Flexibility of the Neck Region of the Rieske Iron-Sulfur Protein Is Functionally Important in the Cytochrome bc 1Complex*

The crystal structure of the mitochondrial cytochrome bc 1 complex suggests that movement of the extramembrane (head) domain of the Rieske iron-sulfur protein (ISP) is involved in electron transfer. Such movement requires flexibility in the neck region of ISP. To test this hypothesis,Rhodobacter sphaeroides mutants expressing His-tagged cytochrome bc 1 complexes with altered ISP necks (residues 39–48) were generated and characterized. Mutants with increased rigidity of the neck, generated by a double-proline substitution at Ala-46 and Ala-48 (ALA-PLP) or by a triple-proline substitution of ADV at residues 42–44 (ADV-PPP), have retarded (50%) or no photosynthetic growth, respectively. However, the mutant with a shortened neck, generated by deleting ADV (ΔADV), has a photosynthetic growth rate comparable to that of complement cells, indicating that the length of the ISP neck is less critical than its flexibility in support of photosynthetic growth. The ΔADV and ALA-PLP mutant membranes have 10 and 30% of the cytochromebc 1 complex activity found in the complement membrane, respectively, whereas the ADV-PPP mutant membrane contains no cytochrome bc 1 complex activity. The loss of cytochrome bc 1 complex activity in the ΔADV membrane is attributed to improper docking of the head domain of ISP on cytochrome b, as indicated by a drastic change in the EPR characteristics of the Rieske iron-sulfur cluster. The loss of cytochrome bc 1 complex activity in the ALA-PLP and ADV-PPP mutant membranes results from the decreased mobility of the ISP head domain due to the increased rigidity of the ISP neck. The ALA-PLP mutant complex has a larger activation energy than the wild-type complex, suggesting that movement of the head domain decreases the activation energy barrier of the cytochromebc 1 complex. Using the conditions developed for the isolation of the His-tagged complement cytochromebc 1 complex, a two-subunit complex (cytochromesb and c 1) was obtained from the ΔADV and ADV-PPP mutants, indicating that mutations at the neck region of ISP weaken the interactions among cytochrome b, ISP, and subunit IV.

The cytochrome bc 1 complex (ubiquinol-cytochrome c reductase) is an essential segment of the energy-conserving electron transfer chains of mitochondria and many respiratory and photosynthetic bacteria (1). The complex catalyzes electron trans-fer from ubiquinol to cytochrome c and concomitantly translocates protons across the membrane to generate a membrane potential and pH gradient for ATP synthesis. The polypeptide composition of the cytochrome bc 1 complex from different sources varies from 3 to 11 subunits. The redox subunits (cytochrome b, cytochrome c 1 , and the Rieske iron-sulfur protein (ISP) 1 ) are conserved in all the cytochrome bc 1 complexes. The proton-motive Q-cycle model (2) has been favored for electron transfer and proton translocation in the complex. The key feature of this model is the presence of two separate ubiquinone-or ubiquinol-binding sites: a ubiquinol oxidation site (Q o ), near the P side of the inner mitochondrial membrane, and a ubiquinone reduction site (Q i ), near the N side of the membrane. The recently solved three-dimensional structure of the mitochondrial cytochrome bc 1 complex not only answered a number of questions concerning the arrangement of the redox centers, transmembrane helices, and Q-like inhibitor-binding sites, but also suggested an unexpected dynamic feature of this complex (3)(4)(5)(6)(7)(8)(9).
Mobility of ISP in the bc 1 crystal was first suggested by observation of a particularly low electron density area, in the intermembrane space portion of the complex, where the extramembrane domains of ISP and cytochrome c 1 reside (4). The mobility of the extramembrane (head) domain of ISP was further substantiated by the anomalous light scattering signals of the [2Fe-2S] cluster observed in native and co-crystals with Q o site inhibitors, such as stigmatellin, 5-n-undecyl-6-hydroxy-4,7-dioxobenzothiazole, and methoxyacetylate stilbene (5,6,8). In a native beef bc 1 crystal, a much weaker anomalous light scattering signal was observed for the [2Fe-2S] cluster compared with that for the heme iron, b H or b L , despite the presence of two irons in the cluster. Furthermore, the anomalous scattering signal of the [2Fe-2S] cluster was strongly enhanced in a co-crystal with stigmatellin, indicating that ISP is in a fixed state in the presence of this inhibitor. On the other hand, the signal diminished in a co-crystal with methoxyacetylate stilbene, suggesting that ISP is in a released state (6). Stigmatellin and methoxyacetylate stilbene are bound to different locations in the Q o pocket, the former being closer to histidine ligands of the [2Fe-2S] cluster and the latter being closer to heme b L .
