Eur. J. Biochem. 209, 475-481 (1992) 0 FEBS 1992

Characteristics of energy-linked proton translocation in liposome reconstituted bovine cytochrome bcl complex Influence of the protonmotive force on the H + /e- stoichiometry

Tiziana COCCO, Michele LORUSSO, Marco D1 PAOLA, Michele MINUTO and Sergio PAPA

Institute of Medical Biochemistry and Chemistry, University of Bari, Italy

(Received May 18, 1992) - EJB 92 0680

A study is presented on the H'je- stoichiometry for proton translocation by the isolated cytochrome bc, complex under level-flow and steady-state conditions. An experimental procedure was used which allows the determination of pure vectorial proton translocation in both conditions in a single experiment. The results obtained indicate an H'/e- ratio of 3 at level-flow and 0.3 at steady-state. The ratios appear to be independent of the rate of electron transfer through the complex. Making use of pyranine-entrapped bcl vesicles, a respiration-dependent steady-state ApH value of 0.4 was determined in the presence of valinomycin. This value could be either decreased by sub- saturating concentrations of the uncoupler carbonyl cyanide m-chlorophenylhydrazone (CCCP) or increased by introducing bovine serum albumin in the assay mixture. The steady-state H' je- ratio appeared to be in linear inverse correlation with the ApH. This indicates that ApH exerts a control on the proton pump of the bcl complex at the steady state.

The effect of valinomycin-mediated potassium-diffusion potential on electron-transfer and proton- translocation activities is also shown. The experiments presented show that the H ' je- ratio is unaffected, both at level flow and steady state, by an imposed diffusion potential up to around 100 mV. At higher potential values the level-flow H ' /e- ratio slightly decreased. Measurements as a function of imposed membrane potential of the rate of electron transfcr at level flow and of the rate of the pre-steady-state reduction of b and c1 cytochromes in the complex indicate activation of electron transfer at potential values of 40 - 50 mV. This activation appears, however, to involve a rate-limiting step which remains normally coupled to proton translocation.

The cytochrome bcl complex of mitochondria catalyzes the transfer of reducing equivalents from quinol to cytochrome c. Electron transfer is coupled to vectorial trans- location of two protons from the matrix (negative phase) to the outer space (positive phase) for every two electrons transferred by the complex. Two additional protons, formally deriving from the scalar oxidation of quinol, are released in the positive phase, thus giving an overall H '/e- stoichiometry of 2. The stoichiometric H'je- coefficient n for vectorial proton translocation was invariably found to be 1 under level- flow conditions, that is under conditions of negligible trans- membrane A p [I - 51. Under these conditions, however, a de- crease of the H+/e- ratio, referred to as the decoupling effect, was produced by treating the isolated bcl complex with dicyclohexylcarbodiimide (cHxN),C [5 - 81 or by controlled

Correspondence to S . Papa, Institute of Medical Biochemistry and Chemistry, University of Bari, Piazza G. Cesare, 1-70124 Bari, Italy

Abbreviations. (cHxN)& dicyclohexylcarbodiimide; DQHZ, duroquinol (durohydroquinone); FeS, Rieske iron-sulphur protein; CCCP, carbonyl cyanide m-chlorophenylhydrazone; dpH, trans- mcmbrane pH gradient; dy, transmembrane electrical potential gradient; Ap, transmcmbrane protonmotive force; HbO,, oxy- hemoglobin.

Enzymes. Ubiquinol -cytochrome-c reductase, bc, complex (EC 1.10.2.2); ferrocylochrome-c: oxygen oxidoreductase, cytochrome c oxidase (EC 1.9.3.1).

proteolytic digestion of its polypeptide subunits [9]. A decrease of the H'je- stoichiometry was also reported to result from exposing the bcl complex isolated from Neurospora crassa and reconstituted in liposomes to a valinomycin-induced potassi- um-diffusion potential [lo]. Conflicting reports have appeared on the effect of Ap on the H' je- ratio for proton translocation by the hcl complex in situ at the steady state [I1 -151. The reported experiments were carried out either in the absence [I1 - 131 or in the presence [14] of the ApH component of the protonmotive force, so that the influence of ApH per se on proton translocation at the steady state was not settled. In this paper a systematic analysis of the H'/e- stoichiometry of the bel complex isolated from bovine heart mitochondria and reconstituted in liposomes is presented. The results show that the H'/e- stoichiometry for vectorial proton translo- cation in liposome-reconstituted bovine bcl complex is 1 (overall H'/e- ratio = 2) under level-flow conditions and appears to be independent either of the rate of electron transfer through the complex or of extended values of transmembrane potential ( A iuy) gencrated by valinomycin-mediated potassi- um diffusion. At the steady state, under conditions where ApH represented the only component of Ap, the H'je- ratio for vectorial H + translocation decreased to about 0.3. Results are also presented on the influence of a valinomycin-induced potassium-diffusion potential on the pathways of electron transfer within the complex.

