Membrane Potential-linked Reversed Electron Transfer in the Beef Heart Cytochrome bcl Complex Reconstituted into Potassium-loaded Phospholipid Vesicles*

The cytochrome bc, complex purified from beef heart mitochondria was incorporated into potassium (K+)- loaded phospholipid vesicles by a cholate dialysis method to study the reverse reaction of electron trans- fer in the complex. The reduction of cytochrome b in the presence of sodium ascorbate was observed on addition of valinomycin to the K+-loaded proteoliposomes in a medium containing no external KCI; it was followed by the gradual oxidation. Nigericin accelerated the reoxidation of reduced cytochrome b, indicating that a K+ diffusion potential (negative inside) induced the reduction of cytochrome b. The extent of the cytochrome b reduction depended on the magnitude of the diffusion potential across the liposomal membranes, and its maximal reduction was attained at more than 210 mV of the diffusion potential. It was cytochrome bw2 that was re- duced during the establishment of the K+ diffusion potential in the presence of ascorbate, and about 90% of cytochrome bw2 was estimated to be reduced. Antimycin A and myxothiazol inhibited the diffusion potential-in-duced reduction of cytochrome b562, and ubiquinone was proved to be essential for the reversed electron transfer. The K+ diffusion potential also induced the par- tial reduction of cytochrome bSBB when cytochrome bas2 had previously been reduced with ascorbate plus tetra-methylp-phenylenediamine. These results

The cytochrome bc, complex purified from beef heart mitochondria was incorporated into potassium (K+)loaded phospholipid vesicles by a cholate dialysis method to study the reverse reaction of electron transfer in the complex. The reduction of cytochrome b in the presence of sodium ascorbate was observed on addition of valinomycin to the K+-loaded proteoliposomes in a medium containing no external KCI; it was followed by the gradual oxidation. Nigericin accelerated the reoxidation of reduced cytochrome b, indicating that a K+ diffusion potential (negative inside) induced the reduction of cytochrome b. The extent of the cytochrome b reduction depended on the magnitude of the diffusion potential across the liposomal membranes, and its maximal reduction was attained at more than 210 mV of the diffusion potential. It was cytochrome bw2 that was reduced during the establishment of the K+ diffusion potential in the presence of ascorbate, and about 90% of cytochrome bw2 was estimated to be reduced. Antimycin A and myxothiazol inhibited the diffusion potential-induced reduction of cytochrome b562, and ubiquinone was proved to be essential for the reversed electron transfer. The K+ diffusion potential also induced the partial reduction of cytochrome bSBB when cytochrome bas2 had previously been reduced with ascorbate plus tetramethylp-phenylenediamine. These results were interpreted well based on the Q cycle scheme which assumed the energy-dependent reduction of ubiquinone at center 0. Dicyclohexylcarbodiimide, which did not perturb the ability of proteoliposomes to generate the K+ diffusion potential, inhibited the energy-dependent reduction of cytochrome b562 without a significant loss in the catalytic activity of the complex. The half-inhibition was brought about by 200 mol of dicyclohexylcarbodiimidd mol of cytochrome cl. These results strongly suggest the coupling of a proton flow with the reversed electron transfer in the bcl complex.
The cytochrome bcl complex catalyzing the electron transfer from ubiquinol to a C-type cytochrome plays a central role in the Education, Science and Culture of Japan. The costs of publication of this * This work was supported in part by grants from the Ministry of article were defrayed in part by the payment of page charges. This with 18 U.S.C. Section 1734 solely to indicate this fact. article must therefore be hereby marked "advertisement" in  process of oxidative phosphorylation taking part in mitochondria, chloroplasts, or other analogous redox systems. The electron transfer mechanism of the bcl complex is described consistently by a Q cycle scheme originally proposed by Mitchell (1) and subsequently refined by Trumpower (2). In the protonmotive Q cycle (2), ubiquinol is oxidized a t center 0. A first electron from ubiquinol is transferred to iron-sulfur protein, which then reduces cytochrome cl. The second electron transferred from ubisemiquinone to cytochrome b566 recycles through the bcl complex; reduced cytochrome b566 gives an electron to cytochrome b562, which then reduces ubiquinone at center i.
