinones involve both protons and electrons: 0 OH 0 OH Their detailed reaction pathways are complicated by the fact that the protons and electrons are involved sequentially, leading to a great number of possible routes of overall reduction or oxidation.Often the dominant route of reduction is different from the dominant route of reoxidation. Two major types of study have contributed to an understanding of these routes, electrochemical and kinetic/physical chemical.350 BIOENERGETICS OF QUINONES Electrochemistry of p-quinones in non-protic solvents is relatively straightfor- ward.4 The only species which are involved are the non-protonated Q, Q-- and Q2- and cyclic voltammograms generally produce two electrochemically reversible waves corresponding to the reactions As the proticity of the solvent is increased, however, the reactions become more Reversible waves are complicated as protonated species become dominant reactants. OH then generally not observed.An example of the electrochemistry of the species ubiquinol-2 (I) in an aqueous buffered medium at pH 7 is presented in fig. 1. It can + 0.5 0 -0.5 - 1 .o voItage/V us SHE FIG, 1.-Electrochemistry of ubiquinol-2. An unstirred solution of 30 pmol dm-3 ubiquinol-2 in an aqueous buffer of 50 mmol dm-3 sodium 2-(N-morpholino)ethane sulphonate + 2 mmol dm-3 EDTA at pH 7 was kept anaerobic with nitrogen. The working electrode was glassy carbon and sweep rate was 120 mV s-l. be seen that the reductive and oxidative waves are not reversible, often being separ- ated by hundreds of millivolts. Both waves are pH-dependent. One interpretation of these data is that the reductive wave is governed by the rate-limiting step of i H+ r d s + e Q QH+ 2 QH. Having formed QH-, rapid reduction to QHZ occurs, either electrochemically : f H + QH.QH- -- QHZP . R. RICH 35 1 or via a chemical dismutation : Q H * e Q * - + H+ QH* + Q'+Q + QH- QH- + H+ QH2 During the oxidative sweep the reaction is limited by removal of the first electron, probably via the route f H + r.d.s. + e QH2 QH- - QH- Again having formed the semiquinone, full oxidation to quinone is rapidly achieved, either electrochemically or through dismutation processes. At more extreme pH values different routes may be dominant, and this has led to a " scheme of squares " diagram as a convenient way of representing the possible routes a ~ a i l a b l e . ~ ~ ~ Kinetic studies of the reactions of quinols with acceptors in solution have also been One such set of studies has involved an examination of the interaction of quinols with the well characterised biological redox protein, cytochrome c.By using standard physical techniques two major routes of equilibration of quinol with the cytochrome could be demonstrated in aqueous buffer in the pH range 4-8 : (1) direct reduction by the anionic quinol, QH- QH2 + QH- + HS QH- + hemeFe3+ + QH. + hemeFe2+ (2) autocatalytic reduction by the anionic semiquinone, Q' QHz + Q --- + 24' + 2H+ QT + hemeFe3+ =+= Q + hemeFe2+. Which of these routes is dominant depends upon the particular quinol used and upon pH, ionic strength and reagent concentrations. A detailed study of the anionic-quinol mediated route of electron transfer from quinol to cytochrome c has been undertaken.1° When a series of substituted p- benzoquinols were used as donors to the cytochrome a relation between kinetic and thermodynamic factors could be found, provided that the physical constants of the quinone system were available.Plotting of this relation involved the following steps. (1) Experimental measurement of the observed rate constant, kobs, for quinol reduction of cytochrome. In doing this, it was necessary to ensure that the semi- quinone-mediated reaction was not contributing to any great extent. This was most easily determined at high quinol/cytochrome ratios where the log plot should indicate pseudo-first-order kinetics (for example fig. 2 illustrates such a test for trimethyl- quinol reduction of cytochrome c at pH 7); (2) Derivation of the true rate constant, kl, for the reaction of anionic quinol reduction of the cytochrome: QH- + Cyt c3+ -% QH* + Cyt c2+.To obtain this rate constant from kobs it was necessary to firstly divide the kobs by a factor of two to allow for the probable very rapid reduction of a second cytochrome c by the semiquinone product of the first reduction.1° Next a rapid protonationl deprotonation equilibrium was assumed for QH2 L QH- + H+352 BIOENERGETICS OF QUINONES final absorbance - - 5 . 0 I time/s FIG. 2.-Reduction of cytochrome c by trimethylquinol in aqueous solution. A solution of 11.6 pmol dm-3 cytochrome c in 50 mmol dm-3 sodium phosphate + 2 mmol dm-3 EDTA at pH 7.0 and 20 "C was used. 2 mmol dm-3 trimethylquinol was added at the points indicated by arrows. The top trace shows the experimental data and the bottom figure is a standard logarithmic plot of these data which indicates pseudo-first-order decay.so that at pH values reasonably below the pKA and hence QH- = KA[QH2][H+]-l (3) Calculation of the free-energy change of the rate-limiting electron-transfer step, to allow the construction of a plot of the logarithm of true rate constant, k,, as a function of the standard free-energy change of the rate-limiting step for the reaction QH- + heme3+ kl_ QH- + hemez+ where standard free-energy change is given by Ei(QH-/QH*) - Ei(heme2 + /heme3 +) and k, is derived from the experimentally measured kobs. Ei(heme2+/heme3+) was taken from literature values of 254 mV for cytochrome c l1 and of 365 mV for solubil- ised cytochrome f.12 E,(QH-/QH.) was calculated using the formula lo 2.303RT EdQH-IQH-) = EdQWQ) + 2 n ~ (PK.- PKA - %KS - lOg1&3 53 P. R. RICH where symbols refer to the reactions KA KB KS Kd QHz QH- + H+ QH- wQ2- + H+ Q H * m Q - + H + Q2- + Q ----LL 2 Q'- and pK = -log,,K. Data for such calculations have been collected in the form of a table in ref. (10). The results of such plots for a series of differently substituted quinols donating to cytochrome c and to soluble cytochrome fare presented in fig. 3. In both cases the 0 - 8 3t 0 0 * 0 ** * * * * -200 -100 0 100 200 300 400 500 AGO of RLS/mV FIG. 3.-Marcus plots of rate constant against free-energy change for quinol reduction of cytochrome c and soluble cytochrome f. Data were obtained as detailed in the text for the reaction ki QH- + heme3+ + QH* + heme2 + , The data are for soluble cytochrome f (0) or cytochrome c (*) and for a series of substituted p- quinols which have been detailed in ref.(10). Data redrawn from ref. (10) and (32). points fall on a line of slope -1/120 mV-I, in accordance with the predictions of the electron transfer theory of Marcus.I4 It is concluded therefore that there is very little steric fitting of the electron-transfer complex between the quinol and the cyto- chrome since there is very little deviation from the line dictated by the overall thermo- dynamic factors. BIOCHEMISTRY OF QUINONES Quinones occur as essential components in the membrane-bound mitochondrial, photosynthetic and bacterial electron-transport chains. The three major types are illustrated in fig. 4.They act as lipophilic electronic connectors between the large354 BIOENERGETICS OF QUINONES multiprotein donors and acceptors. In higher plant chloroplasts, for example, they allow electrical connection between photosystem I1 and the cytochrome complex (the bf complex) which will ultimatedly connect to photosystem I. In mitochondria, they connect the dehydrogenases, which take electrons from metabolites, to the cytochrome complex (in this case termed the bc, complex) which will ultimately cause the electrons to reduce molecular oxygen to water via the cytochrome oxidase muitiprotein catalyst. 0 0 0 FIG. 4.-The common biological quinones : (a) plastoquinone, (b) ubiquinone and (c) menadione. For the biological quinones n is generally between 8 and 10 for ubiquinone and menadione and is 9 for the chloroplast-located plastoquinone.Some commonly used chemical analogues are plasto- quinone-1 and ubiquinone-1 where n = 1. These have the advantage of having significant solubility in the more aqueous solvents. A question of current debate is whether the quinone molecules, which are in sig- nificant excess of the electron-transfer multiproteins, act in a “ liquid state,” i.e. act as freely mobile molecules which are able to diffuse within the lipid membrane, or whether they are in a “ solid state,” i.e. remain attached to the protein donors and acceptors. There are currently proponents of both points of view. The view of the current author is that a “liquid state” of some sort must certainly exist between the donor and acceptor multiproteins, largely based upon the bimolecular collisional nature of the known kinetics of their redox responses and upon the lack of stoichiometric associa- tion between donor and acceptor complexes.Some of the arguments upon which such a notion is based have been summarised re~ent1y.l~ Furthermore, those studies which have directly measured the kinetic response of the active quinone complements in such systems 16*17 suggest that it is the quinone which is providing the observed electronic mobility between donors and acceptors. A further aspect concerns whether the quinone molecules are actually relatively stable in the membrane and act as a “ bucket brigade ” for hydrogen atoms or whether they do physically move in the membrane between the complexes which they connect. Experiments with quinols and quinones in rather hydrophobic, water- saturated media have been performed in the hope of shedding some light on this question.ls It was found that the rates of equilibration of the biological molecules ubiquinone-10 and plastoquinol-9 were extremely slow in such media.This was interpreted as indicative of the problem of forming an appropriate reductant, such asP . R. RICH 355 QH- , through which the electron-transfer process could proceed. From this experi- ment it was suggested that within the biological membrane, in the absence of specific sites through which equilibration could occur, the quinone molecules did not react rapidly enough with each other and instead molecular mobility was favoured as the mechanism of electron and proton movement.Such a notion, coupled with the notion of opposite sides of the membrane for reduction and subsequent reoxidation of the quinone, would provide the quinone with a protonmotive function as originally envisaged by Mit~he1l.l~ An example of how the redox cycling of the quinone in mitochondria could act as such a Mitchellian proton pump is illustrated in fig. 5. In outside 2 H* f Q QHZ dehydrogenase inside n f umarate s uc c inate t 2HS +2H+ FIG. 5,-Schematic representation of the vectorial protonmotive function of ubiquinone in mito- chondrial electron transfer. The scheme represents a function for the quinone as originally envisaged by Mit~he1l.l~ In the scheme succinate dehydrogenase transfers two electrons from succinate to ubiquinone.Two protons are liberated at the dehydrogenase but two are also taken up by the reduced quinone. This quinol then diffuses to the outside of the membrane where the electrons are donated to the bcl complex and the two protons are liberated. Net proton ejection is then 1 H+/e- but there has been no net charge movement. such a scheme it may easily be seen that one proton is transferred across the membrane for each electron which passes through the system. If the donor is a hydrogen-atom donor then net proton ejection is 1 per electron transferred, i.e. H+/e- = 1, although the number of charges translocated per electron transferred, ale-, is zero. Most experimental measurements of the ratio at this site, however, give H+/e- = 2 and q/e- = 1,20*21 and hence a more complex mechanism must be operative. In order to investigate this anomaly further, a look at the detailed chemistry of the interaction of quinols with the acceptor system, the bc complex, is necessary.A WELL-DEFINED BIOLOGICAL ELECTRON-TRANSFER SYSTEM : QUINOL DONATION TO THE bc COMPLEX THE EQUATION FOR STEADY-STATE FLUX It is possible to isolate by detergent extraction from many biological electron- transfer chains which contain quinones a multiprotein complex called generically the bc complex. From chromatophores and mitochondria it is termed the bc, complex and from chloroplasts the bf complex. All contain a remarkably similar set of four redox centres: one c-type cytochrome (for c,), two b-type cytochromes and a 2Fe2S356 BIOENERGETICS OF QUINONES iron sulphur protein usually called the Rieske centre.These may be isolated as reasonably homogeneous multiprotein complexes each containing from 5 (the bf complex)22 to at least eight (the mitochondrial bc, complex)20 distinct proteins. The isolated complexes in aqueous solution are able to accept electrons from added artificial quinols and to donate them to their normal soluble protein acceptors, either cytochrome c for the mitochondrial bc, complex or plastocyanin for the chloroplast bf complex. That the complexes are still operating in a biologically relevant way seems rather likely on the basis that reactivity is still sensitive to the known specific in vivo inhibitors of their functioning. In order to investigate electron donation into the complex, a series of substituted quinols may be used as donors to the complex.Up to the solubility limits of the reagents used the overall reaction may be described by two second-order processes of kobs QH2 -l- bcox ___f bCRED + products k3 bCRED + plastocyanin,, ___+ bcox + plastocyanin,,,. If we ignore the back reactions (acceptable, since the reactions studied are quite exergonic in the isolated system), in the steady state we have rate = kobs[QH2I[bCoxI = k3[PCl[bCREDI where bcox is the oxidised complex and bCRED the reduced complex. Then or Hence, although saturation with one substrate is observable at a fixed concentration of other components, one is merely shifting the rate-limiting step to a different stage. It has not proved possible to cause the internal rate constants of the bf complex to become rate-limiting in these experiments. THE EQUATION FOR QUINOL DONATION INTO THE bc COMPLEX It is possible to measure kobs for quinol donation into the bc complex directly.This may be done in the absence of final acceptor so that addition of quinol causes reduction of the cytochrome c1 or f, and this may be monitored spectrophotometrically. A detailed study by this method of the reduction of cytochrome f of the bf complex by plastoquinol-1 has been undertaken l 3 p Z 3 and the following features were found: (i) Electron transfer from the quinol to the cytochrome f occurs as a non-saturable bi- molecular collision process up to limits of concentration of quinol above which precipitation or micelle formation would occur; (ii) below pH 6 a linear relation between observed rate of the process and [H+]-' occurred, i.e.rate E [QH2][bfox][H+]-'.P. R. RICH 357 Above pH 6, however, the reaction tended rapidly to zero order with respect to protons. At pH values above 9 the reaction rate rapidly diminished. A full experi- mental pH profile 23924 allowed a theoretical curve to be drawn through the points by simulation of a reaction dependent upon anionic quinol, QH- or Q2-, as reductants and upon a pK on the collision site at pH 6.5 such that above pH 6.5 the site becomes uncharged and hence the complex becomes inactive in electron transport. The reactions are therefore ki QH- + @bf - QH. + @bf- k2 Q2- + @bf - QL + @bf-. If we define KX bf + H+ @bf and KA and KB as before then it may easily be shown that This equation was used to simulate a theoretical pH profile using values of pKA = 10.810, pK, = 12.91°, pKx = 6.5 and k, = k2.The curve fits well at low pH but diverges from the theoretical profile at higher pH values. This divergence may arise from quinol lability at high pH, from a value of k2/kl much greater than unity, or from the need to involve extra pK values such as redox pK values of the The equation may be incorporated into the general steady-state flux equation already given so that for unit [bf] Then steady-state flux from quinol to plastocyanin through the bf complex for unit [bf] is given by the ratio An example of experimental and simulated data for the bf-catalysed reduction of plastocyanin by plastoquinol-1 is given in fig. 6. A further point of interest arises in a Marcus plot for a series of quinols acting as donors to the bf complex cytochrome f.Here it was found that a - 1/120 mV-l slope was not obeyedI3 and that in particular the more hydrophobic quinols had ano- malously high rate constants. Arrhenius plots showed that these anomalies were caused primarily by a much smaller than normal activation entropy for the electron transfer. It is concluded that the electron-transfer complex must involve significant hydrophobic binding such that the overall amount of ordering (entropy decrease) required in formation of the appropriate complex is reduced to a low value when the hydrophobic quinols are used. ELECTRON-TRANSFER PATHWAYS THROUGH THE bc COMPLEX Having established the chemistry of interaction of quinols with bc complexes in solution it becomes feasible to study the way in which the four redox components of the complex act to catalyse the overall transfer of electrons from the quinol to the358 BIOENERGETICS OF QUINONES L c 0 0 I 1 I 1 1 2 3 4 plastocyanin/p mol dm-3 FIG.6.