In the beef bc 1 crystal, the distances between heme b L and the [2Fe-2S] cluster and between the [2Fe-2S] cluster and heme c 1 , are 27 and 31 Å, respectively. Although the distance of 27 Å between heme b L and the [2Fe-2S] cluster accommodates well the observed fast electron transfer between these two redox centers, the 31-Å distance between the [2Fe-2S] cluster and heme c 1 is difficult to understand in view of the rapid electron transfer rate observed for these two redox centers (10,11). Movement of the extramembrane domain of ISP, as described below, offers an explanation for this paradox. The [2Fe-2S] cluster is reduced by the first electron of ubiquinol at a position 27 Å from heme b L and 31 Å from cytochrome c 1 . The reduced [2Fe-2S] cluster cannot donate an electron to cytochrome c 1 before the second electron of ubiquinol is transferred to heme b L . It was speculated that either the change of the ubisemiquinone binding position before the reduction of heme b L or the electron transfer from heme b L to b H causes a conformational change in cytochrome b that forces or allows reduced [2Fe-2S] to move close enough to heme c 1 for fast electron transfer (6,8). This model would also explain why ubisemiquinone, a more powerful reductant than ubiquinol, reduces heme b L , but not the [2Fe-2S] cluster, during ubiquinol oxidation.
The ISP structure of beef heart mitochondria has three domains: the membrane-spanning N-terminal domain consisting of residues 1-62 (tail), the soluble C-terminal extramembrane domain consisting of residues 73-196 (head), and the flexible linking domain comprising residues 63-72 (neck). ISP is associated with the complex via the membrane-spanning N-terminal domain (4,7,9). The [2Fe-2S] cluster is located in the rigid head domain as shown in the high resolution structure of the water-soluble fragment of ISP (12,13). The electron density in the neck region is low; the structure was deduced by connecting the C-terminal end of the transmembrane helix to the N terminus of the head domain. No ordered secondary structure was reported in the neck region (4,7).
Although the position of the iron-sulfur cluster changed from a fixed state to a released state upon methoxyacetylate stilbene binding (6), the three-dimensional structures of the head and tail domains of ISP remain the same in these two states, suggesting that a bending of the neck is required for movement of the head domain (7). For the neck region to bend, some flexibility is imperative. The beef ISP neck has 10 amino acid residues with a sequence of SASADVLAMS. This region is highly conserved in all iron-sulfur proteins (Fig. 1). The well conserved alanine residues may provide the needed flexibility. In the crystal structure, the neck region is in close contact with the cd1 helix of cytochrome b, but is not involved in compact docking between subunits, thus leaving enough space for bending (Fig. 2). The neck region is exposed to solvent as indicated by its susceptibility to several proteases, including thermolysin and trypsin (13).
If movement of the head domain of ISP is required for bc 1 catalysis and the neck region of ISP confers the necessary mobility, changing the flexibility of the neck region of ISP should drastically affect the catalytic activity of the bc 1 complex. One way to prove this suggestion is to prepare recombinant mutant ISP, with increased rigidity in the neck, by sitedirected mutagenesis followed by in vitro reconstitution of mutant ISP to an ISP-depleted bc 1 complex. Biochemical and biophysical characterizations of the reconstituted bc 1 com-plexes should reveal the essentialness of neck flexibility. Although the beef cDNA for ISP has been cloned and sequenced (14), the unavailability of reconstitutively active recombinant ISP and the difficulty in preparing fully reconstitutively active ISP-depleted bc 1 complex (15) have prevented us from taking this approach. The four-subunit cytochrome bc 1 complex from the photosynthetic bacterium Rhodobacter sphaeroides is functionally analogous to the mitochondrial bc 1 complex. Since the largest three subunits are homologous to their mitochondrial counterparts and are readily manipulated genetically, this organism is ideal for studying the neck region of ISP by sitedirected mutagenesis.
Herein we report the generation and characterization of three R. sphaeroides mutants expressing His 6 -tagged cytochrome bc 1 complexes with altered ISP neck regions. The length of the neck was shortened by deletion, and its rigidity was increased by proline substitution at various positions. The photosynthetic growth behavior, EPR characteristics of the Rieske [2Fe-2S] cluster, activation energy, and the cytochrome bc 1 complex activity in membranes and the purified state from the complement and mutant strains were examined and compared. The effect of the neck region of ISP on the interaction between cytochrome b and ISP or subunit IV was determined.