In contrast to the electron transfer pathway in the bcl complex, the mechanism by which the electron transfer gives rise to a transmembrane proton electrochemical potential is not clear. According to the original Q cycle (l), it is the vectorial transmembrane electron transfer from cytochrome b566 to cytochrome b562 that generates a membrane potential. However, later studies have shown that electron transfer between the two b hemes can account for only approximately 40% of the membrane potential generated by the bcl complex (3-5). Therefore, the mechanism generating the remaining 60% of the membrane potential has been expected to be elucidated. Robertson and Dutton (6) indicated that this portion of the membrane potential was associated with the electron transfer from cytochrome b562 to ubiquinone at center i in the bcl complex of the photosynthetic bacterium Rhodobacter sphaeroides. On the other hand, recent experiments by Drachev et al. (7) using the bcl complex of R. sphaeroides chromatophores showed that the electrogenic step was the transfer of protons associated with the oxidation and reduction of ubiquinone at both center o and z rather than the vectorial transfer of electrons. Konstantinov (8) also postulated the vectorial movement of protons for the electrogenic mechanism in the mitochondrial bc, complex. The process of energy transduction in the bcl complex has also been investigated by measurement of the back reaction; the transfer of electrons from a component with a more positive redox potential to a component with a more negative potential driven by a n electrochemical potential gradient of protons. The reversed electron transfer in the bcl complex has been studied with mitochondria or submitochondrial particles energized with ATP, and two kinds of redox response of cytochrome b to addition of ATP have been reported, the reduction of both cytochrome b components (b562 and b566) in the absence of antimycin A (9-111, and the oxidation of prereduced cytochrome b562 in the presence of antimycin A (3, 12, 13). But few systematic studies on the reverse reaction of electron transfer in the bcl complex have been carried out by using the purified enzyme reconstituted into phospholipid vesicles.
Here, we report that the K' diffusion potential induces the reduction of cytochrome b components in the bc, complex reconstituted into phospholipid vesicles in the absence of antimy- experiments. The reduction level of cytowas followed at 552-539 nm and at 562-575 nm, respectively. Zkczces A and B were recorded simultaneously using a multiwavelength spectrophotometer. Additions urrowed in the figure were 2 pmol of sodium ascorbate (use), l pg of valinomycin perature was 20 "C.

hfATERIALS AND METHODS
The cytochrome bcl complex was purified from beef heart muscle according to the method of Yu and Yu (14) and stored at -80 "C until Cytochrome c (horse heart, type III), valinomycin, nigericin, and antimycin A were purchased from Sigma. Myxothiazol was a product of Boehringer Mannheim. Safranine, sodium ascorbate, TMPD, and DCCD were obtained from Wako Pure Chemicals (Tokyo, Japan). Asolectin (soybean phospholipids) was purchased from Associate Concentrates (Woodside, N Y ) and purified partially according to Sone et ut. 117). Potassium-load^ cytochrome bc, vesicles were prepared by the cholate dialysis method (18)(19)(20) as follows. Acetone-washed aeolectin at 40 m g / d was sonicated for 10-15 min at 0 "C in a medium containing 10 m~ MOPSiNaOH (pH 7.4), 100 m~ KC1, and 1.5% sodium cholate. The purified bcl complex was added to give 5 p~ cytochrome cl, and the solution was dialyzed against 400 volumes of 10 m~ MOPS/NaOH (pH 7.4), 100 m~ KC1 at 4 "C for 24 h with two changes of the outer medium. m e r reconstitution was completed, the external KC1 was removed by depleted enzyme was incorporated into liposomes in the same way after the reactivation of the enzyme. DCCD treatment of cytochrome b e l vesicles was camed out as follows. Aportion of the pro~oliposomes was mixed with an appropriate amount of DCCD in ethanol, and the mixture was incubated for 16-20 h at 0 "C. Control vesicles were treated in the same way except that ethanol was added in place of the DCCD solution. The final concentration of ethanol was always 1%.