-Electron flow through the bf complex. The bf complex was suspended to 17.5 nmol dm-3 cytochrome f in 50 mmol dmq3 sodium 2-(N-morpholino)ethane sulphonate + 2 mmol dm-3 EDTA at pH 6 and 20 "C. The enzymatic rate was measured (A) on addition of 621 pmol dm-3 plastoquinol-1 at variable plastocyanin ratios ; (B) at increasing plastoquinol-1 concentrations with 5.73 pmol dm-3 plastocyanin (a saturating concentration for 621 pmol dm-3 plastoquinol). Simula- tions (see text) used: pKx = 6.5; pK, = 10.8; pK, = 12.9; k, = k2 = 5.2 x 1O'O dm3 mol-' s-l; k3 = lo6 dm3 mol-I s-'; pH 6.acceptor. Several recent reviews have appeared which have discussed possible wiring diagram^,^^^^^ -27 particularly for the better characterised mitochondrial bc, complex. A particular point of interest lies in the relation of the circuit to the possi- bility of movement of extra protons across the membrane for each electron trans- ferred, so that H+/e- ratios in excess of 1 may be explained. Unfortunately, reduction of the mitochondrial bc, complex by ubiquinol-1, a useful reasonably hydrophilic analogue of the natural hydrophobic donor ubiquinol- 10, is far too rapid to be easily followed, even at quinol concentrations only slightly in excess of those of the bc, complex.In order to overcome this problem trimethylquinol was used as a slower donor which still retained normal inhibitor sensitivities. In this way it was possible to follow kinetically electron flow from a pulse of added quinol intoP. R. RICH 359 the complex. Some typical results for the reduction of the mitochondria1 bc, com- plex are shown in fig. 7. Addition of quinol caused a monophasic reduction of cyto- chrome c1 but only very small transient reduction of cytochrome b-560, which soon decayed to its equilibrium redox state. Two experiments with this system indicated that for each electron observed to enter the complex onto cytochrome c,, a second electron was as rapidly entering an 0.02 2 0.01 0 0 0 0 0 0 0 20 40 60 80 100 120 timels FIG. 7.-Reduction of cytochromes c1 and b-560 by trimethylquinol in solubilised bcl complex.Reaction medium was 50 mmol dm-3 sodium phosphate + 2 mmol dm-3 EDTA at pH 7 and 20 "C, containing 1 mmol dm-3 KCN and 1.1 pmol dm-3 bcl complex. The reaction was initiated by addition of 1.1 pmol dm-3 trimethylquinol. Complete scans from 580 to 540 nm were taken and appropriate wavelength pairs were than chosen to estimate the redox states of the two cytochromes. 0, c1 (551.5-540 nm); +, b-560(562-575 nm). optically invisible component: (1) a subsaturating pulse of quinol (e.g. 1 quinol molecule per 2 bc, complexes) caused a monophasic cytochrome c, reduction but only one cytochrome c1 became reduced for each two electrons added as quinol; (2) the steady-state electron flux through the complex in the presence of excess cytochrome c, when limited by rate of donation into the complex, was twice that predicted from the measured rate of cytochrome c1 reduction by a pulse of quinol.It is most likely that this second electron is entering the essentially optically invisible Rieske centre and that this centre is in rapid equilibrium with cytochrome cl. A further experiment with the system using the specific bc, complex electron- transport inhibitor antimycin A has shed further light on the internal electron-transfer mechanism. In the presence of antimycin A a pulse of quinol caused rapid reduction of only one-half of the cytochrome c,, but also caused rapid reduction of the b-560 (approximately two cytochrome b-560 molecules reduced per rapidly reduced cyto- chrome c1 in this experiment). Two possible wiring diagrams may be drawn to explain such observations [fig. 8(A) and (B)].In both of these schemes quinol interacts with the complex such that one electron enters the Rieske centre/c, region and the second electron enters the cytochromes b. The first reaction must be unable to proceed without the second in order to account for antimycin A inhibition. Antimycin A inhibits the reoxidation of the cytochromes b so that in its presence only a single turnover of the system is possible. The models differ in how, in the uninhibited state, the cytochrome b is re- oxidised. In the first scheme the electron on cytochrome b is delivered to the Rieske centre/c, region and in the second it is reoxidised by quinone.Although this latter360 BIOENERGETICS OF QUINONES scheme would be more in line with current models from several other group^,^*-^^ based upon Mitchell's Q - c y ~ l e , ~ ~ several points make the first scheme worthy of further consideration : (1) Multiple-turnover electron transport should show a phase lag in its onset since some quinone product is needed to reoxidise cytochrome b. This is not observed. (2) In the in vivo system, full reduction of the quinone pool should de- crease the electron-transport rate by causing cytochrome b reoxidation to be rate- limiting. This is also not observed. (3) The ratio of initial rates of cytochrome c1 reduction in the presence and absence of antimycin A on addition of a pulse of quinol should be 1 : 2 in a linear branched scheme [fig.8(B)] and 1 : 1 in the Q-cycle scheme [fig. 8(A)]. It is 1 : 2. The two models, however, may possibly be accommodated in an intermediate wiring system [fig. S(C)]. In such a scheme it is envisaged that the quinone produced during the concerted electron transfers to Rieske centre/c, and to cytochrome b does not have time to leave the complex fully before it acts at an oxidant for cytochrome b-560 and then as a reductant for Rieske centre/cytochrome c,. Such a scheme is somewhat analogous to the " b-cycle " scheme of Wikstrom,26 except that no unique quinone which remains permanently associated with the complex is proposed in fig. In all of the above schemes it is likely that the electron which appears on b-560 8(C). passes firstly through cytochrome b-566.25-27*30 ENERGETICS OF QUINOL-bC COMPLEX ELECTRON TRANSFER Measured midpoint potentials (uersus the standard hydrogen electrode) at pH 7 for The initial overall two-electron reaction of quinol reduction of Rieske centre/cl the relevant couples in the bc, complex are given in table 1. and of cytochrome b-566: QHz + bc, Q + b-ci + 2HS has a value of EA from table 1 of -60 mV at pH 7 per equivalent transferred, or 120 mV of energy made available for a one equivalent process.Hence given, in the steady state, a protonmotive force of 200-250 mV 19925 a 90% reduced quinone pool and 90% oxidised cytochromes c1 and b-566, then sufficient energy could become available at TABLE 1 .-STANDARD POTENTIALS AT pH 7 OF THE REDOX COMPONENTS OF THE MITOCHONDRIAL bC, COMPLEX ~~ component E,/mV reference Rieske centre 280 31 cytochrome c1 280 31 cytochrome b-566 - 30 32 cytochrome b-560 30 32 ubiquinone 65 32 this step to move an extra proton across the membrane against the protonmotive force. The energy for this might possibly be supplemented by electron transfer from b-566 to b-560 (AE; = -60 mV).In a Q cycle [fig. 8(A)] the b-560 would be re-oxidised by the quinone pool, where it is unlikely that net free energy would become available, and so for a net' transfer of one electron through the system from quinol to cytochrome enough energy would be liberated for an extra proton movement across the membrane. If the b-560 were ultimately reoxidised by the Rieske centrelc, region [fig. 8(B) andP . R . RICH 36 1 “1- ,i b-566 le2 b-560 Q t- ,e2 J FIG.8.-Three possible schemes for electron transfer through the mitochondria1 bc, complex. Scheme A represents a typical Q-cycle, scheme B a linear branched model and scheme C an inter- mediate type of operation. (C)] (AE; = -250 mV), then enough free energy could become available for a second proton movement across the membrane. Hence for the two-electron transfer from quinol to cytochrome c enough energy for two-proton transfer would be available. In both cases, therefore, a net extra 1 H+/e- is energetically feasible. Whether the coupling occurs because of the vectorial nature of the reactions (i.e. a Mitchellian ~ c h e m e ) ~ ~ * ~ ~ or whether a proton pump 2 o v 2 6 is activated by the “ liberated ” energy is not clear. POWER PRODUCTION AND EFFICIENCY DURING THE WORKING OF THE bC COMPLEX The above considerations concerning proton stoichiometries have the implicit assumption that operation occurs close to thermodynamic equilibrium at all times such that free-energy transduction from electrical to protonmotive is close to 100% efficient.Although this may be true in certain situations where the throughput capacity of the machine is large when compared with the rate of reaction required, it may be that certain situations exist which require a maximum power output rather than a maximum efficiency of operation. During liver metabolism, for example, we may imagine a situation where maximum efficiency, rather than maximum power delivery, is optimal. In other situations, for example during rapid muscular contrac- tion or during photosynthesis under light-limited condition^,^^^^^ maximum power production is clearly of interest.An example of a chemical reaction which is analysed in terms of efficiency and power production is given in fig. 9. The reaction has been calculated for a bimolecular one-electron-transfer reaction of the form k A - + B-A + B-. --k362 BIOENERGETICS OF QUINONES A generalised rate equation was derived from this second-order equation and this used to derive the rate of decay of the reaction (fig. 9). The Eh values could be derived from the general Nernst equation oxidised Eh = Eo’ + 2 - 3 ; 7 0 g ( Fig. 9(A) illustrates a rate against time plot for a pseudo-first-order reaction where donor is in great excess of acceptor, i.e. a constant power source.1 .o 8 time/s FIG. 9.-Power production and efficiency during an electron-transfer reaction. The method of simulation is discussed in the text. Initial conditions were: ratio of donor to acceptor = 5000 : 1 ; E; (donor) = 100 mV, n = 1, 50% reduced; E( (acceptor) = 150 mV, n = 1,95 % oxidised; all axes normalised. It is assumed in the reaction that the acceptor couple is initially at the “null voltage ” of the system, which is the potential beyond which no energy may be extrac- ted from a biological redox couple. In biology this is generally the ambient potential of the couple H,O/O,, which is around 800 mV at pH 7 and at normal atmospheric contents of oxygen. The efficiency of the process is defined by free energy stored in product free energy stored as + free energy lost efficiency = product as heat free energy stored as product per mole = E,(acceptor) - null voltage free energy lost as heat per mole = E,(acceptor) - E,(donor). Using these relations we may derive a plot of efficiency as a function of time [fig.9(B)]. Power output is then given as rate x efficiency and is given in fig. 9(C). A maximum occurs in this profile. When calculated in this way where rate is not proportional to free-energy difference of reactants and products, then the efficiency at maximum power output varies with several factors. In order to illustrate the generalP. R. RICH 363 principle, however, we may simplify by using an assumption of linear thermodynamic~,3~ such that flux = constant x force i.e. rate = constant x AEh.For such a linear thermodynamic system with a constant power source maximum power output occurs at 50% efficiency. Of course, the question of whether we have linear flux-force relations in such systems is highly questionable and must be ap- proached with caution.36 Nevertheless, the general point emerges that optimal power production must be paid for with decreased efficiencies. Applying this, for example, to the bc, complex reactions, it may be that the figure of 2H+/e- for electron transfer from quinol to cytochrome c represents the maximum equilibrium value for the stoichiometry. Under conditions where maximum power production is required then the overall ratio may be reduced to significantly less than this maximum. I thank the British Petroleum Company p.1.c.for financial support, Miss S. Clarke for expert technical assistance, and Drs R. Hill and D. S. Bendall for many useful dis- cussions W. W. Parson, in The Photosynthetic Bacteria, ed. R. K. Clayton (Academic Press, New York, J. Amesz, Biochim. Biophys. Acta, 1973, 301, 35. B. L. Trumpower, J. Bioenerg. Biomembr., 1981,13, 1. G. Cauquis and G. Marbach, in Proceedings of the First International Symposium on Biological Aspects ofElectrochemistry, ed. G. Milazzo, P. E. Jones and L. Rampazzo (Birkhauser Verlag, Basel, 1971), pp. 205-214. W. J. Albery, in Electrode Kinetics (Clarendon Press, Oxford, 1975), pp. 142-146. P. R. Rich and D. S. Bendall, FEBSLetf., 1979, 105, 189. G. R. Williams, Can. J. Biochem. Physiol., 1963, 41, 231. J. H. Baxendale and H. R. Hardy, Tram.Faraday SOC., 1954, 50, 808. I. Yamazaki and T. Ohnishi, Biochim. Biophys. Acfa, 1966,112,469. lo P. R. Rich and D. S. Bendall, Biochim. Biophys. Acfa, 1980, 592, 506. l1 F. C. Rodky and E. G. Ball, J. Biol. Chem., 1952, 182, 17. l2 D. S. Bendall, H. E. Davenport and R. Hill, in Methods in Enzymology, ed. A. San Pietro l3 P. R. Rich, in Functions of Quinones in Energy Conserving Systems, ed. B. L. Trumpower l4 R. A. Marcus, J. Phys. Chem., 1963, 67, 853. l6 A. Kroger and M. Klingenberg, Eur. J. Biochem., 1973,34, 358. l7 H. H. Stiel and H. T. Witt, 2. Naturforsch., Teil B, 1969, 24, 1588. l8 P. R. Rich, Biochim. Biophys. Acta, 1981, 637, 28. l9 P. Mitchell, Chemiosmotic Coupling in Oxidative and Photosynthetic Phosphorylation (Glynn 2o M. Wikstrom, K. Krab and M. Saraste, Annu. Rev. Biochem., 1981, 50, 623. 21 M. D. Brand, Biochem. SOC. Trans., 1977, 5, 1615. 22 E. Hurt and G. Hauska, Eur. J. Biochem., 1981, 117, 591. 23 P. R. Rich and D. S. Bendall, in Short Reporrs of the First European Bioenergetics Conference 24 P. R. Rich and D. S. Bendall, Biochim. Biophys. Acta, to be submitted. 25 P. Mitchell, J. Theor. Biol., 1976, 62, 327. 26 M. Wikstrom and K. Krab, Curr. Top. Bioenerg., 1980, 10, 51. 27 P. L. Dutton and R. C. Prince, in The Bacteria, ed. R. K. Clayton (Academic Press, New York, 1975), pp. 455-469. (Academic Press, New York, 1971), vol. XXIII, pp. 327-344. (Academic Press, New York), in press. P. R. Rich, FEBS Lett., 1981, 130, 173. Research, Bodmin, 1966). (Patrone Editore, Bologna, 1980), pp. 59-60. 1978), VO~. VI, pp. 523-524. B. L. Trumpower, Biochim. Biophys. Ada, 1981, 639, 129. z9 Q. S. Zhu, J. A. Berden, S. de Vries and E. C. Slater, Biochim. Biophys. Acta, 1982,680, 69. 30 A. R. Crofts, in Short Reports of the Second European Bioenergetics Conference, to be published.364 BIOENERGETICS OF QUINONES 31 J. S. Leigh and M. Erecinska, Biochim. Biophys. Acta, 1975, 387, 95. 32 M. Erecinska and D. F. Wilson, Arch. Biochem. Biophys., 1976, 174, 143. 33 R. Hill, in Photosynthetic Organelles, special issue of Plant and Cell Physiology, 1977, 47. 34 R Hill and P. R. Rich, Annual Report of the A.R. C. Research Group on Photosynthesis (University 35 L. Onsager, Phys. Rev., 1931,37, 405. 36 D. F. Wilson, Biochim. Biophys. Acta, 1980, 616, 371. of Sheffield, 1981), pp. 7-8.