476

MATERIALS AND METHODS

Preparation of cytochrome c reductase and cytochrome c oxidase complexes

The cytochrome c reductase and cytochrome c oxidase complexes were isolated from bovine-heart mitochondria as described by Rieske [16] and by Errede et al. [17] respectively.

Preparation of bcl vesicles

Reconstitution of the bc, complex into phospholipid ves- icles was performed by the cholate dialysis method of Leung and Hinkle [2]. Acetone-washed soybean phospholipids (30 mg) were sonicated in 1 ml 0.1 M K+/Hepcs pH 7.4, 56 mM KCI and 2% potassium cholate. Purified bc, complex (2 mg protein), was then added and the mixture dialyzed against 125 mlO.1 M K'iHepes pH 7.4 and 56 mM KC1, with two changes in 24 h. For proton translocation experiments, the final 2-h dialysis was performed against 1 mM Kf/Hepes, pH 7.4 and 99.6 mM KC1. Where required, the potassium activity of the bc, vesicle suspension was lowered by per- forming the overnight dialysis step against 0.01 M K+/Hepes pH 7.4 and 96 mM KC1 followed by the final 2-h dialysis against 0.001 M K+/Hepes pH 7.4, 1 mM KCl and 0.2 M sucrose. After dialysis the vesicles were passed through a Sephadex G-25 column equilibrated with the medium of the last dialysis step. Only the bulk elution phase was then used in the experimcnts. For the preparation of pyranine-entrapped vesicles, 2 mM pyranine (8-hydroxy-l,3,6-pyrene trisul- phonate) was added to the sonicated detergent/phospholipid/ protein mixture and was also present at lhe same concen- tration in the medium of the first dialysis step.

Measurement of protonmotive activity

For the reductant pulse method [2- 51, spectrophoto- metric determination of cytochrome c reductase activity at 550 - 540 nm and electrometric determination of proton translocation were carried out simultaneously on the same sample in a spcctrophotometric cuvette equipped with a com- bination glass electrode.

For combined methods, the bel vesicles were supplement- ed with duroquinol, cytochrome c and soluble cytochrome oxidase. H + translocation was measured either potentio- metrically with a combination glass electrode or spectrophoto- metrically at 558 - 593 nm by following the signal of externally added phenol red. Oxygen uptake was measured either (a) electrometrically with a Clark oxygen electrode (4004 YSI, Yellow Spring, OH) coated with a high-sensitivity membrane (YSI 57776) in a thermostatically controlled (25 T) all-glass cell also housing the glass electrode or (b) spectrophoto- metrically, following the deoxygenation of human oxyhemo- globin at 577-568 nm ( E = 6.3 mM-' . cm-') as described [18, 191. The correction factor cf, [18, 191 for the Hb prep- aration used was 2.46.

Fluorescence measurements

Respiration-dependent A pH generation was measured in pyranine-containing br, vesicles as described in [20]. After dialysis, the bc, vesicles were passed through a Sephadex G-25 column preequilibrated with the same medium used for rcconstitution and dialysis to remove any external fluorescent dye. Changes of pyranine fluorescence were monitored with

a Perkin-Elmer 650 fluorescence detector. The excitation and emission wavelengths were 460 nm and 520 nm, respectively.

The membrane potential generated by the respiring bcl vesiclcs was monitored following the fluorescence quenching of externally added 5 pM safranin at the excitation and emission wavelengths of 525 nm and 575 nm, respectively [21]. The reaction mixture and the experimental conditions wcre those used for ApH determination.

Potassium-diffusion potential measurements

The potassium-diffusion potential was generated by adding valinomycin to potassium-loaded hc, vesicles sus- pended in a medium containing varying concentrations of KC1. The potential was calculated from the initial K + concen- trations inside and outside the vesicles as described in [lo]. The spectral shift of safranin measured spectrophotometrically at 526- 560 nmwas used as an indicator of the applied potential. A linear relationship was obtained between the calculated membrane potential and the maximal absorbance difference [221.

Spectrophotometric determination of redox changes of the cytochromes

Redox changes of the h and c1 cytocromes were simulta- neously followed at 10 "C with an air-turbine spectrophoto- meter at wavelength couples of 562 ~ 540 nm, 566 - 540 nm and 550 - 540 nm.

