Biochimica et Biophysica Acta, 1018 (1990) 185-189

185

Elsevier BBAEBC 00013

Control processes in oxidative phosphorylation: kinetic constraints and stoichiometry Michel Rigoulet lnstitut de Biochimie Cellulaire et Neurochimie du CNRS, Universit~ de Bordeaux 2, Bordeaux (France)

(Received 1 May 1990)

Key words: Oxidative phosphorylation; Thermodynamic control; Kinetic control; Proton pump stoichiometry; ATPase/ATP synthase; Mitochondrion; (Yeast); (Rat liver); (Bovine heart)

Control processes in oxidative phosphorylation have been studied in three experimental models. (1) In isolated yeast mitochondria, external ATP is a regulatory effector of cytochrome-c oxidase activity. In phosphorylating or uncoupling states, the relationships between respiratory rate and A/~n+, and the respiratory rate and cytochrome-c oxidase reduction level are dependent on this kinetic regulation. (2) In rat liver mitochondria, the response of the respiratory rate to uncoupler addition is age-dependent: liver mitochondria isolated from young rats maintain a greater d/~ n + than liver mitochondria isolated from adults, with the same respiratory rate obtained with the same concentration of uncoupler. This behaviour is linked to redox proton pump properties, i.e., to the degree of intrinsic uncoupling induced by uncoupler addition. (3) The effect of almitrine, a new kind of ATPase/ATPsynthase inhibitor, was studied in mammalian mitochondria. (i) Almitrine inhibits oligomycin-sensitive ATPase - it decreases the A T P / O value without any change in A/~H+; (ii) almitrine increases the mechanistic H + / A T P stoichiometry of ATPase/ATPsynthase; (iii) almitrine-induced changes in H +/ATPase stoichiometry depend on the flux magnitude through ATPase. These results are discussed in terms of the following interdependent parameters: flux value, force, pump efficiency and control coefficient.

Introduction Within the framework of the chemiosmotic theory, the proton electrochemical potential difference, named A/2H+, acts as coupling intermediate between proton pumps embedded in an energy-transducing membrane [1]. Mechanisms as well as thermodynamic control of oxidative phosphorylation have been studied, widely leading to considerable interest in the nature of relationships between flow (respiratory or phosphorylation rates) and forces (A/2H+, free energies of redox or of phosphorylation reactions, i.e., AGo/R or AGp). Since Co~espondence: M. Rigoulet, Institut de Biochimie Cellulaire et Neurochimie du CNRS, Universit~ de Bordeaux 2, 1 rue Camille Saint-Sa~ns, 33077 Bordeaux Cedex, France. Abrreviations: BHM, bovine heart mitochondria, CCCP, carbonylcyanide m-chlorophenyl hydrazone; A pH, transmembrane difference of pH; A~, transmembrane difference of electrical potential; A/XH+, transmembrane difference of proton electrochemical potential; HPLC, high-performance liquid chromatography; YRLM, rat liver mitochondria isolated from 6-week-old rats; ARLM, rat liver mitochondria isolated from 10-week-old rats; TET, triethyltin chloride; AEh, difference in redox potential across the respiratory chain.

such reactions are generally linear and sometimes symmetrical, they have been interpreted in terms of phenomenological linear non-equilibrium thermodynamics [2-4]. It is usually assumed that flow force relationships are independent of the manner in which forces vary, i.e., at a constant AEh, a given reduction in A/~H+ could be associated with a given increase in respiratory rate, whatever the means utilized. However, this prediction is rarely observed. For instance, in rat liver mitochondria, most authors have found that the onset of ATP synthesis is accompanied by a diminution in a/~n+, which is less than that observed when respiration is ihcreased to the same extent with an uncoupler [5-9]. On the other hand, a unique relationship has recently been observed [10]. To explain this discrepancy, it has been proposed that: (i) some experimental conditions inducing systematic deviations in A/~H+ evaluations are a source of erroneous relationships [10,11]; (ii) delocalized A/2H+ is not, in fact, the direct free-energy intermediate and alternative models are necessary (see several proposi. tions in Ref. 12-16). However, the assumption that the linearity in the flow force relationships coincides with the 'Onsager-domain' of non-equilibrium is still in question [17]. In the absence of any indispensable theo-

