Effect of Mutations in the Cytochrome b ef Loop on the Electron Transfer Reactions of the Rieske Iron-Sulfur Protein in the Cytochrome bc₁ Complex

Abstract

Long range movement of the iron-sulfur protein (ISP) between the cytochrome b (cyt b) and cyt c₁ redox centers plays a key role in electron transfer within the cyt bc₁ complex. A series of 21 mutants in the cyt b ef loop of Rhodobacter sphaeroides cyt bc₁ were prepared to examine the role of this loop in controlling the capture and release of the ISP from cyt b. Electron transfer in the cyt bc₁ complex was studied using a ruthenium dimer to rapidly photooxidize cyt c₁ within 1 μs and initiate the reaction. The rate constant for electron transfer from the iron-sulfur center [2Fe2S] center to cyt c₁ was k₁ = 60,000 s⁻¹. Famoxadone binding to the Q_o site decreases k₁ to 5,400 s⁻¹, indicating that a conformational change on the surface of cyt b decreases the rate of release of the ISP from cyt b. The mutation I292A on the surface of the ISP binding crater decreased k₁ to 4,400 s⁻¹, while addition of famoxadone further decreased it to 3,000 s⁻¹. The mutation L286A at the tip of the ef loop decreased k₁ to 33,000 s⁻¹, but famoxadone binding caused no further decrease, suggesting that this mutation blocked the conformational change induced by famoxadone. Studies of all the mutants provide further evidence that the ef loop plays an important role in regulating the domain movement of the ISP to facilitate productive electron transfer and prevent short-circuit reactions.

The cytochrome bc₁ complex (ubiquinol:cytochrome c reductase) is an integral membrane protein in the electron transport chains of mitochondria and many respiratory and photosynthetic prokaryotes (1, 2). The complex contains the Rieske iron-sulfur protein (ISP), cyt c₁, and two b-type hemes (b_L and b_H) in the cyt b subunit (1, 2). The complex translocates four protons to the positive side of the membrane as two electrons are transferred from ubiquinol (QH₂) to cyt c using a widely-accepted Q-cycle mechanism (2). In a key bifurcated reaction, QH₂ binds to the Q_o site located near the outside of the membrane, and transfers its first electron to the Rieske iron-sulfur center ([2Fe2S]), and then to cyt c₁ and cyt c (1,2). The second electron is transferred from semiquinone in the Q_o site to cyt b_L and then to cyt b_H, leading to reduction of ubiquinone in the Q_i site to the semiquinone. This cycle is repeated to reduce the semiquinone at the Q_i site to QH₂. A recent freeze-quench EPR study has shown that QH₂ reduced [2Fe2S] and cyt b_L simultaneously with a half-time of 250 μs, providing direct evidence for the bifurcation mechanism, and indicating that either the semiquinone is very short-lived or the reaction is concerted (3). X-ray crystallographic studies have shown that the conformation of the ISP depends on the presence of inhibitors in the Q_o site as well as the crystal form (4-7). An anomalous signal for [2Fe2S] is found close to cyt b_L in native I4₁22 bovine crystals, but its intensity is small, indicating that the ISP is conformationally mobile (4,7). In beef, chicken and yeast cyt bc₁ crystals grown in the presence of stigmatellin, the ISP is in a conformation with [2Fe2S] proximal to the cyt b_L heme, called the b state (5-8). However, in native chicken or beef P6₅22 crystals in the absence of Q_o site inhibitors, the ISP is in a conformation with [2Fe2S] close to cyt c₁, called the c₁ state (5,6). A novel shuttle mechanism for the ISP has been proposed based on these structural studies (4-7). With the ISP initially in the b state, QH₂ in the Q_o site transfers an electron to the oxidized [2Fe2S] center. The ISP then rotates by 57^o to the c₁ state where reduced [2Fe2S] transfers an electron to cyt c₁. The main features of this mobile shuttle mechanism have been supported by a variety of experimental studies using mutation and/or cross-linking to immobilize the ISP or alter the conformation of the neck region (9-20).

