Evidence for electron equilibrium between the two hemes bL in the dimeric cytochrome bc1 complex.

Structural analysis of the dimeric mitochondrial cytochrome bc1 complex suggests that electron transfer between inter-monomer hemes bL-bL may occur during bc1 catalysis. Such electron transfer may be facilitated by the aromatic pairs present between the two bL hemes in the two symmetry-related monomers. To test this hypothesis, R. sphaeroides mutants expressing His6-tagged bc1 complexes with mutations at three aromatic residues (Phe-195, Tyr-199, and Phe-203), located between two bL hemes, were generated and characterized. All three mutants grew photosynthetically at a rate comparable to that of wild-type cells. The bc1 complexes prepared from mutants F195A, Y199A, and F203A have, respectively, 78%, 100%, and 100% of ubiquinol-cytochrome c reductase activity found in the wild-type complex. Replacing the Phe-195 of cytochrome b with Tyr, His, or Trp results in mutant complexes (F195Y, F195H, or F195W) having the same ubiquinol-cytochrome c reductase activity as the wild-type. These results indicate that the aromatic group at position195 of cytochrome b is involved in electron transfer reactions of the bc1 complex. The rate of superoxide anion (O2*) generation, measured by the chemiluminescence of 2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo[1,2-alpha]pyrazin-3-one hydrochloride-O2* adduct during oxidation of ubiquinol, is 3 times higher in the F195A complex than in the wild-type or mutant complexes Y199A or F203A. This supports the idea that the interruption of electron transfer between the two bL hemes enhances electron leakage to oxygen and thus decreases the ubiquinol-cytochrome c reductase activity.

ubiquinol, is 3 times higher in the F195A complex than in the wild-type or mutant complexes Y199A or F203A. This supports the idea that the interruption of electron transfer between the two b L hemes enhances electron leakage to oxygen and thus decreases the ubiquinolcytochrome c reductase activity.
The cytochrome bc 1 complex (also known as ubiquinol-cytochrome c reductase or Complex III) is an essential segment of the electron transfer chains of mitochondria and many respiratory and photosynthetic bacteria (1). It catalyzes electron transfer from ubiquinol to cytochrome c with concomitant translocation of protons across the membrane to generate a proton gradient and membrane potential for ATP synthesis. The "proton-motive Q-cycle" is the most favored mechanism for electron and proton transfer in this complex (2,3). The central feature of the Q-cycle mechanism is the bifurcation of electrons from ubiquinol at the Qo site. The first electron of ubiquinol is transferred to the "high potential chain," consisting of the Rieske [2Fe-2S] cluster, housed in the iron-sulfur protein (ISP), 1 and heme c 1 , housed in cytochrome c 1 . Then the second electron of ubiquinol is passed through the "low potential chain" consisting of heme b L and heme b H , both housed in the cytochrome b subunit.
Recently, mitochondrial cytochrome bc 1 complexes from beef (4,5), chicken (6), and yeast (7) were crystallized, and their three-dimensional structures were determined. The structural information not only supports the Q-cycle mechanism but also suggests the complex functioning as a dimer. This suggestion stems from the following structural data (4 -7): (i) the intertwining of ISPs in the two bc 1 monomers such that the head domain of ISP in one monomer is physically close to and interacting with the cytochrome b and cytochrome c 1 in the 2-fold symmetry-related other monomer; (ii) the presence of two apparently non-communicating cavities in the dimeric complex, each connecting the Qo pocket of one monomer to the Qi pocket of the other; and (iii) the distance between the Fe atoms of the two hemes b L is only 21 Å, which is approximately the same as that between heme b L and b H in one monomer (Fig. 1). The short distance between the two hemes b L and the presence of several aromatic amino acid residues at the interface of the two cytochrome b proteins has promoted investigator to speculate the existence of electron transfer or equilibrating between the two hemes b L (8 -10). Recently the existence of an intertwined dimer in solution was confirmed (11) in the Rhodobacter sphaeroides bc 1 complex through the formation of a four-subunit (two ISPs and two cytochrome bs) adduct by two inter-subunit disulfide bonds between two engineered cysteine pairs: one at cytochrome b and the head domain of ISP and the other at cytochrome b and the tail domain of ISP. However, evidence for inter-monomer b L -b L electron transfer during bc 1 catalysis is still missing, due to the lack of a suitable assay method.
