Modifications of protein environment of the [2Fe-2S] cluster of the bc1 complex: effects on the biophysical properties of the rieske iron-sulfur protein and on the kinetics of the complex

Abstract

The rate-determining step in the overall turnover of the bc(1) complex is electron transfer from ubiquinol to the Rieske iron-sulfur protein (ISP) at the Q(o)-site. Structures of the ISP from Rhodobacter sphaeroides show that serine 154 and tyrosine 156 form H-bonds to S-1 of the [2Fe-2S] cluster and to the sulfur atom of the cysteine liganding Fe-1 of the cluster, respectively. These are responsible in part for the high potential (E(m)(,7) approximately 300 mV) and low pK(a) (7.6) of the ISP, which determine the overall reaction rate of the bc(1) complex. We have made site-directed mutations at these residues, measured thermodynamic properties using protein film voltammetry to evaluate the E(m) and pK(a) values of ISPs, explored the local proton environment through two-dimensional electron spin echo envelope modulation, and characterized function in strains S154T, S154C, S154A, Y156F, and Y156W. Alterations in reaction rate were investigated under conditions in which concentration of one substrate (ubiquinol or ISP(ox)) was saturating and the other was varied, allowing calculation of kinetic terms and relative affinities. These studies confirm that H-bonds to the cluster or its ligands are important determinants of the electrochemical characteristics of the ISP, likely through electron affinity of the interacting atom and the geometry of the H-bonding neighborhood. The calculated parameters were used in a detailed Marcus-Brønsted analysis of the dependence of rate on driving force and pH. The proton-first-then-electron model proposed accounts naturally for the effects of mutation on the overall reaction.

FIGURE 1.

Schematic hydrogen-bonding network of the [2Fe-2S] cluster and the surrounding amino acids residues. Ser-154 and Tyr-156 are two mutagenesis targets in this study, and the hydrogen atoms binding to the two Nϵ positions of the histidines are likely the two involved in deprotonation (pK_ox1 and pK_ox2) (14–17).

FIGURE 2.

Redox titrations of the bc₁ complex to determine the E_m values of ISP by CD. The redox states of the ISPs (■, wild type; and mutant strains: ●, S154T; ▴, S154C; ▾, S154A) were monitored from the difference in CD spectrum at 500 and 470 nm as described under “Experimental Procedures.” The plots were normalized for the total amplitude of the difference between fully oxidized state and fully reduced state. SHE, standard hydrogen electrode.

FIGURE 3.

Measurement of thermodynamic parameters for redox characteristics of ISP using protein film voltammetry. The redox midpoint potentials (E_m) of the S154T (○), S154C (▵), S154A (▿), Y156F (▶), and Y156W (◀) were calculated by averaging the peak potentials of the reversible cyclic voltammograms and titrated in the pH range of 4–14. The dashed line for the wild type is from Zu et al. (33); other curves were calculated using Equation 1 or the E_acid-based equivalent equation from Ref. , which gave the same fit. SHE, standard hydrogen electrode.

FIGURE 4.

Dependence on pH of Q_o-site turnover. The activities of the mutant bc₁ complex were titrated by measuring the cyt b_H reduction rates (ΔA 561–569 nm) in the presence of antimycin A in the range of pH 5.5–8.5 (the filled symbols) and the electrogenic processes from electrochromic changes of carotenoids (ΔA 503 nm) at pH > 8.5 (the outlined symbols) after flash initiation of the reaction. The reaction conditions are described under the “Experimental Procedures.” The pH dependence from titration data were fitted to the equation suggested by Brandt and Okun (51), and the estimated parameters are given in Table 2. ■, wild type; ●, S154T; ▴, S154C; ▾, ▿, S154A; ▶, Y156F; ◀, ◁, Y156W.

FIGURE 5.

HYSCORE spectra of Y156F (A) and WT (B) are shown. ¹H spectra are displayed in a three-dimensional stacked plot representation and as corresponding contour plots. Parameters of the measurements are listed in the figure. The HYSCORE difference spectrum (C) shows the redistribution and/or disappearance of cross-peaks in a contour plot. Peaks labeled a and b are discussed in the text.

FIGURE 6.

HYSCORE spectra of S154A (A) and WT (B) are shown. ¹H spectra are displayed in a three-dimensional stacked plot representation and as corresponding contour plots. Parameters of the measurements are listed in the figure. The HYSCORE difference spectrum (C) shows the redistribution and/or disappearance of cross-peaks in a contour plot. Peaks labeled a and b are discussed in the text.

FIGURE 7.

Structural changes on mutation. Superimposed images comparing the structures of some mutant ISPs (S154T, red; Y156W, magenta) with that of wild type (CPK color). A, added methyl group in Ser-154 position stretches into the hydrophobic cleft between Ile-162 and Tyr-156, replacing the position of the Cδ of Ile-162 and making the aromatic ring plane of Tyr-156 tilt by 10°. B, volume enclosed by the [2Fe-2S] cluster capturing loops (Cys-129 to Cys-134 and Cys-151 to Tyr-156) expanded on the substitution of tyrosine by tryptophan at the 156 position.

FIGURE 8.

Plausible energy profile of QH₂ oxidation at the Q_o-site. The different intermediate states of the Q_o-site reaction are shown against a free energy scale normalized to the initial state. Dashed vertical arrows show ranges for energy parameters that would allow adequate rates of overall electron transfer with different assumptions (14, 16, 71). For the energy level for the intermediate product state after the first electron transfer, a more stable SQ is shown than in previous work to take into account constraints from Refs. , on maximal occupancy of SQ in the Q_o-site (71).

FIGURE 9.

Dependence of the reaction rates of mutant strains on the driving forces. Measured rates are compared with the theoretical curves generated using software package, Marcus_Bronsted.exe. The data points shown by square symbols are from measurements at pH 7.0 and fall close to the black curve, calculated using parameters for wild type (see supplemental Fig. S2). The data points shown by circles were from kinetic measurements at optimal pH, where the imidazolate form of ISP_ox was saturating. The values for log₁₀k are therefore displaced from those at pH 7, and for most mutant strains are higher. At the optimal pH, the E_m value was also displaced from the value at neutral pH (Fig. 3), so that the driving force for the overall reaction was changed, and the positions are therefore displaced from those of the square symbols with respect to the ΔG scale, to more endergonic values for most strains. The colored curves were calculated using the same values for constant parameters (values for λ, distances, redox properties, and pK values of the quinone system) as used for the black curve, but values for variables appropriate to each optimal condition and appropriate for each strain, using values for k (given by V_opt), and pK_A were from the tables, and E_{m, pHopt} was from Fig. 3. The black symbol represents wild type; the red, S154T; the green, S154C; the blue, S154A; the cyan, Y156F; the magenta, Y156W. Filled square symbols, reaction rates at pH 7; open circles, reaction rates at optimal pH.