An important question regarding the Q-cycle mechanism is how QH₂ at the Q_o site can deliver two electrons sequentially to the high and low potential chains, even though thermodynamics would favor delivery of both electrons to the high potential chain (1,2). A number of mechanisms have been proposed in which the conformations of the ISP and/or the quinone substrates are controlled to favor the reversible oxidation of QH₂ and minimize short circuit reactions (21-29). Studies of the effects of Q_o site inhibitors have indicated that there is a linkage between the occupant of the Q_o site and the conformation and dynamics of the ISP. Stigmatellinforms a hydrogen bond with the His-161 ligand of the [2Fe2S] center, thus increasing its redox potential by 200−250 mV and immobilizing the ISP in the b conformation (5-8, 30). Famoxadone binding to the Q_o site leads to significant conformational changes on the surface of cyt b which trigger a long-range conformational change in the ISP from the mobile state to a state with [2Fe2S] proximal to cyt b (31). The ef loop plays an important role in relaying conformational changes in the Q_o pocket to surface domains that control the binding of the ISP. Moreover, molecular dynamics simulations have shown that residues 263−268 of the ef loop are displaced by up to 2 Å as the ISP rotates from the b state to the c₁ state (32). In this paper, the effects of mutations of ef loop residues on electron transfer from QH₂ to the iron-sulfur center and then to cyt c₁ are studied using the binuclear ruthenium complex, Ru₂D, to rapidly photooxidize cyt c₁ and initiate the reaction (9, 17). The ruthenium photooxidation method provides a unique way to measure the dynamics of the ISP domain movement from the b state to the c₁ state (9, 17, 33).

Experimental Procedures

Materials

A modification of the method of Downard et al. (34) was used to prepare Ru₂D. Succinate, p-benzoquinone and antimycin A were obtained from Sigma Chemical Co., stigmatellin was purchased from Fluka Chemical Co., and N-Dodecyl-β-D-maltoside was obtained from Anatrace. [Co(NH₃)₅Cl]²⁺ was synthesized as described in reference (35), and 2,3-dimethoxy-5-methyl-6-(10-bromodecyl-) 1,4-benzoquinol, Q₀C₁₀BrH₂, was prepared as previously reported (36). Succinate cytochrome c reductase (SCR) was purified as previously described (37).

Generation of R. sphaeroides strains expressing the his6-tagged bc₁ complexes

The Quickchange™ XL site-directed mutagenesis kit from Stratagene was employed for mutagenesis. Plasmid pGEM7Zf(+)-fbcB was used as a template and forward and reverse primers were used for PCR amplification. Template plasmid pGEM7Zf(+)-fbcB was constructed by ligating the fragment between NsiI and XbaI in the plasmid pRKDfbcFBC_6HQ into NsiI and XbaI sites of the pGEM7ZF(+) plasmid. The primers used are listed in the Table 1.

Table 1.