Growth of Bacteria-E. coli cells were grown at 37°C on LB medium. Extra-rich medium (TYP) was used in procedures for the rescue of single-stranded DNA or the purification of low copy number plasmids (20). R. sphaeroides cells were grown at 30°C on enriched Sistrom's medium either semi-aerobically or photoheterotrophically (photosynthetic growth) as reported (21). Antibiotics were added at the following concentrations: ampicillin, 100 -125 mg/liter; tetracycline, 10 -15 mg/ liter for E. coli and 1 mg/liter for R. sphaeroides; kanamycin sulfate, 30 -50 mg/liter for E. coli and 20 mg/liter for R. sphaeroides; and trimethoprim, 85-100 mg/liter for E. coli and 25 mg/liter for R. sphaeroides.
Construction of an R. sphaeroides Strain Expressing the His 6 -tagged Cytochrome bc 1 Complex-A 1.2-kb XbaI-HindIII fragment containing the fbcC and fbcQ genes from pRKDfbcFBCQ, containing fbcF, fbcB, fbcC, and fbcQ, was inserted into a modified pSELECT-1 vector in which the unique Acc65I site was eliminated. The resulting pSELfbcCQ was used as template for site-directed mutagenesis to introduce an Acc65I recognition site right before the stop codon of the fbcC gene. The Altered Sites in Vitro Mutagenesis System from Promega (22) was used for all the site-directed mutagenesis constructions. Two complementary oligonucleotides with His 6 tag coding sequence and the Acc65I overhang attached at the 5Ј-ends (5Ј-GTACGGGC CAT CAC CAC CAC CAT CAC TAA-3Ј and 3Ј-CCCG GTA GTG GTG GTG GTA GTG ATTCATG-5Ј) were synthesized, annealed together by heating up to 70°C and cooling slowly to room temperature, and ligated into the Acc65I site of pSELf-bcCQ to generate pSELfbcC H Q. The 6-histidine insertion was confirmed by DNA sequencing. A 1.2-kb XbaI-HindIII fragment containing fbcC H Q from pSELfbcC H Q was subcloned into an expression vector (pRKD418) containing the fbcFBCQ genes to generate pRKDfbcFB-C H Q, which was then mobilized into R. sphaeroides BC17 by parental conjugation.
Generation of R. sphaeroides Strains Expressing the bc 1 Complexes with Altered ISP-Mutations were constructed by site-directed mutagenesis using the Altered Sites system. Oligonucleotides were synthesized at the Oklahoma State University Recombinant DNA/Protein Core Facility. The oligonucleotides used are GCTGATCAACCAAAT-GAATCCGTCGCAGGCCCTCGCCTCCATCTTCGTCG for the ⌬ADV mutant, CAAATGAATCCGTCGCCGCCGCCGCAGGCCCTCGCCTCC for the ADV-PPP mutant, and TCGGCCGACGTGCAGCCGCTCCCG-TCCATCTTCGTC for the ALA-PLP mutant. The presence of engineered mutations was confirmed twice by DNA sequencing before and after photosynthetic or semi-aerobic growth of the cells.
Enzyme Preparations and Activity Assay-Chromatophore membranes were prepared from BC17 cells harboring complement (sequence is the same as the wild type) or mutant pRKDfbcF m BC H Q as described previously (23) and stored at very high concentration in the presence of 20% glycerol at Ϫ80°C. To purify the His 6 -tagged cytochrome bc 1 complex, the chromatophore suspensions were thawed and adjusted to a cytochrome b concentration of 25 M with 50 mM Tris-Cl (pH 8.0 at 4°C) containing 20% glycerol, 1 mM MgSO 4 , and 1 mM phenylmethylsulfonyl fluoride. DM solution (10%, w/v) was added to the chromatophore suspension to 0.56 mg/nmol of cytochrome b, and the mixture was stirred at 4°C for 30 min and then centrifuged at 27,000 ϫ g for 30 min. The hard precipitates at the bottom of the centrifuge tubes were discarded, and the loose pellets and supernatants were collected. NaCl solution (4 M) was added to a final concentration of 0.1 M, and the suspension was stirred for 1 h at 4°C. This mixture was centrifuged at 200,000 ϫ g for 120 min. The supernatants were collected and stirred with Ni 2ϩ -NTA resin (100 nmol of cytochrome b/ml of resin) for 20 min at 4°C. The mixture was packed into a column, and the effluent was reapplied to the column to maximize the binding of protein to the resin. The column, absorbed with bc 1 complexes, was then subjected to a sequence of washings with TN buffer (50 mM Tris-Cl (pH 8.0 at 4°C) and 200 mM NaCl) containing 0.01% DM, TN buffer containing 5 mM histidine and 0.01% DM, and TN buffer containing 5 mM histidine and 0.5% octyl glucoside. The pure cytochrome bc 1 complex was eluted with TN buffer containing 200 mM histidine and 0.5% octyl glucoside and concentrated using a Centriprep-30 concentrator to a final concentration of 300 M cytochrome b or higher. The purified complex was stored at Ϫ80°C in the presence of 20% glycerol.