Chemicals

Horse heart cytochrome c (type VI), antimycin. valino- mycin, nigericin, soybean phospholipids, carbonyl cyanide m- chlorophenylhydrazone (CCCP) and safranin 0 were from Sigma Chemical Co. ; myxothiazol from Boehringer Mann- heim; duroquinol from K. & K. Laboratories; pyranine from Eastmann Kodak Co. All other reagents were of the highest purity grade commercially available.

RESULTS

In the reductant pulse method, as generally used [2-51, proton translocation coupled to electron flow in bcl vesicles under level-flow conditions is elicited by the addition of quinol to the vesicles. For each electron transferred by the complex from quinol to ferricytochrome c two protons are releascd, one of which is suppressed by uncouplers. We have developed a different procedure (Fig. 1A) in which bcl vesicles are sup- plemented with duroquinol and soluble cytochrome c oxidase (combined method). Electron flow, initiated by the addition of ferricytochrome c, results in oxygen consumption and un- coupler-sensitive acidification of the external medium which attains a steady-state level in about 0.5 min. Once the system becomes anaerobic or upon addition of a bc, complex inhibi- tor, interruption of clcctron flow is accompanied by proton back-flow into the vesicles. By this procedure, which allows measurement of pure vectorial proton translocation (scalar protons released in the oxidation of quinol are taken up in the reduction of oxygen to H 2 0 by the soluble oxidase added in the external medium), the H+/e- ratio can be estimated in the same experiment both at level flow, from the initial rates of proton ejection and oxygen consumption, and at steady state from the initial rates of proton back-flow, ensuing upon intcr-

477

A

0.4pMcyt c #

I

T -I 6 s I- 25, 48pM H '

B

Argon

deox ygenation -3 517 0087 - 568 A nm / 4

Phenol red

558-593nm t 130

-I 10s I-

6.5yM HCI

Fig. 1. Measurement of H+/e- ratio for vectorial proton translocation in bc, vesicles. hc, vesicles (final concentration 0.8 pM cytochrome cl) were suspended in 1.5 m12 mM Hepes pH 7.4 and 100 mM KCI, also containing 2 pg valinomycin, 0.4 pM soluble cytochrome oxidase and 300 pM duroquinol. The reaction was started by the addition of 0.4 pM cytochrome c. In (A), proton translocation and oxygen con- sumption were simultaneously followed by glass and Clark-type elec- trodes, respectively, adapted to a thermostatically controlled (25 "C) gas-tight cell. The oxygen concentration of bc, vesicle suspension was lowered by a stream of argon to a concentration of about 35 pM before ferricytochrome c addition. In (B), separate experiments are shown for oxygen consumption and proton translocation measure- ments, in which the reaction mixture also included 25 pM human hemoglobin and 50 FM phenol red, respectively. The oxygen concen- tration was lowered by a stream of argon until absorbance changes at 577-568 nm showed that Hb was 50% deoxygenated; 1 min later respiration was started. The H'je- ratios at level flow were calculated from the initial rates of proton translocation and electron transfer upon addition of cytochrome c. The H+/e- ratios at steady state were calculated rrom the initial rate of proton back-flow ensuing upon interruption of respiration and the steady-state oxygen consumption rate exhibited just belixe respiration was stopped. Where indicated, 1.2 pM antimycin was added. Figures on the traces represent rates of proton translocation and electron transfer as p M . min For other experimental conditions and details see under Materials and Methods.

ruption of electron flow, and the steady-state rate of oxygen consumption exhibited before respiration is stopped. The values of H ' /e- at level flow and at steady state were, in the measurements illustrated in Fig. 1, around 1.0 and 0.3 respectively. In the presence of 0.5% bovine serum albumin, which reduces passive proton leakage in the mitochondria1 membrane [I 31, the steady-state H+/e- ratio further decreased to around 0.2 (see below) without any effect on the flux rates and ratio at level flow. The H'je- ratio measured at the

ratic I

'"I d? h A * A A %

0.5 0 0 i. 100 m 3N 400

Respiratory rate I p M e - min-1)

Fig. 2. Analysis of the H+/e- ratios as a function of the rate of electron transfer in bcl vesicles. In the reductant-pulse-type experiment (m) bcl vesicles (0.67 pM cytochrome el) were suspended in 2 mM Hepes pH 7.4, 100 mM KCI, 7.5 pM ferricytochrome c, 1 mM KCN and 2 pg valinomycin. Final volume 1.5 ml. The reaction was started by the addition of duroquinol, whose concentration varied over 11 - 100 pM. For the combined method the experimental conditions are those described in the legend to Fig. 1A. The concentration of soluble cytochrome oxidase varied over 0.08-0.8 pM. The level-flow (U) and steady-state ( A 7 A) H+/e- ratios were determined as described in the text and in the legend to Fig. 1. The respiration was stopped by allowing the system to become anaerobic (A) or by the addition of 1.2 pM antimycin (A).