Spectroscopic measurements were camed out with a Unisoku multiwavelength spectrophotometer equipped with a thermostatted cell holder. The catalytic activity of the reconstituted bcl complex was assayed at 20 "C in a medium containing 10 rn MOPS/NaOH (pH 7.41, 100 m~ KC], 50 p cytochrome c, and 25 p~ QzHz. The initial rate of cytochrome c reduction was corrected against non-enzymatic reduction of cytochrome c by ubiquinol in the absence of bc, vesicles. Proteoliposomes reconstituted with the bc, complex exhibited the oxidation control ratio (18) ranging from 7 to 9, which was determined as the ratio of the enzyme activity in the presence and absence of 0.2 pg of valinomycin plus 2 pg of nigericin. The ability of K*-loaded proteoliposomes containing the bel complex to form a K+ diffusion potential was checked spec-trophotome~cally using safranine as an optical probe for the membrane potential (21) as follows. A 5 4 portion of proteoliposomes containing the bc, complex was added to 1 mI of a medium containing 4.8 p~ safranine, 10 rn MOPSNaOH (pH 7.4), and 100 m~ NaCI. The reaction was initiated by adding valinomycin, and the absorbance change of safranine was followed at 525 nm (22) or at 530578 nm (21). A ubiquinone-10 content of the purified bc, complex was estimated according to the reported method (23). The concentration of ubiquinone was determined spectrophotometrically using an extinction coefficient of 12.25 rm-l.cm-l for the difference between the oxidized and reduced quinone at 275 nm (24). Other extinction coefficients used were desao nm (reduced.oxidired) = 19.2 rn-l.cm-l for cytochrome c (14) and Ae553--539 om (redue.&,xidiaed) = 17.5 rn".cm-' for cytochrome c1 (14). Protein was estimated according to Lowry et al. (25).

Dicfusion Potential-Induced Reduction ofCytochrome b in the bcl Complex Reconstituted into K*-loaded Liposomes-
Valinomycin induces a vectorial flow of potassium ions (K+) from inside to outside K+-loaded vesicles, resulting in the formation of a potassium diffusion potential (negative inside) across the liposomal membranes. Fig. 1 shows the response of the reduction level of cytochrome b in the reconstituted bcl complex to the valinomycin-induced K+ diffusion potential. In these experiments, cytochrome c1 and iron-sulfur protein in the be, complex were reduced first with 5 m~ sodium ascorbate in a medium containing no external KC1 (Fig. lA). As shown in  a rapid increase in absorbance at 562 nm relative to 575 nm, the wavelength pair selected to monitor the redox state of cytochrome b components, was observed upon addition of valinomycin. The maximum reduction level was achieved 3 min after addition of valinomycin, followed by its gradual decrease. Nigericin, which collapsed the K+ gradient, accelerated the reoxidation of reduced cytochrome b (Fig. 1C). Moreover, the rapid reduction of cytochrome b was not observed with cytochrome bcl vesicles suspended in a medium containing 100 mM KC1 (Fig. lD). These results clearly indicate that the valinomycin-mediated K' diffision potential induces the reduction of cytochrome b in the ascorbate-reduced bcl complex reconstituted into liposomes. was observed below the potential value of 30 mV, and the reduction level of cytochrome b increased sigmoidally with increasing the diffision potential above 30 mV. The maximal reduction was attained at more than 210 mV (the initial concentration of external KC1 was less than 30 p~) . Fig. 3 shows the spectral analysis of A+-induced reduction of cytochrome b. When a difference spectrum was taken between diffusion potential-imposed and ascorbate-reduced bcl vesicles, absorption maximum appeared at 562 nm (solid line in Fig. 3). On the contrary, a peak was found at 565 nm along with a shoulder around 558 nm in a difference spectrum between Na2Sz04-reduced and diffusion potential-imposed bcl vesicles (the spectrum is not included in Fig. 3). These results indicate that cytochrome b562 in the reconstituted bcl complex is the main component that is reduced upon generation of the K+ diffision potential in the presence of ascorbate.