ISSN:0301-7249    

DOI:10.1039/DC9827400349

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Faraday Discuss. Chem. SOC., 1982, 74, 365-375. Photogenerated Movement of Protons through Bacteriorhodopsin in Relation to the Analysis of Photoelectrical Responses BY CLAUDE GAVACH, PATRICK SETA AND ELIZABETH BIENVENUE Laboratoire AssociC au C.N.R.S. no 330, Physico-Chimie des Systkmes PolyphasCs, B.P. 5051, 34033 Montpellier Cedex, France Received 20th May, 1982 Bacteriorhodopsin, a protein-retinal complex, forms two-dimensional patches located on the membrane of a photosynthetic bacterium. These patches, called purple-membrane fragments, act as light-driven proton pumps converting the light energy into a transmembrane gradient of proton electrochemical potential. Purple-membrane fragments are adsorbed on the surface of bimolecular lipid membranes doped with a lipophilic proton carrier and separating two aqueous solutions of controlled pH.We have measured the photocurrents associated with the photogenerated proton movements through the bacteriorhodopsin. In particular the influence of the light energy, the pH of the aqueous media and the applied voltage have been investigated, the main result being a reversal in the direction of proton pumping. In order to account for these results, a model of phototransport of protons through the bacterio- rhodopsin is proposed. According to the model, the pathway for proton movement is composed of two chains of amino acids acting as proton-conducting wires which connect the aqueous solutions to sites able to exchange protons with the photoactive heart. Undoubtedly the photoactive heart contains the Schiff base of the retinal.From this transport model it appears that the degree of protonation of the donor sites which exchange protons with the photoactive heart plays a central role in the working of the pump. A kinetic expression of the stationary photocurrent is derived from the model and accounts for the experimental data. Halobacterium halubium is an organism that lives in environments containing high salt concentrations. When this halophilic bacterium is grown under illumination at low oxygen pressure it synthesizes specialized regions of plasma membrane: these are purple patches composed of a rhodopsin-like protein, bacteriorhodopsin, which has a retinal chromophore covalently bound to one of its amino acids. The purple- membrane fragments, which can be extracted from the bacteria, are formed of a two- dimensional crystalline lattice of bacteriorhodopsin which remains stable for long periods at room temperature and under drastic changes of hydration and ionic strength of the medium.Other than the chlorophyll-based systems, bacteriorhodopsin is the only pigment-protein complex known to utilize light energy for biosynthetic purposes; it acts as a light-driven psmp translocating protons from the cell interior to the external medium, thereby generating a gradient of electrochemical potential via protons across the cytoplasmic membrane. The molecular mechanism of the phototransfer of protons through the bacterio- rhodopsin is still not fully understood, although biochemical and biophysical studies have revealed that under illumination bacteriorhodopsin undergoes a photochemical cycle composed of at least five intermediates identified by distinct absorption bands in the visible spectrum-l Raman spectroscopy shows that, during this photocycle, the Schiff base formed by the covalent bonding of a lysine residue to the retinal becomes366 PROTON MOVEMENTS THROUGH BACTERIORHODOPSIN deprotonated,2 and ff uorescence measurements have revealed conformational changes in the bacteriorhodopsin during the phot~cycle.~ The phototransfer of protons through the bacteriorhodopsin molecule has been studied by measuring variations in pH in aqueous fragment suspensions 4-11 or with purple-membrane fragments in- corporated in lipid vesicles.12-16 An alternative method consists of recording the photoelectrical responses generated by various synthetic membranes containing bacteriorhodopsin ; purple-membrane fragments are fixed on the surface of either thick membranes 17-22 or planar lipid bilayer membranes (b.l.m.).23-28 In this paper we present a study of the photocurrents generated by purple-membrane fragments adsorbed on the surface of a b.1.m.doped with a lipophilic proton carrier with which the b.1.m. support becomes permeant towards protons. We also propose an analysis to account for the photoelectrical response in relation to the proton- transfer reactions which take place inside the bacteriorhodopsin under steady illumin- ation and under an applied voltage. THEORETICAL CONSIDERATIONS Transport models have already been proposed29-31 on the basis that, due to the absorption of light, only the proton of the Schiff base is translocated to an acceptor site; thereby only one proton per excited bacteriorhopsin molecule can be ejected during one photocycle.Yet experimental studies have demonstrated that at high ionic strength more than one proton is pumped during the p h o t ~ c y c l e . ~ * ~ ~ ~ ~ ~ - ~ ~ In the transport model proposed here these data are taken into consideration. The general basis remains the existence in the bacteriorhopsin molecule of a proton pathway formed by a chain of amino-acid residues between which proton-transfer reactions take place. The amino acids which are involved are not identified, despite our knowledge of the amino-acid sequence in bacteriorhodop~in.~~ We shall define several distinct parts of the pathway which have specific functions in the photo- transport of protons. Our transport model will involve the minimum number of components necessary to account for the observed facts.DESCRIPTION OF THE MODEL OF PROTON PHOTOTRANSPORT The central component of this model (fig. 1) is the photoactive heart (p.a.h.) composed of the Schiff base of the retinal and a proximate tyrosine group, which both eject protons upon light absorption. According to Kalinsky et aZ.36 the tyrosine group is deprotonated over the same timescale as the Schiff base and should catalyse its deprotonation. The physical reason why light absorption by the retinal induces this primary proton transfer is not clearly understood. Lewis 37 has suggested that the primary action of light is to produce a significant redistribution of electrons in the retinal.This charge redistribution in the chromophore induces charge-density- assisted bond rearrangement (such as proton translocation) in the protein structure which is stabilized by subsequent alterations to the retinal structure such as a 13 cis isomerisation. Whatever the physical origin, the first step in our transport model is a transfer of protons from the photoactive heart to deprotonated sites nearby. Reprotonation of the p.a.h. in the last steps of the photocycle involves proton-transfer reactions from protonated sites again in its close vicinity. The p.a.h. is not directly accessible from the aqueous media. The existence of internal proton-donor sites is postulated following the observation that the reprotonation of the Schiff base takes place even with highly dehydrated purple-membrane fragment^.^^ The existence of internal proton-acceptor sites is deduced from the following facts: (i) the release of protonsC .GAVACH, P . SETA AND E . BIENVENUE 367 Frc. 1 .-Schematic representation of the proton pathway through a purple-membrane fragment (p.m.f.) adsorbed on the surface of a bimolecular lipid membrane (b.1.m.) doped with a lipophilic proton carrier. S1, S1, S,, S, are amino-acid residues able to exchange protons; S, and S j are usually donor and acceptor sites, respectively. The photoactive heart (p.a.h.) includes the Schiff base of the retinal and presumably other amino-acid residues. The lower part of the figure shows the electrical potential profile in the p.m.f.-b.1.m. system according to the constant-field approximation.U is the applied voltage between the two aqueous phases. The potential profiles in the diffuse layers are not drawn [(a) is the cytoplasmic side and (b) the outer side of the membrane]. into the aqueous solution is slower than the deprotonation of the Schiff b a ~ e ; ~ * l ' (ii) the yield of deprotonation of the Schiff base is not affected by the absence of aqueous The mechanism of proton translocation through the bacteriorhodopsin molecule has already been discussed. Merz and Zundel 39 suggested that proton conduction takes place along a chain of hydrogen bonds between the hydroxyl groups of identified tyrosines. This process represents a particular case of the most general mechanism of proton translocation through biomembranes, i.e.along a chain of hydrogen bonds between the side groups of amino acids, such as the hydroxyl groups of serine, threonine and tyrosine and the carboxyl groups of aspartic and glutamic acids.40 On the other hand, Kayalar 41 proposed an alternative mechanism based on keto-enol tautomerism in hydrogen-bonded peptide groups. In the model proposed here we postulate that there exist only two sites, Si and S,, present near to the p.a.h. which are able to exchange protons with the p.a.h.; accord- ing to their degree of protonation these two sites can act as either donor or acceptor sites. This assumption is suggested from the results reported here which reveal a reversal of the direction of proton pumping at pH 8 and under an applied voltage.Si will be termed the donor site and S, the acceptor site. This only means that when proton pumping takes place in the normal direction, at the end of each photo- cycle, more protons have been transferred from Si to the p.a.h. than in the opposite direction and more protons have been transferred from the p.a.h. to S j than vice versa. Once Si has been deprotonated it is protonated again by proton translocation through the chain of proton-transfer reactions taking place between the 1 - - - i couples of exchanging proton sites. The protonation of the external S1 site is carried out by the368 PROTON MOVEMENTS THROUGH BACTERIORHODOPSIN heterogeneous transfer of a proton from the closest-approach plane of protons (0’) in the aqueous solution.On the other side of the p.a.h. the pathway is completed by a symmetrical translocation of protons from S j to the last site, S,. With intact cells and with purple-membrane fragments in suspension or incorporated in liposomes the ultimate step is the heterogeneous proton-transfer reaction between S, and the closest- approach plane in the second aqueous side 0”. In the case of purple-membrane fragments fixed on bilayer membranes the last step is the transport of protons through the lipid bilayer by the lipophilic carrier. KINETIC EQUATIONS AND EXPRESSION OF THE PHOTOCURRENT Because the proton uptake from the aqueous side takes place over the same time- scale as the reprotonation of the Schiff base, we assume that the chain of proton- transfer reactions forming the part of the pathway between S i and 0’ is a reversible process.This means that each proton-transfer reaction is fast whatever its direction. As a consequence of this assumption, the expression of the degree of protonation, 8, of each site may be deduced through conditions of equilibrium. In particular, the expression of the degree of protonation of S, is given by where Kl is the equilibrium constant of proton association with S1, Ayll is the electrical potential difference between S1 and 0’ and c z i is the proton concentration at 0’. The degree of protonation of S1 (the site nearest the p.a.h.) can be deduced from the degrees of protronation of all intermediate sites : where Ki is the product of the successive equilibrium constants, Ayl, the total electrical potential difference between Si and 0’, andf = F//RT. Flash photolysis has revealed that absorption by bacteriorhodopsin in the light-adapted form (BR568) gives rise to a photoreaction cycle composed of the following intermediates : BR568~K5~o-fL5503M412-fN530-Jo64Q-fBR568.Reaction (1) is the only light-induced reaction required in the cycle. The intermediate K decays thermally back to BR via the four additional intermediates L, M, N and 0. Since the Schiff base of the retinal is deprotonated in M we shall assume that proton transfer from the p.a.h. to S j (and eventually to S i ) is achieved by reaction (3) (fig. 2), and in the absence of agreement concerning the step during which reprotonation of the Schiff base takes place we shall assume that the reprotonation of the p.a.h.with protons transferred from Si or S j is accomplished by reaction (5). Authors giving spectroscopic measurements have considered that the photo- reactions are ~nbranched,~~ unidirectional and first-order. We shall keep these assumptions only for reactions in which proton transfer does not take place. The kinetics of reaction (1) BR + Rv-tK (1) vl = @&nBR&. (11) will follow the expression for the rate of a photochemical reaction 43 v1 is the rate of reaction (1), @ its quantum yield, e the absorption coefficient of BRSa, B the light energy and nsR the number of bacteriorhodopsin molecules in the formC. GAVACH, P. SETA A N D E. BIENVENUE 369 590 2 4 L550 0640 FIG. 2.-Proton-transfer reactions between the photoactive heart and the donor (S,) and acceptor (S,) sites occurring in the bacteriorhodopsin photocycle.BR568. For the other photoreactions during which protons are not exchanged between the p.a.h. and the donor or acceptor sites [reactions (2), (4) and (6)] K-+L M-+N O+BR the expressions for the reaction rates will be very simple: v2 = k2nK v6 = k8,. 0 4 = k 4 n ~ k2, k4 and k6 are the rate constants, and nK nM and no the numbers of bacteriorhodop- sin molecules in the forms K, M and 0, respectively. For reactions coupled by proton exchange between the p.a.h. and S l or S, we assume that the reaction rate will be proportional to the degree of protonation or deprotonation of the site involved : (VIII) (5b)370 PROTON MOVEMENTS THROUGH BACTERIORHODOPSIN Since the rate of proton release into the aqueous phase is much slower than the rate of deprotonation of the Schiff base,ll at least one proton-transfer reaction in the last part of the pathway (from S , to S,) is a rate-determining step, eventually taking place under only one intermediate form of the bacteriorhodopsin.In order to establish an expression for the photocurrent it is necessary to calculate the flux of electrical charges through one of the planes of the system p.m.f.-b.1.m.- aqueous solutions (p.m.f. is the purple-membrane fragment). Let us consider the geo- metical plane 0’ at the boundary between the p.m.f. and the adjoining aqueous solution. The photocurrent has two components: on one hand the capacitive current due to the variation of the diffuse charges located in the aqueous side and compensat- ing for a part of the charges born by the p.m.f., and on the other hand the electrical current due to the flux of protons through 0’. The latter has two origins: the proton transfer occurring at each photocycle between Si and the p.a.h.and on the other side the variation of the degree of protonation of all the sites which are accessible from the aqueous side. At steady state only the proton flux due to the proton-transfer reaction between the p.a.h. and Si remains. In this case the expression for the photocurrent, Istat, is reduced to the simplest form: EXPERIMENTAL Experimental details and a description of the apparatus are given in previous paper^.^^.^' The only improvements here are the use of an electromechanical shutter (Steinhel) with a rise- time of ca.1 ms and the use of a digital-storage oscilloscope (Tectronix 468) with which the responses obtained at a given applied voltage were averaged 32 times. The pH values of the aqueous solutions were measured before and after illumination using a combined micro-glass electrode. RESULTS AND DISCUSSION Fig. 3 shows the variations with time of the photocurrent recorded when both aqueous solutions were at pH 8 and for applied voltages +0.06 and -0.06 V. In both cases the current first increases to a peak value and afterwards decreases to a steady-state value. At the beginning of illumination the photocurrent is always positive (corresponding to a flow of positive electrical charges from the p.m.f. to the b.1.m.).This means that in the first instant of the illumination, the sum of the dis- placement currents of the phototransferred protons inside the bacteriorhodopsin molecule is positive. At the steady state, when the photocurrent is only due to the proton transfer between Si and the p.a.h. at pH 8, this photocurrent is positive when the applied voltage is positive and negative when the applied voltage is negative (fig. 4). When the pH is 6 in both aqueous solutions (fig. 5), the stationary photocurrent remains positive when the applied voltage takes negative values. In fig. 4 and 5 the plotted points correspond to measured values of the stationary current. Note that at both pH 8 and pH 6, the steady-state photocurrent varies linearly with applied voltage. For a given value of the applied voltage the variation of the stationary photocurrent with light energy follows a hyperbolic function (see insets). Combining eqn (VII), (IX) and (X) one obtains the following expression for the steady-state photocurrent : where the bars refer to the steady state.C. GAVACH, P .SETA AND E . BIENVENUE 37 1 I * time FIG, 3.--Recording of the variation in photocurrent with time; the two aqueous solutions contain mol dm-3 NaCl (pH 8). Light energy 1.20 x W. 4 (lo-'/€) I w-' x E=120 x10-6w U l v - 2 FIG. 4.-Dlots of the stationary photocurrent as a function of applied voltage; pH 8 ; [NaCl] = lo-' mol d ~ n - ~ ; points are experimental data and curves calculated values from eqn (XIII); h = 9.0 x lo-'; a = 0.1; A = 5.5 x lo-''; p = 5.6 x 10"'; o = 1.0.372 PROTON MOVEMENTS THROUGH BACTERIORHODOPSIN FIG. mol I / 0 2 (ro-9q)W I -5 0 5 u/ ld2V 5.-Plots of the stationary photocurrent as a function of applied voltage; pH 6; [NaCI] = lod2 dm-3; points are experimental data and curves calculated values from eqn (XIII); h = 9.0 x a = 0.06; 1 = 7.2 x lo-''; j~ = 5.6 x lo-"; cr = 1.3.Introducing into this relation the expressions for AN and f i L given by eqn (A8) and (A9) (see Appendix), one obtains: When the coefficients A , B, vL, vN and ei are constant, eqn (XII) indicates a linear vari- ation of 1/1,,,, as a function of I/& with a positive extrapolated value of l/Zstat. Note that, for a given voltage, the variation of l/Zstat as a function of 1/Q follows this law of variation (see insets of fig.4 and 5). The same law of variation has been found experimentally by Bamberg et aZ.25 at pH 7 under short-circuit conditions. This result indicates that the coefficients A , B, vL, vN and ei have constant values. Thus outside the saturation range, higher values of light energy only increase the number of intermediates, f i L and tiM, [see eqn (A8) and (A9)] and do not influence the degrees of protonation of the donor and acceptor sites. This conclusion is in accordance with our assumption concerning Bi [cf. eqn (I)]. In order to establish the mathematical expression of Istat as a function of the applied voltage, U, eqn ( 1 ) is introduced into eqn (XII) with the assumption that A9 is a fraction of the applied voltage. One obtains the general expression with nFk,,K,c;: p=- nFk3b A = A VN AVL h = B/A, a = Ki&: and a = - Aqi/U.U is taken with reference to the side without bacteriorhodopsin. (XIII)C . GAVACH, P . SETA AND E. BIENVENUE 373 At pH 8 and pH 6 the plots of I,,,, as a function of Ucalculated using eqn (XIII) and the numerical values of the parameters, which have been determined as indicated in the Appendix, fit the experimental data satisfactorily (fig. 4 and 5). The agreement is excellent at pH 8. Note that the value of the parameter cc is ca. 0.1, This result suggests that Si (and consequently the p.a.h.) are located very close to the plane 0‘ (fig. 1). This fact has been confirmed by structural data 44 obtained from neutron-diffusion spectra. The numerical values of the product 0 = K&+ are, respectively, 1.0 at pH 8 and 1.3 at pH 6,O’ being the plane at the boundary between the aqueous solution and the p.m.f.This small change is in accordance with the results of Lozier for the small change in the decay of 0640 between pH 6 and 9. cgk is related to the proton con- centration in the bulk aqueous solution, c,?, by means of a classical Boltzmann’s- law expression : cg+ = cH? exp( - Fy//RT) (XIV) ly being the diffuse-layer potential, which to a first approximation can be expressed using the Gouy-Chapman theory as a function of the total charge q born by the bacteriorhodopsin molecule 2RT F y/ = -argsinh In this equation E is the dielectric constant of the solvent and c the totai concentration of the 1-1 electrolyte present in the solution. When the proton concentration in the bulk increases from lo-* to mol dm-3, the charge born by the bacteriorhodopsin will increase because of the protonation of several amino-acid residues; therefore the potential ly will increase and the ratio c$’+/c$+ will decrease. From the calculated values of K&’+ at pH 8 and pH 6 it is possible to estimate the variation of the diffuse- layer potential when the pH is increased from 6 to 8.We have v/ (PH 6) - y/ (PH 8) = 0.110 V with a value of mol dm-3 for the ionic strength. This value of A y seems reason- able, taking into account a charge density on the 0’ plane of ca. 5pC cm-2: this allows us to calculate an absolute value of ly of ca. 300 mV. This result reveals the role of the variation in surface potential due to the change in pH which accompanies the proton-pumping mechanism.In conclusion, the proposed model for the phototransport of protons through bacteriorhodopsin successfully accounts for the experimental data. In this model the rate constants of the intramolecular proton-transfer reactions are considered to be independent of the electric field. This behaviour is completely different from the gating processes observed in excitable membranes in which proteins form ionic channels. This fact can be explained by the very rigid structure of the bacterio- rhodopsin molecule. APPENDIX DERIVATION OF EQN ( X I I ) AND (XIII) At the steady state the variation with time of the number of each photointermediate, which can be expressed using equations of the set of eqn (11)-(IX), is zero.We have374 PROTON MOVEMENTS THROUGH BACTERIORHODOPSIN The sum of the numbers of each photointermediate is equal to the total number of bacterio- rhodopsin molecules present in the light-adapted form, n, n = ZBR i- i i K 4- fiL i- fiM + EN + fio. (A7) The resolution of the set of eqn (Al)-(A7) gives expressions for the numbers of each photo- intermediate at the steady state. In particular, the expressions of iiL and fiN are with DETERMINATION OF THE NUMERICAL VALUES OF THE PARAMETERS I N EQN ( X I I I ) For a given light energy eqq (XIII) can be rewritten in the form d’ exp (fa U) - p’ 1 + (T exp ( f a U ) Istat = d d II and p‘ = - with A’ = - b + h 8 t - h ’* Initially a curve Istat = f (U) for a given pH and a given light energy is considered; an arbitrary value of a between 0 and 1 is chosen.Considering three experimental points (for instance U = 0, U = 0.08 and U = -0.08 V) it is possible to calculate a first set of values of the parameters d‘, p‘ and o. Introducing these numerical values into eqn (A14) it is possible to check if they fit the other experimental points. If not, another value of o! is chosen which will in turn give other values of A’, p’ and-o. The calculations are carried on until a fit is obtained. For a given pH and a given electrical potential, eqn (XIII) can be rewritten S M I s t a t = - & + h in the form (A15) which givesC. GAVACH, P. SETA A N D E. BIENVENUE 375 The values of M and h are deduced directly from the straight line showing the variations of l/lStat with l / 8 (cf. insets of fig.4 and 5). The parameters h and p keep the same values for pH 6 and pH 8; 1 and Q are fitted at each pH, but their ratio is kept constant, because il and o are both proportional to &’+. The latter value is a function of the charge born by the bacteriorhodopsin [eqn (XIV) and (XV)], and thus depends on the pH. We are indebted to the C.N.R.S. for a grant. R. H. Lozier and W. Niederberger, Fed. Proc. , Fed. Am. SOC. Exp. Bwl., 1977,36, 1805. A. Lewis, J. Spoonhower, R. A. Bogomolni, R. H. Lozier and W. Stoeckenius, Proc. Natl. Acad. Sci. USA, 1974,71,4462. R. A. Bogomolni, L. Stubbs and J. Lanyi, Biochemistry, 1978, 17, 1037. D. Oesterhelt and W. Stoeckenius, Proc. Natl Acad. Sci. USA, 1973,70, 2853. A. Danon and S. R. Caplan, Biochim. Biophys. Acta, 1976, 423, 133.E. P. Bakker, H Rottenberg and S . R. Caplan, Biochim. Biophys. Acta, 1976, 440, 557. 440,68. M. Renard and M. Delmelle, Biophys. J., 1980,32, 993. R. H. Lozier, W. Niederberger, R. A. Bogomolni, S. B. Hwang and W. Stoeckenius, Biochim. Biophys. Acta, 1976, 440, 545. lo M. Eisenbach, H. Garty, E. P. Bakker, G. Kemperer, H. Rottenberg and S . R. Caplan, Bio- chemistry, 1978, 17, 4691. l1 R. Govindjee, T. G. Ebrey and A. R. Crofts, Biophys. J., 1980, 30, 231. l2 E. Racker and W. Stoeckenius, J. Biol. Chem., 1974,249, 662. l3 M. Eisenbach, E. P. Bakker, R. Korenstein and S. R Caplan, FEBS Lett., 1976, 71, 228. l4 M. Hape and P. Overath, Biochem. Biophys. Res. Commun., 1976,72, 1504. l5 M. Hape, R. M. Taether, P. Overath, A. Knobling and D. Oesterhelt, Biochim.Biophys. Acta, l6 S. B. Hwang and W. Stoeckenius, J. Membr. Biol., 1977, 33, 325. l7 L. A. Drachev, V. N. Frolov, A. D. Kaulen, A. Liberman, S. A. Ostrumov, U. Plakunova, Is L. A. Drachev, A. D. Kaulen, S. A. Ostrumov and V. P. Skulachev, FEBS (Fed. Eur. Biochem. l9 L. A. Drachev, A. D. Kaulen and V. P. Skulachev, FEBS Lett., 1978, 82, 161. 2o M. C. Blok, K. I. Hellingwerf and K. Van Dam, FEBS Lett., 1977, 76, 45. 21 M. C. Blok and K. Van Dam, Biochim. Biophys. Acta, 1978, 507, 48 22 J. I. Korenbrot and S. B. Hwang, J. Gen. Physiol., 1980, 76, 649. 23 Z. Dancshazy and B. Karvaly, FEBS Lett., 1976, 72, 136. 24 T. R. Hermann and G. W. Rayfield, Biophys. J.. 1978,21, 111. 25 E. Bamberg, H. J. Apell, N. Demcher, W. Sperling, H. Stieve and P. Lauger, Biophys. Struct. 26 P. Seta, B. d’Epenoux and C. Gavach, Bioelectrochernistry and Bioenergetics, 1980, 7, 539. 27 P. Set% P. Ormos, B. d’Epenoux and C. Gavach, Biochim. Biophys. Acta, 1980,591, 37. 28 A. Fahr, P. Lauger and E. Bamberg, J. Membr. Biol.? 1981, 60, 51. 29 W. Stoeckenius, R. H. Lozier and R. A. Bogomolni, Biochim. Biophys. Acta, 1979,505, 215. 30 I. A. Kozlov and V. P. Skulachev, Biochim. Biophys. Acta, 1977, 463, 29. 31 P. Lauger, Biochim Biophys. Acta, 1979, 552, 143. 32 D. Ort and W. Parson, Biophys. J., 1979, 25, 341. 33 R. A, Bogomolni, R. A. Backer, R. H. Lozier and W. Stoeckenius, Biochemistry, 1980, 19, 34 D. Kuschmitz and B. Hess, Biochemistry, 1981, 20, 5950. 35 Y. A. Ovchinnikov, N. G. Abdulaev, M. Y. Feigina, A. V. Kiselev and N. A. Lobanov, FEBS 36 0. Kalisky, M. Ottolenghi, B. Honig and R. Korenstein, Biochemistry, 1981, 20, 649. 37 A. Lewis, Proc. Natl. Acad. Sci. USA, 1978, 75, 549. 38 R. Korenstein and B. Hess, Nature (London), 1977, 270, 284. 39 H. Merz and G. Zundel, Biochem. Biophys. Res. Commun., 1981,101, 540. 40 J. F. Nagle and H. J. Morowitz, Proc. Natl. Acad, Sci. USA, 1978, 75, 298. 41 C. Kayalar, J. Membr. Biol., 1979, 45, 37. 42 R. H. Lozier, R. A. Bogomolni and W. Stoeckenius, Biophys. J., 1975, 15, 955. 43 E. L. Simmons, J. Chem. Phys., 1977,66, 1413; J. Phys. Chem., 1971,75, 588. 44 D. M. Engelman and G. Zaccai, Proc. Natl. Acad. Sci. USA, 1980,77, 5894. ’ R. A. Bogomolni, R. A. Baker, R. H. Lozier and W. Stoeckenius, Biochim. Biophys. Acta, 1976, 1977,465, 415. A. Y. Semenou and V. P. Skulachev, J. Biol. Chem., 1976,251,7059. SOC.) Proc., 1974, 39, 43. Mech., 1979, 5 , 277. 2152. Lett., 1979, 100, 219.