0005-2728/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

186 retical justification, a 'black box' description of oxidative phosphorylation (in terms of near equilibrium domain) is not satisfactory. The aim of this paper is to gain further insight into the interdependence between flux value, force size, pump efficiency and control distribution within the metabolic network.

tion was monitored with a potassium-sensitive electrode. At a given steady state, oxygen consumption or ATPase activity were measured and the charge efflux catalyzed by this pump was determined from the initial rate of K + efflux following addition of a specific inhibitor to this pump - either 0.2 /~M antimycin or 5 #M myxothiazol (respiratory chain) and 100/~M oligomycin or 100/xM TET (ATPase), respectively.

Materials and Methods

Mitochondria preparation Cell culture conditions, mitochondria preparation and other methods used in yeast mitochondria studies have been detailed elsewhere [19-21]. Bovine heart mitochondria were isolated according to a mechanical method [22]. Male Wistar rats were killed by cervical dislocation and their liver mitochondria were isolated according to Ref. 23. The protein concentration was estimated by the biuret method using bovine serum albumin as a standard. Respiration assay and A TP / O measurement The oxygen consumption rate was measured polarographically at 30 °C using a Clark electrode connected to a microcomputer giving an on-line display of rate values. Two different respiratory media were used (i) for bovine heart mitochondria: 0.25 M sucrose, 1 mM EGTA, 5/~M rotenone, 10 mM succinate, 3 mM Tris-Pi, 10 ~tM RbC1, 0.2 nmol valinomycin per mg protein and 10 mM Tris-HC1 (pH 7.2) and (ii) for rat liver mitochondria: isolation medium (see above) supplied with 10 mM succinate, 5 /~M rotenone, 10 mM Tris-Pi, 10 #M RbC1 and 0.2 nmol valinomycin per mg protein (pH 7.2). ATP/O stoichiometries were determined as described in Ref. 21. Measurements of ApH, Aq and Pi, ADP and A T P concentrations Matrix space, ApH and AXO were determined as previously described [20] with a slight modification: ApH was estimated from BHM and RLM, using the distribution of [14C]DMO (5,5-dimethyl[2-14C]oxazoli dine-2,4-dione) instead of [3H]acetate. ATP and ADP were measured in protein-free neutralized extracts by HPLC [21], and Pi was measured according to Summer [24]. Determination of K + / 0 or of K +/A TP The technique for measuring the electrical charge upon O (K+/O) and upon ATP (K+/ATP) ratios at steady state was as described by Murphy and Brand [25]. This method is based on strict equality of the rate of charge efflux from mitochondria, catalyzed by a proton pump and the involved carriers (i.e., ADP/ATP exchange) to the charge influx, catalyzed by valinomycin, at a given steady state. The potassium concentra-

Mammalian mitochondria ATPase activity was estimated by HPLC analysis of protein-free neutralized extracts [21]. Almitrine ((bisallylamino-4,6-S-triazinyl-2)-I-(bis-p- fluorobenzhydryl)-4piperazine(bismethanesulfonate)) was a gift from Servier Laboratory, France. Results and Discussion

Flow/force relationships depend on the kinetic properties of the components of oxidative phosphorylation We have already presented variations of redox states of cytochromes, control on respiration rate by cytochrome-c oxidase and protonmotive force size as a function of the respiratory rate of yeast mitochondria under three different types of steady-state [21]. (a) Respiration is stimulated by ATP synthesis with ethanol as respiratory substrate. This has been performed at a saturating ADP concentration and different concentrations of Pi (conditions 1). (b) The respiratory rate is stimulated by a proton-leak increase which is obtained by adding different amounts of CCCP in the presence of oligomycin and carboxyatractylate, for blocking ATP synthesis and the A D P / A T P carrier respectively (conditions 2). (c) Experimental conditions as in (b) except for the addition of 0.4 mM ATP (conditions 3). Under conditions 1, cytochromes a+a3, highly oxidized in state 4, remained poorly reduced when the respiratory rate increased. In contrast, under conditions 2, when the respiratory rate arose, the reduction level of cytochrome-c oxidase largely increased. The kinetic behaviour of cytochrome-c oxidase depends of the way in which the oxygen consumption flux is modulated: uncoupling leads to a large increase of its reduction level (up to 30%), whereas during phosphorylation, and at the same respiratory rate, cytochrome-c oxidase remains slightly reduced (less than 10%). Under conditions 3, the dependence of the cytochrome-c oxidase reduction level on the respiratory rate was the same as that observed under conditions 1. It should be noted that ATP addition did not change the respiratory rate whatever CCCP concentration was used. A theoretical framework, introduced by Higgins [26], reformulated later and developed by Kacser and Burns [27], and Heinrich and Rapoport [28], respectively, has