Ubiquinol-cytochrome c reductase activity was assayed at room temperature in a Shimadzu UV2101PC spectrophotometer (24). The purified cytochrome bc 1 complex was diluted with TN buffer containing 0.01% DM to a final concentration of cytochrome b of 5 M. 3 l of diluted bc 1 complex was added to an assay mixture (1 ml) containing 25 mM sodium/potassium phosphate buffer (pH 7.4), 0.3 mM EDTA, 125 M cytochrome c, and 25 M Q 0 C 10 BrH 2 . Activity was determined by measuring the reduction of cytochrome c (by following the increase in absorbance at 550 nm) using a millimolar extinction coefficient of 18.5 cm Ϫ1 . Nonenzymatic oxidation of Q 0 C 10 BrH 2 , determined under the same conditions in the absence of enzyme, was subtracted. 30 M potassium cyanide was added to the assay mixture to inhibit the oxidase activity when the bc 1 activity in chromatophores was determined.
Determination of the Activation Energy of the Cytochrome bc 1 Complex-The activation energy of the bc 1 complex was determined in both chromatophore membrane and purified preparations. This is essentially assaying steady-state enzyme activity at various temperatures. The temperatures of the assay mixture were controlled (Ϯ0.1°C) by a Shimadzu TCC controller installed on a Shimadzu UV2101PC spectrophotometer. Activity was measured from 9 to 30°C at 3°C intervals. The activation energy was calculated from an Arrhenius plot.
Other Biochemical and Biophysical Techniques-Protein concentration was measured by the method of Lowry et al. (25). Cytochrome b (26) and cytochrome c 1 (27) were determined according to published methods. SDS-polyacrylamide gel electrophoresis was performed according to Laemmli (28) using a Bio-Rad Mini-Protean dual-slab vertical cell. Western blotting was performed using rabbit polyclonal antibodies against cyto- chrome c 1 , ISP, and subunit IV of the R. sphaeroides bc 1 complex. The polypeptides separated on the SDS-polyacrylamide gel were transferred to polyvinylidene difluoride membrane for immunoblotting. Goat antirabbit IgG conjugated to alkaline phosphatase or protein A conjugated to horseradish peroxidase was used as the second antibody.
EPR spectra were recorded with a Bruker ER 200D apparatus equipped with a liquid N 2 Dewar at 77 K. Instrument settings are detailed in the figure legends.

RESULTS AND DISCUSSION
Characterization of the His 6 -tagged Cytochrome bc 1 Complex-The cytochrome bc 1 complex, which was overexpressed in R. sphaeroides BC17 cells by a low copy number expression vector (pRKD418) containing the fbcFBCQ genes, was routinely purified from chromatophore preparations by DM solubilization followed by DEAE-Bio-Gel A and DEAE-Sepharose CL-6B column chromatography. Although this purification scheme produces an enzyme complex of high purity and activity, it is time-consuming and gives low yields. This procedure requires extensive washing of the DEAE-Bio-Gel A column (usually 20 column volumes) to remove contaminating proteins, chlorophyll, and other pigments from the absorbed bc 1 complex. This step takes at least 8 h.
To speed up the preparation of the bc 1 complex from complement and mutant cells, a His 6 tag was genetically engineered into the C terminus of the cytochrome c 1 subunit to allow the use of Ni 2ϩ -NTA affinity agarose in a one-step purification. This construction was achieved by ligating annealed His 6 -tag coding oligonucleotides, with the Acc65I overhang attached at the 5Ј-ends, into the Acc65I site created right before the stop codon of the fbcC gene in pSELfbcCQ plasmid to generate pSELfbcC H Q. The fragment containing the fbcCQ km genes in pRKDfbcFBCQ km was replaced with the fragment containing the fbcC H Q genes from pSELfbcC H Q to generate pRKDfbcFB-C H Q. R. sphaeroides BC17 cells harboring pRKDfbcFBC H Q have photosynthetic and respiratory growth behaviors similar to those of untagged cells. This construction adds 9 amino acid residues (GTGHHHHHH) to the C terminus of cytochrome c 1 in the expressed bc 1 complex.