steady state was independent of the method used to stop the respiration. Antimycin was routinely used to interrupt electron flow, instead of either anaerobiosis, which requires pre-lowering of the oxygen concentration, o r myxothiazol whose inhibitory effect took several seconds to develop. Anti- mycin is reported to have an uncoupling effect [23] ; separate controls, in which a potassium-diffusion potential was im- posed and the rate of pH change in the vesicle suspension was measured, showed, however, that antimycin at the concen- trations used did not affect the rate of proton diffusion. It should be noted that even if an uncoupling effect were exerted by antimycin, this would tend to artifactually enhance the rate of proton back-flow with consequent over-estimation of the H+,/e- ratio. In parallel experiments respiratory rates were determined spectrophotometrically, following HbOz deoxy- genation, and H + translocation was also monitored spectro- photometrically with phenol red on a separate sample (Fig. 1B). The H+/e- ratios obtained with the spectropho- tometric measurements under level-flow and steady-state con- ditions, agreed substantially with those obtained in these ex- periments with the potentiometric measurements. Further- more the same results were obtained when proton translo- cation was followed spectrophotomelricdy and electron transfer measured simultaneously with a Clark electrode adapted to the spectrophotometric cuvette (not shown).

Fig. 2 presents data from three different sets of exper- iments where the H+/e- ratio was measured as a function of the rate of electron flow. For reductant pulse experiments the rate was varied by changing the duroquinol concentration, for the combined method the rate of electron flow was changed by varying the activity of added cytochrome oxidase. Differently from what was found with cytochrome oxidase vesicles (cf. [19]), the H+/e- ratio either under Icvel-flow conditions or at the steady-state did not vary with the rate of electron flow, at least within the span examined. The H'je- ratios for H' release were constantly 2 and 1 at level flow (reductant pulse and combined method, respectively) and 0.3 at the steady state (combined method).

478

t c y t c C

I 0; 0 8 16 24 0 8 16 21

CCCP(nM) CCCP ( n M )

Fig. 3. Effect of CCCP on respiration-linked proton translocation and H+/e- ratios in bc, vesicles. The experimental conditions are those described in the legend to Fig. IA. (A) Control (-); +4 nM CCCP (----); (B) initial rates of electron transfer (0 ) and proton translocation (0) under level-flow conditions; (C) level-flow ( A ) and sleady-state ( A ) H+/e- ratios. Antimycin (1.2 pM) was used to stop rcspiration.

Influence of ApH on the H+/e- ratio

In Fig. 3 the effect of sub-saturating concentrations of CCCP (cdrbonyl cyanide m-chlorophenylhydrazone) is shown. Lower concentrations of uncoupler (up to 6 nM), while ineffective on the initial rates of proton ejection and electron flow at level flow (Fig. 3 A, B), caused a decrease of the steady-state extent of proton release and a definite increase of the rate of proton re-entry ensuing upon addition of anti- mycin. Under these conditions the H+/e- ratio measured at level flow (with the combined method) remained equal to 1, whereas the steady-state H' je- ratio increased from around 0.3 in the control to values almost double (Fig. 3 C). Concen- trations of nigericin ranging over 15 - 50 ng/ml produced simi- lar results (not shown). At higher concentrations of CCCP, a decline of the Ht /e - ratios both at level flow and steady state occurred (Fig. 3 C). An analogous experiment in which respiration was stopped by allowing the system to become anaerobic gave superimposable results (not shown).

Fig. 4 shows the measurement of transmembrane ApH generated by respiration in hc, vesicles. Initiation of electron flow by the addition of cytochrome c to hcl vesicles sup- plemented with duroquinol and soluble cytochrome oxidase resulted in generation of ApH (alkaline inside) as revealed by increased fluorescence of pyranine entrapped into the bcl proteoliposome. 'The magnitude of the dpH at the steady state was less than 0.1, increased to about 0.4 in the presence of valinomycin and was completely abolished by the K +/H +

exchanger nigericin. The respiration-dependent transmem- brane potential, as revealed by the fluorescence signal of safranin, appeared to be completely collapsed by the concen- tration of valinomycin used in the present experiments (not shown). The ApH generation followed first-order kinetics and no appreciably fast fluorescence increase of pyranine was seen upon addition of an amount of KOH which caused the extra- vesicular pH to vary by 0.2 (cf. [20]).

