Effect of Mutations of Arginine 94 on Proton Pumping, Electron Transfer, and Superoxide Anion Generation in Cytochrome b of the bc1 Complex from Rhodobacter sphaeroides*

Background: The bc1 complex catalyzes electron transfer from ubiquinol to cytochrome c and translocates protons across the membrane. Results: Substitution of Arg-94 decreased the electron transfer activity and proton pumping capability and increased the superoxide production. Conclusion: Arg-94 is an important amino acid for the proton pumping capability of the bc1 complex. Significance: The result is significant for understanding the catalytic mechanism of the bc1 complex. Proton transfer involving internal water molecules that provide hydrogen bonds and facilitate proton diffusion has been identified in some membrane proteins. Arg-94 in cytochrome b of the Rhodobacter sphaeroides bc1 complex is fully conserved and is hydrogen-bonded to the heme propionate and a chain of water molecules. To further elucidate the role of Arg-94, we generated the mutations R94A, R94D, and R94N. The wild-type and mutant bc1 complexes were purified and then characterized. The results show that substitution of Arg-94 decreased electron transfer activity and proton pumping capability and increased O2̇̄ production, suggesting the importance of Arg-94 in the catalytic mechanism of the bc1 complex in R. sphaeroides. This also suggests that the transport of H+, O2, and O2̇̄ in the bc1 complex may occur by the same pathway.

nism (20). Arg-94 in cytochrome b of Rhodobacter sphaeroides corresponds to Arg-79 in yeast. To further elucidate the role of this important amino acid in cytochrome b of R. sphaeroides, we report herein detailed procedures for generating mutations of Arg-94 in cytochrome b and characterize the electron transfer activity, proton pumping capability, pre-steady-state reduction kinetics of hemes (b and c 1 ), and superoxide anion production in purified bc 1 complexes. We found that substitution of Arg-94 decreased electron transfer activity and proton pumping capability and increased superoxide anion production. The transport of H ϩ , O 2 , and O 2 . may follow the same pathway.
Generation of R. sphaeroides Cytochrome bc 1 Mutants-Mutations were constructed using the QuikChange site-directed mutagenesis kit (Stratagene) with supercoiled double-stranded pGEM7Zf(ϩ)-fbcB as a template. Forward and reverse primers were used for PCR amplification ( Table 1). The pGEM7Zf(ϩ)-fbcB plasmid (23) was constructed by ligating the NsiI-XbaI fragment from pRKD418-fbcFBC 6H Q into the NsiI and XbaI sites of the pGEM7Zf(ϩ) plasmid. The NsiI-XbaI fragment from the pGEM7Zf(ϩ)-fbcB m plasmid was ligated into the NsiI and XbaI sites of the pRKD418-fbcFBK m C 6H Q plasmid to generate the pRKD418-fbcFB m C 6H Q plasmid. A plate-mating procedure (24) was used to mobilize the pRKD418-fbcFB m C 6H Q plasmid in Escherichia coli S17 cells into R. sphaeroides BC17 cells. The presence of the engineered mutations was confirmed by DNA sequencing before and after semi-aerobic growth of the cells. The expression plasmid pRKDfbcF m BC H Q was puri-fied from an aliquot of semi-aerobically grown culture using the Qiagen plasmid miniprep kit. Because R. sphaeroides cells contain four types of endogenous plasmids, the isolated plasmids lacked the purity and concentration needed for direct sequencing. Therefore, a 2-kilobase pair DNA segment containing the mutation sequence was amplified from the isolated plasmids by PCR. The PCR products were purified with an extraction kit from Sigma and then sequenced. DNA primers were purchased from Invitrogen. DNA sequencing was carried out at the Recombinant DNA/Protein Core Facility of Oklahoma State University.
Growth of Bacteria-E. coli cells were grown at 37°C in LB medium. R. sphaeroides BC17 cells (24) were grown photosynthetically at 30°C in enriched Siström's medium containing 5 mM glutamate and 0.2% casamino acids. Photosynthetic growth conditions for R. sphaeroides were essential as described previously (25). The concentrations and antibiotics used were as follows: 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 Specific Electron Transfer Activity Assay-Chromatophore membranes were prepared from frozen cell paste, and cytochrome bc 1 complexes with a His 6 tag placed at the C terminus of cytochrome c 1 were purified from chromatophores as described previously (25) and stored at Ϫ80°C in the presence of 10% glycerol. Protein concentrations were determined by absorbance at 280 nm using a converting factor of 1 A 280 ϭ 0.56 mg/ml. The concentrations of cytochromes b and c 1 were determined spectrophotometrically using published molar extinction coefficients (26 -28).