w e estimated the extent of the cytochrome 6562 reduction using ascorbate and TMPD as electron donors. As reported with the purified complex from beef heart mitochondria (26), cytochrome b562 in the reconstituted bcl complex can be fully reduced with ascorbate plus TMPD (data not shown, but see Fig.  8). By comparing a difference spectrum between A+-imposed minus ascorbate-reduced bcl vesicles with that between the (ascorbate and TMPD)-reduced minus ascorbate-reduced vesicles, about 90% of cytochrome b562 was estimated to be reduced during the formation of the diffision potential. This result is important (see "Discussion"). Effect of Electron Dansfer Inhibitors-The effect of electron transfer inhibitors on the A+-induced reduction of cytochrome b562 was examined with the reconstituted bcl complex supplemented with either antimycin A or myxothiazol (Fig. 4). In these experiments, the reconstituted bcl complex was prereduced with 5 m~ sodium ascorbate for 30 min at 20 "C. Sodium ascorbate alone reduced cytochrome b very slowly in the reconstituted bcl complex (see Fig. lD) as well as in the purified enzyme (data not shown). About 10% of total cytochrome b was finally in the reduced form during incubation for 30 min in the presence of 5 m~ ascorbate. Spectral examination indicated that the cytochrome b species reduced by ascorbate was cytochrome b562 (data not shown). When antimycin A at a molar ratio of 2.1 moVmol cytochrome c1 was added to the reconstituted bcl complex, generation of the diffusion potential promoted the reoxidation of cytochrome b562 prereduced by ascorbate (Fig. 4B). On the contrary, myxothiazol at a ratio of 8.3 moVmol cytochrome c1 lowered the reduction level of cytochrome b562 to 12% of the control level (Fig. 4C). Either 0.5 nmol of antimycin A or 2.0 nmol of myxothiazol did not affect the time course of safranine response (data not shown), thus eliminating the possibility for those inhibitors to perturb the formation of the diffusion potential. Effect of Ubiquinone-When the ubiquinone (Q)-depleted bcl complex (0.03 mol of Qldmol of cytochrome cl) was incorporated into K+-loaded liposomes, only 7% of total cytochrome b was reduced upon generation of the K+ diffusion potential (Fig.  5). Therefore, ubiquinone-2 (Q2) was added to ascorbate-reduced proteoliposomes reconstituted with the Q-depleted bcl complex before generation of the diffusion potential to examine the effect of ubiquinone. The addition of Qz restored the A+induced reduction of cytochrome b562 up to the 42% reduction of total cytochrome b at a molar ratio of 0.75 mol of Qz/mol of cytochrome cl. On the other hand, Qz externally added to proteoliposomes reconstituted with the native bcl complex enhanced the energy-dependent reduction of cytochrome b562 maximally by 10% (Fig. 5). Although the maximum level of the recovered cytochrome b562 reduction in the Q-depleted bcl complex is lower than that observed with the native complex, this result indicates that ubiquinone is a n essential component for the reverse transfer of electrons. With increasing concentra-  1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  tions of exogenous Qz higher than 1 moVmol cytochrome c l , the extent of cytochrome b562 reduction decreased in both the native and the Q-depleted bcl complex (Fig. 5). Since the oxidized form of ubiquinone acts as an oxidant of reduced cytochrome b components as reported previously (24), the apparent decrease in the reduction level of cytochrome b562 would be ascribed to the reoxidation of reduced cytochrome b562 by Qz, Effect of DCCLLDCCD has been reported to inhibit only the proton translocation but not the electron transfer activity of the cytochrome bcl complex from beef heart (27) and yeast mitochondria (191, and the cytochrome bsf complex from spinach chloroplasts (28).
We, therefore, investigated the effect of DCCD on the energy-linked reduction of cytochrome b562 (Figs. 6 and 7). When the reconstituted bcl complex was incubated for 16 h at 0 "C with DCCD at a molar ratio of 444 moVmol cytochrome c l , the diffusion potential-induced reduction of cytochrome b662 was lowered to 38% of that observed in control vesicles (Fig. 6A). As shown in Fig. 7, the energy-dependent reduction of cytochrome bssz decreased with increasing concentration of DCCD in a dose-dependent manner, an apparent half-inhibition concentration being 200 mol of DCCD/mol of cytochrome cl. Contrary to the inhibitory effect of DCCD on the reversed electron transfer, it exerted no effect on the ability of  (Figs. 6B  and 7). Furthermore, the enzyme activity was inhibited by only 10% during a 16-h incubation at 0 "C with DCCD at a molar ratio of 481 moVmol cytochrome c1 (Fig. 7).
Energy-linked Reduction of Cytochrome b5,,F1atmark and Pedersen (11) reported that energization of brown adipose tissue mitochondria from guinea pig with ATP caused the reduction of cytochrome b566 as well as cytochrome bS6z in the presence of ascorbate plus TMPD, suggesting that ATP was capable of inducing reduction of both the cytochrome b components. batwTMPD-reduced bcl vesicles, corresponding to state d uersus a .
