Visualizing changes in electron distribution in coupled chains of cytochrome bc(1) by modifying barrier for electron transfer between the FeS cluster and heme c(1)

Abstract

Cytochrome c(1) of Rhodobacter (Rba.) species provides a series of mutants which change barriers for electron transfer through the cofactor chains of cytochrome bc(1) by modifying heme c(1) redox midpoint potential. Analysis of post-flash electron distribution in such systems can provide useful information about the contribution of individual reactions to the overall electron flow. In Rba. capsulatus, the non-functional low-potential forms of cytochrome c(1) which are devoid of the disulfide bond naturally present in this protein revert spontaneously by introducing a second-site suppression (mutation A181T) that brings the potential of heme c(1) back to the functionally high levels, yet maintains it some 100 mV lower from the native value. Here we report that the disulfide and the mutation A181T can coexist in one protein but the mutation exerts a dominant effect on the redox properties of heme c(1) and the potential remains at the same lower value as in the disulfide-free form. This establishes effective means to modify a barrier for electron transfer between the FeS cluster and heme c(1) without breaking disulfide. A comparison of the flash-induced electron transfers in native and mutated cytochrome bc(1) revealed significant differences in the post-flash equilibrium distribution of electrons only when the connection of the chains with the quinone pool was interrupted at the level of either of the catalytic sites by the use of specific inhibitors, antimycin or myxothiazol. In the non-inhibited system no such differences were observed. We explain the results using a kinetic model in which a shift in the equilibrium of one reaction influences the equilibrium of all remaining reactions in the cofactor chains. It follows a rather simple description in which the direction of electron flow through the coupled chains of cytochrome bc(1) exclusively depends on the rates of all reversible partial reactions, including the Q/QH2 exchange rate to/from the catalytic sites.

Fig. 1

Ribbon model of the crystal structure of the heme domain of cytochrome c₁ subunit of Rba. capsulatus cytochrome bc₁. Heme c₁ (red sticks) is axially ligated by histidine and methionine (greens sticks). Position A181 (magenta spheres) is located within the sixth axial ligand domain. Position Y152 (orange line) is located between the two cysteine residues that form disulfide bond (yellow sticks).

Fig. 2

Optical redox difference spectra of cytochrome bc₁ isolated from wild type (A), A181T mutant (B) and A181T/C144A/C167A mutant (C). Solid and dashed lines correspond to dithionite minus ferricyanide and ascorbate minus ferricyanide spectra, respectively.

Fig. 3

Potentiometric dark equilibrium titration of heme c₁ in isolated wild-type cytochrome bc₁ (closed black circles), A181T mutant (open grey squares) and A181T/C144A/C167A triple mutant (open black triangles). The titrations were performed at pH 7, and experimental data were fit to the Nernst equation for a one-electron couple. The E_m values obtained are shown in the figure and also listed in Table 1.

Fig. 4

Flash-activated cytochrome c oxidation and re-reduction in chromatophores containing wild-type cytochrome bc₁ (A, C) and the A181T mutant (B, D). Kinetic transients at 550–540 nm were recorded at pH 7 (A, B) or pH 6 (C, D) with the Q pool half-reduced, in the absence of inhibitors (black) and in the presence of antimycin (red), myxothiazol (blue) or stigmatellin (green).

Fig. 5

Difference in midpoint potential between the FeS cluster and heme c₁ and its influence on the post-flash distribution of electrons in the c-chain in the presence of antimycin. (A) pH-dependence of E_m of the FeS cluster (dotted line) and heme c₁ in wild type (solid line) and in A181T mutant (dashed line). Horizontal dashed line for A181T is assumed from the similar values of E_m measured at pH 7 and 9 (Table 1). (B) The experimentally determined fraction of cytochrome c re-reduced in the wild type (open squares) and A181T mutant (open circles) in the presence of antimycin (antimycin phase) is plotted vs. difference in E_ms between the FeS cluster and heme c₁. Solid line represents the changes in the fraction of cytochrome c re-reduced simulated for the conditions when antimycin is present by the model described in Fig. 6D.

Fig. 6

Analysis of post-flash electron distribution in wild type and A181T cytochrome bc₁. Schemes in A, B compare wild type (WT) with A181T for a given set of conditions. Black and white squares represent reduced and oxidized cofactors, respectively. Incomplete reduction is represented in a grey scale. Dotted squares mark the cofactors of c-chain that are observable in the flash experiments. (A) Single flash generates two oxidizing equivalents per cytochrome bc₁ (white squares c) at the time where hemes b are oxidized (white squares b_L and b_H) and the FeS cluster and heme c₁ are reduced (black squares FeS and c₁, respectively). (B) Within tens of milliseconds electrons redistribute to reach equilibrium in which the reduction levels of cofactors vary depending on experimental conditions. In the absence of inhibitors, unperturbed electron flow out of the b-chain upon oxidation of two QH₂ secures complete re-reduction of all cofactors in the c-chain in both WT and A181T (black squares in no inhibitor panel). In the presence of antimycin, the level of reduced hemes b available for reverse reaction is higher and the equilibrium is reached before the c-chain is fully reduced. In WT, incomplete oxidation of second QH₂ (grey squares in antimycin panel) leaves FeS partially oxidized, which leads to redistribution of electrons in the entire c-chain, as observed by flash at the level of cytochromes c (note that intensity of squares in c-chain should be reduced, which for simplicity is not shown). In A181T, incomplete oxidation of second QH₂ faces additional barrier of potential difference between FeS and heme c₁ which shifts equilibrium toward reduced FeS at the expense of reduced heme c₁ (represented as black square FeS and light grey square c₁). This increases probability of reverse reaction and decreases probability of forward reaction at the Q_o site, and the level of reduced heme b_L may decrease (note that in this case the reduction level of heme c₁ determines the reduction level of heme b_L, as represented by light grey squares c₁ and b_L). In the presence of myxothiazol (myxothiazol panel), electron from pre-reduced FeS cluster redistributes among the cofactors of the c-chain, but in WT the oxidation of FeS cluster by cytochromes c is more prominent than in A181T (note that for simplicity, for WT a complete electron transfer from FeS cluster to heme c is shown, white square FeS and black squares c). In the presence of stigmatellin (stigmatellin panel) only cytochromes c are in equilibrium thus full extent of flash-oxidized cytochromes c is preserved. Panel (C) shows simulations of the traces for the reduction of the observable experimentally cytochromes c in WT (dashed line) and A181T (solid line) obtained from the model schematically presented in panel (D). The time constants for partial reactions denote k_0f/k_0b — association/dissociation rate of Q to/from the Q_o site, k_1f/k_1b — same as k_0f/k_0b but for QH₂, k_2f/k_2b — two-electron oxidation/reduction of QH₂/Q in the Q_o site, k_3f/k_3b — rate constants for movement of the FeS head domain to/from cytochrome c₁ position, k_4f/k_4b — rate constants for electron transfer from FeS to heme c₁ or reverse reaction, k_5f/k_5b — rate constant for electron transfer from heme c₁ to heme c or reverse reaction. Dotted line in the antimycin panel in C shows the trace simulated for A181T assuming that QH₂ oxidation at the Q_o site was concerted but irreversible (i.e., k_2b = 0 M^-1s^-1). This illustrates the prediction of how the system would respond if at least one reaction was irreversible (a case not supported experimentally).