This results from a leakage of the second electron of ubiquinol, from the low potential chain of the Q cycle electron transfer pathway, to interact with molecular oxygen. The electron-leaking site is thought to be located at the reduced cytochrome b L or ubisemiquinone of the Qo site (12-15, 17, 18). The amount of electron leakage is believed to be proportional to the concentration of reduced cytochrome b L or ubisemiquinone at the Qo site. Thus, if the inter-monomer b L -b L electron transfer is interrupted, one should see an increase in the rate of O 2 . generation, because electrons that normally shuttle between the two * This work was supported by Grant GM30721 from the National Institutes of Health and by the Agricultural Experiment Station (Projects # 1819 and # 2372), Oklahoma State University. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ To whom correspondence should be addressed. b L hemes will accumulate at the b L heme of one monomer, thus enhancing the chance for leakage and reaction with oxygen.
In the three-dimensional structure of mitochondrial bc 1 complex, several pairs of aromatic residues are located at the dimer interface between the two hemes b L (4). These aromatic pairs may facilitate the inter-monomer heme b L -b L electron transfer. If this is indeed the case the existence of inter-monomer heme b L -b L electron transfer in the bc 1 complex can be revealed by comparing the rates of O 2 . generation by the wild-type bc 1 complex with those of mutant complexes having these aromatic pairs replaced with non-aromatic residues. An increase in O 2 .
generation, as a result of increasing the concentration of reduced b L or ubisemiquinone at the Qo site, by the mutant bc 1 complex, would indicate inter-monomer heme b L -b L electron transfer, involving aromatic amino acid residues. Herein we report procedures for generating R. sphaeroides mutants expressing His 6 -tagged bc 1 complexes with mutations at three highly conserved aromatic residues (Phe-195, Tyr-199, and Phe-203) located between the two b L hemes of the dimeric complex. The rate of superoxide anion generation, the effect of oxygen on the activity, and the EPR characteristics of the cytochromes b L and b H in purified complexes from wild type and mutant strains are examined and compared.
Generation of R. sphaeroides Strains Expressing the His 6 -tagged Cytochrome bc 1 Complexes with Mutations of Aromatic Residues Located at the Dimer Interface between Two Hemes b L -Mutations were constructed by the QuikChange site-directed mutagenesis kit from Stratagene using a supercoiled double-stranded pGEM7Zf(ϩ)-fbcFB as template and a forward and a reverse primer for PCR amplification. The pGEM7Zf(ϩ)-fbcFB plasmid (20) was constructed by ligating the EcoRI-XbaI fragment from pSELNB3503 into EcoRI and XbaI sites of the pGEM7Zf(ϩ) plasmid. The primers used are given in Table I. The BstEII-XbaI fragment from the pGEM7Zf(ϩ)-fbcFB m plasmid was ligated into the pRKD418-fbcFB KmBX C H Q plasmid to generate the pRKD418-fbcFB m C H Q plasmid. A plate-mating procedure (21) was used to mobilize the pRKD418-fbcFB m C H Q plasmid in Escherichia coli S17-1 cells into R. sphaeroides BC17 cells. The presence of engineered mutations were confirmed by DNA sequencing of the 962-bp BstEII-XbaI fragment before and after photosynthetic growth of the cells as previously reported (21). DNA sequencing and oligonucleotide syntheses were performed by the Recombinant DNA/Protein Core Facility at Oklahoma State University.
Growth of Bacteria-E. coli cells were grown at 37°C in LB medium (sodium chloride, SELECT peptone 140, and SELECT yeast extract, autolyzed low sodium). For photosynthetic growth of the plasmid-bearing R. sphaeroides BC17 cells an enriched Siström's medium containing 5 mM glutamate and 0.2% casamino acids was used. Photosynthetic growth conditions for R. sphaeroides were essentially as described previously (21). Cells harboring the mutated cytochrome b gene on the pRKD418-fbcFB m C H Q plasmid were grown photosynthetically for one or two serial passages to minimize any pressure for reversion. The inoculation volumes used for photosynthetic cultures were at least 5% of the total volume. Antibiotics were added to the following concentrations: ampicillin (125 g/ml), kanamycin sulfate (30 g/ml), tetracycline (10 g/ml for E. coli and 1 g/ml for R. sphaeroides), and trimethoprim (100 g/ml for E. coli and 30 g/ml for R. sphaeroides).