187

B

A

o

?z

o

~20G

2oo



O. c

E E 0

E

o

~0 100

I00

o

o

I

I

iO0

200

~

~p mV

o

l

0

I00

I 200

~p rnV

Fig. 1. Relationships between respiratory rate and protonmotive force in YRLM (A) and in ARLM (B). Measurements of respiratory rate (J0) and protonmotive force (Ap) were performed as described in Materials and Methods in standard medium with either: 0.5 mM ADP and various Pi concentrations (0.5-8 mM) (e); or 10 mM Pi, 5 #M carboxyatractyloside, 25 # g / m l oligomycin and different CCCP concentrations (0-0.25 #M) (•).

been used to quantify the relative contribution of different enzymatic steps to the control of oxidative phosphorylation in mammalian [29-32] and yeast [33] mitochondria. One of the main parameters in control analysis, CeJ, corresponds to the flux control coefficient of enzyme, Ej, on flux J. Kacser and Burns [27] have elaborated a possible determination of this parameter by inhibitor titration studies. This approach was applied for measuring the flux control coefficient of cytochrome-c oxidase using potassium cyanide as previously described [33]. Flux control coefficient of cytochrome-c oxidase increased only slightly when the respiratory rate was stimulated, either by ADP + Pi or by CCCP, in the presence of ATP (from 0.08 in state 4 to 0.15 at maximal rate). In contrast, this coefficient significantly increased up to 0.44, when the flux was modulated by CCCP, in the absence of ATP. Uncoupler depresses A/Ill+ m o r e than ADP + Pi at equivalent respiratory rates. However, in the presence of ATP, uncoupler addition leads to a flow/force relationship identical to that observed during oxidative phosphorylation. In conclusion, in yeast mitochondria, in parallel to the fact that the respiratory rate is not dependent on A/in+ only, the redox state of cytochromes is not dependent on the respiratory rate; indeed, at the same respiratory rate, cytochrome a + a 3 is more reduced in the presence of uncoupler than in the presence of ADP + Pr This indicates various types of kinetic behaviour in this complex. It is worth noting that for a given value of oxygen uptake, this higher degree of cytochrome a + a 3 reduction is paralleled by a lower value of A/in+. Under uncoupling conditions, added ATP changes the reduction level of cytochromes a + a3; this effect is due to external ATP, since it occurs when translocase is inhibited by carboxyatractylate. Our data support the

view that cytosolic ATP is a kinetic effector to cytochrome-c oxidase. If one assumes that ATP increases the enzymatic capacity of cytochrome-c oxidase without stimulating respiration, the following events should occur: (i) a decrease of the control of the respiratory rate by this complex; (ii) a change in the distribution of control in one or many other steps. Thus, a unique relationship between a flux and its associated forces can be expected only if the conditions used for modifying the forces do not change the kinetic structure of the system. Age-dependent change in the redox proton pump intrinsic uncoupling Liver mitochondria, isolated from rats aged either 6 weeks (YRLM) or 10 weeks (ARLM), were used. In YRLM, there was a unique steep-dependence of the respiratory rate on the magnitude of A/in+; in contrast, in ARLM, two relationships between respiratory rate and A/Ill+ w e r e observed: uncoupler decreases A/XH+ more than ADP + Pi at an equivalent respiratory rate. It should be noted that, under phosphorylation conditions, the dependence of the respiratory rate on A/Ill÷ was the same in both kinds of mitochondrion. Moreover, the A T P / O ratio was identical in both types of mitochondrion. Thus, the only difference observed concemed the response to CCCP, i.e., for the same respiratory rate obtained with the same concentration of CCCP, YRLM maintained a greater A/in÷ than ARLM. Two non-exclusive hypotheses can be proposed: (i) at a given CCCP concentration, the same H ÷ back-flow is obtained for a A/Ill÷ value which is greater in YRLM than in ARLM; (ii) CCCP could induce an intrinsic uncoupling of the respiratory chain as previously described by Luvisetto et al. [35]. In such a case, intrinsic