ISSN:0301-7249    

DOI:10.1039/DC9827400365

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Faraday Discuss. Chem. Soc., 1982, 74, 377-388 Proton-coupled Energy Transducti on by Biological Membranes Principles, Pathways and Praxis BY DOUGLAS B. KELL AND G. DUNCAN HITCHENS Department of Botany and Microbiology, University College of Wales, Aberystwyth, Dyfed SY23 3DA Received 7th April, 1982 A brief outline of certain features of the chemiosmotic hypothesis of the mechanism of free-energy transfer between the reactions of electron transport and adenosine triphosphate synthesis catalysed by biological membranes is given. Pulses of electron transport induced by the addition of small quantities of oxygen to suspensions of the bacterium Paracoccus denitrifcans lead to vectorial H+ movements into the aqueous phase external to the organisms, where they may be detected with a glass pH electrode.The stoichiometry of the number of protons translocated into the bulk phase external to the organisms, per oxygen atom reduced, is essentially unchanged when the amount of oxygen reduced is varied, in a manner inconsistent with the predictions of the chemiosmotic-coupling hypothesis. These and other observations lead to the view that the energy-coupling proton-transfer processes utilised in reactions such as electron-transport phosphorylation are confined to the membrane phase. Mechanisms which most easily account for this are discussed. It is now well established that localised and specific proton-transfer processes play an important role in enzymatic catalysis.1-8 In addition it has become clear that free- energy transfer in biological membrane systems is often effected by means of a current of " energised " protons. The classical example of this is electron-transport phos- phorylation, in which the passage of pairs of electrons down a thermochemical gradient (e.g.from reduced nicotinamide adenine dinucleotide, NADH, to dioxygen; AGO' = -210 kJ mol-') is coupled to the otherwise endergonic synthesis of adenosine triophosphate, ATP, from adenosine diphosphate, ADP, and inorganic phosphate (AGO' = +31 kJ m ~ l - l ) . ~ The two sets of reactions are catalysed by spatially separ- ate enzyme complexes embedded in a contiguous, fluid mosaic lo lipoprotein " coup- ling " membrane of molecular thickness, and the problem with which we are con- cerned is " how is the free energy released by electron transport coupled to the syn- thesis of ATP?".Current thinking l1 contends that the activity of certain components of the electron- transport chain is more or less tightly coupled to an initial proton translocation across the coupling membrane (fig. l), and that the " energised " protons so translocated (" pumped ") pass to the ATPase enzymes, so as to provide the free energy to drive the synthesis of ATP. Controversy attaches both to mechanisms by which the protons are pumped by the electron-transport chain and used by the ATPase l1 and to the path- way taken by the " energised " coupling protons in electron-transport phosphoryl- ation. In the present article we confine our questioning to the latter problem. Recognising the high mobility of protons in aqueous media l2-I6 Mit~hell,l'-~l in his chemiosmotic hypothesis, proposed that the unit of energy coupling was the entire volume that the coupling membrane surrounded, so that the coupling protons378 ext race1 lular bulk cytoplasmic membrane intracel luhr bulk RESPIRATION-DRIVEN H+ TRANSLOCATION FIG.1 .-Proton-coupled reactions in electron transport phosphorylation. The figure gives a diagrammatic representation of a bacteria! cytoplasmic membrane separating the intracellular and extracellular spaces, and containing lipoprotein electron transport and ATPase complexes. It is supposed that the general ion and proton permeability of the coupling membrane is low, and that electron transport and ATP synthesis are coupled to vectorial proton translocation. In the present case we wish to distinguish direct and localised pathways of proton transfer (a) from proton movements in the bulk aqueous phase external to the bacteria (b).We use the symbol H+ to denote protons of unspecified degrees of hydration. were in energetic equilibrium with the electrochemical potential of protons in the bulk aqueous phases that the coupling membrane served to separate. In this way he could describe their free energy in terms of macroscopic thermodynamics as a measurable “ proton-motive force,” Ap, given by the equation: Ap = A,iiH+/F = A~+v - 2.303 RT ApH/F (1) where APH+ is the electrochemical potential difference of protons between the 2 bulk phases that the membrane separates, and A~+v and ApH are, respectively, the electrical transmembrane potential and pH gradient between these p h a s e ~ .~ ~ , ~ ~ R,T and P have their usual thermodynamic meanings. According to this hypothesis, the proton- motive force equilibrates reversibly with the reaction catalysed by the membrane ATPase in the steady state. However, much experimental evidence 24-26 indicates that the energy-coupling process in electron-transport phosphorylation is much more localised than this, and following the original proposals of William~,~’-~* a number of workers have adopted the view that the “ energised ” coupling protons are retained on the surfaces of the coupling membrane,25*31’33 so that they are out of equilibrium with the bulk phase electrochemical proton potentials even in the steady state. Since one of the apparently most persuasive pieces of evidence for the chemi- osmotic interpretation of processes such as electron-transport phosphorylation is the observation of electron-transport-linked proton ejection into the bulk aqueous phase external to suspensions of respiratory bacteria,34 we decided to reinvestigate this process in the bacterium Paracoccus denitrificans.It is concluded that those H+ that are ejected into the bulk aqueous phase during electron transport are not kineticallyD. B . KELL AND G . D . HITCHENS 379 competent to drive otherwise endergonic reactions such as ATP synthesis under any circumstance studied. EXPERIMENTAL P . denitrijicans (NCIB 8944) was grown and maintained as described.35 Cells from mid-log phase cultures were washed 3 times and resuspended, at ca.3 (mg dry weight) ~ m - ~ , in a 6 cm3 reaction medium in a thermostatted vessel at 30 "C containing 150 mmol dm-3 KCl + 0.25 mmol dm-3 glycylglycine (PH 6.5). Carbonic anhydrase was added to a concentration of 80 p g ~ m - ~ . Vigorous stirring was effected 35 and potentiometric measurements were carried out 36 by means previously described. H+ movements in the aqueous phase external to the organisms were calibrated with an- aerobic HCl and KOH, and oxygen pulses were delivered as air-saturated saline, in the manner described by Scholes and Mitchell. 34 The endogenous steady-state respir- ation rate of the cell was ca. 75 (ng-atom 0) min-l (mg dry weight)-' as measured using a Clark-type oxygen e l e c t r ~ d e . ~ ~ In our hands the relationship 1 (mg dry weight of cells) = 5.59 x lo8 organisms = 2.0 mm3 internal volume was obtained [cf.ref. (25) and (38)]. Chemicals and biochemicals were obtained as previ~usly.~~ RESULTS Scholes and Mitchell 34 demonstrated that the addition of pulses of 02, as air- saturated 150 mmol dm-j KCl(235 pmol dm-3 O2 at 30 "C), to weakly buffered, anoxic suspensions of Micrococcus (now Paracoccus) denitrijicans elicited the vectorial ejection of protons into the bulk aqueous phase external to the organisms, where they could be detected with a sensitive glass electrode (see fig. 1). We have repeated such experiments, with the resulting traces shown in fig. 3. Fig. 3 shows that in the absence of compounds such as SCN- (see fig. 2) the rate of H+ ejection is rather slow ( t , z 5 s) compared with the calculated time of O2 reduction (< 1 s, see Experimental), and the apparent stoichiometry of H+ translocated per 0 atom reduced, the +H+/O ratio, is rather small, ca.2.5, compared with the expected value 34*39-42 of ca. 8. When the experiment is repeated [fig. 3(b)] in the presence of 100 mmol dm-3 KSCN, how- ever, the ejection of the protons to, and their decay from, the bulk aqueous phase are much more rapid, and the +H+/O ratio observed is increased approximately three- fold to a value of ca. 7.5 [cf. ref. (34) and (39)-(42)]. When carbonyl cyanidep-tri- fluoromethoxy phenylhydrazone, FCCP, is present [fig. 3(c) and (d)] essentially no H+ movements are observable, indicating presumably that all protons previously seen to be ejected had been translocated back across the membrane (fig.2), and were not just the products of scalar chemical reactions, since it is believed 4 3 9 4 4 that, inter aZia,45 one role of FCCP lies in accelerating greatly the passage of H+ back across the coupling membrane (fig. 2). These data are in substantial quantitative agreement with those obtained by Scholes and Mitchell.34 According to the interpretation of these data given by Scholes and Mitchell,34 it is proposed that in the absence of compounds such as SCN- a large transmembrane potential [eqn (l)] is set up by the translocation of a small fraction of the pumped H+ . It is assumed that this electrical potential causes the H+ to pass back across the membrane electrophoretically from the outer bulk aqueous phase before it can be observed, since the half-response time of our glass electrode and recording system is ca.0.5 s. Transmembrane electrophoretic cotransport, with the pumped H+, of SCN' ions (which, unlike Cl-, cross the cytoplasmic membrane of this organism fairly rapidly)46 would act to disspate this bulk-to-bulk phase membrane potential and thus380 RESPIRATION-DRIVEN H+ TRANSLOCATION 0 SCN- L- o e.t.c. - :N ' + Fccp- I out FIG. 2.--cOnventional explanation of d e role of compounds such as SCN- and FCCP in affecting the apparent stoichiometry of respiration-driven proton translocation. It is supposed that trans- membrane H+ translocation driven by the electron-transport complex (e.t.c.) sets up a large membrane potential, A y , between the two bulk phases that the membrane separates.Electrophoretic movement of SCN- in response to the Ay allows more H+ to be pumped into the bulk. FCCP is a lipophilic weak acid which can cross the membrane in both neutral and anionic forms, thus catalysing the electrogenic passage of pumped H+ back across the coupling membrane. E.- I 1 I I I 1 I I I I 1 I I time (1 min intervals) FIG. 3.-Respiration-driven H + translocation in P. denitrificans. Respiration-driven H + trans- location was measured as described in the Experimental section. All reaction media contained, in a final volume of 6 cm3 at 30 "C, 150 mmol dm-3 KCI, 0.25 mmol dm-3 glycylglycine (pH 6.5), 480 pg carbonic anhydrase and 1.05 x 1O'O cells. In addition, traces (b) and(d) contained 100 mmol dm-3 KSCN. In traces (c) and(d) 2 pmol dm-3 FCCP was also present.At the arrows, 50 mm3 air- saturated KCI (23.5 ng-atom 0) was added to the closed reaction vessel. At no time did the pH change exceed 0.05 pH units.D. B . KELL AND G . D . HITCHENS 381 120 100 80 60 40 20- 0 allow measurement, by extrapolation to the half-life of O2 reduction, of the true stoichiometry of H+ translocation, in this case ca. 7.5. The amount of oxygen added in each case (fig. 3) was the same, and it is assumed (but see Discussion) that large amounts of ATP are not produced when SCN- is absent. The foregoing explanation (fig. 2) of the role of compounds such as SCN- was challenged by the work of Archbold et aZ.,47 of Conover and Azzone 48 and of Could and Cramer.49 The latter authors, who carried out experiments with Escherichia coli in the absence of SCN-, showed that when the cell/O, ratio was made very high, i.e.when the calculated membrane potential was energetically insignificant (see later), the measured +H+/O ratio remained much lower than its limiting stoichiometric value, obtained in the presence of SCN-, of ca. 4. Further, when a second 0, pulse was added immediately following the first one the stoichiometry of H+ ejection caused by the second pulse was the same as that caused by the first. This result would not be expected, according to the conventional view, since the membrane potential should be so large after the first pulse that no H+ at all should be seen to enter the outer bulk aqueous phase in response to the second O2 pulse. We have therefore carried out experiments of a similar nature, simply by varying the size of a single O2 pulse, in P.den itr ificans . Fig. 4 is a plot of the number of H+ translocated into the bulk phase external to 1 I I 1 1 I I I 1 0 - C) O O - - 0 0 - 0 0 0 0 - - 0 0 0 0 - 1 I 1 I I I 1 I I 0 10 20 30 40 8 8 ng-atom oxygen added FIG. 4.-Effect of the size of the oxygen pulse on respiration-driven H+ translocation in P. dentri- ficans. Respiration-linked H+ movements were measured as described in the legend to fig. 3, trace (a), except that either 1.09 x 10" (0) or 3.26 x 10" (a) cells were present. The size of the oxygen pulse was varied as indicated. the organisms as a function of the size of the oxygen pulse, obtained under the same conditions as the trace in fig. 3(a). It should be stressed that, throughout the range of O2 pulses used, t+ for H+ ejection was similar to that observed, 1 s, in fig.3(a). This slow t+ cannot be attributed to a protonmotive reversal of the reaction catalysed by the membrane ATPase since this reaction is extremely slow in P . denitri~kans.~~*~l It is clear that the build-up of a membrane potential, which should in principle have stopped the increase in the number of protons translocated at an added oxygen382 RESPIRATION-DRIVEN H+ TRANSLOCATION concentration below even that in fig. 3(a) (see above), has no significant effect upon the stoichiometry of H+ ejection into the bulk aqueous phase at any cell/O, ratio examined (fig. 5). If we treat the organisms as 1 pm diameter spherical-shell capacitors, with a mem- brane capacitance of 1 pF cm-2,52 it is possible to calculate the maximum bulk-phase x oxygen atoms per cell FIG.5.-Lack ofeffect of the size of the oxygen pulse on the stoichiometry of respiration-driven H+ translocation in P. denitrificans. Measurements of respiration-driven H + translocation were carried out as described in the legend to fig. 4, except that ca. 5 x 10" cells were present. The number of cells and the size of the oxygen pulse were varied to give the oxygen/cell ratios indicated. The maximum possible membrane potential that could have been built up, AyVmax: was calculated as described in the text. transmembrane potential, Aymax, built up by electrically uncompensated H + trans- location from the formula 49 Avrnax = en/C (2) where n is the number of protons translocated across the membrane capacitance of a single bacterial cell of total capacitance C, and e is the elementary electrical charge.By varying the size of the 0, pulse and/or the number of cells, Aymax will also be varied, and may be made arbitrarily small, according to eqn (2), since, given the constancy of e and of C, Atymax depends only on the number of oxygen atoms (and hence H+ trans- located) per bacterial cell. As Atymax tends to zero the orthodox view 34 would have it that there is nothing to stop the ejection of H+ equal to the true limiting stoichio-D . B . KELL AND G . D. HITCHENS 383 rnetry, as seen when SCN- is present (fig. 31, into the bulk phase external to the organ- isms. Fig. 5 shows a plot of the H+ translocated as a function of the oxygen/cell ratio added during the pulse, in which the Atymax, calculated using eqn (2) with a value of C of 3 x F,49 was as low as 12 mV.It may be observed (fig. 5) that there is no significant increase in the -+H+/O ratio as the 02/ce11 ratio is decreased to very low levels. Whilst the difference in the absolute values of the +H+/O ratios observed in the experiments of fig. 4 and 5 is due to variations between batches of cells, this varia- tion makes no difference to the present analysis and interpretation, which is indepen- dent of the absolute values measured. Care was taken to ensure that both the cell and O2 concentrations were varied in experiments such as those shown in fig. 5, so as to ensure that imperfect mixing did not constitute a potential artefact in these measure- m e n t ~ .