Enzyme Preparations and Activity Assay-Chromatophores were prepared, from which the His 6 -tagged cytochrome bc 1 complexes were purified, as previously reported (22). To assay cytochrome bc 1 complex activity, chromatophores or purified cytochrome bc 1 complexes were diluted with 50 mM Tris-Cl, pH 8.0, containing 200 mM NaCl and 0.01% dodecyl maltoside to a final concentration of cytochrome b of 3 M. Appropriate amounts of the diluted samples were added to 1 ml of assay mixture containing 100 mM Na ϩ /K ϩ phosphate buffer, pH 7.4, 300 M EDTA, 100 M cytochrome c, and 25 M Q 0 C 10 BrH 2 . Activities were determined by measuring the reduction of cytochrome c (the increase of absorbance at 550 nm) in a Shimadzu UV 2101 PC spectrophotometer at 23°C, using a millimolar extinction coefficient of 18.5 for calculation. The non-enzymatic oxidation of Q 0 C 10 BrH 2 , determined under the same conditions in the absence of enzyme, was subtracted from the assay.
To measure the effect of oxygen on bc 1 activity, a Thunberg cuvette was used. 1 ml of assay mixture was placed in the main chamber, and a 10-l aliquot of enzyme (5-10 pmol) was in the side arm. The cuvette was evacuated and flushed with argon five times. The reaction was started by mixing the bc 1 complex solution and the assay mixture.
Measurement of Superoxide Anion Generation-Superoxide anion generation by the cytochrome bc 1 complex was determined by measuring the chemiluminescence of MCLA-O 2 .

adduct (23), in an Applied
Photophysics stopped-flow reaction analyzer SX.18MV (Leatherhead, England), by leaving the excitation light off and registering light emission (24 Other Biochemical and Biophysical Techniques-Protein concentration was determined by the method of Lowry et al. (25). Cytochrome b (26) and cytochrome c 1 (27) contents were determined according to reported methods. SDS-PAGE was performed according to Laemmli (28) using a Bio-Rad Mini-Protean dual slab vertical cell. Samples were digested with 10 mM Tris-Cl buffer, pH 6.8, containing 1% SDS, and 3% glycerol in the presence of 0.4% ␤-mercaptoethanol for 2 h at 37°C before being subjected to electrophoresis.
EPR spectra of b cytochromes were recorded at 8.5 K on a Bruker EMX EPR spectrometer equipped with an Air Products flow cryostat. The instrument settings are detailed in the figure legend. Refined coordinates for the crystal structure of native bovine mitochondrial bc 1 (29) (PDB code 1NTM) were used for distance determination.

Involvement of the Phe-195 of Cytochrome b in Electron
Transfer Activity of the Cytochrome bc 1 Complex-Three aromatic amino acid residues in cytochrome b, Phe-195, Tyr-199, and Phe-203, were selected for mutation to test the hypothesis that inter-monomer b L -b L electron transfer occurs during bc 1 catalysis, and such electron transfer is facilitated by aromatic residues located between the two b L hemes. Alignment of more than 40 sequences of cytochrome b reveals that Phe-195 is fully conserved except in Rhodopseudomonas virids (in which Phe is replaced by Tyr); Tyr-199 is highly conserved, although it is replaced by Phe in many cases; and Phe-203 is less conserved, being substituted with Met, Ser, Leu, and Trp in some species (30). The selection of these three residues was based on the three-dimensional structure of the four-subunit cytochrome bc 1 complex of R. sphaeroides ( Fig. 2A) constructed by using coordinates from bovine cytochromes b and c 1 , ISP, and subunit VII (31). The distances between the symmetry pairs of aromatic residues Phe-195, Tyr-199, or Phe-203 are 4.5, 7.7, and 6.8 Å (Fig. 2B), respectively, when measured from edge to edge of the phenyl rings of the amino acid residues. They are, edge to edge, 3.8, 3.7, and 3.3 Å apart, respectively, in the corresponding residues in the bovine complex (Phe-179, Phe-183, and Phe-187 in bovine). The distances from heme b L to Phe-195, Tyr-199, or Phe-203 are 8.8, 4.9, and 10.3 Å, respectively, when measured from the iron center to the edges of the aromatic ring of the amino acid residues. They are 7.5, 8.6, and 10.2 Å, respectively, in the corresponding bovine enzyme. Recently a relative low resolution structure of cytochrome bc 1 complex from Rhodobacter capsulatus was reported (32). The distances obtained among these aromatic amino acids pairs are surprisingly closed to those deduced from the model of the complex of R. sphaeroides.