To assay ubiquinol-cytochrome c reductase activity, purified cytochrome bc 1 complexes were diluted with 50 mM Tris-Cl (pH 8.0) containing 200 mM NaCl and 0.01% n-dodecyl-␤-Dmaltopyranoside to a final cytochrome b concentration of 1 M unless specified otherwise. 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), 0.3 mM EDTA, 100 M cytochrome c, and 25 M Q 0 C 10 BrH 2 . The specific electron transfer activities were determined by measuring the reduction of cytochrome c (the increase in the absorbance at a wavelength of 550 nm) with a Shimadzu UV-2401 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 during specific activity calculations. Although the chemical properties of Q 0 C 10 BrH 2 are comparable with those of Q 0 C 10 H 2 , the former is a better substrate for the cytochrome bc 1 complex (22).
Preparation of Electron Transfer Complex-inlaid PL Vesicles-Protein-PL vesicles were prepared by the cholate dialysis method of Kagawa and Racker (21). The wild-type or mutant bc 1 complex was mixed with 1 ml of asolectin micellar solution to give an asolectin (milligrams)/protein (milligrams) ratio of 40. The asolectin micellar solution was prepared by sonicating 200 mg of acetone-washed asolectin in 4 ml of 50 mM sodium phosphate buffer (pH 7.4) containing 2% sodium cholate and 100 mM KCl in an ice-water bath. Sonication was performed in  1 Complex-PL Vesicles-Proton translocation coupled to electron flow through the bc 1 complex-PL vesicles was measured at room temperature using an Accumet Model 10 pH meter and Model 13-620-96 combination pH electrode. 25 nmol of Q 0 C 10 BrH 2 was added to the 1.6-ml reaction mixture containing 150 mM KCl, 4 M ferricytochrome c, 1 M valinomycin, and an appropriate amount of bc 1 complex-PL vesicles (30 -50 l). Electron flow was initiated by the addition of 5 nmol of ferricyanide, which oxidizes the cytochrome c and thus provides an electron acceptor for the complex. Electron flow under conditions in which no transmembrane ⌬pH formed was measured in an identical manner except that the protonophore CCCP was present at a concentration of 2 M to make the vesicles permeable to protons. Proton pumping (H ϩ /e Ϫ ) was calculated as the ratio of the decrease in pH upon ferricyanide addition to bc 1 complex-PL vesicles before and after treatment with CCCP.
Fast Kinetics Study-Measurements were performed in an Applied Photophysics SX.18MV stopped-flow spectrometer with a photodiode array scan between 600 and 500 nm. The reaction was started by mixing equal volumes of solutions A and B at room temperature. For the experiment to determine electron transfer rates between ubiquinol and heme b or c 1 , solution A contained 50 mM Tris-Cl (pH 8.0) at 4°C, 200 mM NaCl, 0.01% n-dodecyl-␤-D-maltopyranoside, and 12.0 M cytochrome bc 1 complex (based on cytochrome c 1 ). Solution B was the same as solution A except that the bc 1 complex was replaced with 240 M Q 0 C 10 BrH 2 . Reductions of cytochromes b and c 1 were determined from the absorption changes at 560 -580 nm and at 551-539 nm, respectively. When an inhibitor was used, the cytochrome bc 1 complex was treated with a 5-fold molar excess of inhibitor over heme c 1 for 5 min at 4°C prior to the experiment.
Because the concentration of ubiquinol used was much higher than that of the cytochrome bc 1 complex, the reactions between bc 1 and ubiquinol were treated as pseudo first-order reactions. Time traces of the reaction were fitted using a firstorder rate equation to obtain the pseudo first-order rate constants (k 1 ) by KaleidaGraph.
For the experiment to determine proton production rates, solution A contained 150 mM KCl, 30 M Q 0 C 10 BrH 2 , 2 M valinomycin, 100 M pyranine, and bc 1 complex-PL vesicles (35 l/ml of solution A). Solution B contained 150 mM KCl and 8 M ferricytochrome c. Solutions A and B were both adjusted carefully to pH 7.0 with 2 mM KOH. Proton pumping was determined from the absorption changes at 457 nm.