~ ~ Essentially similar data to those shown in fig. 4 and 5 were obtained when all K+ salts were substituted by the corresponding sodium or choline salts (data not shown), indicating that electrically compensating cation movements into the cells were not the cause of the observable H' movements. As an alternative approach to assessing the role of compounds such as SCN- in stimulating the apparent +H+ /O ratio, we chose to study, for reasons i ~ ~ t i m a t e d , ~ ' * ~ ~ the effect of the more lipophilic, membrane-permeable tetraphenylborate (TPB -) anion on respiration-driven proton translocation. At the concentrations used, this compound had no effect upon the steady-state respiration rate of these organisms (data not shown).Fig. 6 shows the effect of low concentrations of sodium tetraphenylborate on the apparent +H+/O ratio in P. denitrzlficans. There is initially an essentially linear in- I I I I 1 1 I I 6 - - 5 - - 4- - a Y ) ' - f 1 - - I I 1 I I I I 1 20 40 60 80 [Na tetraphenylborate]/pmol dm-3 FIG. 6.-Effect of sodium tetraphenylborate on the apparent stoichiometry of respiration-driven H + translocation in P. denitrificans. Measurements were carried out as described in the Experimental section, except that all potassium salts were replaced by the corresponding sodium salts. Sodium tetraphenylborate was added to the concentration indicated. The number of cells present was 1.29 x 10'' and the amount of oxygen added was either 14.1 ng-atom (a) or 42.3 ng-atom (0).A11 respiration-linked pH changes were abolished by 5 pmol dm-3 FCCP.384 RESPIRATION-DRIVEN H+ TRANSLOCATION 0- 2 % w/v Triton X-100 fraction number FIG. 7.-Lack of a Donnan potential (positive inside) across the cytoplasmic membrane of P. denitri- ficans. Cells were harvested and resuspended at 12.3 (mg dry weight) cm-3 in 0.1 mol dm-3 tris acetate, pH 7.3. 1 cm3 of this suspension was placed in the upper chamber of a flow dialysis cell as described." At time zero, 20 pmol dm-3 KS''CN (60 mCi mmol-') was added to the upper chamber and the flow started. A decrease in radioactivity corresponds to uptake by the cells. At the point indicated, the detergent Triton X-100 was added to a final concentration of 0.2% w/v to disrupt the cytoplasmic membrane. No release of KSCN is observed, indicating that no concentrative uptake of SCN- had taken place.Identical results were obtained in the absence of cells. crease in the +H+/O ratio as the TPB- concentration is increased, over at least a three-fold change in 0,-added per cell. If the mechanism by which this was occurring was by simple outward electrophoresis of intracellular TBP- (as in the conventional explanation, fig. 2), the " extra " H+ observed should be accompanied by a similar number of TBP- ions. However, assuming that there is no pre-existing Donnan potential across the bacterial cytoplasmic membrane (see fig. 7), we may calculate, from the known intracellular volume of the cells (see Experimental), the maximum number of TPB- ions that could move from the intracellular bulk phase to the extra- cellular bulk phase in response to electrogenic H+ pumping.Under the conditions used (fig. 6) this is equal to 0.46 ng-ion for each added 10 pmol dm-3 TPB-. The " extra " H+ observed under these conditions is, especially with the larger 0, pulse, greatly (>20 times) in excess of this, as the TPB- concentration is raised from zero to 50 pmol dm-3. Had there been a pre-existing Donnan potential (positive inside) across the bacterial cytoplasmic membrane, the concentration of free TPB- inside the bacteria prior to the 0, pulse would have been greater than that outside by a factor given by the Nernst equation :D. B. KELL AND G. D. HITCHENS 385 However, the experiment displayed in fig. 7 shows that, since no concentrative uptake of the permeant SCN- ion is observed in these cells, no such Donnan equili- brium exists across the cytoplasmic membrane of P .denitrijkans, the lower limit of detection under these experimental circumstances being ca. 20 Thus, since bulk-phase intracellular TPB- molecules should have been very much depleted at very low -+H+/O ratios under the conditions described if they were passing electro- phoretically from one bulk phase to another in response to the primary H+ movements, the independence on O2 concentration of the TPB- stimulation of the apparent +H+/O ratio, in addition to the evidence described, would seem to negate the view that the sole effect of ions such as SCN- and TPB- on the appearance of the true number of H+ in the bulk phase during +H+/O measurements lies in their ability to collapse a bulk-to-bulk phase membrane potential.DISCUSSION The chief question to which we wish to address ourselves in the present submission concerns the pathway taken by the protons pumped across the bacterial cytoplasmic membrane in response to a pulse of respiratory activity. In confirmation of the original findings of Scholes and Mitchell 34 it was observed (fig. 3) that the addition of a small pulse of dioxygen, as air-saturated KCI, to a suspension of anoxic P. denitri- ficans elicited the vectorial ejection of protons into the bulk aqueous phase external to the organisms. Both the rate and extent of H+ ejection were markedly stimulated by the addition of -100 mmol dm-3 KSCN, such that the stoichiometry of protons trans- located per oxygen atom reduced, the -+H+/O ratio, attained under these conditions a value approaching its accepted 3 4 9 3 9 - 4 2 limiting stoichiometry.Various authors [e.g. ref. (42), (56) and (57)] have discussed possible reasons why even this value may be an understimate of the " true " stoichiometry, but for our present purposes we wish to know what happened to the " missing " protons when this experiment was performed in the absence of KSCN. It is worth mentioning at the outset that, owing to the relatively low electrical capacitance of the coupling membrane when compared with the differential buffering capacitance of the system, for every electrically uncompensated H+ moved across the membrane into the external aqueous phase there will be a much greater increase in the transmembrane potential than in the pH gradient.18 Thus, in the experiment depicted in fig, 3(a) in the absence of KSCN, a large transmembrane potential will be set up, and this may, for instance, be rapidly used for ATP synthesis.If this membrane potential is thus dissipated, the residual pH gradient will be thermodynamically too small to drive further ATP synthesis, and thus the ejected protons will remain for a relatively long time in the external aqueous phase. A similar pattern would also be seen 58 if a leak of protons back across the coupling membrane that is not coupled to ATP synthesis is also highly non-ohmic, so that at high values of the protonmotive force the decay rate is particularly rapid [cf. e.g. ref. (59)]. However, this type of phenomenon fails to explain the slow half-life of ejection of H+ into the bulk phase [fig.3(a)], since, if this type of phenomenon alone were acting to lower the apparent +H+/O ratio in the absence of KSCN, the apparent rate of H+ ejection should be the same when KSCN is present. Under conditions in which the calculated value of Ay is very small, the extrapolated +H+/O ratio should be increased to its true stoichio- metric value of ca. 8, a phenomenon which is not observed. In any event, it has been demonstrated with phospholipid vesicles,36 including those containing the proton pump bacteriorhodopsin,60 that the rate of decay of a pH gradient across the phos- pholipid membrane is directly proportional to the magnitude of the pH gradient, even386 RESPIRATION-DRIVEN H+ TRANSLOCATION when this is initially caused by artificial means to exceed 4 pH units.36 Thus a non- ohmic leak of protons back across the coupling membrane seems an unlikely explan- ation of the present data.It is known from the work of several groups with related systems [e.g. ref. (61)-(65)] that ATP synthesis may be driven by a transmembrane field alone, in the absence of observable bulk-phase proton movements. Controversy remains as to whether this is a thermodynamic or a kinetic phenomenon,26 although it is worth noting that under such conditions ATP synthesis stops immediately upon cessation of electron trans- port.66 Hanselmann 38 showed that, upon the initiation of respiration in P. denitriJicans, ATP synthesis began following a variable lag of between 0 and 2 s.For technical rea- sons it was not possible for us to measure ATP synthesis on this timescale, nor, unfor- tunately, was an inhibitor of the ATP synthase enzyme itself found which was active in intact cells. Thus it would seem that some uncharacterised factors are operating to slow the expression of respiration-linked H+ movements in the bulk phase for times that are extremely long compared with those expected from simple diffusion alone.67 Various mechanisms 2s*68 have been proposed to account for this, but as yet none has We have proposed elsewhere 32933 that, in addition to the proton pumps which are coupled to the activity of the electron transport chain and ATPase complexes, there exist in such coupling membranes proteinaceous devices whose role is to act as elements in specific networks (protoneural networks) for energised proton transfer along the membrane surfaces. It is envisaged 32*33 that the transmembrane field acts to change .the benefit of extensive and rigorous experimental support. protoneural proteins FIG. 8.-Proposed model of energy coupling in electron-transport phosphorylation. The diagram shows an energy coupling membrane containing an electron-transport complex(e. t. c.), an ATPase and protoneural proteins whose role is, upon initiation of electron transport (b), to change their conforma- tion in a coherent fashion and effect passage of the pumped protons to the ATPase. In the resting state (a), in which no electron transport is taking place, the proton electrochemical potentials at the membrane surfaces are in equilibrium with those in the bulk.It is proposed that compounds such as SCN- and TPB- inhibit the conformational transitions of the protoneural proteins.D. B . KELL AND G . D. HITCHENS 387 their conformation in a coherent fashion between proton-conducting and non- conducting states, so that only under the latter conditions do protons pumped across the coupling membrane come into equilibrium with those in the bulk aqueous phases that the coupling membrane separates. Such a minimal proposal would serve to explain all the data presented here, as well as many others in the literature [e.g. ref. (24)-(26), (32), (33) and (68)]. The role of TPB- in increasing the apparent +H+/O ratio, in addition to a transmembrane electrophoretic moment in response to electro- genic proton transport, is viewed as an inhibition of the transition of the protoneural networks between their non-conducting and conducting states (fig.8). In summary, electron transport in P. denitrijicans is demonstrably coupled to H+ translocation across the bacterial cytoplasmic membrane. However, under no con- ditions examined did those translocated protons which could be observed as changes in pH in the bulk aqueous phase external to the organisms appear to be kinetically competent to return across the membrane, and so effect free-energy transfer. It is suggested that there are controls over current flow between local devices in membranes which cannot be understood from studies of bulk-phase phenomena. D. B. K.expresses thanks to Drs S. J. Ferguson, J. B. Jackson and H. V. Wester- hoff and to Professors J. G. Morris and R. J. P. Williams for many lively and stimul- ating discussions. This work was supported by the S.E.R.C. W. P. Jencks, Catalysis in Chemistry and Enzymology (McGraw-Hill, New York, 1969). D. M. Blow, J. J. Birktoff and B. S. Hartley, Nature (London), 1969, 221, 337. S. Doonan, C. A. Vernon and B. E. C. Banks, Prog. Biophys. Mol. Biol., 1970,20, 247. I. D. Campbell, S. Lindskog and A, I. White, J. Mol. Biol., 1975,98, 597. A. Fersht, Enzyme Structure and Mechanism (W. H. Freeman, Reading, Mass., 1977). J. R. Knowles and W. J. Albery, Acc. Chem. Res., 1977, 10, 105. C. Walsh, Enzymatic Reaction Mechanisms (W. H. Freeman, Reading, Mass., 1979). C. W. Wharton and R.Eisenthal, MolecuIar Enzymology (Blackie, Glasgow, 1981). A. L. Lehninger, Ber. Bunsenges. Phys. Chem., 1980,84,943. lo S. J. Singer and G. L. NicoIson, Science, 1972, 175, 720. l1 P. D. Boyer, B. Chance, L. Ernster, P. Mitchell, E. Racker and E. C. Slater, Annu. Reu. Biochem., l2 M. Eigen, Angew. Chem., Int. Ed. Engl., 1964,3, 1. l3 Discuss. Faraday SOC., 1965, 39. l4 R. P. Bell, The Proton in Chemistry (Chapman and Hall, London, 2nd edn, 1973). l5 Proton Transfer Reactions, ed. E. F. Caldin and V. Gold (Chapman and Hall, London, 1975). l6 Comprehensive ChemicaIKinetics, ed. C. H. Bamford and C. F. H. Tipper (Elsevier, Amsterdam, f 7 P. Mitchell, Nature (London), 1961, 191, 144. lB P. Mitchell, Biol. Rev., 1966, 41, 445. l9 P. Mitchell, Chemiosmotic Coupling and Energy Transduction (Glynn Research Ltd, Bodmin, 1977, 46, 955.1978), vol. 8. 1968). P. Mitchell, Eur. J. Biochem., 1979, 95, 1. 21 P. Mitchell, Chem. Br., 1981, 17, 14. 22 H. Rottenberg, Meth. Enzymol., 1979, 55, 547. 23 D. G. Nicholls, Bioenergetics (Academic Press, London, 1982). 24 G. F. Azzone, S. Massari and T. Pozzan, Mol. Cell. Biochem., 1977, 17, 1. 25 D. B. Kell, Biochim. Biophys. Acta, 1979, 549, 55. 26 A. Baccarini-Melandri, R. Casadio and B. A. Melandri, Curr. Top. Bioenerg., 1981, 12, 197. 27 R. J. P. Williams, J. Theor. Biol., 1961, 1, 1. 29 R. J. P. Williams, FEBSLett., 1978, 85, 9. 30 R J. P. Williams, Biochim. Biophys. Acta, 1978, 505, 1. 31 K. van Dam, A. C. H. A. Wiechmann, K. J. Hellingwerf, J. C. Arents and H. V. Westerhoff, in Proceedings of the IIth FEBS Meeting, Copenhagen, ed.P. Nicholls, J. Mraller, P. Jsrgensen and A. Moody (Pergamon Press, Oxford, 1977), vol. 45, pp. 121-132. R. J. P. Williams, J. Theor. Biol., 1962, 3, 209.388 RESPIRATION-DRIVEN H+ TRANSLOCATION 32 D. B. KeH, D. J. Clarke and J. G. Morris, FEMS Microbiol. Lett., 1981, 11, 1. 33 D. B. Kell and J. G. Morris, in Vectorial Reactions in Electron and Ion Transport in Mitochondria and Bacteria, ed. F. Palmier], E. Quagliariello: N. Siliprandi and E. C. Slater (EIsevier/North- Holland, Amsterdam, 1981), pp. 339-347. 34 P. Scholes and P. Mitchell, J. Bioenerg., 1970, 1, 309. 35 J. E G. McCarthy, S. J . Ferguson and D. B Kell, Biochem J.: 1981, 196, 311. 36 D. B. Kell and J. G. Morris, J, Biochem Biophys. Methods, 1980, 3, 143.37 M. K. Phillips and D. B. Kell, FEMS Microbiol. Lett., 1981, 11, 1 1 1 . 38 K. W. Hanselmann, Ph.D. Thesis (University of Zurich, 1974). 39 C. W. Jones, Symp. SOC. Gen. Microbiol., 1977, 27, 23. 40 H. G. Lawford, Can. J. Biochem., 1978, 56, 13. 41 A, H. Stouthamer, Trends Biochem. Sci., 1980, 5, 164. 42 P. M. Vignais, M-F. Henry, E. Sim and D. B. Kell, Curr. Top. Bioenerg., 1981, 12, 115. 43 P. Scholes and P. Mitchell, J. Bioenerg., 1970, 1, 61. 44 S. G. A. McLaughlin and J. P. Dilger, Physiol. Rev., 1980, 60, 825. 45 D. B. Kell, Trends Biochem. Sci., 1982, 7, I . 46 D. B. Kell, P. John and S. J. Ferguson, Biochem. J., 174, 257. 47 G. P. R. Archbold, C. L. Farrington, S. A. Lappin, A. M. McKay and F. H. Malpress, Biochem. J., 1979, 180, 161. 48 T. E. Conover and G. F. Azzone, in Mitochondria and Microsomes, ed. C. P. Lee, G. Schatz and G. Dallner (Addison-Wesley, New York, 1981), pp. 481-518. 49 J. M. Gould and W. A. Cramer, J. Biol. Chem., 1977, 252, 5875. S. J. Ferguson, P. John, W. J. Lloyd, G. K. Radda and F. R. Whatley, FEBS Lett., 1976, 62, 272. '' S. J. Ferguson, Biochem. Soc. Trans., 1977, 5, 582. 52 K. S. Cole, Membranes, Ions and Impulses (University of California Press, Berkeley, 1969). 53 P. Mitchell, J. Moyle and R. Mitchell, Methods Enzymol., 1979, 55, 627. 54 I. A. Skulskii, N-E. L. Saris, M. V. Savina and V. V. Glasunov, Eur. J. Biochem., 120,263. 55 D. B. Kell, S. J. Ferguson and P. John, Biochim. Biophys. Acta, 1978, 502, 11 1. 56 M. Wikstrom and K. Krab, Curr. Top. Bioenerg., 1980, 10, 51. 57 E. Heinz, H. V. Westerhoff and K. van Dam, Eur. J. Biochem., 1981,115, 107. 58 H. V. Westerhoff, personal communication. 59 M. C. Sorgato and S. J. Ferguson, Biochemistry, 1979, 18, 5737. 6o J. C. Arents, H. van Dekken, K. J. Hellingwerf and H. V. Westerhoff, Biochemistry, 1981, 20, 61 D. R. Ort, R. A. Dilley and N. E. Good, Biochim. Biophys. Acta, 1976, 449, 108. 62 D. A. Harris and A. R. Crofts, Biochim. Biophys. Acta, 1978, 502, 87. 63 H. T. Witt, Biochim. Biophys. Acta, 1979, 505, 355. 64 J. W. Davenport and R. E. McCarty, Biochinz. Biophys. Acta, 1980,589, 353. 65 C. Vinkler, M. Avron and P. D. Boyer, J. Biol. Chem., 1980,255, 2263. 66 W. S. Chow, S. W. Thorne and N. K. Boardman, in Light-transducing Membranes, ed. D. W. 67 M. Gutman, D. Huppert, E. Pines and E. Nachliel, Biochim. Biophys. Acta, 1981, 642, 15. 68 H. V. Westerhoff, A. L. M. Simonetti and K. van Dam, Biochem. J., 1981, 200, 193. 51 14. Deamer (Academic Press, New York, 1977), pp. 253-268.