When each of these three aromatic residues was replaced with alanine, the resulting mutants (F195A, Y199A, and F203A) grew photosynthetically at a rate comparable to that of the complement (wild-type) cells. Chromatophores prepared from these mutants have, respectively, 80, 100, and 100% of the bc 1 activity found in the complement chromatophores (Table  II). When cytochrome bc 1 complexes prepared from these three mutant chromatophores were assayed for ubiquinol-cytochrome c reductase activity, the Y199A and F203A mutant complexes had the same activity as the complement complex, and the F195A mutant complex had about 78% (see Table II). These results indicate that Phe-195 of cytochrome b is involved in inter-monomer electron transfer in the dimeric bc 1 complex, but residues Tyr-199 or Phe-203 are not.
Mutants with double and triple alanine substitutions in these three aromatic residues of cytochrome b were also generated and characterized. These are: F195A/Y199A, F195A/ F203A, Y199A/F203A, and F195A/Y199A/F203A. All but the Y199A/F203A mutant complex have about 78% of the activity found in the complement complex (see Table II). The Y199A/ F203A has the same activity as that of the complement complex. Because the extent of bc 1 activity decrease in the double or triple alanine substitution mutants containing the F195A mutation is the same as that observed for the F195A mutant complex, aromatic residues Tyr-199 and Phe-203 apparently play no complementary or auxiliary role to residue Phe-195 in inter-monomer electron transfer.
Because the extent of activity change upon the replacement Phe-195 with alanine is a relatively small, 22% decrease, special attention was paid during data collection. The experiments were not only repeated by three times but also repeated by different investigators. We are very confident that the difference in activity is real and not an experimental artifact. In fact the small change in activity observed upon replacement of aromatic amino acid with an alanine at residue position 195 was expected, because the inter-monomer electron transfer is not in the main path of electron transfer.
Absorption spectral analysis of mutant complexes of F195A, F195A/Y199A, F195A/F203A, and F195A/Y199A/F203A indicates that the amounts and absorption properties of cytochromes b and c 1 in these mutant complexes are the same in the complement complex. Western blot analysis using antibodies against R. sphaeroides ISP and subunit IV also indicate that these mutant complexes have the same amount of ISP and subunit IV as does the complement complex. Thus the decrease in bc 1 activity in the F195A mutant complex is not due to mutational effects on the assembly of the bc 1 protein subunits into the chromatophore membrane or to changes in the binding affinity of protein subunits in the complex.