Measurement of Superoxide Anion Generation-Superoxide anion generation by the cytochrome bc 1 complex was determined by measuring the chemiluminescence of the MCLA-O 2 .
adduct (29) in an Applied Photophysics SX.18MV stopped-flow spectrometer by leaving the excitation light off and registering the light emission (30). Reactions were carried out at 23°C by mixing solutions A and B at 1:1. Solution A contained 100 mM Na ϩ /K ϩ phosphate buffer (pH 7.4), 1 mM EDTA, 1 mM KCN, 1 mM NaN 3 , 0.1% bovine serum albumin, 0.01% n-dodecyl-␤-Dmaltopyranoside, and 5.0 M wild-type or mutant bc 1 complex. Solution B was the same as solution A except that the bc 1 complex was replaced with 150 M Q 0 C 10 BrH 2 and 4 M MCLA.

RESULTS AND DISCUSSION
Photosynthetic Growth Behaviors of the Wild-type and Mutant bc 1 Complexes-Because the cytochrome bc 1 complex is absolutely necessary for photosynthetic growth of R. sphaeroides, a mutant with a substitution at a critical position will not grow photosynthetically, whereas mutants with substitutions at noncritical positions will grow. Thus, observing the photosynthetic growth behavior, one can determine whether substitutions are at critical positions. In mid-log phase, semiaerobically dark-grown wild-type and mutant cells were inoculated into enriched Siström's medium and subjected to anaerobic photosynthetic growth conditions. Table 2 shows that mutants could grow, indicating that substitutions of Arg-94 are noncritical to the complex. These mutants grew at a decreased rate compared with the wild-type bc 1 complex, suggesting that substitutions of Arg-94 affect the function of the cytochrome bc 1 complex and thus retard the growth of R. sphaeroides.
Proton Translocation Activity of the Wild-type and Mutant bc 1 Complexes-The bc 1 complex catalyzes electron transfer from ubiquinol to cytochrome c and concomitantly translocates protons of ubiquinol across the membrane. The two protons of ubiquinol are released via two pathways. The ejection of the first proton is controlled by the protonation and deprotonation of the histidine ligands of the [2Fe-2S] cluster. The histidine ligands take up a proton from the substrate, ubiquinol, upon reduction of the [2Fe-2S] cluster and release it to the intermembrane space when oxidized by cytochrome c 1 , as was suggested recently (31). Glu-295 in cytochrome b is important for the release of the second proton, as proposed previously (31)(32)(33). Glu-295 is completely conserved in mitochondrial cytochrome b (34), and the importance of the residue in proton transfer is indicated by mutagenesis studies because alteration of glutamine abolishes ubiquinol oxidation in R. sphaeroides (35). Additionally, recent kinetic studies showed that protonation of a group with a pK a of 5.7 blocked catalysis, and this effect was attributed to Glu-295 (36). The second proton of ubiquinol is first transferred to Glu-295 to form the neutral acid and is then released and delivered to heme propionate A by rotation of the side chain of Glu-295. The subsequent proton release is mediated by a hydrogen-bonded water chain stabilized by cytochrome b residues (Fig. 1) (20, 31). a The enzyme-specific electron transfer activity of the purified bc 1 complexes is expressed as micromoles of cytochrome c reduced per min/nmol of cytochrome b at room temperature.
The proton pumping property has been well demonstrated in PL vesicles embedded with the purified mitochondrial bc 1 complex (37)(38)(39)(40). Fig. 2 shows a typical proton translocation activity assay for PL vesicles embedded with the R. sphaeroides bc 1 complex. After pH equilibrium was reached for the mixture containing Q 0 C 10 BrH 2 , valinomycin, ferricytochrome c, and bc 1 complex-PL vesicles, an aliquot of ferricyanide solution (arrow 1) was added to initiate electron flow from Q 0 C 10 BrH 2 to cytochrome c. After pH equilibrium was attained, the protonophore CCCP was added (arrow 2) to render the liposome membrane freely permeable to protons, and then a second aliquot of ferricyanide was added (arrow 3). The ratio of the pH changes produced by the addition of equal amounts of ferricyanide before (x) and after (y) the addition of CCCP was taken as a measure of the H ϩ /e Ϫ ratio for the proton translocation activity of the bc 1 complex. The protons released after CCCP addition were the "scalar" protons only, as no contribution from the accumulation of "vectorially" translocated protons was possible. Table 3 shows the proton translocation activity (H ϩ /e Ϫ ) of the wild-type and mutant bc 1 complexes. The proton translocation activity of R94N showed a small decrease compared with the wild-type bc 1 complex, whereas that of R94A and R94D exhibited a significant decrease. H ϩ /e Ϫ reflects only the final ratio of proton translocation to electron transfer. To further study the proton translocation rate, a kind of dye, pyranine, was used in a stopped-flow experiment. Fig. 3 shows that the magnitude of proton translocation rate of the wild-type and mutant bc 1 complexes can be ranked as WT Ͼ R94N Ͼ R94D Ͼ R94A. These results indicate that a positive correlation exists between the changes in proton translocation activity (H ϩ /e Ϫ ) and the proton translocation rate.