188 uncoupling must be greater in ARLM than in YRLM, at a given CCCP concentration• In order to test the first alternative, mitochondrial permeability to protons was determined as the initial rate of isoosmotic swelling of non-respiratory mitochondria [36]. Indeed, when mitochondria were suspended in an isoosmotic potassium acetate solution, in the presence of a non-limiting concentration of valinomycin, the swelling rate was only controlled by proton permeability. The relationship between the initial swelling rate and CCCP concentration was identical in YRLM and ARLM. The K + / O ratio was measured at different steady states, corresponding to different CCCP concentrations, according to Murphy and Brand [25]. In YRLM, the K * / O value was constant whatever the respiratory rate, indicating that intrinsic uncoupling of the respiratory chain did not increase• In contrast, in ARLM, the K ÷ / O ratio decreased drastically, whereas the CCCP concentration increased, indicating a reduction of the proton pumping efficiency of the respiratory chain. Thus, flow/force relationships appeared to be linked to properties of the redox proton pump. CCCP, which is a protonophore, induces a secondary effect on ARLM, i.e., it increases the degree of intrinsic uncoupling. In contrast to what is observed in yeast, ATP addition does not modify the flux/force relationships in ARLM. It would be of interest to measure the effect of ATP in redox proton pump slipping in yeast mitochondria (work underway). However, the mechanism by which some protonophoric uncouplers induce a molecular slipping of the respiratory chain remains an open question.

Flux-dependent almitrine-induced increase in stoichiometry of charge translocation by A TPase / A TPsynthase The mechanism of action of almitrine was investigated in three kinds of mitochondrion isolated from yeast, rat liver and bovine heart. Although RLM were more almitrine-sensitive (at least 10-times) than BHM or yeast mitochondria, this drug exhibited similar effects in all types of mitochondrion. The first study on yeast mitochondria [19] displayed the following features: (i) atmitrine at a concentration of 30 /~M inhibited oligomycin-sensitive ATPase activity with 50% inhibition; (ii) no direct effect on respiration could be evidenced; (iii) during oxidative phosphorylation, the ATP/O ratio decreased largely without any change in the magnitude of A/ill.; (iv) the higher the ATP synthesis and respiratory fluxes are, the larger is the almitrine-induced decrease in the ATP/O ratio. Thus, almitrine appears to act essentially on the ATPase complex, but its effect is very different from that of classical ATP synthase inhibitors such as oligomycin or aurovertin D, which decrease ATP synthesis and respiratory rates in such a manner that the A T P / O ratio remains constant. Recently, a new type of uncoupler of oxidative phosphorylation has been described in the litera-