Although the three pairs of aromatic residues are all located at the dimer interface of the cytochrome b subunits, they are not the only residues contributing to the stability of the dimer. As observed in the bovine complex, at least 23 residue pairs from a cytochrome b subunit make contact at the dimer interface, most are from the transmembrane helices of A and E, from the AB loop and from the N-terminal ␣ helix. In addition, the ISP transmembrane helix and the head domain make sub- a F and R in the parentheses denote forward and reverse primers, respectively. stantial contributions to holding the dimer together. As the cytochrome b and ISP subunits are highly conserved, it is reasonable to believe that the interaction at the dimer interface would also be conserved. Because the structural effect of these mutations is probably a creation of small cavities, it is not surprising to find that even the triple mutation does not disturb the structural integrity of the bc 1 dimer. In bovine cytochrome b the three interfacial aromatic pairs display three entirely different contact geometries: the Phe-179 pair (195 in R. sphaeroides bc 1 ) has an on-edge interaction with the two phenyl rings aligned roughly parallel to the membrane plane; with an angle between the two planes of 37°; the phenyl rings of the Phe-183 pair (199 in R. sphaeroides bc 1 ) are stacked on top of each other with an angle of 0°; and the Phe-187 pair (203 in R. sphaeroides bc 1 ) is partially stacked, and these two phenyl rings are oriented normally to the membrane plane with an angle between the two rings of 34.7°. In biological electron transfer complexes such as the cytochrome bc 1 complex (4) and photosynthetic reaction centers (33), donors and receptors of electron transfer machines are always observed in an on-edge arrange-ment, as determined by x-ray crystallography. Presumably the on-edge interaction between partners provides more efficient electron transfer than other orientations. The observed on-edge interaction for the Phe-179 pair in the bovine bc 1 structure strongly supports our mutational data indicating that this particular pair mediates lateral electron transport between bc 1 monomers. In addition, Phe-179 has the shortest distance to the Qo site, which may also be advantageous for its role as an electron transfer mediator between monomers.
Essentiality of the Aromatic Group in the Phe-195 of Cytochrome b-To establish that the loss of ubiquinol-cytochrome c reductase activity in the F195A mutant complex results from the loss of an aromatic ring at this position of cytochrome b, mutants with conservative substitution at Phe-195 (F195Y, F195W, and F195H), were generated and characterized. These three mutants grew photosynthetically at a rate comparable to that of the complement cells and, in bc 1 complexes in chromatophore membranes or in the purified state, have the same ubiquinol-cytochrome c reductase activity as that of the complement complex (see Table II). These results confirm the es- . Therefore, one way to confirm that the aromatic group in Phe-195 participates in inter-monomer b L -b L electron transfer during bc 1 catalysis is to compare superoxide anion radical generation by the complement and F195A mutant complexes. Although the rate of superoxide anion production by the cytochrome bc 1 complex can be determined by measuring the decrease in rate of cytochrome c reduction in the presence of superoxide dismutase under conditions of continuous turnover of the bc 1 complex, the small rate of superoxide anion formation, compared with the normal rate of cytochrome c reduction, compromises the accuracy of this method. MCLA has a high sensitivity for O 2 . in the neutral pH range (34,35). The MCLA chemiluminescence method, which has been widely used to detect O 2 . (14, 36 -38), is 95 times more sensitive than the non-enzymatic oxidation of ubiquinol by cytochrome c, is eliminated. This method enables us to accurately evaluate changes in the rate of superoxide anion generation by various bc 1 complexes. Fig. 3 shows actual tracings of superoxide generation by wild-type and F195A mutant bc 1 complexes. MCLA chemiluminescence induced by bc 1 complex reaches peak intensities after ϳ0.06 s at room temperature and then decays. No detectable luminescence is detected when bc 1 complex is omitted from the enzyme-containing solution or Q 0 C 10 BrH 2 is omitted from the substrate-containing solution. Addition of superoxide dismutase to either the substrate or enzyme solution completely abolishes luminescence, indicating that O 2 . is responsible for the luminescence observed. Maximum peak height induced by the F195A mutant complex (0.13 V) is about three times higher than that reached by the wild-type complex.