" " , the difference spectrum between NazS204-reduced and ascor-8 shows the effect of the diffusion potential on the reconstituted bcl complex when ascorbate and TMPD were used as electron donors. In contrast to brown adipose tissue mitochondria, in which ascorbate plus TMPD led to the partial reduction of cytochrome b562 (9% reduction of total cytochrome b ) (ll), cytochrome b562 in the reconstituted bcl complex was fully reduced by the ascorbate/l"PD system (Fig. a), as reported with the purified complex from beef heart mitochondria (26). After the complete reduction of cytochrome b562 with ascorbate plus TMPD, addition of valinomycin induced a further increase in absorbance at 565 minus 575 nm (Fig. 8A). And nigericin caused the slow reoxidation of the cytochrome b component which had been reduced on generation of the diffusion potential (Fig. &I). As shown in Fig. 8B, a characteristic spectrum of cytochrome b566 with a peak at 566 nm and a shoulder around 558 nm was obtained when the difference spectmm was taken before and after addition of valinomycin (solid line). These results indicate that the K+ diffusion potential induced the reduction of cytochrome b566 in the reconstituted bcl complex in the presence of ascorbate and TMPD. The extent of cytochrome b566 reduction was estimated to be about 30% of the cytochrome.

DISCUSSION
Energy-linked reduction of cytochrome b has been studied with mitochondria or submitochondrial particles prepared from different sources (9)(10)(11). Originally, cytochrome b566 was reported to be reduced when ATP was added under anaerobic conditions to beef heart submitochondrial particles (9) or pigeon heart mitochondria (10) in which cytochrome b562 had been fully reduced with a substrate such as succinate. Later, it was shown that energization of brown adipose tissue mitochondria from guinea pig with ATP led to the antimycin A-sensitive reduction of cytochrome b562 as well as cytochrome b566 in the presence of ascorbate plus TMPD (111, suggesting that ATP was capable of inducing reduction of both cytochrome b components (b562 and b566). We confirmed the energy-dependent reduction of cytochrome b562 and b566 using ascorbate and ascorbate1 TMPD, respectively, in the reconstituted system with energy provided by a K+ diffusion potential (negative inside). Another type of energy-linked redox response of cytochrome b is the oxidation of prereduced cytochrome b562 in the presence of antimycin A (3,12,13). Beattie and Villalobo (19) demonstrated, using proteoliposomes reconstituted with the yeast bcl complex, that the K+ diffusion potential (negative inside) in the presence of antimycin A promoted the oxidation of cytochrome b prereduced by the Q2/Q2H2 mixture. Therefore, we believe that the results presented here are the first report demonstrating and characterizing the energy-linked reduction of cytochrome b in the reconstituted proteoliposome system.
The present study has shown that ubiquinone is an essential component for the energy-linked reduction of cytochrome b562 (Fig. 5), which is inhibited by electron transfer inhibitors like antimycin A and myxothiazol (Fig. 4). Cytochrome b566 is also promoted to be reduced on generation of the K+ diffusion potential in the presence of ascorbate and TMPD (Fig. 8). These results can be consistently interpreted in terms of the Q cycle scheme which assumed the energy-dependent reduction of ubiquinone at center 0. The pathway of the reversed electron transfer in the bcl complex can be summarized as follows (Fig.  9B).
Step 2: the next step is the reduction of ubiquinone to ubiquinol at center 0. Electrons to reduce ubiquinone are supplied from the iron-sulfur center which is reduced by ascorbate via cytochrome c l . The membrane potential is required to enable electrons to move at center o from iron-sulfur protein with a more positive redox potential of +280 mV (29) to the ubiquinonelubisemiquinone couple with a more negative potential as low as -230 mV or to the ubisemiquinonel ubiquinol couple with a potential of +190 mV (30). This step is sensitive to myxothiazol. There was no sign of the transient reduction of cytochrome b566 even when the A+-induced reduction of cytochrome b562 in the presence of ascorbate was investigated with a stopped-flow and rapid-scan technique.2 Moreover, antimycin A inhibited the A+-induced reduction of cytochrome b562 (Fig. 4). We, therefore, conclude that ubisemiquinone, a possible intermediate during ubiquinone reduction, is not likely to give an electron to cytochrome b5% in the reverse reaction of electron transfer.
Step 3: reduced ubiquinone moves from center o to center i. Step 4: ubiquinol reduces cytochrome b562 at center i, and antimycin A inhibits this reaction.
Step 5: the diffusion potential can induce electron transfer from cytochrome b562 to cytochrome b566 when cytochrome b562 is in the fully reduced form (Fig. 8).