ISSN:0301-7249    

DOI:10.1039/DC9827400377

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

GENERAL mscussroN Prof. L. I. Krishtalik (Academy of Sciences of the U.S.S.R., Moscow) (communi- cated): In connection with Dr. Williams' paper I would like to touch on two problems. The first one is the use of Marcus' cross-relation for the comparison of the rates of self- exchange and cross-reaction for cytochrome c and iron hexacyanides. The Marcus cross-relation formula is based on the estimate of the reorganization energy for cross- reaction as an arithmetical mean of the reorganization energies for both self-exchange reactions. This approximation works well in the case of two particles of similar radii but for ions of significantly different dimensions the situation is more complicated. Thus for cytochrome c radius 16.5 A and Fe(CN), radius 4.5 A the solvent reorganiz- ation energy E, calculated by the Marcus formula is for self-exchange reactions 5.5 and 20.0 kcal, respectively, but for cross-reaction it is 16.9 kcal, i.e.4.15 kcal higher than the arithmetical mean. Further, this calculation is valid for the model of two spheres in contact while (as was discussed by Dr. Williams) the hexacyanide ion can penetrate to some depth into the cytochrome globule. In the event that the ion is fully immersed in the protein the medium reorganization energy may be evaluated with the aid of the Kharkats formu1a.l For an ion touching the external surface of the globule such a calculation gives E, = 9.9 kcal, with immersion only 0.5 A deeper E, = 8.3 kcal and the next step again 0.5 A deeper leads to E, = 7.0 kcal. We see that at such an arrangement E , is much lower than the arithmetical mean.Some inter- mediate configurations are of course possible with E, value accidentally close to the arithmetical mean. It is clear that for reliable analysis of the problem the real geometry of the reacting complex must be taken into account. The hexacyanide penetration can affect not only the solvent reorganization but also the inner sphere reorganization, particularly the protein reorganization, and this factor is also to be allowed for. The other problem is the question on the influence of the charged lysine residues surrounding the haem edge. Their electrostatic attraction and hydrogen bonding to some anionic groups seems to play an important role for a proper binding of cyto- chrome c to the corresponding enzymes. But the same factors act in just the opposite way in the case of reaction between two identical molecules.From this point of view one can expect the relatively low total rate of self-exchange reaction in comparison to reactions with other natural (and, probably, some synthetic) reaction partners, the difference being mainly due to the difficulty of formation of the reaction complex. So I wonder if any data which can answer this question are available. Yu. I. Kharkats, Elektrokhimiya, 1976, 12, 1866. Dr. R. J. P. Williams (Oxford Uniuersity) said: We accept the points made by Prof. Krishtalik with certain reservations: (I) the radius used for cytochrome c may not be meaningful since the protein is very anisotropic; (2) the geometry of the [Fe(CN),I3- cytochrome c [Fe"] will be known with some accuracy soon from our n.m.r.data. We also need the [Fe(CN),I4- cytochrome c [Felll] geometry. They are not exactly the same of course. Given these problems we believe that any theory can only be a guide to experiment. We The charges of the lysines are very important.390 GENERAL DISCUSSION shall therefore use amino-acid substitutions to provide empirical knowledge. When this is available we shall look very carefully at the helpful points made. Mr. D. van Lith (Interuniversitair Reactor Instituut, Delft) said : Dr. Williams has mentioned in his contribution four possible mechanisms of electron transfer in cytochrome c, namely transfer through (1) the edge of the haem group, (2) the peptide bonds, (3) the aromatic residues of the amino acids and (4) transfer by a tunnelling mechanism.As he indicated, none of these mechanisms has until now been proved nor can be excluded, although in his opinion migration through the peptide bonds is improbable. I would like to point out that another possibility, namely transfer through the water molecules, cannot be excluded either. In fact experimental evidence is available which proves that water plays an essential role in the electrical properties of many proteins and other biomacromolecules.' Powell and Rosenberg demonstrated that for such different compounds as DNA, collagen, lecithin and cytochrome c the d.c. conductivity increases exponentially with the level of hydration. Cytochrome c for instance shows an increase of 5 orders of magnitude in d.c.conductivity on increasing the water content from 0 to 10%. Recently Lewis has shown that for bovine serum albumin it is the mobility which is increasing exponentially with water content and not the number of charge carriers as King and Medley had earlier s~pposed.~ In our laboratory we have observed that the water content affects considerably the migration of excess charge carriers in DNA and collagen.' We used short pulses of high-energy electrons to create excess charges and detected the resulting conductivity by measuring the microwave power absorption in DNA and collagen as a function of water con- centration. We found that to observe any radiation-induced conductivity at all, a critical amount of water (0.45 and 0.80 g of water per g dry collagen and DNA, respectively) has to be present.In our opinion this demonstrates that at least at the temperatures where we did our measurements (between -20 and - 135 "C) the water is an essential requirement for the migration of excess charges, and on the basis of different arguments we think that the hydration layer is actually the phase where the charge migration is taking place. So, my question to Dr. Williams is, should one not include transfer through the water molecules as a fifth possible mechanism of electron transfer in cytochrome c? R. Pethig, in Dielectric and Electronic Properties of Biological Materials (John Wiley and Sons, Chichester, 1979). M. R. Powell and B. Rosenberg, Bioenergetics, 1970, 1,493. T. J. Lewis and R. Toomer, J. Chem. SOC., Faraday Trans.I , 1981, 77, 2087. G. King and J. A. Medley, J. Colloid Sci., 1949, 4, 1. D. van Lith, M. P. de Haas, J. M. Warman and A. Hummel, Biopolymers, in press. Dr. R. J. P. Williams (Oxford University) said: I am aware of some of the work referred to in the question but I thought that it had been shown that the charge carriers in the hydrated biopolymers were ions not electrons. Generally inorganic chemists consider that water is a poor medium for long-range electron transfer. All the evidence we have for electron transfer between small molecules and cytochrome c shows that there are preferred sites from which the molecules transfer electrons and that there is little or no water between the molecules in these sites and the iron of the cytochrome. In conclusion we do not think that water is an important part of the electron-transfer path.Dr. S. Larsson (University of Lund) said: I would like to make some theoretical remarks concerning electron-transfer pathways. First it may be worthwhile toGENERAL DISCUSSION 391 remember that what we call pathways for the electron (or electron hole) between two metal ions are not pathways in the ordinary sense of the word. Since we are dealing with resonance transfer between metal ions, the pathway may be defined as a collec- tion of suitable orbitals on the bridge between the metal ions which contribute to the formation of two supermolecular orbitals. The main components of the latter are metal components which have the same sign in one and different signs in the other.At the activated state of Marcus theory the density of the supermolecular orbitals is equally divided between the two metal ions and the energy difference is at a minimum which we may denote A. The electron goes from one metal ion to the other in ca. a.u. ( ~ 2 0 0 cm-' z 0.03 eV) and proportional to At for A < a.u. A may be accurately calculated only for small systems and small distances between the metal ions. For larger systems qualitative differences between various types of bridges may be demonstrated in the extended Huckel approximation (fig. 1). n-bonded systems such as polyacetylene s without affecting very much the electronic density on the bridge. Roughly the rate constant is independent of A if A > 1 FIG. 1 .-See text. and dipyridine with a varying number of -C=C- groups inserted between the rings, may transfer more than 25 A, whereas t~ transfer in the same systems (between Cu+ and Cu2+) only takes place up to ca.15 A. In water the transfer distance is ca. 7 A and in open space ca. 5 8, in this approximation. In a biological system the bridge may frequently be divided into components with empty gaps in between. One may then derive the equation Oi may be called the transfer capability of component i. qi is the coupling matrix element for overlapping orbitals on either side of the gaps between components i - 1 and i. Since \Oil are usually around unity (a.u.-l) for t~ as well as n transfer for molecular groups up to 7 8, or more between the end atoms, it is mainly the gaps which392 GENERAL DISCUSSION cause the decrease of IAI.ql and qN+l are the gaps between the transition metal ions and the bridge components 1 and N, respectively. For normal bonding distances ql and qN+' are ca. 0.1 a.u. Gaps 5 A wide may cause a decrease in A by three orders of magnitude. A 5 gap is thus enough to cause non-adiabaticity. Even if the edge- to-edge distance can be brought down to 5 A in the cytochrome interaction, electron transfer is difficult unless some atom or molecular group is placed in the gap. Dr. G. R. Moore and Dr. R. J. P. Williams (Oxford University) (communicated): The present state of the n.m.r. studies of cytochrome c is defined in the paper. We can define reasonably accurately the sites of [Fe(CN),]3-/4- binding (our fig. 11) and the sites of [Fe(edta)(H,O)]- binding. These two anions have different selectivities: [Fe(CN)6]3-'4- is preferred by the lower site while [Fe(edta)(H,O)]- is preferred by the upper site. The Fe-Fe distances of both sites are >10 A indicating that some functional groups of cytochrome c form a bridge to assist electron transfer.We do not yet know what groups these are. From planned studies of different mito- chondrial cytochromes c and chemically modified cytochromes we hope to be able to comment on the effects that different groups have upon the electron-transfer process.' The precise geometries of the protein-protein complexes that form the biological electron-transfer systems are not known. Reasonable speculation has provided some models that are supported by various pieces of experimental evidence. In these systems the Fe-Fe distances are all >10 A, and even in the system with closest approach of haem groups there is an edge-to-edge separation of 8.5 A.Thus, in the real biochemical systems, amino-acid side-chains must play some role in the electron transfer. All that can be said at present is that Phe82, a residue of cytochrome c in contact with the haem group, appears to be the most likely group to participate in a bridge to another protein. G. Williams et al., FEBS Lett., 1982, 150, 293. Dr. N. Sutin (Brookhaven National Laboratory) said: The question of the distance dependence of electron-transfer rates in metalloproteins has recently been addressed by Professors H. Gray and S. Isied and their coworkers. Using ingenious synthetic techniques they have succeeded in covalently attaching a Ru(NH,),-residue to histi- dine-33 of ferricytochrome c and measuring the intramolecular electron-transfer rate from the ruthenium(I1) to the haem iron(II1).The rate constant obtained by Gray et al.' is 22 s -', while lsied et al., report a slightly higher value. If the interaction energy decreases exponentially with distance [eqn (l)] and the intramolecular electron-transfer rate constant is given by eqn (2) k,, = 1013 exp[-2p(r -ro)]exp(-AG*/RT) then a value of /3 = 0.5 A-' is calculated for (r - ro) = 15 A. Although the value of r is not known with certainty, major structural changes due to the binding of the Ru- (NH3), residue are unlikely since the histidine-33 is on the less flexible side of the cyto- chrome molecule, and the Ru(NH3), residue does not significantly alter the redox potential of the haem centre. In any event, the estimated value of p seems of a reason- able magnitude for protein mediation of the electronic interaction of the two metal centres. J.R. Winkler, D. G. Nocera, K. M. Yocom, E. Bordignon and H. B. Gray, J. Am. Chem. SOC., 1982, 104, 5798. S. S. Isied, G. Worosila and S. J. Atherton, J. Am. Chem. SOC., 1982, 104, 7659.GENERAL DISCUSSION 393 Dr. R. J. P. Williams (Oxford University) said: Unfortunately very little is yet known about the flexibility of His-33. It is on the surface of cytochrome c. Attach- ing groups to the surfaces of proteins can cause major changes of structure even when the protein fold is stable. One example is the movement of a tyrosine in carboxy- peptidase, where a distance change of 15 8, can be observed.It is then essential to check the structure of the modified protein. For this very reason we have preferred the determination of binding sites of reagents to unmodified proteins (see our paper and the discussion after Dr. Larsson's remark). The data we have define binding sites quite accurately. Prof. E. M. Kosower (TeZ Aviv University) said: Are the conformational fluc- tuations found for cytochrome c in water altered by interaction with phospholipid? Dr. R. J. P. Williams (Oxford University) said: We do not know if the conform- However, the binding of cytochrome ational fluctuations are altered by lipid binding. c to lipids is quite weak and we suspect that there would be no effect.Dr. J. F. Holzwarth (Fritz-Haber Institut, Berlin) said: We have investigated a series of electron-transfer reactions between Cyt ~ ( I I ) and very fast electron acceptors. A short summary of our results is given in table 1. TABLE 1 .-OXIDATION OF CYTOCHROME C(I1) k,,,/dm3 mol-I s-' at different ionic strengths /mol dm-3 oxidant 10- 3(Tris) lO-'(NaCl) 5 x 10-'(NaCl) IrCIi- 1.1 x 1O'O 3.7 x 1 0 9 1.4 x 109 Fe( CN): - 3.4 x 109 1.8 x 107 1 . 1 x 107 Fe(phen)(CN) 4 2.6 x 109 4.4 x lo8 3.4 x lo8 Fe(phen)z(CN) t 1.4 x 10' 8.5 x lo8 2.3 x 109 Os(bipy); + 4.6 x lo6 2.4 x 107 1.6 x lo8 pH = 7.1, tris buffer c = lo-' mol dm-'; T = 298 K; [horse-heart Cyt ~ ( I I ) ] = mol dm-3 in HzO. One result which struck us most was the impossibility of causing Os(bipy)\+ to accept an electron from Cyt ~ ( I I ) with a completely diffusion-controlled rate constant.This is in contrast to other measurements which showed clearly that O~(bipy)$+/~+ is the fastest homonuclear electron-transfer system known (activation barrier at high ionic strength approximately zero).l Our explanation for this behaviour is that the part of Cyt ~ ( I I ) which is capable of electron transfer carries a positive charge which cannot be completely shielded by high concentrations of anions. J. F. Holzwarth and H. Jurgensen, Ber. Bunsenges. Phys. Chem., 1974, 78, 526. Dr. G. R. Moore (Oxford University) (communicated) : In the paper by Moore et al. presented at this Discussion kinetic parameters published by Ohno and Cusanovich are given for, the reaction of horse-heart cytochrome c with [Fe(CN),]3-/[Fe(CN),]4-, The parameters were obtained by an analytical procedure that indicated saturation kinetics resulting from precursor complex formation.Butler et aZ.2 have shown that the procedure of Ohno and Cusanovich is incorrect and that the correct procedure394 GENERAL DISCUSSION provides no evidence of saturation kinetics. However, the studies of Stellwagen and Shulman clearly indicate that complex formation occurs prior to electron exchange, and thus the following scheme is the minimal acceptable scheme: Cyt"' + Fe4- F= (Cyt"' - Fe4-) === (Cyt" * Fe3-) F= Cyt" + Fe3-. Butler et a1.2 conclude that at Z = 0.1 mol dm-3 (NaCl) and at 25 "C, pH 7 the pre- cursor association constants are <200 dm3 mol-'. Under these conditions the rate of reduction of Cyt"' is 3.5 x lo4 dm3 mol-1 s-l and the rate of oxidation of Cyt" is 9.2 x lo6 dm3 mol-1 s-l. N.Ohno and M. A. Cusanovich, Biophys. J., 1981,36, 589. J. Butler, D. M. Davies and,A. G. Sykes, J. Inorg. Biochem., 1981, 15,41. E. Stellwagen and R. G. Shulman, J. Mof. Biol., 1973, 80, 559. Dr. N. Sutin (Brookhaven National Laboratory) said: Contrary to a common belief, two low-spin configurations are not necessarily required for rapid electron transfer between two metal complexes. A good illustration of this is provided by the Co(bipy)i/2+ exchange, for which the rate constant is in excess of lo* dm3 mo1-' s-'-the reactants in this exchange are both high spin with ( T C ) ~ ( O * ) ~ and (x)5(~*)2 configurations, respectively.The rapid exchange is a consequence of the low solvent barrier resulting from the large sizes of the reactants, and the relatively low inner- sphere barrier resulting from back-bonding (although in this case the back-bonding introduces an additional contribution to the inner-sphere barrier from intraligand bond-distance changes) and the fact that a x-electron is transferred in the exchange.' The rapid rate of this reaction should be contrasted with the slow rate of the Co- ( b i ~ y ) ~ + / ~ + exchange in which the reactants have (~C)~(O*)~ and ( T C ) ~ configurations: the exchange in the Co(bip~);+/~+ system is accompanied by an electronic rearrange- ment involving the o* electrons. Consequently the latter exchange, unlike the Co- (bipy)+3l2 + exchange, features a relatively large inner-sphere barrier.The important criterion is evidently not that both reactants be low spin, but that the nuclear configur- ation change accompanying the electron transfer be small. D. Szalda, C. Creutz, D. Mahajan and N. Sutin, to be published. Dr. B. R. Eggins (Ulster Polytechnic) said: The paper by Dr. Rich makes a basic assumption that the primary quinone species carrying out reduction of cytochromes is the hydroquinone monoanion, QH-, which may not be justified. The quinone/hydroquinone redox system is best considered using the scheme of squares referred to by the author. Q + Q" + Q2- 11 11 11 QH+ s QH' + QH- 11 11 It QHI+ + QH;+ + QH2. One needs to plot a mechanistic route through the square to give the order of electron and proton transfers.Under the author's conditions (pH < 7) hydroquinone exists mainly as the neutral molecule (pK, = 9.83). Thus the route assumed by the author would be HeeH. However, to give the full thermodynamic picture one needs to know the six E" values for the electron transfers and the six pK, values for proton transfers. A further complexity must also be considered. QH' may disproportionate by homogeneousGENERAL DISCUSSION 395 electron transfer, as predicted by Hawley and Feldberg.' This may occur in one of two ways: QH;+ + QH' = QH+ + QH2 Keg M lo2' (1) or 2QH' = QH+ + QH- Keq z 3 x lo8. (2) Both are thus thermodynamically favourable with (1) preferred. This fits our con- clusions about QH2 oxidation on platinum in acetonitrile including a 30 mV per pH unit dependence of &(ox) on concentration of added proton donors.We,, Cauquis and Parker have studied the Q/QH, system extensively in aprotic solvents. Between us we have observed all nine species in the square electrochemically and spectroscopically except for QH', which has only been observed directly by pulse radiolysis. An approximate scale of redox potentials for the benzoquinone systems in acetonitrile is as follows : QH$+/QH;+ QH;+ / Q H ~ QH'/QH - QH + /QH. Q/Q*+ Q- + / ~ 2 - I ___ I I I I -I 0 +1.8 +1.3 ca. +0.2 ca. -0.1 -0.51 -1.1 VusSCE(aq). A reversible wave for QH2/QH;+ and an irreversible wave for QH;+/QH2,+ have recently been reported by Parker in CH2C12 containing FS03H at -50 "C. Some pK, values have been measured for aqueous or mixed aqueous solvents, e.g.QH2/ QH-/Q2- ; QH'/Q'- and QH+/Q. Others may be estimated from thermochemical cycles. In acetonitrile the favoured initial step for QH2 oxidation is QH, + QH;+ + e- ; especially at pH < 7. It may be different in water. It may be different again in the special biological " soup " created by proteins and lipids. Even if the appropriate E" and pK, values for the solvent mixture are known, this is not in itself sufficient. One also needs to know the most favourable transition state. Albery5 has given a detailed analysis of the situation for a four species square, and one restricted to the cases where proton transfer is relatively fast for the full nine-species square. Considering the first electron and proton transfers from QH2 one can see that there are four possible situations: QH' == QH- The route taken will depend on the lowest transition state, to determine which one needs to know the rate constants for each electron and proton transfer.Hale and Parsons have applied the Marcus theory to the polarographic reduction of Q to QH, in aqueous buffer solution at pH 4. They conclude that the route is HeeH, contrary to Vetter who suggested HeHe for pH < 5 from polarisation'curve data but in agreement with our conclusions for the reduction of Q/H+ in acetonitrile. Neither Albery nor Parsons considered the disproportionation route. In fact, it may only be important for the oxidation of QH2 but not for the reduction of Q. In order to distinguish between the two alternative routes, eH and He, more data are needed on oxidation potentials for hydroquinones, E(QH,/QHi+), and on the pK of QH;+ species.It is difficult to reconcile the existing data as not only does Rich use oxidation potentials ' rather than the standard reduction potentials but Hale and Parsons ti use AG in kcal mol-' on a scale of AG(Q/QH,) = 0, at pH 0. However,396 GENERAL DISCUSSION they do not appear to have allowed for the fact that at pH 4, AG(Q/QH,) = -5.4 kcal mo1-I (EQ,Q,, = +0.236 V). However, from Parsons’ data the energy required to reach the transition state for both routes (i) or (iv) appears to be ca. 80-90 kJ mol-l. More and better data are needed. M. D. Hawley and S. W. Feldberg, J. Phys. Chem., 1966,70, 3459. J. Q. Chambers, in The Chemistry of the Quinoid Compounds, ed. S .Patai (Wiley, London, 1974), part 2, chap. 14; B. R. Eggins, Faraduy Discuss. Chem. SOC., 1974,56,276; B. R. Eggins and J. Q. Chambers, J . Electrochem. SOC., 1970, 117, 186. J. Bessard, G. Cauquis and D. Serve, Electrochim. Acta, 1980, 25, 1187. 0. Hammarich and V. D. Parker, Acta Chem. Scand., Ser. By 1982,36,63 and references therein. W. J. Albery and M. L. Hitchmann, Ring-Disc Electrodes (Clarendon Press, Oxford, 1971), chap. 5. J. M. Hale and R. Parsons, Trans. Faruduy SOC., 1963, 59, 1429. ’ K. J. Vetter, Electrochemical Kinetics (Academic Press, New York, 1967), p. 483. Dr. P. R. Rich (University of Cambridge) (partly communicated) : Spectrophoto- metric measurements have shown that the mechanism of reduction of cytochrome c by quinols in aqueous buffers around pH 7 appears to occur by two major pathways, the anionic quinol- and the semiquinone-mediated routes.Either of these may be made to be dominant by appropriate adjustment of experimental conditions. When the reaction is initiated by addition of excess quinol, a predominantly semiquinone- mediated process produces an autocatalytic reaction, whereas a predominantly quinol-mediated process produces a pseudo-first-order profile (see fig. 2 of my paper). All points in fig. 3 were obtained under conditions where the semiquinone mediated route was negligible. The conclusion that the anionic quinol was the active reductant was surmised from the relation between log kobs and pH which gave a slope of 1 . Assuming protonation/deprotonation reactions to be rapid, such a plot indicates that the anionic quinol, QH-, is important (cytochrome c has no redox pK in the pH range used).This conclusion was strengthened by the data of fig. 3, which show that the relative rates of reduction by a series of differently substituted quinols conform to the Marcus prediction when one uses the free-energy changes of rate-limiting electron transfers from the anionic quinol to the cytochrome. As pointed out here by Dr. Eggins, the routes of redox reaction of quinone systems at electrode surfaces are extremely varied. Most reactions contained in the scheme of squares may be observed under appropriate conditions and interpretation of data becomes more complex. I have limited my description of the electrodic reactions to only a narrow range of conditions which are relevant to the biological processes. Under such restricting conditions we find that some analogy may be drawn between the electrodic process and the quinol-protein interaction in solution.The data of fig. 1 of my paper were obtained under such a condition. Prof. J . M . SavCant (University of Paris) said: This a question concerning the paper by Drs. Eddowes and Hill. It is seen on the free-energy diagram on fig. 6 that the thermodynamic and kinetic parameters of the adsorption-desorption process onto the electrode surface covered with 4,4’-bipyridine are the same for both the Fetrr and Fe” oxidation states. This could be regarded as expected in view of the rather large distance between the metallic centre and the assumed binding sites acting in the adsorption process.However, as pointed out by Dr. Williams in his paper, changes in the oxidation state of iron may induce changes in the surrounding protein at rather large distances from the iron atom. My question is thus: Was the equality of the adsorption parameters for Fe”’ and Felt a preliminary assumption in the treatment of the authors’ kinetic data or does this equality derive from experimental measurements?GENERAL DISCUSSION 397 Dr. R. Parsons (C.N.R.S., Meudon) said: It seems to me surprising that a simple Langmuir model works so well for this quite complicated system. The coverage of cytochrome c is relatively high, so that perhaps the neglect of adsorption of other com- ponents is reasonable. However, it seems less reasonable to neglect the inter- actions between these highly polar molecules as well as the potential dependence of the rate constants for adsorption and desorption.Is there any possibility to say anything about the orientation of the molecules from the kinetics of the process? The adsorption-desorption process seems in fact to be quite fast compared with the times required to adsorb polymers found by Stromberg e f al. using ellipsometry and Grant et al. using radiotracers.2 This might suggest that orientation is of little importance for electron transfer. R. R. Stromberg, E. Passaglia and D. J. Tutas, Ellipsometry (NBS Miscellaneous Publication no. 256, Washington D.C., 1964). ’ W. H. Grant, L. E. Smith and R. R. Stromberg, Faraday Discuss. Chem. Soc., 1975,59, 209. Prof. W. J. Albery (Imperial College, London) said: I would like to answer the questions of Prof.Savkant and Dr. Parsons. In our collaborative work with Drs. Eddowes and Hills, the adsorption of cytochrome c was studied by Dr. Hillman at Imperial College using a.c. ring-disc a p p a r a t ~ s . ~ - ~ In this technique the ring electrode can be set to monitor either the reduced or the oxidixed form of the cytochrome c. Hence the adsorption of both forms can be measured separately. For each substance FIG. 2.