The spectral properties and bc 1 complex activity in the His 6tagged chromatophores are similar to those in untagged chromatophores. The effectiveness of DM in the solubilization of tagged and untagged cytochrome bc 1 complexes from their respective chromatophores is comparable; ϳ95% of the cytochrome b present in chromatophores was solubilized when 0.56 mg of DM/nmol of cytochrome b was used. About 50% of the cytochrome bc 1 complex in chromatophores was recovered from the Ni 2ϩ -NTA column. The His 6 -tagged bc 1 complex has purity, activity, and cytochrome content similar to those of the untagged enzyme complex. The yield of the purified His 6 -tagged bc 1 complex was twice that of untagged bc 1 preparations. The total time for purification of the His-tagged complex was ϳ6 h as compared with 2 days for the conventional purification. The molecular mass of the His 6 -tagged bc 1 complex, determined by sedimentation velocity or sedimentation equilibrium, is 220 kDa, indicating that the isolated complex is in a dimeric state. The EPR characteristics of the [2Fe-2S] cluster in the purified His 6 -tagged bc 1 complex are the same as those in the untagged complex (data not shown).
Characterizations of the Cytochrome bc 1 Complexes Containing an Altered ISP Neck-To establish that flexibility of the neck region of ISP is essential for the head domain movement required for bc 1 catalysis, mutants expressing bc 1 complexes with increased ISP neck rigidity were generated and characterized. The R. sphaeroides ISP neck is composed of residues 39 -48 with the sequence NPSADVQALA; Ala-42, Asp-43, Val-44, Ala-46, and Ala-48 are the conserved amino acid residues. We focused our mutational studies on these 5 residues. The flexibility of the ISP neck is expected to decrease when proline residues are introduced because proline has a conformational constraint due to the cyclic nature of its pyrrolidine side chain. Since the ADV residues are located in what appears to be the most flexible part of the neck, deletion of these residues is expected to affect the movement or the positioning of the head of ISP.
Three R. sphaeroides mutant strains were generated: ⌬ADV, in which the ADV residues (residues 42-44) are deleted; ALA-PLP, in which Ala-46 and Ala-48 are substituted with prolines; and ADV-PPP, in which Ala-42, Asp-43, and Val-44 are substituted with prolines. All mutations were constructed by site-directed mutagenesis using a 3.5-kb EcoRI-HindIII R. sphaeroides DNA carrying the fbcFBC genes in pSELECT-1 plasmid (pSELNB3503) as template. A 2.5-kb EcoRI-PinAI fragment containing the kanamycin-resistant gene in pRKDfbcFBC 6H Q km was replaced with a 1.7-kb EcoRI-PinAI fragment containing the mutated ISP gene from pSELNB3503 to generate pRKDfbcF m BC H Q plasmid, which was then mobilized into R. sphaeroides BC17 cells. Mutations in pRKDfb-cF m BC H Q harbored in E. coli S17-1 (before conjugation) and photosynthetically or semi-aerobically dark-grown R. sphaeroides cells (after conjugation) were confirmed by DNA sequencing after the targeted DNA fragment was amplified by polymerase chain reaction.
The ⌬ADV mutant cells grew photosynthetically at a rate similar to that of complement cells. The ALA-PLP mutant was also capable of photosynthetic growth, but at a maximal doubling rate of ϳ50% that of the complement strain. The ADV-PPP mutant was unable to grow photosynthetically (Table I, first column), but could grow under semi-aerobic conditions. These results indicate that the ISP neck flexibility is more critical than its length in supporting photosynthetic growth.
To investigate whether mutations at the neck region of ISP affect cytochrome bc 1 complex activity, ubiquinol-cytochrome c reductase activity in chromatophores from the ⌬ADV and ALA-PLP strains and in the intracytoplasmic membrane (ICM) of the ADV-PPP strain was assayed and compared with that of the complement strain. To study subunit association of cytochrome bc 1 complexes in the membrane, Western blot analyses of chromatophores and ICM of mutants were performed and compared with those of the complement strain. Chromatophores from the ⌬ADV and ALA-PLP mutant cells had 10 and 30% of the ubiquinol-cytochrome c reductase activity found in the complement chromatophores, respectively. As expected, ICMs of the ADV-PPP mutant had no ubiquinol-cytochrome c reductase activity because the cytochrome bc 1 complex is an obligatory enzyme complex for photosynthetic growth, and this mutant was unable to grow photosynthetically. These results indicate that shortening the length or decreasing the flexibility of the ISP neck drastically decreases the cytochrome bc 1 complex activity in membranes. It should be noted that the cytochrome bc 1 complex activities in the photosynthetic chromatophore membrane and in semi-aerobic ICM from complement cells are the same. When membranes from these three mutants and the complement cells were subjected to Western blot analysis with antibodies against R. sphaeroides cytochrome c 1 , ISP, and subunit IV, stoichiometric amounts of these three subunits were detected in all three mutant membranes (Fig. 3). Absorption spectral analysis also revealed that the ratio of cytochrome b to c 1 /c 2 in all these mutant membranes was similar to that in the complement membrane. These results indicate that these mutations did not affect the assembly of ISP protein into the membrane. However, we observed an apparently elevated level of the bc 1 complex in the ⌬ADV mutant membrane, as indicated by the elevated level of cytochrome b (19 nmol/mg of protein), relative to that found in membranes from complement cells (16 nmol/mg of protein). It should be noted that membranes from complement cells already possess three times the amount of cytochrome b found in wild-type R. sphaeroides, presumably due to a gene dosage effect. Therefore, the increased level of expression could be a regulatory response compensating the lowered electron transfer activity in the ⌬ADV mutant. This explains why 10% bc 1 (specific) activity observed in the mutant complex is sufficient to support the photosynthetic growth.