Proton transfer pathways involving internal water molecules that provide hydrogen bonds and facilitate proton diffusion have been identified in other membrane proteins such as bac-teriorhodopsin (41) and cytochrome c oxidase (42). Although the mechanisms for proton translocations are different for different proteins, the underlying principle for bringing protons to the active sites is similar. For Arg-94, the nitrogen atom of the guanidyl easily forms a hydrogen bond with H 2 O, and the repulsion function of the positive charge to H ϩ is also favorable for H ϩ movement. For Ala, the carbon atom of the methyl does not easily form a hydrogen bond with H 2 O, and there is no positive charge on its side chain. In addition, as a hydrophobic amino acid, the hydrophobic microenvironment is also unfavorable    for H ϩ movement. Thus, the proton translocation activity of the R94A mutant is inhibited. The structure of Asn is similar to that of Arg except for a small difference in molecular weight, so the proton translocation activity of the R94N mutant is similar to that of the wild type. The low proton translocation activity of R94D may be due to the negative charge of the side chain of Asp exerting an attractive force to H ϩ , which is unfavorable for H ϩ movement. The low proton translocation activity of the mutants indicates that some of the "second H ϩ " of Q 0 C 10 BrH 2 cannot be pumped into the exterior of the vesicle and also suggests that Arg-94 in cytochrome b of the bc 1 complex is an important amino acid for proton pumping. Fig. 4 shows the time course traces of heme b reduction by Q 0 C 10 BrH 2 in the wild type and mutants in the absence (panel A) and presence (panel B) of antimycin A. Antimycin A is a Q N site inhibitor that blocks electron transfer from heme b H to ubiquinone and prevents reduction of heme b H by Q 0 C 10 BrH 2 through the Q N site.

Effect of Mutations on Cytochrome b Reduction by Ubiquinol-
In the case of the native complex (Fig. 4A), the extent of heme b reduction of mutants decreased relative to the wild type. It should be noted that heme b reduction is mostly heme b H (43). The heme b H reduction by ubiquinol in the bc 1 complex, according to the Q-cycle mechanism, is affected by (i) the forward reduction through the Q P site to heme b L and then to b H , (ii) the reoxidation of reduced heme b H by ubiquinone, and (iii) the "back door" reduction through the Q N site. Therefore, it is difficult to explain the differences in the extent of heme b reduction among the wild type and mutants.
In the presence of antimycin A (Fig. 4B), heme b reduction by Q 0 C 10 BrH 2 occurred only through the Q P site via heme b L , so this case is simplified. Relative to the wild type, the heme b reduction extent of R94N, R94D, and R94A decreased by 2.4, 77.71, and 84.93%, respectively. This decrease implies that some electrons may deviate from the low potential electron transfer chain and leak from the reduced heme b L or ubisemiquinone at the Q P site during bc 1 catalysis. The results also indicate that the extent of the electron leak is WT Ͻ R94N Ͻ R94D Ͻ R94A.
Whether in the absence or presence of antimycin A, the rate constants of heme b reduction of the mutants all decreased relative to the wild type, and the extent of the decrease can be described as R94N Ͻ R94D Ͻ R94A (Fig. 4). The decrease in the rate constants of heme b reduction suggests that the transfer rate of the second electron of Q 0 C 10 BrH 2 is decreased for mutants.
The addition of antimycin A (Fig. 4) increased the extent and decreased the rate of heme b reduction in the wild type and mutants, and these results are consistent with a previous report (44). However, the degree of this effect varied in these complexes. Relative to the native case (Fig. 4A), the rate of heme b reduction of the wild type, R94N, R94D, and R94A was decreased by 52, 43, 50, and 58%, respectively. Relative to the native case (Fig. 4A), the extent of heme b reduction of the wild type, R94N, R94D, and R94A was increased by 83, 255, 25, and 56%, respectively. Fig. 5 shows the time course traces of heme c 1 reduction by Q 0 C 10 BrH 2 in the wild type and mutants in the absence (panel A) and presence (panel B) of antimycin A. In both the absence and presence of antimycin A, the rate constants of heme c 1 reduction of all mutants decreased relative to the wild type, and the extent of the decrease is R94N Ͻ R94D Ͻ R94A. The decrease in the rate constants of heme c 1 reduction suggests that the transfer rate of the first electron of Q 0 C 10 BrH 2 is decreased for mutants.