In the experiment illustrated in Fig. 5 the correlation be- tween the steady-state ApH and the H+/e- ratios was exam-

N;g I "Ialinomycin

t cyt c

i 20s I-

Fig. 4. Internal akalinization in respiring pyranine-containing be, ves- icles. Pyranine-containing bc, vesicles (0.15 pM final concentration cytochrome cl) were suspended in the same medium used for sonica- tion and dialysis (0.1 M Hepes, 56 mM KCl, pH 7.4) also containing 0.1 pM soluble cytochrome oxidase and 200 pM duroyuinol. The reaction was started by the addition of 0.16 pM ferricytochrome c. Final volume 3.5 ml. Where indicated, 2 Fg valinomycin and 1 pg nigericin were added. Fluorescence changes were calibrated with microliter additions of 6 M KOH in the presence of both valinomycin (Val) and nigericin (Nig).

h

J - 1 2 0 s k

H%- ra t to

B ''I

0 0.2 0.4 0.6

JIJH

Fig. 5. Relationship between steady-state ApH and H+/e- ratio in bcl vesicles. (A) The respiration-linked transmembrane ApH generation was followed as described in the legend to Fig. 4 in the presence of valinomycin. (a) Control; (b) + 4 n M CCCP; (c) +0.5% bovine serum albumin. Respiration was stopped by adding 0.2 pM antimycin. (B) The steady-state H'/e- ratios were measured as shown in Figs 1A and 3 in the absence ( 0 ) or in the presence of either CCCP (1 - 6 nM) (El) or 0.5% bovine serum albumin (0). The Hi/e- ratios are plotted as function of the dpH values determined under thc samc conditions.

ined. The extent of steady-state ApH was modulated by in- troducing in the medium either 0.5% bovine serum albumin or low concentrations of CCCP (1 - 6 nM) giving a decrease of the steady-state ApH without appreciably affecting the rate of respiration-dependent ApH generation. In the experiments in which bovine serum albumin was present, the concentration of valinomycin was doubled since the ionophore could be partly bound by albumin [13]. Separate controls showed that albumin did not limit the valinomycin-mediated K+ counter migration. Albumin increased and CCCP decreased the steady-state ApH value causing opposite changes of the ap- parent rate of dpH collapse ensuing upon addition of anti- mycin. Results similar to those produced by CCCP were also obtained with sub-saturating concentrations of nigericin (not shown). Fig. 5B shows a statistical evaluation of the steady- state H'je- ratio as a function of dpH. An inverse linear correlation does appear to exist between the steady-state

479

I I

0 fiu 120 180 ~ l 'P ( rnW1

Fig. 6. Influence of valinomycin-mediated potassium-diffusion potential on proton translocation and H+/e- ratios in bel vesicles. Proton trans- location coupled to electron transfer was followed either with the combined method (see Fig 1A) or with the reductant pulse method In the combined method experiment, bcl vesicles (08 pM cytochrome~~) were suspended in 2 m M Hepes pH 7.4, 300pM d iroquinol, 0.4 pM soluble cytochrome oxidase and different concen- tration rdtios of 100 mM KCl and 100 mM LiCl which resulted in membrane potential values indicated in the figurc. Then 2 pg valinomycin was added, followed by ferricytochrome c addition Res- piration was stopped by the addition of 1.2 pM antimycin. Under conditions of induced dy, valinomycin addition caused a small deflec- tion of the pH trace towards alkalinity, which was taken into account to correct the iates of respiration-dependent proton translocation. Initial rates of proton translocation (0) and electron transfer ( 0 ) were determined electrometrically (see Fig IA). (n) Level-flow H + / e- rdtio, (A) steady-state H+/e- ratio In the reductant-pulse-typc experiment (see legend to Fig 2 and under Matenals and Methods) bc, vesicles (0 67 pM cytochrome c I ) werc suspcnded in 2 inM Hepes pH 7 4, 7 5 pM ferricytochrome L , 1 mM KCN and various concen- trdlion ratios of 100 mM KC1 and 100 mM LiCl ARm vahomycin (2 pg) addition, thc reaction was started by adding t 1 pM duraquinol (m) Level-flow H+/e ratio, measured with reductant pulses

dpH and the Ht/e- ratio, which by extrapolation gives at ApH = 0 a value close to 1. Analysis of a set of these exper- iments showed that, in the range of the respiration rates exam- ined in the experiments reported in Fig.2, the measured ApH did not vary significantly.