Effect of Mutation on the Rieske Iron-Sulfur Cluster and on Assembly of ISP into the Cytochrome bc 1 Complex-Since the bc 1 complex activity decreased by 90, 70, and 100% in the ⌬ADV, ALA-PLP, and ADV-PPP mutant membranes, respectively, with no decrease in the amount of ISP, it is important to determine whether the activity loss resulted from improper assembly of ISP into the complex or from the fact that the ISP head domain is less mobile. We addressed this question by comparing EPR characteristics of the [2Fe-2S] cluster in complement and mutant membranes because EPR signals from the iron-sulfur protein, especially the g x signature, are very sensitive to changes in its microenvironments. Also, we compared the subunit stoichiometry in the purified His 6 -tagged mutant complexes with that found in the complement complex.
When the [2Fe-2S] cluster was reduced by a small excess of ascorbate, the complement chromatophore or ICM had a spectrum that was essentially the same as that previously reported for the chromatophores from wild-type R. sphaeroides, with resonance at g x ϭ 1.80 and g y ϭ 1.9 (Fig. 4, spectra A and D). The g z ϭ 2.02 signal of the [2Fe-2S] cluster could not be resolved in membrane preparations because it was shielded by many other signals.
The [2Fe-2S] cluster in ⌬ADV chromatophore membranes showed no detectable g x signal (totally broadened) and a very small g y signal (Fig. 4, spectrum B), indicating that the microenvironments of the iron-sulfur cluster have been drastically altered in this deletion mutant. In the bc 1 crystal structure, the iron-sulfur cluster sits at the tip of the head domain of ISP, and this tip of ISP fits into the concave hydrophobic surface of the Q o pocket located in the cytochrome b subunit (4). Changing the microenvironments of the [2Fe-2S] cluster by mutation of residues in the docking interface of ISP and cytochrome b, such as Leu-132 (29) and Gly-133 (30) in ISP and Ile-292 (31) in cytochrome b, resulted in a loss of the g x signal of the [2Fe-2S] cluster and a decrease in electron transfer activity. Since the neck region is spatially separated from the docking interface of ISP and cytochrome b, the change of microenvironments of the [2Fe-2S] cluster in the ⌬ADV mutant complex, indicated by the g x /g y signal change, is probably due to the improper docking of the head domain of ISP on cytochrome b as a result of the shortened neck. Loss of cytochrome bc 1 complex activity (90%) in the ⌬ADV membrane is therefore attributed to improper assembly of ISP into the complex. It should be noted that the FIG. 3. Western blot analysis of chromatophore membranes from mutant and wild-type complement strains. Membrane samples containing 75 pmol of cytochrome b were loaded into each well and subjected to SDS-polyacrylamide gel electrophoresis. The gel was transferred electrophoretically to a polyvinylidene difluoride membrane. Polyclonal antibodies raised against subunits of R. sphaeroides bc 1 complex (cytochrome c 1 (Cyt c 1 ), ISP, and subunit IV (Sub IV)) were used to detect these three subunits. The antibody's titer for cytochrome c 1 is much higher than those for subunit IV and ISP. To reduce the nonspecific background reaction, the membrane was cut into two pieces so that the upper part, containing proteins with a molecular mass Ͼ23 kDa, including cytochrome c 1 , was developed with a horseradish peroxidase system. The bottom section, containing the low molecular mass proteins, was developed with an alkaline phosphatase system. Chromatophore pastes were partially reduced by addition of 5 mM ascorbate. The samples were incubated on ice for ϳ20 min and frozen in liquid nitrogen. EPR spectra were recorded at 77 K with the following instrument settings: microwave frequency, 9.336 Hz; microwave power, 20 milliwatts; modulation amplitude, 20 G; modulation frequency, 100 kHz; time constant, 0.1 s; and scan rate, 20 G/s. drastic decrease in the amplitude of the g y signal in the ⌬ADV mutant chromatophores is not due to a destabilizing effect on the oxidized [2Fe-2S] cluster as reported for the T134R, T134H, or T134G mutation in ISP of Rhodobacter capsulatus (29) since the signal was not increased in EPR measurements of ⌬ADV chromatophores prepared by including 20 mM ascorbate to keep the [2Fe-2S] cluster in the reduced state. Such a decrease in the EPR signal is not caused by a labile [2Fe-2S] cluster as the result of ADV deletion because when the membrane was prepared by a gentler method, such as treating freshly grown cells with lysozyme and an appropriate amount of detergent to disrupt the membrane, no increase in EPR signals and the membrane bc 1 activity was observed. However, when the mutant membrane was incubated with 100 M stigmatellin, a small but distinctive g x ϭ 1.78 signal showed up, similar to that of the stigmatellin-treated complement membrane (data not shown). No bc 1 activity was detected in this inhibitor-treated ⌬ADV membrane, which further confirmed the typical response of g x signal to Q o site inhibitor. Therefore, the decrease in the EPR signal in the mutant membrane is not due to the destruction of the [2Fe-2S] cluster during the preparation of the membrane, but rather an intrinsic property of the mutant. Similar results were observed when cell pastes were used in EPR analysis (data not shown).