ture. For instance, general anaesthetics, such as chloroform or halothane, inhibit ATP synthesis, stimulate mitochondrial ATPase activity and state 4 respiration like pure protonophoric uncouplers, but have no measurable effect on A/~n. magnitude [37]. Consequently, by comparing the effect of almitrine to that of different kinds of inhibitor and uncoupler, it appears clear that almitrine corresponds to a new type of mitochondrial energy-transduction inhibitor. Two hypotheses may be suggested for explaining the mechanism whereby almitrine decreases the A T P / O ratio: either (i) it increases intrinsic uncoupling of H÷-ATPase, also called slip [38,39]; or (ii) it changes the real H÷/ATP ratio of ATPase/ATPsynthase. A change in proton pump stoichiometry could be interpreted as an increase in slipping if coupling efficiency between proton flux and the chemical reaction is decreased, regardless of the direction of the reaction (i.e., ATP synthesis and ATP hydrolysis)• In contrast, experimental proof of real mechanistic change in the stoichiometry is provided by the observation of the same degree of stoichiometry in forward and in reverse chemical reactions• To investigate whether almitrine modifies the q+/ATP ratio of the ATPase/ATP synthase complex, we tested the action of this drug on the ATP-consuming process. A very simple system is the energy-dependent potassium salt uptake, which can be followed by valinomycin-induced mitochondrial swelling [40]. A1mitrine does not change the extent of swelling of BHM or of RLM either in potassium acetate or potassium phosphate when the proton pump is the respiratory chain. However, energy-linked swelling, supported by oligomycin-sensitive ATPase, is largely increased by almitrine. Moreover, almitrine slightly decreases ATPase activity. Thus, in the presence of almitrine, a lower rate of ATP hydrolysis leads to the maintenance of a greater transmembrane salt gradient, indicating a better efficiency of such a proton pump. This is corroborated by electrical charge/ATP ratio measurements. In experiments performed without almitrine, the K + / A T P value was 2.7 + 0.3; therefore, stoichiometry of the ATPase proton pump can be estimated to be 1.7 + 0.3. A 2-fold increase in this ratio was observed at an almitrine concentration of approx. 45 #M. It is generally proposed that ATP synthase works through a reaction cycle in which conforrnational transitions of the complex, corresponding to energy span, occur (see Ref. 41 for a review). In this view, the energy-dependent conformational cycle is coupled to a proton flux through the complex. As a working hypothesis, one may suggest that almitrine modifies at least a given state of enzyme into the cycle in such a way that the energy span between this state and the following one is increased. Indeed, in the presence of almitrine, more energy is needed for synthesizing 1 mol

189 of ATP, at a given A/~H+ , and reciprocally, more energy is delivered by hydrolyzing 1 mol of ATP. Whatever the mechanism may be by which almitrine acts on the complex, it is likely that inhibition of ATPase/ATP synthase activity is linked to the change in H+/ATP stoichiometry. Such dependence between the increase in the H+/ATP ratio and inhibition of ATPase activity could be explained in the context of the above-mentioned hypothesis, since an increase in the energy span between two steps of the cycle reduces the probability of a transition between the two steps.

Concluding remarks Oxidative phosphorylations cannot be considered as a system in which flows are only determined by thermodynamic force values. Indeed, flux, force, pump efficiency and kinetic control appear as interdependent parameters within the metabolic network. As an example, at a given force value, flux depends on pump efficiency and/or on kinetic structure. In turn, it has recently been reported that the efficiency of oxidative phosphorylation (ATP/O) is flux value-dependent, at a given force (A/2n+) [42]. It would be of great interest to investigate the physiological consequences of variable stoichiometry on cell adaptability to various types of constraint.

Acknowledgements The author is very grateful to Bernard Gu&in for stimulating discussions.

References 1 Mitchell, P. (1961) Nature (London), 191,144-148. 2 Caplan, S.R. (1971) Curr. Top. Bioenerg. 4, 1-65. 3 Rottenberg, H., Caplan, S.R. and Essig, A. (1970) in Membrane and Ion Transport (Bittar, E.B., ed.), Vol. 1, 165-191, John Wiley, New York. 4 Caplan, S.R. and Essig, A. (1977) Curr. Top. Membr. Transp. 9, 145-175. 5 Padan, E. and Rottenberg, H. (1973) Eur. J. Biochem. 40, 431-437. 6 Azzone, G.F., Pozzan, T., Massari, S. and Bragadin, M. (1978) Biochim. Biophys. Acta 501, 296-306. 7 Holian, A. and Wilson, D.F. (1980) Biochemistry 19, 4213-4221. 8 Kuster, U., Letko, G., Kunz, W., Duszynsky, J., Bogucka, K. and Wojtczak, L. (1981) Bioehim. Biophys. Acta 636, 32-38. 90'Shea, P.S. and ChappeU, J.B. (1984) Biochem. J. 219, 401-404. 10 Woelders, H., Putters, J. and Van Dam, K. (1986) FEBS Left. 204, 17-21.