The main question in the present work is the effect of DCCD on the reversed electron transfer in the reconstituted bcl complex. We observed that DCCD prevented the reversed electron transfer from cytochrome c1 to cytochrome ,5562 without a significant loss in the catalytic activity of the reconstituted bcl complex (Figs. 6 and 7). Similar results have been reported by Beattie and Villalobo (19), in which the energy-linked oxidation of prereduced cytochrome b in the presence of antimycin A was inhibited by the DCCD treatment of the reconstituted bcl com-Y. Orii, and T. Miki, unpublished observation. One possible model incorporating the proton flow into the Q cycle is shown in Fig. 9, which is a modified version of the scheme proposed by Konstantinov (8). This model postulates the reversible flow of protons between the outer medium and center o (Fig. 9, A and B 1. Important features of the proton flow are as follows. First, protons move electrogenically from center o to the outer phase in the catalytic reaction (Fig. 9A). Second, the membrane potential promotes the uptake of protons from the outer medium in the reverse reaction (Fig. 9B), and the reduction of ubiquinone at center o (Step 2 ) is tightly coupled to the membrane potential-induced uptake of protons. Third, DCCD specifically blocks the reversible flow of protons between center o and the outer medium without affecting any step of the electron transfer pathway. According to this model, the effect of DCCD on the bcl complex can be interpreted as follows. In the reverse reaction (Fig.  9B), DCCD inhibits the A+-dependent uptake of protons and, consequently, blocks the reverse transfer of electrons because the reversed electron transfer from the iron-sulfur cluster to ubiquinone is closely coupled to the A+-dependent uptake of protons. In the catalytic reaction (Fig. 9A), however, DCCD dose not inhibit the electron transfer activity of the bcl complex (Fig. 7) in spite of its specific effect on the proton translocation activity of the complex (19). These observations are interpreted well by the Q cycle when the pathway of the proton flow in the DCCD-treated bcl complex is different from that in the intact complex. We, therefore, postulated in the catalytic reaction (Fig. 9A) that protons produced by the Q H 2 oxidation at center o might go back electrophoretically from center o to the inner space through the lipid phase (dotted line in Fig. 9A) because the normal electrogenic pathway for the proton ejection is blocked by DCCD. In this case, the enzyme activity of the bcl complex would not be affected by the DCCD treatment because DCCD does not block any step of the electron transfer pathway.
Generation of the K+ diffusion potential induced the reduction of about 90% of cytochrome b562 (Fig. 2), which was interpreted to be reduced by ubiquinol (Step 4). But Q2H2 can reduce only 58% of cytochrome b562 (35% of total cytochrome b ) in reaction of the reconstituted bcl complex with 10 p~ Q2H2 (data not shown), in line with the result observed in the purified complex from beef heart mitochondria (34). These results sug-+ + + + + + gest that the electron transfer step from ubiquinol to cytochrome b562 is somewhat an up-hill reaction. Using the bcl complex of R. sphaeroides chromatophores, Drachev et al. (7) also showed that the transfer of protons associated with the reduction of ubiquinone at center i was the electrogenic step (see also Ref. 38). We, therefore, assumed the reversible proton flow between the inner phase and center i in analogy with Step 2. The energy-dependent reduction of cytochrome bse2 (Step 4 ) seems to proceed along with the inward flow of protons.
It has been generally accepted (3,351 that membrane potential-induced reversed electron transfer from cytochrome b562 to b566 (Step 5) is due to equilibria shifts associated with membrane potential-induced change in electron distributions between the two cytochrome b components which are arranged transmembranously in the bcl complex (36). The main point of our present work is that the reversed transport of electrons is closely coupled to the reverse flow of protons, and one possible model is presented based on the Q cycle (Fig. 9). This model also accounts for the specific effect of DCCD on the bcl complex. Beattie and co-workers have reported that DCCD binds to the cytochrome b subunit of the bcl complex purified from various sources (19, 281, and the binding site of DCCD has been reported to be Asp16o in the cytochrome b subunit of the yeast bcl complex (28). The aspartate residue is present in the a-helical segment which lies on the outer surface of mitochondrial (liposomal) membranes (37). These results suggest that the cytochrome b subunit is significantly involved in the proton flow through the bcl complex, presumably by forming a proton channel, and that DCCD might block the outlet part of the proton channel (inlet of the proton uptake). Further studies are required for elucidating the role of the cytochrome b subunit in proton translocation coupled t o the electron transfer reaction of the bcl complex.