Primers used during mutagenesis operations

Mutant (in the cytochrome b)	Primers
H276A (D252)	5′-GCCGAACTACCTCGGCGCCCCCGACAACTACATCG-3′
	3′-CGGCTTGATGGAGCCGCGGGGGCTGTTGATGTAGC-5′
P277A (P253)	5′-CGAACTACCTCGGCCACGCCGACAACTACATCGAGG-3′
	3′-GCTTGATGGAGCCGGTGCGGCTGTTGATGTAGCTCC-5′
D278A (D254)	5′-CTACCTCGGCCACCCCGCCAACTACATCGAGGCGAAC-3′
	3′-GATGGAGCCGGTGGGGCGGTTGATGTAGCTCCGCTTG-5′
D278N	5′-CTACCTCGGCCACCCCAACAACTACATCGAGGCGAAC-3′
	3′-GATGGAGCCGGTGGGGTTGTTGATGTAGCTCCGCTTG-5′
D278E	5′-CTACCTCGGCCACCCCGAAAACTACATCGAGGCGAAC-3′
	3′-GATGGAGCCGGTGGGGCTTTTGATGTAGCTCCGCTTG-5′
D278I	5′-CTACCTCGGCCACCCCATCAACTACATCGAGGCGAAC-3′
	3′-GATGGAGCCGGTGGGGTAGTTGATGTAGCTCCGCTTG-5′
D278H	5′-CTACCTCGGCCACCCCCACAACTACATCGAGGCGAAC-3′
	3′-GATGGAGCCGGTGGGGGTGTTGATGTAGCTCCGCTTG-5′
N279A (N255)	5′-CTCGGCCACCCCGACGCCTACATCGAGGCGAACCC-3′
	3′-GAGCCGGTGGGGCTGCGGATGTAGCTCCGCTTGGG-5′
Y280A (Y256)	5′-GGCCACCCCGACAACGCCATCGAGGCGAACCCGC-3′
	3′-CCGGTGGGGCTGTTGCGGTAGCTCCGCTTGGGCG-5′
L286A (L262)	5′-CATCGAGGCGAACCCGGCCTCGACGCCCGCGCAC-3′
	3′-GTAGCTCCGCTTGGGCCGGAGCTGCGGGCGCGTG-5′
L286E	5′-CATCGAGGCGAACCCGGAGTCGACGCCCGCGCAC-3′
	3′-GTAGCTCCGCTTGGGCCTCAGCTGCGGGCGCGTG-5′
L286I	5′-CATCGAGGCGAACCCGATCTCGACGCCCGCGCAC-3′
	3′-GTAGCTCCGCTTGGGCTAGAGCTGCGGGCGCGTG-5′
S287A (N263)	5′-CGAGGCGAACCCGCTCGCCACGCCCGCGCACATCG-3′
	3′-GCTCCGCTTGGGCGAGCGGTGCGGGCGCGTGTAGC-5′
A290S (A266)	5′-CCGCTCTCGACGCCCTCGCACATCGTGCCGG-3′
	3′-GGCGAGAGCTGCGGGAGCGTGTAGCACGGCC-5′
H291A (H267)	5′-GCTCTCGACGCCCGCGGCCATCGTGCCGGAATGG-3′
	3′-CGAGAGCTGCGGGCGCCGGTAGCACGGCCTTACC-5′
I292A (I268)	5′-CTCGACGCCCGCGCACGCCGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGCGGCACGGCCTTACCATGAAG-5′
I292L	5′-CTCGACGCCCGCGCACCTGGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGGACCACGGCCTTACCATGAAG-5′
I292M	5′-CTCGACGCCCGCGCACATGGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGTACCACGGCCTTACCATGAAG-5′
I292V	5′-CTCGACGCCCGCGCACGTCGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGCAGCACGGCCTTACCATGAAG-5′
I292E	5′-CTCGACGCCCGCGCACGAGGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGCTCCACGGCCTTACCATGAAG-5′
I292R	5′-CTCGACGCCCGCGCACCGCGTGCCGGAATGGTACTTC-3′
	3′-GAGCTGCGGGCGCGTGGCGCACGGCCTTACCATGAAG-5′

After mutagenesis, NsiI-XbaI fragment from the pGEM7Zf(+)-fbcBm was ligated into NsiI and XbaI sites of the pRKDfbcFB_bpKmC_6H plasmids which contained Kanamycin-resistant gene in the BstI and pinA sites (38). Strains containing recombinant plasmid will be sensitive to antibiotics kanamycin. The pRKDfbcFBmC_6HQ plasmid in E. coli S17−1 cells was mobilized into R. sphaeroides BC-17 through conjugation (38). The engineered mutations were confirmed by DNA sequencing before and after photosynthetic growth as previously reported (38), which was performed by the Recombinant DNA/Protein Core Facility at Oklahoma State University.

Mutant cytochrome bc₁ was purified as described by Xiao et al.(39). The steady-state activity of bc₁ complexes was determined as described by Liu et al.(40).

Flash Photolysis Experiments

Flash photolysis experiments were carried out on 300 μL solutions contained in a 1-cm glass semimicrocuvette using the detection system described by Heacock et al. (41). A Phase R model DL1400 flash lamp-pumped dye laser using coumarin LD 490 produced a 480 nm light flash of < 0.5 μs duration. Samples typically contained 5 μM cyt bc₁, 20 μM Ru₂D, 5 mM [Co(NH₃)₅Cl]²⁺, in 20 mM sodium borate, pH 9.0, 0.01% dodecylmaltoside. The [Co(NH₃)₅Cl]²⁺ was used as a sacrificial electron acceptor. The cyt bc₁ was treated with 10 μM Q_oC₁₀BrH₂, 1 mM succinate, and 50 nM SCR to completely reduce [2Fe2S] and cyt c_1, and reduce cyt b_H by 30%. The photooxidation and reduction of cyt c₁ was monitored at 552 nm, while the reduction of cyt b_H was monitored at 561 − 569 nm. Under the conditions used the reoxidation of cyt b_H by Q in the Q_i site was much slower than reduction because of the low concentration of oxidized Q.