If ISP is indeed improperly assembled into the bc 1 complex in the ⌬ADV mutant chromatophore, the mutant complex is expected to be less stable than the wild type. In other words, the binding affinity of ISP for other subunits in the ⌬ADV mutant complex is probably weaker than that in the wild-type complex. To confirm this speculation, chromatophore membranes from the ⌬ADV and complement cells were treated with dodecyl maltoside at 0.55 mg/nmol of cytochrome b. Although the amounts of cytochromes b and c 1 , ISP, and subunit IV solubilized from these two chromatophore membranes were the same, no bc 1 complex activity was detected in the detergentsolubilized membrane fraction from ⌬ADV. When dodecyl maltoside-solubilized chromatophore membranes from ⌬ADV and complement cells were individually applied to a Ni 2ϩ -NTA column and analyzed for subunit composition in the histidineeluted fractions, the His 6 -tagged ⌬ADV complex was found to contain only cytochromes b and c 1 (Fig. 5, lane 5). ISP and subunit IV were detected in the unbound fraction by Western blot analysis (Fig. 3, lane 7), whereas the complement complex contained cytochromes b and c 1 , ISP, and subunit IV (Fig. 5,   lanes 2 and 3). Although the lack of ISP in the ⌬ADV complex was expected, the lack of subunit IV in the complex was rather surprising. Perhaps residues ADV of the ISP neck are involved in the packing of ISP and subunit IV with cytochrome b and c in the complex.
EPR characteristics of the Rieske [2Fe-2S] cluster in the ADV-PPP membrane (ICM) (Fig. 4, spectrum E) are the same as those of the [2Fe-2S] cluster in complement chromatophores, and binding of the Q o site inhibitor stigmatellin induces an upfield shift in the g x signal that is similar to that observed with complement chromatophores (data not shown). These results indicate that the environments of the iron-sulfur cluster are not changed by this mutation (ADV-PPP). Although the increased rigidity of the ISP neck did not affect the docking of the ISP head domain on the cytochrome b protein, the complex was unstable as evident from the dissociation of ISP and subunit IV from cytochromes b and c 1 when this mutant complex was purified from the DM-solubilized membrane (Fig. 5, lane  6). The complete lack of cytochrome bc 1 complex activity in the ADV-PPP membrane may be attributed to weak binding of ISP and subunit IV with cytochromes b and c 1 as well as to a lesser head domain mobility of ISP during bc 1 catalysis.
EPR characteristics of the Rieske [2Fe-2S] cluster in the ALA-PLP chromatophore membrane (Fig. 4, spectrum C) are the same as those of the [2Fe-2S] cluster in complement chromatophores, indicating similar iron-sulfur cluster environments, i.e. the head domain of ISP is properly docked on cytochrome b in this mutant chromatophore. When the complex in the ALA-PLP chromatophore membrane was solubilized with dodecyl maltoside and applied to a Ni 2ϩ -NTA affinity gel, four subunits, corresponding to cytochromes b and c 1 , ISP, and subunit IV with unit stoichiometry were recovered in the histidine eluate (Fig. 5, lane 4), indicating that this mutation does not affect the binding affinity of ISP for other subunits of the complex. However, the cytochrome bc 1 complex activity in the chromatophore membrane and in the purified complex from the ALA-PLP mutant was 30% of that found in corresponding preparations from complement cells. These results, together with the fact that the ALA-PLP cytochrome bc 1 complex differs from the complement complex only in the increased rigidity of the ISP neck, suggest that the 70% loss of bc 1 activity in the ALA-PLP complex results from decreased mobility of the ISP head domain. Apparently, head domain movement of ISP is required for bc 1 complex activity.