-Adsorption of ferricytochrome c (0) and ferrocytochrome c (A) measured by a.c. ring disc. The results were obtained at the half-wave potential and b , is the bulk concentration of ferricyto- chrome c. the adsorption obeys a Langmuir isotherm and the results plotted according to eqn (1 7) of ref.(1) are shown in fig. 2. There is no significant difference between the ferri- cytochrome c and ferrocytochrome c results. Dr. Parsons is quite right; the frac- tional coverage of the surface can be over half a monolayer; however, the treatment takes into account the competition between reduced and oxidised forms.398 GENERAL DISCUSSION W. J. Albery, M. J. Eddowes, H. A. 0. Hill and A. R. Hillman, J. Am. Chem. SOC., 1981, 103, 3905. W. J. Albery, J. S. Drury and A. P. Hutchinson, Trans. Faraday SOC., 1971, 67, 2414. W. J. Albery, A. H. Davis and A. Mason, Faraday Discuss. Chem. SOC., 1973,56, 317. W. J. Albery and A. R. Hillman, J. Chem. SOC. Faraday Trans. I , 1979,75, 1623. Prof. K. Niki (Yokohama National University) said: I should like to address these points to Drs.Eddowes and Hill. (1) Concerning the gold electrode surface, what kind of surface preparation do you need to reproduce your results and what type of characterization of this surface do you use? Quite recently, Prof. Kuwana mentioned that whenever he used a glassy carbon electrode, which was polished by alumina only, he observed using s.e.m. that many alumina particles were anchored in the electrode. With this electrode he observed electrocatalytic reduction of catechol and ascorbic acid. When the alumina particles were carefully removed from the electrode by ultrasonic washing, these electrochemical reactions turned out to be irreversible. Taniguchi et a1.l reproduced your results by using Sigma chemical cytochrome c (type VI) without further purifi- cation.(2) We find that the adsorption of cytochrome c on a well defined gold single crystal surface is so strong that the adsorbed film is removable only when we repeat the potential sweep between hydrogen evolution and oxygen evolution potentials for half an hour in 0.02 mol dm-3 NaF supporting electrolyte. On the other hand, the adsorption of 4,4‘-bipyridyl is reversible and we can easily wash it off from the elec- trode. In the present paper measurable adsorption kinetics of cytochrome c is assumed, but this assumption seems very unlikely in view of the strong and irreversible adsorption mentioned above. (3) In the case of cytochrome c3, which is extracted from Miyazaki strain, the heterogeneous rate constant is ca. 1 cm s-l and cc = 0.5 while the homogeneous intermolecular electron-transfer reaction rate is 103-104 dm3 mol-’ s-‘.These results are in agreement with the relation derived from Marcus theory. Electron transfer at the electrode in this case must be mediated by the strongly adsorbed cytochrome c3, the reacting molecule interacting rather weakly with this ‘‘ modified electrode.” Under such conditions it is reasonable to expect agreement with Marcus theory. I. Taniguchi et al., J. Electroanal. Chem., 1982, 131, 397. Dr. H. A. 0. Hill (Oxford University) said: In reply to Prof. Niki we have shown [see ref. (48) of our paper and references therein] that 4,4’-bipyridyl adsorbed on a polished gold electrode greatly enhances the rate of direct electron transfer to horse- heart cytochrome c, the heterogeneous rate constant being cm s-l.For the promotion of electron transfer to be effective the cytochrome must be pure. Other materials, including 4,4’-bipyridyl and related compounds, must be as pure as possible. It is possible that polishing exposes planes or creates defects that are essential for the promotion of the electron transfer. Gold is not the only material that can be used. Electrochemically clean platinum, palladium and copper adsorb 4,4‘-bipyridyl and quasi-reversible electron transfer to horse-heart cytochrome c ensues. We prefer to use bis-l,2-(4-pyridyl)ethylene as promotor since it appears to be more effective at lower concentrations. Pyrazine, 4-phenylpyridine, 2,2’-bipyridyl and bis-1,2-(4- pyridy1)ethane are not effectiLVe. How do the promotors work? Let me emphasize that there is no evidence that either promotor functions by acting as a conventional Even so, the electrodes “age”, for reasons that are not yet clear.GENERAL DISCUSSION 399 mediator.(Some confusion has resulted from the relationship between 4,4‘-bipyridyl and its di-alkylated derivatives, the viologens. The latter, which have, of course, quite different redox properties, can act as conventional mediators.) The reaction does not proceed : 4,4’-bipy + e [4,4’bipy]- : The promotors are adsorbed on the metal surface and their adsorption behaviour appears to be approximately Langmuirian. We have proposed that the 4,4’-bipyridyl- coated surface allows rapid, although relatively weak, adsorption of both oxidised and reduced cytochrome c on to the electrode surface and have suggested that such adsorp- tion is a prerequisite for fast electron transfer. We thus envisage a situation where the 4,4’-bipyridyl is sandwiched between the gold surface and the cytochrome c.The 4,4’-bipyridyl is adsorbed quite strongly (AGads = -35 kJ mol-l), although the adsorption is impeded by the presence of metal oxides on the surface. Surface- enhanced resonance Raman studies confirm this adsorption at gold, the mode of adsorption of 4,4’-bipyridyl being different from that of bis-l,2-(4-pyridyl)ethylene. We have suggested that the promotors act as a bridge between the metal surface and the lysines which surround the exposed haem edge of the cytochrome, since modifi- cation of the lysines with concomitant loss of charge results in no such electron transfer. [It might also be necessary that a “ conjugated ” bridge is required in view of the lack of promotion observed with bis- 1,2-(4-pyridyl)ethane.l Some of these lysines are known to be required for binding of cytochrome c to its natural partners, cytochrome c oxidase, cytochrome c peroxidase and cytochrome c reductase. We have proposed that the 4,4’-bipyridyl-modified electrode in some ways mimics that surface presented to cytochrome c by its natural partners.In reply to Prof. Savkant: how can oxidised cytochrome c be distinguished from the reduced form if the same binding site is used by both cytochrome reductase and cytochrome oxidase? If, for example, three lysines, in a roughly triangular disposition, are required for binding then a change in the distances between these three centres could result as a consequence of a change in oxidation state of the iron.This would be “ recognised ” by the ‘‘ partner ” proteins; the 4,4’-bipyridyl-modified electrode would not necessarily respond to such subtle changes, and indeed the electrochemical properties of the two redox forms can be interpreted by assuming that they have the same adsorption properties, This proposal requires that not only does the presence of 4,4’-bipyridyl lead to the adsorp- tion of cytochrome c but also that the latter is preferentially oriented with the haem edge presented to the surface. In reply to Dr. Parsons: It is possible that the large dipole moment, ca. 300 D, associated with cytochrome c at pH 7, may, through interaction with the electric field gradient, lead to a set of preferred orientations as the cytochrome approaches the electrode. Of course, the cytochrome c’s natural partners do not present a binding site composed of 4,4’-bipyridyl! Instead, the available evidence is that the lysine groups ion-pair with carboxylate species.Consequently it is not surprising that horse- heart cytochrome c gives rise to quasireversible electrochemical behaviour at an oxidised pyrolytic graphite surface. If it is indeed the case that to achieve rapid, direct electron transfer one must create a surface that mimics the surface of a given protein’s natural partner, there may be only three types of surface required: one with negatively charged groups (or with equivalently oriented dipoles, as with 4,4’-bipyridyl) ; with positively charged groups or with a “ hydrophobic ” surface exposed.Many interesting redox enzymes and400 GENERAL DISCUSSION proteins are acidic and carry a negative charge at pH 7. Although it is not necessarily the case that the overall charge is a guide to the charge at the binding site that is proximal to the point of entry (if such exists) of the electron, it is obviously of interest to create surfaces that belong to the second class. We have not yet achieved this by modifying surfaces with positively charged organic compounds but we have achieved it “ surreptitiously.” The electrochemistry of the eight-iron ferredoxin from Clostri- dium pasteurianum is “ slow ” at a 4,4’-bipyridyl-modified gold electrode. It becomes quasireversible, however, in the presence of divalent metal ions perhaps by forming a more extended bridge: Au t NN -+ Mg2+ -0OC-Fd where NN indicates 4,4’-bipyridyl.The results at an oxidised pyrolytic graphite sur- face are even more striking, Mg2+ (or Ca2+ or Mn2+) giving rise to electrochemically quasireversible and chemically reversible electron transfer. It remains to be seen how general a solution this provides to the problem of attaining rapid electron transfer be- tween these ubiquitous proteins and readily available electrodes. Drs. Armstrong, Busby, Creighton, Uosaki and Walton have contributed hitherto unpublished results to this comment. K, Uosaki and H. A. 0. Hill, J. Electroanal. Chem., 1981, 122, 321. Prof. L. I. Krishtalik (Academy of Sciences of the U.S.S.R., Moscow) (communi- cated): Dr.Hill has explained that the electron-transfer rate in electrochemical reac- tions of cytochrome c is rather low. For electrochemical redox reactions the re- organization energy (for the solvent reorganization) is approximately half the value of the same quantity for homogeneous reactions. Hence it follows that the electro- chemical rate constant must be the lower limit for any homogeneous redox reaction (both recalculated into quasimonomolecular characteristics). How to relate then the “ electrochemical ” rate constant 50 s-’ to the rate constant for the reaction in the cytochrome c complex with Fe(CN):l4- equal to 2500 s-’? Is it possible to assume that the experimental value for electrode reaction is a result of some averaging over different orientations of adsorbed cytochrome including the unfavourable ones, or may the low electron-transfer rate be ascribed to the influence of the bipyridyl layer on the tunnelling probability ? Prof.M. J. Weaver (Purdue University, West Lafayette) said : The comparison of the energetics of corresponding electrode reactions occurring uia ‘‘ attached-reactant ” (inner-sphere) and “ unadsorbed ” (outer-sphere) pathways, and with the energetics of the corresponding homogeneous redox pathways, requires careful consideration of the appropriate statistical formalism needed to relate the measured rate parameters to the desired electron-transfer barriers. Hill and coworkers take the frequency factor for outer-sphere pathways to be the collision frequency 2, and for surface-attached pathways to be the usual unimolecular frequency kT/h.While conventional, this approach suffers from the implication that the motion towards the transition state is different for unattached versus attached reactants. Instead of the collision model, a more satisfactory approach is to employ a “ pre- equilibrium ” formalism whereby the frequency factor is determined by activation within a “precursor state” with the reactant sufficiently close to the electrode surface (or “ coreactant ”) so that electron tunnelling is facilitated once the appro- priate transition-state geometry has been achieved. Such a model has recently been applied to outer-sphere electron transfer in homogeneous solution 1*2 (see papers byGENERAL DISCUSSION 401 Brunschwig et al.and Friedman and Newton in this Discussion), and also to electro- chemical proce~ses.~.~ Since the pre-equilibrium model treats outer- as well as inner- sphere reactions in terms of unimolecular activation within a preassembled precursor state, the kinetics of both processes at electrodes as well as in homogeneous solution can be described in terms of first-order (s-l) rate constants. Instead of the arbitrary frequency kT/h the unimolecular activation frequency can be described in terms of the proper combination of vibrational and solvent reorientational m0des.l Such an approach yields markedly different overall frequency factors for outer- sphere reactions than are obtained using the collisional m ~ d e l . l - ~ Consequently the free energies of activation AG$ derived from a given set of rate data using the pre- equilibrium formalism will differ noticeably from those obtained using the collisional model. In addition, the relative values of AGt for corresponding inner- and outer- sphere processes, and for corresponding electrochemical and homogeneous reactions estimated from the appropriate rate constants using these two formalisms will be different .4 It is therefore suggested that comparisons of reaction energetics for related electron-transfer steps, such as those discussed by Hill et al., should be based on the pre-equilibrium rather than the collisional description of the frequency factor.B. S. Brunschwig, J. Logan, M. D. Newton and N. Sutin, J. Am. Chem. Soc., 1980,102, 5798.R. A. Marcus, Int. J. Chem. Kinet., 1981, 13, 865. M, J. Weaver, Inorg. Chem., 1979, 18,402. J. T.Aupp and M. J. Weaver, J. Electroanal. Chem., in press. Prof. W. J.,Albery (Imperial College, London) said: In our derivation of the free energy profile for the cytochrome c reaction1 we chose to use the conventional frequency factors of kT/h for first-order processes and of lo4 cm s-l for encounters of reactant molecules with the electrode.2 This procedure has the advantage that we are using a widely accepted connection between the measured rate constants and the derived free energies of activation. Confusion could result if each author chose his own frequency factor. Furthermore, I do not believe that the conventional choice of frequency factors necessarily implies that the motion towards the transition state is " different for unattached versus attached reactants." I agree with Prof. Weaver that for the electrode reaction one should break the process down into an association constant and a frequency factor.In our treatment we considered that the localis- ation of the reactant close to the plane of the electrode would, for ordinary molecules, have a statistical value of ca. 3 x cm. Combining this with the conventional frequency factor of kT/h one obtains a value of cu. 18 x lo4 cm s-l. This is larger than the collisional frequency of lo4 cm s-l introduced by Marcus.* However, a power of 10 is usually neither here nor there in these types of comparison and we have not so far thought it worthwhile to be the only man in step and use lo5 cm s-l.Turning to the Marcus value for the collision frequency of 10" dm3 rno1-l s-I used for second-order homogeneous reactions, again this can be considered as a statistical pre- equilibrium association of (1/55.5) multiplied by kT/h. Finally I do agree with Prof. Weaver that in the long run we should aim to replace the ubiquitous kT/h with the proper solvent frequency factor vs, but I doubt whether the value of vs is well enough established to be used generally at this time. W. J. Albery, M. J. Eddowes, H. A. 0. Hill and A, R. Hillman, J. Am. Chem. Soc., 1981,103, 3904. R. A. Marcus, J. Phys. Chem., 1963, 67, 853. W. J. Albery and M. L. Hitchman, Ring-Disc Electrodes (Clarendon Press, Oxford, 1971), p. 164. W. J. Albery, Electrode Kinetics (Oxford University Press, Oxford, 1975), p.112.402 GENERAL DISCUSSION Prof. H. L. Friedman (State University of New York, Stony Brook) said: When reading Dr. Heremans’ paper it should be remembered that there is some uncertainty in interpreting activation volumes A V as differences in volume between the initial and transition states. More generally transition-state theory does not apply (cf. dis- cussion of contribution by Hupper, Kanety and Kosower). Then there is at least one slow activation process whose pressure derivative also will contribute to A W . So it seems desirable in every case to make at least some estimate as to the applicability of transition-state theory. Of course the same consideration applies to AH3 and A S . Dr. K. Heremans (University ofleuven) said : The applicability of the transition- state theory to the interpretation of A P has been considered recently in two experi- mental papers.Karplus and McCammon have interpreted the experiments of Wagner Instead of a difference in physical volume between transition state and reactants, A V3 is interpreted as being dominated by the interaction of the reacting species with the solvent environment, i.e. the hydrophobic protein interior. Jonas et aL3 have considered the conformational isomerization of cyclohexane. For this reaction AH3 is independent of solvent and temperature, while A V3 is solvent- and pressure-dependent. The authors interpret their results on the transition-state theory and using stochastic models for unimolecular reactions in the condensed phase.The observed A P correlates well with the shear viscosity of the solvent and its pressure dependence. From the estimated A V for the transition state (based on the molar volume difference between cyclohexene and cyclohexane), the authors calculate the collision contribution to A V which is solvent- and pressure-dependent. Both these reactions are unimolecular and non-ionic. The extension of the inter- pretation to electron transfer reactions where two nuclear reactive modes and two potential energy surfaces are present is at present a ~hallenge.~ With these reservations in mind, an estimate of the collision contribution to the observed A V3 can be made from the effect of pressure on the viscosity of water (A V = 1 cm3 mol-I). This is small compared with the experimental values for A V but not small compared with the calculated A V for the inner- and outer-sphere rearrange- ments of ligands and solvent. For ionic reactions the assumed value for the A W electrostriction holds.on the aromatic-ring rotation in proteins. M. Karplus and J. A. McCammon, FEBS Left., 1981, 131, 34. G. Wagner, FEBS Lett., 1980, 112, 280. D. L. Hasha, T. Eguchi and J. Jonas, J. Am. Chem. SOC., 1982,104, 2290. B. L. Tembe, H. L. Friedman and M. D. Newton, J. Chem. Phys., 1982,76, 1490. Prof. B. E. Conway (University of Ottawa) said: I believe that one can go con- siderably further with the interpretation of volumes of activation, A V , than just the comparison of experimentally evaluated numbers. Thus, at least with simple redox reactions, especially when they are symmetrical as in the case of Fe(CN):-/ Fe(CN);-, it is possible to measure (e.g.electrochemically) the overall A V and have a good idea of the intrinsic volume change in the activation process from half the volume change associated with bond length changes (small in the case exemplified). Beyond the first hydration co-sphere, a dielectric polarization volume change can also be calculated quite well from the treatments of Frank and Booth (for dielectric saturation at low fields) as described by Desnoyers et al.,4 or an upper limit to this AV3 contribution can be rather reliably calculated. Hence the remaining inner hydration co-sphere volume change can be estimated by difference.l Approaches along these lines could lead to considerable progress in interpretation of activationGENERAL DISCUSSION 403 volume and reaction volume changes in more complex redox systems in the water solvent, for which data on electrostriction, compressibility and dielectric behaviour are k n ~ w n .~ ' ~ B. E. Conway and J. C. Currie, J. Electrochem. SOC., 1978, 125, 258. H. S. Frank, J. Chem. Phys., 1955,23,2033. F. Booth, J. Chem. Phys., 1951, 19, 391; 1327; 1615. J. E. Desnoyers, R. E. Verrall and B. E. Conway, J. Chem. Phys., 1965,43,243. Dr. K. Heremans (University of Leuven) said: As our paper shows it is important to realize that the experimental activation volumes for redox cross-reactions have a kinetic and a thermodynamic contribution. The experimental numbers can therefore not be advocated as a test of the Marcus theory as is sometimes done.' It is possible, as indicated by Prof.Conway, to calculate the intrinsic volume change for the acti- vation process in simple redox reactions. The experimental results obtained from electrochemical studies for the Fe(CN);-/Fe(CN):- system are, however, highly dependent on the experimental conditions. I refer to the work of Sat0 and Yamada on the effect of the supporting electrolyte on A P for the above reaction. It seems that ion association is important in these reactions, so that unless allowance is made for this effect calculations on intrinsic volume changes seem questionable. F. B. Ueno, Y. Sasaki, T. Tuo and K. Saito, J. Chem. Sac., Chem. Commun., 1982, 328. M. Sat0 and T. Yamada, in High Pressure Science and Technology, ed.B. Vodar and Ph. Marteau (Pergamon Press, Oxford, 1980), vol. 2, pp. 812-814. Prof. H. L. Friedman (State University of New York, Stony Brook) said : Substantial uncertainties are generated when one evaluates the '' thermodynamic part " of A W or A S on the basis of the equations [Heremans et al. eqn (3) and (9), eqn (3) and (4) of Brunschwig et aZ.] K i j = A exp[-wij(r)/kT) (1) (2) wij(r) = eiej/Dsr(l + Icr) where A is a coefficient to be discussed below, Ki is the mass-action constant for form- ation of an i, j pair at separation Y, D, the solvent dielectric constant and IC the Debye reciprocal shielding length. Three different problems connected with these equations applied in the range up to ca. 1 mol dm-3 ionic strength should be mentioned.(1) These equations may be considered as approximations for the properties of the primitive model for ionic solutions; charged hard spheres in a continuum.lP2 In fact the corresponding pair correlation function gi j ( r ) = ex~[-wij(r)/kr] (3) (4) with wij given by wi j(r) = eieje-K(r - N)/D,r(l + K a ) where all of the charged hard spheres have diameter a is quite successful for primitive model aqueous 1-1 electrolytes but not for electrolytes of higher charge type.4 Eqn (4) in the case r = a gives eqn (2). (2) The accurately calculated correlation functions for the primitive model, together with adjustable ion radius parameters, give reasonably satisfactory excess free energies for aqueous 1-1 electrolytes but are less satisfactory for higher charge types and especially for excess energies4 and volumes. It is quite clear that to represent these thermodynamic derivatives of the free energy in a realistic way one404 GENERAL DISCUSSION needs a richer model in which there is some representation of specific ion-ion inter- a c t i o n ~ .~ Even for the free energy, specific terms are needed to account for the properties of electrolyte mixtures as found by Reilly et aL6 (3) For a primitive model electrolyte the mass-action constant for forming an ion pair whose centre-to-centre distance is in the range from “ contact ” at R i j to Ri + dh is exactly R , + 611 Kij = i,, gi j(r)4nr2dr. Assuming that dh is small enough so that g,,(R, j ) z gij(Rij + dh) this equation leads to the A factor in eqn (1) used by Brunschwig et al., although they do not justify their particular choice of dh, nor its assumed independence of temperature.On the other hand the A factor in eqn (1) used by Heremans et al. corresponds to replacing eqn (5) by R i i Ki j = gij(Ri j ) / 47trzdr 0 a choice which seems very peculiar to me, although in many cases its numerical con- sequences may not be very different from those of eqn (4). €3. L. Friedman, J. Chem. Phys., 1960, 32, 1 134. J. S. Hoye and G. Stell, Faraday Discuss. Chem. Soc., 1977, 64, 16. K S Pitzer, Acc. Chem. Res., 1977, 10, 371. J. P. VaIleau, L. K. Cohen and D. N. Card, J . Chem. Phys., 1980,72, 5942. P. S. Ramanathan and H. L. Friedman, J. Chem. Phys., 1970,54, 1086, P. J. Reilly, R. H. Wood and R. A. Robinson, J . Phys. Chern., 1971, 75, 1305.’ R. M . Fuoss, J. Am. Chem. SOC., 1958,80, 5059. Dr. K. Heremans (University of Leuven) said: There are two thermodynamic contributions to AVt [Heremans et al. eqn (13)]. The equilibrium volume change for the reaction which is obtained experimentally and the volume change for the for- mation of the precursor complex. The latter is calculated from eqn (1 1). Ideally one would like to obtain this value from experiment but in practice this is almost never possible. I agree with Prof. Friedman that the equation used is only applicable in the limit of high dilution. The calculated volume changes are small. (See table 1 in Heremans et al.) Since we are working at I = 0.2 mol dm-3, the actual volume changes will be even smaller. We therefore do not expect substantial uncertainties from this part of the theory. The question also arises whether the improved equations can be used for the interaction of a protein with a small molecule. The overall dipole . and local dipole at the binding site of the protein may substantially contribute to the interaction of the small molecules with the protein.’ The important result from our study, that the observed AVZ can be either positive or negative depending on the overall AV of the reaction, is certainly not affected. W. H. Koppenol, Biophys. J., 1980, 29, 493. Dr. Sutin (Brookhaven National Laboratory) (partly communicated) : I agree with Prof. Friedman’s comment that his eqn (2) [eqn (4) of Brunschwig et al.] is an approxi- mation: fortunately the prevailing electrolyte in most of the systems studied is of the 1-1 variety (H+, Li+ or Na+ perchlorate or trifluoromethylsulphonate) for which there is some justification for the use of eqn (2) subject to the equal ionic radius assumption. When the latter assumption is not valid, the more general equation presented else- where ’ should be considered-this equation reduces to Friedman’s eqn (4) when o2 =GENERAL DISCUSSION 405 03, as is the case for exchange reactions. As noted by Brunschwig et al., the temper- ature dependence of eqn (2) does not adequately account for the entropies of acti- vation of the reactions. Justification of the particular choice made for 6h (0.8 A, 6r of Brunschwig et al.) is presented elsewhere; the value of 6h is determined by the degree of adiabaticity of the reaction (as well as by other factors) and typical values of 6h range from ca. 0.3 to 2 A. There could be some temperature dependence to 6h and this would make a contribution to AS*. ' N. Sutin and B. S. Brunschwig, ACS Symp. Ser., 1982, 198, 105. Prof. W. J. Albery (Imperial College, London) said: Dr. Kell has done a good demolition job on the control of the H+ flux by the membrane potential. How- ever, I would be interested to know what is his explanation of the results for curves (a) and (b) in his fig. 3. Dr. D. B. Kell (University College of Wales, Aberystwyth) said: It is evident that one must conclude that two parallel processes take place simultaneously: ejection of H+ into the bulk phase and ejection of H+ into a phase which is not in electrochemical equilibrium with the external bulk aqueous phase. The presence of SCN- blocks the latter pathway so that only the former can operate. It may be mentioned that Tedeschi has proposed that, for experiments corresponding to trace (a) of fig. 3, none of the H+ observable in the bulk phase have been transported electrogenically from the internal bulk phase. ' H. Tedeschi, Biochim. Biophys. Acta, 1981, 639, 157.