Results

The kinetics of electron transfer within R. sphaeroides cytochrome bc₁ were studied using the ruthenium dimer Ru₂D to photoinitiate the reaction (9). Laser flash photolysis of a solution containing Ru₂D and cyt bc₁ with cyt c₁ and [2Fe2S] initially reduced resulted in rapid photooxidation of cyt c₁, followed by electron transfer from from [2Fe2S] to cyt c₁ with a rate constant of k₁ = 60,000 s⁻¹, monitored at 552 nm (Figure 1) (9). The oxidant-induced reduction of cyt b_H monitored at 561 − 569 nm has a rate constant k₂ = 2,300 s⁻¹ (Figure 1), which is rate-limited by electron transfer from QH₂ to [2Fe2S] with rate constant k₂ followed by rapid electron transfer from [2Fe2S] to cyt c₁, and from the semiquinone to cyt b_L and cyt b_H (Scheme 1) (9). Binding famoxadone to the Q_o site decreases the rate constant k₁ from 60,000 s⁻¹ to 5,400 s⁻¹, indicating that the dynamics of rotation of the ISP from the b state to the c₁ state was significantly decreased (Figure 1) (33). Since famoxadone binding affected the conformation of the ef loop in cyt b, it was suggested that the ef loop might play a role in controlling the conformation of the ISP protein (31, 33). In order to examine this conformational link further, a series of mutants in which ef loop residues were substituted with Ala were prepared. These included the ef loop residues H276, P277, D278, N279, Y280, L286, S287, A290, H291, and I292 (Figures 2, 3). The steady-state activity and the values of the rate constants k₁ and k₂ were measured for each of these mutants, both in the presence and absence of famoxadone. A number of the mutants caused relatively small decreases in the rate constant k₁, but the largest changes were seen for the mutants D278A, Y280A, L286A, and I292A (Table 2). Additional mutants at residues D278, L286 and I292 were prepared to further explore the roles of these residues.

Figure 1.

Electron transfer within wild-type R. sphaeroides cyt bc₁ following photooxidation of cyt c₁ (9). The 300 μL sample contained 5 μM cyt bc₁, 20 μM Ru₂D, 5 mM [Co(NH₃)₅Cl]²⁺, in 20 mM sodium borate, pH 9.0 with 0.01% dodecylmaltoside. The cyt bc₁ was treated with 10 μM Q_oC₁₀BrH₂, 1 mM succinate, and 50 nM SCR to completely reduce [2Fe2S] and cyt c₁, and reduce cyt b_H by about 30%. The sample was then excited with a 480 nm laser flash to photooxidize cyt c₁ within 1 μs. (Top two traces) The 552 nm transient indicates that cyt c₁ was photooxidized within 1 μs, and then reduced in a biphasic reaction with rate constants of 60,000 s⁻¹ and 2,000 s⁻¹. The rate constant for the reduction of cyt b_H measured at 561 − 569 nm was 2,300 s⁻¹. (Bottom two traces) Addition of 30 μM famoxadone decreased the rate of reduction of cyt c₁ to 5,400 s⁻¹ and eliminated reduction of cyt b_H.

Scheme 1.

Figure 2.

X-ray crystal structure of bovine cyt bc₁ with bound famoxadone (31). The cyt c₁ and cyt b_L hemes are colored red, the [2Fe2S] center is represented by a CPK model, the ISP is blue, and cyt b is grey. Residues 252−268 in the ef loop are colored orange while residues 269−283 in the PEWY sequence and the ef helix are red. Residues 136−152 in the cd1 helix are green and residues 163−171 in the neck-contacting domain are colored yellow. The residues in cyt b that were mutated are shown as sticks and labeled with R. sphaeroides sequence numbering.

Figure 3.

Sequence alignment of residues near the Q_o pocket of the cyt b subunit in cyt bc₁ complexes from bovine (BT) and R. sphaeroides (RS). Residues with a high sequence similarity are in boldface. Helices determined crystallographically are indicated by gray rectangles. Residues in direct contact with the ISP are indicated with a black dot. Adopted from reference 21.

Table 2.

Effect of the mutations in the ef loop of cyt b on the kinetics of electron transfer within R. sphaeroides cyt bc₁.