FIG. 3. Tracings of superoxide generation in wild-type and mutant F195A cytochrome bc 1 complexes. To measure the superoxide anion production during the pre-steady-state reaction of the reduction of the bc 1 complex by ubiquinol, stopped-flow assays were carried out at 23°C in an Applied Photophysics stopped-flow reaction analyzer SX 18MV by mixing 1:1 solutions A and B. Solution A consisted of 100 mM Na ϩ /K ϩ phosphate buffer, pH 7.4, containing 1 mM EDTA, 1 mM KCN, 1 mM NaN 3 , 0.1% bovine serum albumin, 0.01% dodecyl maltoside, and 9 M bc 1 complexes. Solution B was the same as solution A except that the bc 1 complex was replaced with 50 M Q 0 C 10 BrH 2 and 4 M MCLA. For each sample, eight kinetic traces were averaged. For control, either bc 1 complexes or Q 0 C 10 BrH 2 were omitted from the above system, or 300 units/ml superoxide dismutase was added to the system.   Table IV shows the effect of oxygen on the cytochrome bc 1 activity in purified complexes from wild type and mutants F195A, Y199A, and F203A. The complement, Y199A and F203A mutant complexes catalyzed electron transfer, from ubiquinol to cytochrome c, at a rate of 2.5 mol of cytochrome c reduced per minute per nanomole of cytochrome b at 23°C under aerobic conditions. Removal of oxygen from the assay system increases the activities of these four complexes by 4% (2.6 mol cytochrome c reduced per minute per nanomoles of cytochrome b). It should be noted that this activity increase is not due to the inhibition of the contaminated cytochrome c oxidase in the bc 1 preparation, because addition of sodium azide to the assay mixture, under aerobic conditions, does not increase the rate of cytochrome c reduction. Addition of superoxide dismutase to the assay mixture causes a 5.2% activity decrease in the complement, Y199A and F203A cytochrome bc 1 complexes. . . The disruption of electron transfer between the two b L hemes in dimeric cytochrome bc 1 complex enhances (from 5.2 to 12%) the electron leakage and decreases (2.61 to 2.39 mol of cytochrome c reduced per minute per nanomole cytochrome b) the normal electron transfer rate during oxidation of ubiquinol and reduction of cytochrome c. Because the activity of F195A mutant complex is not restored to the level of the activity of the complement complex under the anaerobic conditions, activity loss due to disruption of electron transfer between the two hemes b L cannot attributed entirely to electron leak to oxygen. About half of the activity loss is due to disruption of electron transfer between the two hemes b L . In the presence of inter-monomer electron transfer, the electron at Qo (either at b L or ubisemiquinone) of one monomer can be oxidized by the two hemes b H of the complex. When inter-monomer electron transfer is interrupted, as in the F195A mutant complex, the electron at heme b L can only be oxidized by heme b H of the same monomer; this accounts for a 9% decrease of activity. Another explanation for activity decrease in the F195A mutant complex is the relatively more oxidized state of hemes b L in dimeric bc 1 complex with inter heme b L electron transfer than in those without. This more oxidized state of heme b L would facilitate electron trans-   a To measure the effect of oxygen on bc 1 activity, 1 ml of assay mixture containing 100 mM Na ϩ /K ϩ phosphate buffer, pH 7.4, 300 M EDTA, 100 M cytochrome c, and 25 M Q 0 C 10 BrH 2 was placed in the main chamber, and a 10-l aliquot of enzyme with cytochrome b of 1 M was in the side arm of a Thunberg cuvette. Complete anaerobic conditions were achieved by evacuating and flushing with argon five times. The reaction was started by tipping the bc 1 complex solution in the side arm into the main chamber. The measurement and calculation of activity were as described under "Experimental Procedures." The data presented are mean values Ϯ S.D. from five experiments. fer from ubisemiquinone at Qo site. The fact that disruption of electron transfer between two b L hemes of dimeric complex causes a decrease of activity indicates that both monomers are not function independently, some sort of negative cooperativity does exist (9).
Effect of F195A on EPR Characteristics of the b Cytochrome-Although evidence presented above clearly demonstrates that the loss of bc 1 complex activity in mutants, F195A, F195A/ Y199A, F195A/F203A, and F195A/Y199A/F203A, correlates with the leakage of electrons that normally shuttle between the two b L hemes through Phe-195, whether or not mutation F195A affects the microenvironments of the cytochromes b is unknown. To test this possibility, EPR characteristics of the b cytochromes in cytochrome b F195A and wild type were examined after samples were reduced with sodium ascorbate to eliminate the overlapping signal from cytochrome c 1 . As shown in Fig. 4, this mutant has EPR characteristics identical to those of the wild-type complex, with features at g ϭ 3.53 and g ϭ 3.77 previously assigned to cytochrome b H and b L , respectively, and a g ϭ 4.29 signal thought to be due to nonspecifically bound iron (III) (40). These data indicate that substitution of Phe-195 with alanine has no significant effect on the environments of cytochrome b heme. This mutation also has no effect on the midpoint potential of cytochromes b.