ISSN:0301-7249    

DOI:10.1039/DC9827400389

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

LIST OF POSTERS 1 . Activation energy and frequency factor effects in electrochemical proton transfer from the unhydrated H30+ ion and the fully hydrated H 9 0 i ion B. E. Conway and D. F . Tessier University of Ottawa, Canada barriers T. R. Knowles Free University of Berlin, West Germany 3. Electron transfer during electrosorption W. Schmickler and A. Kornyshev University of Dusseldorf, West Germany and Academy of Sciences of the U.S.S.R., Moscow H . Killesreiter Institut fur Erdolforschung, Hannover, West Germany D. Oelkrug and R. W. Kessler Tiibingen University, West Germany and Daimler Benz A.G., Stuttgart, West Germany Localization of charge during the transition state of the solute-the cation localization energy T . B. Truong Universitk de Paris-Sud, France D. Meyerstein Ben-Gurion University of the Negev, Israel G.Scherer and F . Willig Fritz-Haber-Institut, Berlin, West Germany by ruthenium(i1) species in aqueous solution G. Daramole, J. F. Ojo, 0. Olubuyide and F. Oriafor University of Ife, Nigeria A scheme of wide relevance in chemistry M. Chanon and M. L. Tobe FacultC des Sciences de Saint Je'rGme, Marseille, France and University College, London 11. 1- and 2-electron transfer reactions between aromatic and quinoidic systems K. Kemnitz, U. Nickel and W. Jaenicke University of Erlangen-Niirnberg, West Germany 12. Electron-transfer kinetics of an organic two-step redox system determined by e.s.r. line- broadening measurements G. Grampp and P. Pluschke University of Erlangen, West Germany electronically excited 3,5-dinitroanisole C.A. G. 0. Varma State University , Gorlaeus Laboratories, Leyden, The Netherlands 14. Electron transfer from [2H] hydroxyalkyl radicals to substituted nitrobenzenes via addition-elimination S . Steenken Max-Planck-Institut fur Strahlenchemie, Mulheim, West Germany 15. Charge transfer in the dibenzocarbazole-pyridine hydrogen-bonded system: A time- resolved picosecond spectroscopy study M. M. Martin, N. Ikeda, T. Okada and N. Mataga Universite' Paris-Sud, Orsuy, France and University of Osaka, Toyonaka, Japan 2 . Hydrogen evolution reaction mechanisms : quantum calculation of the activation 4. Charge transfer from dyes into molecular crystals controlled by spatial relations 5 . Electron transfer from chemisorbed aromatic hydrocarbons to metal-oxide surfaces 6.Charge transfer to a solvent state VIII. 7. New powerful single electron redox reagents in aqueous solutions 8. Experimental test of the energy dependence of electron transfer for the redox ions 9. Kinetics and mechanism of the reduction of the halopenta-amminecobalt(iI1) complexes 10. Double activation induced by single electron transfer. 13. Influence of solvent structure on the efficiency of electron transfer between OH- and16. Application of single-electron-transfer theory to radiationless decay M. R. V. Sahyun Central Research 3M, St. Paul, U.S.A. 17. The efficiency of charge separation in solution A. Harriman The Royal Institution, London 18. Temperature dependence of electron-transfer reactions. Reductive quenching of *RuLg+ luminescence by aromatic amines J. E.Baggott and M. J . Pilling University of Oxford 19. An evaluation of the distance dependence on charge-transfer reactions in the excited state S . M. B. Costa, A. L. Macanita, M. J. Prieto and M. I. Viseu Instituto Superior Tecnico Lisbon, Portugal 20. Intramolecular electron- and proton-transfer reactions studied after photoionisation by an intense monosecond laser pulse J. A. Delaire, M. Castella and J. Faure Uniuersitk de Paris-Sud, Orsay, France 21. The recombination of hydrogen and hydroxyl ions in aqueous solution: temperature dependence and kinetic isotope effect J. F. Holzwarth, W. Frisch and A. J . Kresge Fritz-Haber-Institut, Berlin, West Germany and University of Toronto, Canada 22. Prototropic charge transport in water B. Halle and G.Karlstrom University of Lund, Sweden 23. Proton transfers in hydrogen-bonded systems S . Scheiner Southern Illinois University, Carbondale, US. A. 24. Mechanisms of proton transfer reactions in the polar solvents acetonitrile and benzo- nitrile R. Suttinger, F. Strohbusch, D. B. Marshall and E. M. Eyring BASF Ludwigshafen, University of Freiburg, West Germany and University of Utah, Salt Lake City, U.S.A. D. M. Goodall, C. B. Moore and W. Natzle University of Yorkand University of Cali- fornia, Berkeley, U.S.A. Application to proton transfer and other bond-making/bond-breaking processes J. R. Murdoch, J. Nonnella, M. Berry and D. E. Magnoli University of California, Los Angela, U.S.A. 27. On the theory of charge-transfer enzymatic reactions L. I . Krishtalik Academy of Sciences of the U.S.S.R., Moscow 28.Spectroelectrochemical study of the electron transfer in oxidoreductases M. Comtat and H. Durliat Paul Sabatier University, Toulouse, France 29. Electron structure and electron transfer in biological systems S. Larsson University of Lund, Sweden 30 The pH dependence of the midpoint redox potential of some bacterial c-type cytochromes F. A. Leitch Royal (Dick) School of Veterinary Studies, Edinburgh 31. pH dependence of the redox equilibrium of D. gigas cytochrome c3: electron transfer mechanisms H. Santos, J. J. G. Moura, I. Moura, J. LeGall and A. V . Xavier Centro de Quimica Estrutural, Lisbon and University of Georgia, U.S.A. 32. Do the pH dependent properties of cytochromes c2 have physiological significance? G. W. Pettigrew and G. R. Moore University of Edinburgh and University of Oxford 33. The interaction of cytochrome c with metal hexacyanides C. G. S. Eley, G. R. Moore, G. Williams and R. J . P. Williams University of Oxford 34. Etude par saut de temperature du transfert d’electrons au sein du centre actif de l’enzyme multifonctionnel flavocytochrome b2 de la levure hansenula anomala M. Tegoni, F. Labeyrie, M. C . Silvestrini and M. Brunori CNRS, Gif-sur- Yvette, France and Rome University, Italy 25. Proton-transfer reactions induced by single-photon vibrational excitation 26. The Marcus equation and “ strong overlap ” reactions.

ISSN:0301-7249    

DOI:10.1039/DC9827400407

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

INDEX OF NAMES* * The page numbers in heavy type indicate papers submitted for discussion. Albery, W. J., 102, 111, 245, 286, 291, 294, 296, 397,401,405 Amouyal, E., 147 Baggott, J. E., 86 Barker, H. A., 311 Bernas, A., 147, 193 Bienvenue, E., 365 Bixon, M., 17, 103 Booty, T. E., 107 Bormans, M., 343 Brocklehurst, B., 202 Brooke, D. N., 215 Bruhn, H., 129 Brunschwig, B. S., 113 Caldin, E. F., 200, 215, 281 Chanon, M., 182,308 CharlC, K-P., 141 Conway, B. E., 98, 185, 267, 270,402 Creutz, C., 113 Dagnall, S. P., 215 Eddowes, M. K., 331 Eggins, B. R., 98, 394 Eley, C. G. S., 311 Friedman, H. L., 73, 88, 104, 184, 198,402,403 Gavach, C., 365 Gerischer, H., 193 Gerritzen, D., 229, 279 Goodall, D. M., 275, 284 Grand, D., 147 de Haas, M. P., 196 Hennig, J., 229 Heremans, K., 343, 402, 403, 404 Hill, H.A. O., 331, 398 Hitchens, G. D., 377 Holzwarth, J. F., 104, 129, 179, 184, 185, 186, 187,188, 393 Huang, Z-X., 311 Huppert, D., 161 Jortner, J., 17, 94, 106, 110, 111, 178, 193, 271, Kanety, H., 161 Kell, D. B., 377, 405 Kosower, E. M., 161,197, 199,202,308,309,393 Kreevoy, M. M., 257, 273, 308 Krishtalik, L. I., 95,205,268,272,274,293, 309, Kuznetsov, A. M., 31, 49, 89, 97, 177, 281 Larsson, S., 390 283, 306 389,400 Limbach, H-H., 229,279,281,282,284,285,288, van Lith, D., 390 Macartney, D. H., 113 Mak, M. K. S., 215 Marcus, R. A., 7, 83, 85, 86, 93, 103, 104, 110, 199,272,307 Markvart, T., 91 Martin, M. M., 202 Meyerstein, D., 188, 203 Mialocq, J-C., 147 Moore, G. R., 311, 392, 393 Murdoch, J. R., 297 Newton, M. D., 73, 88, 95, 101, 108, 110, 111, 178,308 Nigam, S., 129 Niki, K., 398 OstoviC, D., 257 Parsons, R., 397 Rich, P. R., 349, 396 Roberts, R. M. G., 257 Robinson, M. N., 311 Rosseinsky, D. R., 105, 106, 107 Rumpel, H., 229 Sahyun, M. R. V., 85 Salmon, G. A., 191 SavCant, J-M., 57, 87, 90, 96, 98, 99, 101, 102, Schmickler, W., 88, 272 Seta, P., 365 Sham, T-K., 113 Snauwaert, J., 343 Sutin, N., 113, 178, 180, 182, 183, 392, 394, 404 Tessier, D., 57 Tonge, J. S., 107 Truong, T. B., 186 Tucker, P., 106 Ulstrup, J., 31, 93, 94, 100, 271 Vandersypen, H., 343 Varma, C. A. G. O., 83, 84, 194 Vennereau, P., 276 Weaver, M. J., 101, 180, 400 Wells, C. F., 294 Williams, G., 311 Williams, R. J. P., 107, 183, 185, 188, 311, 389, 390, 392,393 Willig, F., 141, 189 291, 294 396

ISSN:0301-7249    

DOI:10.1039/DC9827400409

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

GENERAL DISCUSSIONS OF THE FARADAY SOCIETY/FARADAY DISCUSSIONS OF THE CHEMICAL SOCIETY Date 1907 1907 1910 191 1 1912 1913 1913 1913 1914 1914 1915 1916 1916 1917 1917 1917 1918 1918 1918 1918 1919 1919 1920 1920 1920 1920 1921 1921 1921 1921 1922 1922 1923 1923 1923 1923 1923 1924 1924 1924 1924 1924 1925 1925 1926 1926 1927 1927 1927 1928 1929 1929 1929 1930 1931 1932 Subject Osmotic Pressure Hydrates in Solution The Constitution of Water High Temperature Work Magnetic Properties of Alloys Colloids and their Viscosity The Corrosion of Iron and Steel The Passivity of Metals Optical Rotatory Power The Hardening of Metals The Transformation of Pure Iron Methods and Appliances for the Attainment of High Temperatures in a Laboratory Refractory Materials Training and Work of the Chemical Engineer Osmotic Pressure Pyrometers and Pyrometry The Setting of Cements and Plasters Electrical Furnaces Co-ordination of Scientific Publication The Occlusion of Gases by Metals The Present Position of the Theory of Ionization The Examination of Materials by X-Rays The Microscope : Its Design, Construction and Applications Basic Slags : Their Production and Utilization in Agriculture Physics and Chemistry of Colloids Electrodeposition and Electroplating Capillarity The Failure of Metals under Internal and Prolonged Stress Physico-Chemical Problems Relating to the Soil Catalysis with special reference to Newer Theories of Chemical Action Some Properties of Powders with special reference to Grading by Elutria- The Generation and Utilization of Cold Alloys Resistant to Corrosion The Physical Chemistry of the Photographic Process The Electronic Theory of Valewy Electrode Reactions and Equilibria Atmospheric Corrosion. First Report Investigation on Oppau Ammonium Sulphate-Nitrate Fluxes and Slags in Metal Melting and Working Physical and Physico-Chemical Problems relating to Textile Fibres The Physical Chemistry of Igneous Rock Formation Base Exchange in Soils The Physical Chemistry of Steel-Making Processes Photochemical Reactions in Liquids and Gases Explosive Reactions in Gaseous Media Physical Phenomena at Interfaces, with special reference to Molecular Atmospheric Corrosion.Second Report The Theory of Strong Electrolytes Cohesion and Related Problems Homogeneous Catalysis Crystal Structure and Chemical Constitution Atmospheric Corrosion of Metals.Third Report Molecular Spectra and Molecular Structure Colloid Science Applied to Biology Photochemical Processes The Adsomtion of Gases by Solids tion Orientation Volume Trans. 3* 3* 6* 7* 8* 9* 9* 9* 10* 10* 11* 12* 12* 13* 13* 13* 14* 14* 14* 14* 15* 15* 16* 16* 16* 16* 17* 17* 17* 17* 18* 18 19* 19 198 19 19* 20* 20* 20 * 208 20* 21 * 21 22 22 23 * 23 * 24 24 25" 25 * 26* 26 27 28 1932 The Colloib Aspect of TexGle Materials 29412 FARADAY DISCUSSIONS OF THE CHEMICAL SOCIETY Date 1933 1933 1934 1934 1935 1935 1936 1936 1937 1937 1938 1938 1939 1939 1940 1941 1941 1942 1943 1944 1945 1945 1946 1946 1947 1947 1947 1947 1948 1948 1949 1 949 1949 1950 1950 1950 1950 1951 1951 1952 1952 1952 1953 1953 1954 1954 1955 1955 1956 1956 1957 1958 1957 1958 1959 1959 1960 1960 1961 1961 1962 1962 1963 1963 Subject Liquid Crystals and Anisotropic Melts Free Radicals Dipole Moments Colloidal Electrolytes The Structure of Metallic Coatings, Films and Surfaces The Phenomena of Polymerization and Condensation Disperse Systems in Gases : Dust, Smoke and Fog Structure and Molecular Forces in (a) Pure Liquids, and (b) Solutions The Properties and Functions of Membranes, Natural and Artificial Reaction Kinetics Chemical Reactions Involving Solids Luminescence Hydrocarbon Chemistry The Electrical Double Layer (owing to the outbreak of war the meeting The Hydrogen Bond The Oil-Water Interface The Mechanism and Chemical Kinetics of Organic Reactions in Liquid The Structure and Reactions of Rubber Modes of Drug Action Molecular Weight and Molecular Weight Distribution in High Polymers (Joint Meeting with the Plastics Group, Society of Chemical Industry) The Application of Infra-red Spectra to Chemical Problems Oxidation Dielectrics Swelling and Shrinking Electrode Processes The Labile Molecule Surface Chemistry (Jointly with the SociCt6 de Chimie Physique at Colloidal Electrolytes and Solutions The Interaction of Water and Porous Materials The Physical Chemistry of Process Metallurgy Crystal Growth Lipo-proteins Chromatographic Analysis Heterogeneous Catalysis Physico-chemical Properties and Behaviour of Nuclear Acids Spectroscopy and Molecular Structure and Optical Methods of Investi- gating Cell Structure Electrical Double Layer Hydrocarbons The Size and Shape Factor in Colloidal Systems Radiation Chemistry The Physical Chemistry of Proteins The Reactivity of Free Radicals The Equilibrium Properties of Solutions on Non-electrolytes The Physical Chemistry of Dyeing and Tanning The Study of Fast Reactions Coagulation and Flocculation Microwave and Radio-frequency Spectroscopy Physical Chemistry of Enzymes Membrane Phenomena Physical Chemistry of Processes at High Pressures Molecular Mechanism of Rate Processes in Solids Interactions in Ionic Solutions Configurations and Interactions of Macromolecules and Liquid Crystals Ions of the Transition Elements Energy Transfer with special reference to Biological Systems Crystal Imperfections and the Chemical Reactivity of Solids Oxidation-Reduction Reactions in Ionizing Solvents The Physical Chemistry of Aerosols Radiation Effects in Inorganic Solids The Structure and Properties of Ionic Melts Inelastic Collisions of Atoms and Simple Molecules High Resolution Nuclear Magnetic Resonance The Structure of Electronically Excited Species in the Gas Phase Fundamental Processes in Radiation Chemistry was abandoned, but the papers were printed in the Transactions) Systems Bordeaux) Published by Butterworths Scientific Publications, Ltd Volume 29 * 30 30 31* 31 * 32.32* 33* 33* 34. 34* 35 * 35. 35* 36* 37 * 37 * 38 39 40* 41 * 42 * 42 A 42 B Disc. 1* 2 Trans. 43" Disc. 3 4* 5 6 7 8* Trans. 46' Disc. 9 Trans. 47 Disc. 10 11 12* 13 14 15 16* 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33* 34 35 36Date 1964 1964 1965 1965 1966 1966 1967 1967 1968 1968 1969 1969 1970 1970 1971 1971 1972 1972 1973 1973 1974 1974 1975 1975 1976 1977 1977 1977 1978 1978 1979 1979 1980 1980 1981 1981 1982 1982 FARADAY DISCUSSIONS OF THE CHEMICAL SOCIETY Subject Chemical Reactions in the Atmosphere Dislocations in Solids The Kinetics of Proton Transfer Processes Intermolecular Forces The Role of the Adsorbed State in Heterogeneous Catalysis Colloid Stability in Aqueous and Non-aqueous Media The Structure and Properties of Liquids Molecular Dynamics of the Chemical Reactions of Gases Electrode Reactions of Organic Compounds Homogeneous Catalysis with Special Reference to Hydrogenation and Bonding in Metallo-organic Compounds Motions in Molecular Crystals Polymer Solutions The Vitreous State Electrical Conduction in Organic Solids Surface Chemistry of Oxides Reactions of Small Molecules in Excited States The Photoelectron Spectroscopy of Molecules Molecular Beam Scattering Intermediates in Electrochemical Reactions Gels and Gelling Processes Photo-effects in Adsorbed Species Physical Adsorption in Condensed Phases Electron Spectroscopy of Solids and Surfaces Precipitation Potential Energy Surfaces Radiation Effects in Liquids and Solids Ion-Ion and Ion-Solvent Interactions Colloid Stability Structure and Motion in Molecular Liquids Kinetics of State Selected Species Organization of Macromolecules in the Condensed Phase Phase Transitions in Molecular Solids Photoelectrochemistry High Resolution Spectroscopy Selectivity in Heterogeneous Catalysis Van der Waals Molecules Electron and Proton Transfer Oxidation 413 Volume 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 * 66 67 68 69 70 71 72 73 74 * Not available; for current information on prices, etc., of available voiumes, please contact the Marketing Officer, Royal Society of Chemistry, Burlington House, London WI V OBN stating whether or not you are a member of the Society.

ISSN:0301-7249    

DOI:10.1039/DC9827400411

出版商:RSC    

年代:1982

数据来源: RSC