Cyt bc1 Mutanta	Specific Activityb	k1c (s−1)	Famoxadone k1 (s−1)	k2d(s−1)	Δ CAe(Å)
WT	2.35	60,000	5,400	2,300
H276A (D252)	1.56	42,000	2,800	2,200	3.3
P277A (P253)	1.10	35,000	4,500	1,700	3.0
D278A (D254)	1.43	26,000	2,300	2,300	2.9
D278N	1.88	35,000	2,500	2,400
D278E	1.60	51,000	6,700	2,900
D278I	2.41	38,000	4,300	2,000
D278H	1.49	36,000	4,100	1,500
N279A (N255)	2.43	37,000	4,300	1,900	1.5
Y280A (Y256)	1.34	7,900	3,200	2,800	0.5
L286A (L262)	0.78	33,000	35,000	740
L286E	0.88	17,000	12,000	1,900
L286I	2.40	18,000	2,900	2,300
S287A (N263)	2.16	55,000	5,700	1,000	0.7
A290S (P266)	2.65	49,000	5,600	1,900	0.9
H291A (H267)	2.54	51,000	4,100	2,000	1.4
I292A (I268)	0.81	4,400	3,000	350	1.4
I292L	1.30	10,000	4,700	1,500
I292M	0.92	6,200	2,800	1,700
I292V	0.92	33,000	3,500	1,300
I292E	0.12	-	-	-
I292R	0.56	10,000	2,500	1,700

^aThe corresponding residues of bovine cyt b are given in parenthesis.

^bSpecific activity is expressed as micromole of cyt c reduced/min/nmol cyt b at 25 °C using chromatophores. The error limits are ± 15%.

^cThe rate constant k₁ for electron transfer from [2Fe2S] to cyt c₁ was measured at 552 nm in a solution containing 5 μM cyt bc₁, 20 μM Ru₂D, 5 mM [Co(NH₃)₅Cl]²⁺, in 20 mM sodium borate, pH 9.0, 0.01% dodecylmaltoside. The cyt bc₁ was treated with 10 μM Q_oC₁₀BrH₂, 1 mM succinate, and 50 nM SCR to completely reduce [2Fe2S] and cyt c₁, and reduce cyt b_H by about 30%. Famoxadone (30 μM) was added where indicated. The error limits are ± 20%.

^dThe rate constant k₂ for electron transfer from QH₂ to [2Fe2S] was measured from the rate of cyt b_H reduction at 561−569 nm under the conditions described for k₁ above. The error limits are ± 20%. Addition of famoxadone eliminated that 561−569 nm transients for all of the mutants, indicating that famoxadone bound to all the mutants in the Q_o pocket.

^eΔCA is the displacement of the α carbon of the indicated residue in bovine cyt bc₁ induced by famoxadone binding (31).

The mutation I292A decreased the steady-state activity from 2.35 to 0.81, k₁ from 60,000 s⁻¹ to 4,400 s⁻¹ and k₂ from 2,300 s⁻¹ to 350 s⁻¹ (Figure 4A). Famoxadone binding had a relatively small additional effect on k₁, decreasing it only to 3,000 s⁻¹. All of the mutations at I292 led to significant decreases in k₁ and k₂ (Table 2). The I292L mutation decreased k₁ to 10,000 s⁻¹, while the I292M mutation decreased k₁ to 6,200 s⁻¹ (Figures 4B, 4C). Famoxadone binding had a rather small effect on these two mutants, decreasing k₁ to 4,700 s⁻¹ and 2,800 s⁻¹, respectively. The largest effect was observed for the I292E mutation, which decreased the steady-state activity to 0.12, and eliminated all electron transfer transients in the ruthenium photooxidation experiment. Famoxadone binding eliminated the reduction of cyt b_H observed at 561 nm for all of the mutants, indicating that famoxadone binding was not affected.

Figure 4.

Electron transfer from [2Fe2S] to cyt c₁ in R. sphaeroides cyt bc₁ mutants following photooxidation of cyt c₁. The conditions were the same as described in Figure 1. (A) I292A cyt bc₁. The rate constant k₁ is 4,400 s⁻¹ in the absence of inhibitor, and 3000 s⁻¹ in the presence of 30 μM famoxadone. (B) I292L cyt bc₁. The rate constant k₁ is 10,000 s⁻¹ in the absence of inhibitor, and 4,700 s⁻¹ in the presence of famoxadone. (C) I292M cyt bc₁. The rate constant k₁ is 6,200 s⁻¹ in the absence of inhibitor, and 2,800 s⁻¹ in the presence of famoxadone.