Influence of electrical membrane potential on proton translocation and redox reactions in the camplex

Addition of valinomycin to potassium-loaded bcl vesicles exposed to lower external concentrations of K + generated a diffusion potential (negative inside), which was monitored by measuring safranin absorbance changes [ 10,22J. The induced diffusion potential was stable for several minutes and decayed only slowly. Addition of nigericin caused a mpid and complete collapse of the potassium-diffusion potential. Fig. 6 rqcirts averaged data from three different expe rha t s in which the influence of valinom ycin-induced potassium-diffusion poten- tial on electron flow and H+/e- ratio in bc, vesicles was examined. The initial rate of e- transfer at level flow (mea- sured by the combined method) was first slightly enhanced up to a value of diffusion potential of about 50 mV, then a small dccrease of the activity was observed. The proton translo- cation activity followed the reductase activity profile, being somewhat more inhibited at higher vahes of induced poten- tial. As a consequence, the level-flow H+/e- ratio, unaffected by diffusion Ay up to 90 mV, showed a small decline at the higher values of induced potential. Analogous experiments

562-540nm I f - - - - - - - /f

t

t DOHL

Ant A i 5 s c

Fig. 7. Influence of potassium-diffusion membrane potential on pre- steady-state reduction kinetics of b and c , cytochromes. bcl vesicles (0.7 pM cytochrome el) were suspended in 2 mM Hepcs pH 7.4, 1 mM KCN and various concentration ratios of KC1 and LiCl to give the indicated potentials. Final volume, 1.8 ml. Valinomycin (2 pg) was added, followed by 33 pM duroquinol, 0.1 pM ferricytochrome c, 3 pM antimycin and 40 pM ferricyanide. (-) 0 mV; (----) 102 mV.

carried out with the reductant pulse method indicated, simi- larly, a decay of the level-flow H+/e-ratio from 2 in the range 0-90 mV to 1.82 0.04 (n = 6 ) at 180 mV. The steady- state H+/e- ratio was practically unaffected by the electrical membrane potential, ranging at all the values tested around 0.3.

The influence of potassium-diffusion Ayi on the redox reactions of the cytochromes within the complex was then tested. In Fig. 7 pre-steady-state reduction kinetics of h and ci cytochromes in the absence and in presence of applied diffusion A I ~ are shown. By increasing the potassium-diffusion membrane potential both the rate and the extent of reduction of b cytochromes increased, whereas the ferricyanide-induced antimycin-promoted extra-reduction of b cytochromes de- creased. The effect on cytochrome c1 reduction was, on the other hand, rather limited; its reduction rate was enhanced by 10 - 15% at a potential value of about 50 mV, then remained almost constant. In the presence of myxothiazol (Fig. 8A ~ C) the same effect of the membrane potential on the reduction of b cytochromes was observed, i.e. a progressive stimulation of both the rate and the extent of their reduction. By applying a mathematical procedure which allows the calculation of individual absorbance changes of cytochrome b566 and b562 [24], the pre-steady-state extent of reduction of b566 appeared to be mostly stimulated by the membrane potential (Fig. 8C). In the presence of antimycin (Fig. 8D), intermediate values of membrane potential (50 mV) stimulated appreciably the rate of reduction of b and c1 cytochromes, whereas at higher poten- tial values a decrease of b cytochrome reduction rate, mostly that measured at 562 - 540 nm, was observed. The extent of reduction of the cytochromes did not vary significantly.

DISCUSSION

Reconstitution of isolated bcl complex in liposomes allows measurement of the flux rates in a system whch is assumed to be free of secondary ion transport systems and, furthermore, exhibits a much lower proton conductance than that of the

480

A B !i- I

=L 0 0 0 20 40 60 Bo 100

ArP lmVl D 20 40 60 80 100

A@ ImY)

Fig. 8. Influence of potassium-diffusion membrane potential on pre- steady-state reduction of d and c1 cytochromes in the presence of site- specific inhibitors. hcl vesicles were suspended in the reaction mixture described in the legend to Fig. 7 with various concentration ratios of KCl and LiCl to give the indicated potentials. (A-C) Experiments in which 3 pM myxothiazol was present. (-) 0 mV; (----j 102 mV. The duroquinol-dependent absorbance changes measured at 562- 540 nm and at 566-540 nm (A, B) were used to determine the respective percentage reduction of the individual hemes b562 ( A ) and bSe6 (A) (C). In the experiment shown in (D) 3 pM antimycin instead ofmyxothiazol was present. Rates of absorbance changes on addition of duroquinol are reported measured at 562 - 540 nm ( j, 566 - 540 nm (0) and 550- 540 nm (0).