11 Woelders, H., Van der Velden, T. and Van Dam, K. (1988) Biochim. Biophys. Acta 934, 123-134. 12 Williams, R.J.P. (1961) J. Theor. Biol. 1, 1-17. 13 Rottenberg, H. (1978) FEBS Lett. 94, 295-297. 14 Kell, D.B. (1979) Biochim. Biophys. Acta 549, 55-99. 15 Westerhoff, H.V., Melandri, B.A., Venturoli, G., Azzone, G.F. and Kell, D.B. (1984) Biochim. Biophys. Acta 768, 257-292. 16 Slater, E.C., Berden, J.A. and Herweijer, M.A. (1985) Biochim. Biophys. Acta 811, 217-231. 17 Wanders, R.J.A. and Westerhoff, H.V. (1988) Biochemistry 27, 7832-7840. 18 Gu6rin, B., Labbe, P. and Somlo, M. (1979) Methods Enzymol. 55, 149-159. t9 Rigoulet, M., Ouhabi, R., Leverve, X., Putod-Paramelle, F. and Gu6rin, B. (1989) Biochim. Biophys. Acta 975, 325-329. 20 Beauvoit, B., Rigoulet, M. and Gu6rin, B. (1989) FEBS Lett. 244, 255-258. 21 Rigoulet, M., Gu6rin, B. and Denis, M. (1987) Eur. J. Biochem. 168, 275-279. 22 Azzone, G.F., Colonna, R. and Ziche, B. (1979) Methods Enzymol. 55, 46-50. 23 Klingenberg, M. and Sleneska, W. (1959) Biochem. Z. 331, 486495. 24 Summer, J.B. (1944) Science 100, 413-418. 25 Murphy, M.P. and Brand, M.D. (1987) Nature 329, 170-172. 26 Higgins, JJ. (1965) in Control of Energy Metabolism (Chance, B., Estabrook, R.W. and Williamson, J.R., eds.) pp. 13-46, Academic Press, New York. 27 Kacser, H. and Bums, J.A. (1973) in Control of Biological Processes (Davies, D.D., ed.), pp. 65-104, Cambridge University Press~ London. 28 Heinrich, R. and Rapoport, T.A. (1974) Eur. J. Biochem. 42, 89-95. 29 Groen, A.K., Wanders, RJ.A., Westerhoff, H.V., Van der Meer, R. and Tager, J.M. (1982) J. Biol. Chem. 257, 2754-2757. 30 Gelterich, F.N., Botmensack, R. and Kunz, W. (1983) Biochim. Biophys. Acta 722, 381-391. 31 Tager, J.M., Wanders, R.J.A., Groen, A.K., Kunz, W., Bohnensack, R., Kuster, U., Letko, B., Bohne, G., Duszynski, J. and Wojtczak, L. (1983) FEBS Lett. 151, 1-9. 32 Doussi&e, J., Ligeti, E., Brandolin, G. and Vignais, P.V. (1984) Biochim. Biophys. Acta 766, 492-500. 33 Mazat, J.P., Jean-Bart, E., Rigoulet, M. and Gu6rin, B. (1986) Biochim. Biophys. Acta 849, 7-15. 34 Reference deleted. 35 Luvisetto, S., Pietrobon, D. and Azzone, G.F. (1987) Biochemistry 26, 7332-7338. 36 NichoUs, D.G. and Lindberg, O. (1973) Eur. J. Biochem. 37, 523-530. 37 Rottenberg, H. (1983) Proc. Nail. Acad. Sci. USA 80, 3313-3317. 38 Pietrobon, D., Azzone, G.F. and Walz, D. (1981) Eur. J. Biochem., 117, 389-394. 39 Pietrobon, D., Zoratti, M. and Azzone, G.F. (1983) Biochim. Biophys. Acta 723, 317-321. 40 Chappell, J.B. and Crofts, A.R. (1965) Biochem. J. 95, 393-402. 41 Senior, A.E. (1988) Physiol. Rev. 68, 177-231. 42 Ouhabi, R., Rigoulet, M. and Gu6rin, B. (1989) FEBS Lett. 254, 199-202.

Control processes in oxidative phosphorylation: kinetic constraints and stoichiometry.

Control processes in oxidative phosphorylation have been studied in three experimental models. (1) In isolated yeast mitochondria, external ATP is a r...
502KB Sizes 0 Downloads 0 Views