The mutation L286A decreased the steady-state rate by 3-fold, k₁ to 33,000 s⁻¹, and k₂ to 740 s⁻¹. Most surprisingly, famoxadone binding did not significantly affect the rate constant k₁, but it did completely inhibit reduction of cyt b_H. The L286E mutant decreased k₁ to 17,000, while addition of famoxadone only decreased k₁ to 12,000. The conservative mutation L286I decreased k₁ to 18,000, but famoxadone binding decreased k₁ by a larger factor, to 2,900.

The D278A mutation decreased the steady-state activity from 2.35 to 1.43, and the rate constant k₁ from 60,000 s⁻¹ to 26,000 s⁻¹, but did not affect the rate constant k₂. Famoxadone binding decreased the rate constant k₁ to 2,300 s⁻¹. The mutants D278N, D278E, D278I, and D278H did not have as large an effect on k₁ or k₂ as the D278A mutation. The Y280A mutation decreased the steady-state activity from 2.35 to 1.34, and the rate constant k₁ from 60,000 s⁻¹ to 7,900 s⁻¹, without significantly affecting the rate constant k₂. Famoxadone did bind well to this mutant, as indicated by the complete inhibition of reduction of cyt b_H, but it only decreased the rate constant k₁ from 7,900 s⁻¹ to 3,200 s⁻¹.

Discussion

Changes in the conformation of the ISP play an essential role in the mechanism of cyt bc₁, and it is important to understand how the dynamics of these conformational changes control electron transfer within the complex. The ruthenium photooxidation technique provides a unique method to measure the rate constants for electron transfer from the [2Fe2S] center to cyt c₁, as well as from QH₂ to the [2Fe2S] center. A number of experiments have indicated that the measured rate constant k₁ for electron transfer from [2Fe2S] to cyt c₁ is rate-limited by a conformational gating mechanism (33). If k₁ was rate-limited by true electron transfer from [2Fe2S] to cyt c₁, then Marcus theory would predict a large dependence on the driving force of the reaction, which depends on the difference in redox potentials of [2Fe2S] and cyt c₁. However, k₁ is independent of the protonation state of His-161, which significantly affects the redox potential of [2Fe2S] (33). Moreover, the rate constant is not affected by ISP mutations which decrease the redox potential of [2Fe2S] by up to 160 mV (33). These studies indicate that the rate constant k₁ provides a direct measure of the dynamics of the conformational change of the ISP from the b-state and/or the mobile state to the c₁ state (33, 42).

A number of mechanisms have been proposed for the bifurcated electron transfer reaction at the Q_o site in which the first electron is transferred from QH₂ to [2Fe2S], while the second electron is transferred from semiquinone to cyt b_L (21-29). An important requirement for these mechanisms is that they account for the reversibility of the reaction, as well as minimization of short circuit reactions, such as the delivery of both electrons from QH₂ to [2Fe2S]. Double gating mechanisms have been proposed in which QH₂ only reacts at the active Q_o site when the ISP is in the b-state and both [2Fe2S] and cyt b_L are oxidized (23, 25-29). In another proposal, the semiquinone anion moves from the distal site near [2Fe2S] to a position near oxidized cyt b_L to favor electron transfer to the latter (22, 27). Concerted mechanisms have also been proposed in which QH₂ transfers both electrons to [2Fe2S] and cyt b_L simultaneously without formation of a semiquinone intermediate (23, 26, 28, 29). A recent freeze-quench EPR experiment has shown that QH₂ reduces [2Fe2S] and cyt b_L simultaneously with a half-time of 250 μs for a fast phase which accounts for about 10% reduction of each redox center (3). No semiquinone was detected at the Q_o site during oxidation of QH₂, consistent with earlier studies. It is apparent that either the semiquinone intermediate is very short-lived, or the bifurcation reaction involves a concerted mechanism.