inner mitochondria1 membrane 125 - 271. The measurement of the H+/e- ratio has been carried out both at level flow and at the steady state. The procedure used (Fig. 1) allows determi- nation of thc H+/e- ratio both at level flow and steady state in a single experiment. At level flow the H+/e- ratio for vectorial H + translocation was found to be exactly 1 as is generally accepted. This provides an internal direct validation of the Hf / e - value measured at the steady state which amounted to about 0.3. In particular the lower value found for the steady-state H ' /e- ratio cannot be due to insufficient valinomycin-mediated K + counter-migration limiting H +

translocation at the steady state, since the same valinomycin concentration gave, at level flow, the expected stoichiometry of 1 H+/e- . Furthermore, doubling the valinomycin concen- tration did not cause the measured Hf /e- ratios to vary. The same results were obtained using spectrophotometric ap- proaches for determining the rates of H f translocation and electron transfer instead of the conventional potentiometric measurements (Fig. 1). The results obtained by using the spectrophotometric approach for the determination of proton translocation (the overall spectrophotometric response time is around 200 ms> rule out the possibility of significant delay between the inhibitor addition and the detection of proton influx into the vesicles. Finally it can be noted that the H + / e- ratio measured at the steady state is independent of the procedure used to stop respiration. The rcsults obtained indi- cate that the pH difference, set up across the membrane by respiration, controls the rate of proton translocation at the steady state causing a drop of the H+/e- ratio (cf. [15]). We have checked that ApH is the only component of Ap under our experimental conditions (valinomycin present) and found a value of 0.4. This value could be either decreased by sub-

saturating concentrations of CCCP or nigericin or increased by adding bovine serum albumin. In any case, a linear inverse correlation was found between the steady-state ApH and the H'je- ratio (Fig. 5).

It has been suggested [20] that an electroneutral H+/K+ exchanger is operating in the proteoliposome. This could cause a decrease of the steady-state dpH primarily established by respiration. Since, in the absence of A y , the H + influx is proportional to dpH (ohmic response), the lower the steady- state ApH is, the slower will be the H + influx, from which the steady-state H+/e- ratio is determined. The present obser- vations show, however, that under conditions in which the transmembrane ApH was decreased, the initial rate of H +

influx was enhanced. Conversely, upon enlarging ApH by bovine serum albumin addition, a significant decrease of the rate of H + influx and of the steady-state H'/e- ratio was observed. It is, therefore, evident that the effect of ApH on the steady-state H+/e- stoichiometry is unrelated to H+/K+ exchange in the proteoliposomal membrane but rather results directly from decoupling of the proton pump of the bcl com- plex.

It cannot be excluded that respiration would have pro- duced a larger ApH than that observed. One possibility of under-estimation is related to the experimental evaluation of ApH, since calibration of fluorescence changes of liposomal pyranine is done by adding externally alkali, in the presence of both valinomycin and nigericin, to all the vesicles in the sample including those eventually not incorporating the en- zyme. This, however, would not affect the conclusions present- ed.

When exposed to a valinomycin-mediated potassium-dif- fusion potential, the hcl complex exhibited a level-flow H '/ e- ratio fairly constant at 1 up to a potential values of around 100 mV, then the ratio decreased slightly. We cannot exclude that a higher degree of decoupling could occur at higher values of potential. In our experimental system a potassium gradient over more than three orders of magnitude could not be im- posed on the vesicles. These observations differ from those of Bechmann and Weiss [lo] who found a 50% drop of the H+/e- ratio in Neurospora bcl complex at applied diffusion potential around 50 mV. The reasons for this difference are not clear. It could reside in a peculiarity of the fungal complex or in the different experimental procedures used. These authors [lo] measured the H +/e- stoichiometry only with the reductant pulse method. That the imposed potential is acting on and felt by the bcl complex in our experiments is clearly demonstrated by its effect on thc pre-steady-state reduction kinetics of h and c1 cytochromes. Either in the absence or in the presence of myxothiazol the applied potential promoted the reduction of b cytochromes, mostly that of h566. This observation, besides indicating that the substrate duroyuinol feeds electrons to the complex mainly at the quinone reduction site [28], further confirms the localization at the positive side of cytochrome b5h6 [4, 291 whose reduction is promoted by the imposed potassium-diffusion potential. The reduction of b562 appears to be much less, if at all, stimulated by the imposed potential. This could be due to its suggested localiza- tion towards the negative side of the membrane [30]. The rate of electron transfer through the complex, together with the pre-steady-state reduction kinetics in the presence of anti- mycin, indicate an activation of clectron transfer at potential values of 40-50 mV (cf. [lo].) It is to be noted that the activation of electron transfer at low potential values appears to involve a rate-limiting step which is still coupled to the proton translocation process, since at these potential values

the H+/e- ratio is definitely unaffected. We can, tentatively, localize the activation step at the quinol oxidation site. A diffusion potential positive outside might promote migration of the semiquinone anion produced in the oxidation of sub- strate quinol towards the outer side of the membrane where it reduces cytochrome b566 [4,31]. The small inhibition exerted by higher values of imposed Ay on steady-state respiration could be due to depression of forward electron flow from heme b566 to heme b5,, and from the latter to the FeS center and cytochrome cl, this last step being directly involved in electrogenic proton pumping [4].