Effect of Mutations on Cytochrome c 1 Reduction by Ubiquinol-
Relative to the values for the native case, the addition of antimycin A decreased the rate of heme c 1 reduction by 18, 9, 9, and 73% in the wild type, R94N, R94D, and R94A, respectively (Fig.  5). These results are consistent with previous reports (18,43) that antimycin A has a significant effect on the reduction rate of heme c 1 in cytochrome bc 1 complexes. This inhibitive effect has been attributed to the long-range effect of antimycin A on the Q P site when binding to the Q N site (18,43). In other words, the effect of antimycin A on the heme c 1 reduction rate does not occur through the low potential redox component but through the cytochrome b protein subunit.
In the Q-cycle mechanism, the oxidation of ubiquinol and reduction of ubiquinone or ubisemiquinone occur mainly in the Q P and Q N sites of cytochrome b. In the mutant bc 1 complexes, the Q P and Q N sites are functional because all of the mutant complexes were sensitive to antimycin A (Figs. 4B and 5B) and stigmatellin (a Q P site inhibitor) (data not shown). The changes in electron transfer activity or the changes in heme b and c 1 reduction should all be related to the mutation at Arg-94, but they are not related to the impairment of the Q P or Q N site.
Substitution of Arg-94 decreased the proton translocation rate, slowed down the decomposition rate of Q 0 C 10 BrH 2 by feedback inhibition, and then slowed down the transfer of the two electrons of ubiquinol. A decrease in the first electron transfer induced the heme c 1 reduction rate decrease and the electron transfer activity decrease ( Table 2). A decrease in the second electron transfer induced the heme b reduction rate decrease (Fig. 4).
The decreases in the proton translocation rate, the heme b and c 1 reduction rate, and the electron transfer activity all changed in the same order of R94N Ͻ R94D Ͻ R94A for the three mutants. This positive correlation also supports the relationship discussed above. In addition, the decreases in the heme b reduction rate and the electron transfer activity are also related to the production of superoxide anion, as discussed below.
Effect of Mutations on Superoxide Anion Production by the Cytochrome bc 1 Complex-The above results show that some electrons may leak from the low potential electron transfer chain during bc 1 catalysis for mutants. What is the fate of these leaked electrons? One of the possible pathways is a reaction with molecular oxygen to form superoxide anions. If this is the pathway, one should see more superoxide anion production by mutants than the wild-type complex because more electrons leak from the mutant complexes. Fig. 6    catalysis. The result also indicates that the extent of the electron leak is WT Ͻ R94N Ͻ R94D Ͻ R94A, which is the same as the result from the heme b reduction analysis above. There is an inverse relationship between O 2 . production and the heme b generation as well as to the decreased proton pumping (discussed above). Under anaerobic conditions, there is no O 2 . generation. Thus, one can deduce that the difference between activities in a mutant and the wild type under anaerobic conditions should be less than that under aerobic conditions. This is indeed the case. Fig. 7 shows that, compared with the wild type, the electron transfer activities of R94N, R94D, and R94A decreased by 49, 65, and 96%, respectively, under aerobic conditions, whereas they decreased by 41, 61, and 94%, respectively, under anaerobic conditions. The difference is significant (p Ͻ 0.05, Student's t test) between aerobic and anaerobic conditions for every mutant complex. These data also show that the effect of the superoxide anion generation on the decrease in the electron transfer activity of mutants is R94N Ͼ R94D Ͼ R94A, and the superoxide anion generation has a smaller effect on the decrease in the electron transfer activity of mutants compared with the proton pumping decrease. in the hydrophobic environment of the Q P pocket through bifurcated oxidation of ubiquinol by transfer of its two elec-trons to a high potential electron acceptor, an iron-sulfur cluster, and a low potential heme b L or molecular oxygen. The protein subunits, at least those surrounding the Q P pocket, play a role either in preventing the release of O 2 . from its production site to aqueous environments or in preventing O 2 from getting access to the hydrophobic Q P pocket (45). The increase in superoxide anion production by mutants (Fig. 6)  movement. In this study, only one residue was mutated and studied, so this suggestion remains to be further investigated.  The blue arrow indicates protons that return to the matrix for mutants; the purple arrow indicates oxygen molecules that are transferred to the Q P quinol oxidation pocket from the cytosol solvent. Q N is the quinone reduction pocket. QH 2 , ubiquinol; TM, transmembrane region; ISP, Rieske iron-sulfur protein.