It has been proposed that a structural linkage between the quinol substrate in the Q_o binding site and the conformation of the ISP might play an important role in the mechanism of bifurcated electron transfer (21). Unfortunately, it has not been possible to experimentally detect QH₂, semiquinone, or Q in the Q_o binding pocket. Nevertheless, Q_o site inhibitors have many similarities to the natural ubiquinol substrate, and it has been suggested that the effects of these inhibitors on the conformational linkage between cyt b and the ISP might provide insight into the catalytic mechanism of the bifurcation reaction (21). X-ray crystallographic studies have shown that Q_o site inhibitors including stigmatellin, UHDBT, famoxadone, and JG-144 displace both the cd1 helix and the PEWY sequence in the ef helix outward to expand the Q_o pocket and form a binding crater which captures the ISP in the b-state conformation (21, 43). Stigmatellin binding leads to the largest changes in the Q_o pocket, and forms a hydrogen bond with the His-161 ligand of the reduced [2Fe2S] center, thus increasing its redox potential by 200 − 250 mV, and completely immobilizing the ISP in the b conformation (5-8, 30). It has been proposed that QH₂ might bind in a conformation similar to that of stigmatellin, with a hydrogen bond to His-161 to facilitate proton-coupled electron transfer to [2Fe2S] (22).

Famoxadone binds somewhat deeper in the Q_o pocket than stigmatellin, and does not form a hydrogen bond with His-161 on the ISP (31). Even though famoxadone does not contact the ISP, its binding leads to extensive conformational changes on the surface of cyt b which are correlated with the capture of the ISP from the loose state to the b state (31). These changes include one domain between residues 160−175 (bovine numbering) that contacts the neck region of the ISP, another domain between residues 262 and 268 at the end of the ef loop which forms the docking crater for the ISP, and the middle of the ef loop (residues 252−256) which connects the Q_o pocket with the other two surface domains (Figures 2, 3) (31). Famoxadone binding to the Q_o site decreases the rate constant k₁ for electron transfer from [2Fe2S] to cyt c₁ from 60,000 s⁻¹ to 5,400 s⁻¹ (33). Therefore, famoxadone does not completely lock the ISP in the b state, but instead significantly decreases the rate of release of the ISP from the b state to the c₁ state.

In order to further explore the role of the ef loop in regulating the dynamics of the ISP, a series of mutants at residues in the ef loop were constructed. R. sphaeroides cyt b residues 276 − 280 correspond to bovine residues 252 − 256 which are in the middle of the ef loop on the surface of cyt b (Figures 2, 3). These residues undergo large displacements upon famoxadone binding, suggesting that they might relay conformational changes in the Q_o pocket to the ISP docking crater. Surprisingly, mutation of residues H276, P277, D278, and N279 led to relatively small changes in k₁ and k₂, suggesting that if these residues do help relay conformational changes, that function is not sensitive to substitution of individual amino acid side changes. The D278A mutation led to the largest decrease in k₁ for this group, from 60,000 s⁻¹ to 26,000 s⁻¹, but no significant change in rate constant k₂. Famoxadone binding to this mutant decreased k₁ to 2,300 s⁻¹, which is similar to the percent reduction observed in wild-type cyt bc₁, suggesting that this mutant does not modulate the effect of famoxadone. The mutation Y280A led to a substantial decrease in k₁ to 7,900 s⁻¹, while famoxadone binding only caused a further decrease to 3,200 s⁻¹. This mutation might cause a conformational change similar to that of famoxadone, limiting the additional effect of famoxadone binding.

R. sphaeroides cyt b residues 286 − 292 correspond to bovine residues 262 − 268 in the ef loop which forms part of the ISP docking crater. Mutation of residues S287, A290, and H291 had relatively little effect on k₁, which is surprising in the case of H291 since the histidine side chain moves 9.3 Å upon famoxadone binding (31). However, mutations at I292 located in the ISP docking crater near the [2Fe2S] center had a large effect on electron transfer activity (Table 2). The I292A mutation decreased k₁ to 4,400 s⁻¹ while famoxadone binding only further decreased it to 3,000 s⁻¹. The mutations I292L and I292M also led to substantial decreases in k₁, while famoxadone binding caused relatively small additional decreases in k₁. It appears that these mutations significantly decrease the rate of release of the ISP from cyt b, but famoxadone binding leads to a relatively small additional decrease in the rate of release. The I292A mutation led to a decrease in the rate constant k₂ for QH₂ oxidation to 350 s⁻¹, indicating that this mutation had an effect on the conformation of the QH₂ reaction site. Substitution of I292 with a basic arginine residue was tolerated with a decrease in k₁ to 10,000 s⁻¹ and only a modest change in k₂. In contrast, substitution of I292 with an acidic glutamate nearly completely eliminated both steady-state activity and electron transfer activity measured by the ruthenium photooxidation method. It is apparent that I292 plays an important role in controlling the conformation of the ISP.