REFERENCES 1.

2.

3. 4.

5.

6.

7.

8.

9.

10.

11.

12. 13.

Papa, S. , Guerrieri, F., Lorusso, M., Tzzo G., Boffoli, D., Capuano, F., Capitanio, N. & Altamura, N. (1980) Biochem. 1. 192,203-218.

Leung, K. H. & Hinkle, P. C. (1975) J . Biol. Chem. 250, 8467- 8471.

Guerrieri, F. & Nelson. B. D. (1975) FEBS Left . 54, 339 - 342. Papa. S., Lorusso, M., Boffoli, D. & Bellomo, E. (1983) Eur. J .

Lorusso, M., Gatti, D., Boffoli, D., Bellomo, E. &Papa, S . (1983)

Degli Esposti, M., Saus, J . , Timoneda, J., Berloli, E. & Lenaz, G.

Nalecz, M. J . , Casey, R. P. & Azzi, A. (1983) Biochim. BiophyJ.

Clejan, L., Bosch, L. G. & Bcattie, D. S. (1984) J . Biol. Chem.

Lorusso, M., Cocco, T., Boffoli, D., Gatti, D., Meinhardt, S. W., Ohnishi, T. &Papa, S . (1989) Eur. J . Biochem. 179, 535-540.

Bechmann, Ci. & Weiss, IT. (1991) Eur. J. Biochem. 195, 431- 438.

Murphy, M. P. & Brand, M. D. (1988) Eur. J . Biochem 173,645 - 651.

Hafner, R. P. & Brand, M. D. (1991) Biochem. .I. 275,75-80. Luvisetto, S., Conti, E., Buso. M. & Azzone, G. F. (1991) J . Rid.

Biochem. 137,405-412.

Eur. J . Biochcm. 137,413 -420.

(1982) FEBS Lett. 147, 101 - 105.

Acta 724, 75-82.

259, 11169-11172.

Chem. 266,1034- 1042.

48 1

14. Brown, G. C. (1989) .I. Bid. Chem. 264, 14704-14709. 15. Rich, P. R. & Heathcote, P. (1983) Sixth International Congress

16. Rieske, J . S. (1967) Methods Enzymol. 10, 239-245. 17. Errede, B., Kamen, M. 0. & Hatefi, Y. (1978) Methods Enzymol.

18. Papa, S., Capuano, F., Markcrt, M. & Altamura, N. (1980) FEBS

19. Capitanio, N., Capitanio, G., De Nitto, E., Villani, G. & Papa,

20. Wrigglesworth, J. M., Cooper, C. E., Sharpe, M. A. & Nicholls,

21. Singh, A. P. & Nicholls, P. (1985) J . Biochem. Biophys. Metho&

22. Akerman, K. E. 0. & Wikstrom, M. K. F. (1976) FEBS Lett. 68,

23. Al-Shawi, M. K. & Brand, M. D. (1981) Biochem. J . 200, 534 -

24. Papa, S . , Lorusso, M., Izzo, G. & Capuano, F. (1981) Biochem.

25. Krishnamoorthy, G. & Hinkle, P. C. (1985) Biochemistry 23,

26. O’Shea, P. S., Petrone, G., Casey, R. P. & Azzi, A. (1984) Bio-

27. Dcamer, D. W. & Nicholls, J. W. (1983) Proc. Nut1 Acud. Sci.

28. Zweck, A., Beckmann, G. & Weiss. H. (1984) Eur. J . Biochem. 183, 199 -203.

29. Ohnishi, T., Schagger, H., Meinhardt, S. W., Lo Brutto, R., Link, T. A. & von Jagow, G. (1989) J . Biol. Chern. 264; 735-744.

30. Konstanlinov, A. A. & Popova, E. (1987) in Cytochrome system: molecular biology and bioenergetics (Papa, S . , Chance, B. & Ernster, L., cds) pp. 751 -765, Plenum Press, New York.

31. Papa, S.; Lorusso, M., Cocco, T., Boffoli, D. & Lombardo, M. (1990) in Highlights in ubiquinvne reseurch (Lenaz, G., Barnabci, O., Rabbi, A. & Battino, M., eds) pp. 122-135, Taylor & Francis Ltd, London.

on Photosynthesis, Abstracts pp. 363 - 366, Brussels.

52,40 - 47.

Lett. 1 I I 243 - 248.

S. (1991) FEBSLett. 288, 179-182.

P. (1990) Biochem. J . 270, 109-118.

11,95-108.

191 -197.

546.

J. 194, 395- 406.

1640- 1645.

chem. J . 219.719-726.

USA 77,2038 - 2042.