Effect of Mutation on the Activation Energy of the Cytochrome bc 1 Complex-If the head domain movement of ISP is required for bc 1 activity, it is important to know whether this step contributes to the activation energy barrier of the bc 1 -catalyzed reaction. This question can be addressed by comparing the activation energies of the cytochrome bc 1 complexes, in chromatophores or in the purified state, of ALA-PLP mutant and complement cells. An increase in activation energy for the ALA-PLP complex would indicate that the head domain movement of ISP contributes to the activation energy barrier. This deduction is based on the fact that the only difference between the ALA-PLP mutant complex and the complement complex is the decreased ISP head mobility of the former. Fig. 6 shows Arrhenius plots of bc 1 complex activity in ALA-PLP mutant and complement chromatophores. Since the concentrations of electron donor (ubiquinol) and electron acceptor (cytochrome c) used in the bc 1 activity assay mixture were at the saturation level, diffusion limitation of cytochrome c was avoided, and no production inhibition was observed. An activation energy of 24.7 kJ/mol was obtained for the complement complex and 69 kJ/mol for the ALA-PLP mutant, indicating that a decrease in head domain movement in ISP increases the activation energy of the bc 1 complex. Similar results were obtained with purified bc 1 complexes. The activation energy obtained for the complement cytochrome bc 1 complex was the same as that obtained for the beef heart mitochondrial bc 1 complex under similar assay conditions (32,33).
Although head domain movement of ISP is thought to lower the activation energy barrier of the bc 1 complex, based on the observation of an increased activation energy for the ALA-PLP complex, the complexity of the kinetics of the bc 1 complex prevents us from ruling out other steps as contributors to the activation energy barrier of this reaction. There are several distinct catalytic sites in the bc 1 complex: one for ubiquinol oxidation at the Q o site, one for ubiquinone reduction at the Q i site, one for reduction and reoxidation of ISP as well as cytochrome c 1 , and one for reduction of cytochrome c. Any one of these steps may contribute to the activation energy barrier of the overall reaction.
Based on flash kinetic studies of the bc 1 complex in R. sphaeroides chromatophore membranes showing a large activation energy for cytochrome b L or b H reduction by ubiquinol-10 (ϳ32 kJ/mol), Crofts and Wang (34) proposed that the activation energy barrier of the bc 1 complex activity was associated with the oxidation of QH 2 , either during QH 2 oxidation to an intermediate (Q . ) or during further oxidation of the semiquinone by ISP at the Q o site. The formation of a transient ubisemiquinone radical at the Q o site (E m for Q . /Qϭ -300 -400 mV) may be the largest activation energy barrier of this reaction (34). Failure to detect a Q o site inhibitor-sensitive ubisemiquinone radical in beef cytochrome bc 1 complexes 2 as well as in R. sphaeroides bc 1 complexes (34) under oxidant-induced reduction conditions (35) raises questions concerning this hypothesis.
Link (36) proposed a "proton-gated affinity change" mechanism for cytochrome bc 1 complex catalysis in which a stable semiquinone species is present at the Q o site and the first step for Q oxidation is the deprotonation of ubiquinol. Brandt and Okun (37) have shown that deprotonation of ubiquinol (⌬G ϭ 24 kJ/mol at pH 7) accounts for most of the activation barrier by measuring the pH dependence of the activation energy during steady-state turnover of bovine and yeast cytochrome bc 1 complexes. They found that it decreased linearly from pH 5.5 to 9.0. This deprotonation hypothesis has been questioned re-cently by Trumpower and co-workers (38). They have demonstrated that the midpoint potential difference between the [2Fe-2S] cluster and the electron donor ubiquinol is the driving force for the electron transfer at center P. Mutational study of Ser-183 and Tyr-185 in the S. cerevisiae Rieske iron-sulfur protein indicated that elimination of the hydrogen bond from the hydroxyl group of Ser-183 to S-1 of the [2Fe-2S] cluster or of Tyr-185 to S-␥ of Cys-159 lowers the midpoint potential of the [2Fe-2S]cluster, which in turn slows down the intracomplex electron transfer from ubiquinol to cytochrome c. Ubiquinolcytochrome c reductase activity decreases as the potential of ISP declines from approximately ϩ280 to ϩ100 mV, which confirms that oxidation of ubiquinol by ISP is the rate-limiting partial reaction in the bc 1 complex and that the rate of this reaction is extensively influenced by the midpoint potential of the [2Fe-2S] cluster. These investigators suggested that deprotonation of ubiquinol is not the rate-limiting step in the ubiquinol oxidation catalyzed by the cytochrome bc 1 complex.