The effects of mutation of L286 at the tip of the ef loop are particularly interesting. The mutation L286A decreases the rate constant k₁ to 33,000 s⁻¹ and k₂ to 740 s⁻¹. However, famoxadone binding does not lead to any further decrease in k₁, (Table 2) suggesting that this mutation might block the famoxadone-induced conformational change in the wild-type protein which decreases the rate constant k₁ by such a large factor. The L286E mutation also led to a significant decrease in the rate constant k₁, to 17,000 s⁻¹, while famoxadone binding did not cause much further decrease. The importance of L286 is also illustrated in experiments by Darrouzet and Daldal (14, 19). They found that insertion of one alanine at residue 46 in the neck region of the ISP decreased the rate of the domain movement from the b state to the c₁ state to a half time of 10 ms, which was slow enough to measure in R. capsulatus chromatophores. A second revertant mutation substituting Leu for Phe at residue 286 restored rapid domain movement, which was too fast to measure in chromatophores (19). This finding is particularly remarkable given that cyt b residue 286 is more than 25 Å from the neck region of the ISP. Revertant mutations were also found in the ISP neck region for a primary mutation at T288 in the ef loop (24). A possible mechanism for this linkage between the ef loop and ISP neck region is suggested by the effect of famoxadone binding, which causes correlated conformational changes in both the ef loop region and the cyt b residues 179 −187 which contact the ISP neck region. It appears likely that the conformational linkage between the ISP neck region and the cyt b ef loop is mediated by the cyt b protein using structural linkages similar to those caused by famoxadone binding.

It has been proposed that the conformation of the ISP plays an important role in several different stages of the mechanism of cyt bc₁ (21). First, it is proposed that binding QH₂ to the active Q_o pocket displaces the cd1 helix and the ef helix outwards to capture the oxidized ISP in the b state conformation. The formation of a hydrogen bond between QH₂ and His-161 would stabilize the active site conformation, and allow proton-coupled electron transfer from QH₂ to oxidized [2Fe2S]. After the second electron is transferred from semiquinone to cyt b_L and cyt b_H, oxidized Q would leave the Q_o binding pocket and the cd1 helix and the ef helix would relax to their original positions, releasing the ISP and allowing it to rotate to the c₁ position and transfer an electron to cyt c₁ (21). An important question about this mechanism is whether ISP is released from the b state after the first electron is transferred to [2Fe2S] and the hydrogen bond to His-161 is broken, or remains locked in the b state until cyt b_H is reduced. Famoxadone binding is informative in this regard, because it leads to many of the same conformational changes in the ISP binding crater as stigmatellin, but does not form a hydrogen bond to His-161. The rate of release of the ISP from the b-state is decreased from 60,000 s⁻¹ to 5,000 s⁻¹ by famoxadone binding, but the ISP is not completely locked in the b-state (33). Once the ISP is released from the b state and rotates to the cyt c₁ state, there may be conformational linkages that prevent its return to the b state under conditions that would lead to short-circuit reactions. When cyt b_H is reduced in the presence of antimycin, an electron is not transferred from cyt b to the ISP on a long time scale, suggesting that oxidized ISP does not return to the b state in the absence of QH₂. The ef loop presents a barrier to the rotation of the ISP between the b and c₁ states (19, 32), and this barrier might regulate the rotation in both directions to facilitate productive electron transfer and prevent short-circuit reactions.

Abbreviations used

cyt

cytochrome

ISP

Rieske iron-sulfur protein

[2Fe2S]

Rieske iron-sulfur center

Ru₂D

[Ru(bpy)₂]₂(qpy)(PF₆)₄

qpy

2,2’:4’,4”:2”,2”’–quaterpyridine

Q₀C₁₀BrH₂

2,3-dimethoxy-5-methyl-6-(10-bromodecyl-) 1,4-benzoquinol

SCR

succinate cytochrome c reductase