Magnetic interactions sense changes in distance between heme b(L) and the iron-sulfur cluster in cytochrome bc(1)

Abstract

During the operation of cytochrome bc(1), a key enzyme of biological energy conversion, the iron-sulfur head domain of one of the subunits of the catalytic core undergoes a large-scale movement from the catalytic quinone oxidation Q(o) site to cytochrome c(1). This changes a distance between the two iron-two sulfur (FeS) cluster and other cofactors of the redox chains. Although the role and the mechanism of this movement have been intensely studied, they both remain poorly understood, partly because the movement itself is not easily traceable experimentally. Here, we take advantage of magnetic interactions between the reduced FeS cluster and oxidized heme b(L) to use dipolar enhancement of phase relaxation of the FeS cluster as a spectroscopic parameter which with a unique clarity and specificity senses changes in the distance between those two cofactors. The dipolar relaxation curves measured by EPR at Q-band in a glass state of frozen solution (i.e., under the conditions trapping a dynamic distribution of FeS positions that existed in a liquid phase) of isolated cytochrome bc(1) were compared with the curves calculated for the FeS cluster occupying distinct positions in various crystals of cytochrome bc(1). This comparison revealed the existence of a broad distribution of the FeS positions in noninhibited cytochrome bc(1) and demonstrated that the average equilibrium position is modifiable by inhibitors or mutations. To explain the results, we assume that changes in the equilibrium distribution of the FeS positions are the result of modifications of the orienting potential gradient in which the diffusion of the FeS head domain takes place. The measured changes in the phase relaxation enhancement provide the first direct experimental description of changes in the strength of dipolar coupling between the FeS cluster and heme b(L).

Figure 1

Changes in distances between the FeS cluster and hemes c₁ and b_L in cytochrome bc₁. The movement of the FeS head domain changes the length of interspin vector r between the center of the FeS cluster plane (black rhombus) and the Fe metal of heme b_L from ∼2.6 nm (Q_o position) to ∼3.6 nm (c₁ position). (a) Addition of ascorbate produces paramagnetic form of the FeS cluster and heme b_L (black) and diamagnetic form of heme c₁ (gray). (b) Sodium dithionite reduces all cofactors leaving the paramagnetic FeS cluster (black) and diamagnetic hemes (gray). Magnetic interactions (dashed arrows) are only possible between the paramagnetic centers (heme b_L and the FeS cluster) so they are present exclusively in (a) but not in (b). These long-range interactions depend on r (the widths of the arrows are proportional to the strength of the interaction) and can be detected by relaxation measurements when one of the centers is a fast-relaxing and the other is a slow-relaxing species. A, B, C, and D refer to four crystal positions of FeS head domain used in simulations of dipolar curves (see Experimental Procedures and Supporting Information). (c) compares the relaxation behavior of heme b_L (squares) and the FeS cluster (circles) expressed as P_1/2 (mW) determined from power saturation CW EPR experiments at X-band at g = 3.78 (g_z transition of heme b_L) and g = 1.9 (g_y transition of the FeS cluster).

Figure 2

Q-band EPR spectra of the reduced FeS cluster in the isolated native cytochrome bc₁. Absorption-like ED EPR spectra were recorded by measuring the magnetic field dependence of echo amplitude exited by a pulse sequence π/2−τ−π. Spectra obtained for three values of τ, 200, 600, or 1000 ns, are shown for dithionite- (black) and ascorbate-reduced (gray) samples. X-band CW EPR spectrum of FeS cluster is added to the top of the figure to compare the extent of Zeeman splitting of the rhombic spectrum at 9.35 (X-band) and 33.6 GHz (Q-band). g_z, g_y, and g_x denote the transitions related to the principal values of the g-tensor. CW EPR X-band spectrum is shifted to align g_y field position to the field value of g_y transition at Q-band.

Figure 3

Effect of temperature and reduction state of heme b_L on spin−lattice relaxation rate of the FeS cluster in stigmatellin-treated cytochrome bc₁. (a) shows a typical Q-band inversion−recovery curve, registered at 12 K. (b) shows the temperature dependence of spin−lattice relaxation rate in ascorbate- and ditionite-reduced cytochrome bc₁ (squares and circles, respectively). Fits in (b) (solid lines) assume dominant Orbach process. Estimated Orbach energies of the FeS cluster for ascorbate- and dithionite-reduced samples are 128 and 133 K, respectively.

Figure 4

Anisotropy of the ESE decay curves of the FeS cluster. Electron spin echo decays were measured at Q-band at 15 K in native, noninhibited cytochrome bc₁ reduced with dithionite. The curves were registered at spectral positions corresponding to g_z (gray solid), g_y (black solid), and g_x (gray dotted) transitions.

Figure 5

Influence of relaxation of heme b_L on phase relaxation rate of FeS cluster in the stigmatellin-treated native cytochrome bc₁. (a) Electron spin echo decay registered at Q-band at 19.5 K in dithionite-reduced (gray) and ascorbate-reduced (black) cytochrome bc₁. (b) Temperature dependence of phase relaxation rate defined as 1/T_M in dithionite-reduced (open circles) and ascorbate-reduced (closed circles) cytochrome bc₁. Solid lines represent a trend that was approximated by fitting the third order of polynomial to the dithionite-reduced sample or the sum of third order polynomial and Lorentzian function to the ascorbate-reduced samples.

Figure 6

Comparison of the effect of inhibitors on the temperature dependence of phase relaxation rate of the FeS cluster in native cytochrome bc₁. Shown are data for ascorbate-reduced (closed symbols) and dithionite-reduced (open symbols) samples of noninhibited cytochrome bc₁ (a) and enzyme inhibited with myxothiazol (b), antimycin (c), or antimycin plus myxothiazol (d). For comparison, in (a) the dashed gray line shows data for ascorbate-reduced, stigmatellin-treated cytochrome bc₁ reproduced from Figure 5b; in (b−d) dashed lines show data for ascorbate-reduced noninhibited cytochrome bc₁ reproduced from (a).

Figure 7

Comparison of the effect of inhibitors on the temperature dependence of phase relaxation rate of the FeS cluster in the FeS motion knockout. Shown are data for ascorbate-reduced (closed symbols) and dithionite-reduced (open symbols) samples of noninhibited cytochrome bc₁ (a) and enzyme inhibited with myxothiazol (b), antimycin (c), or stigmatellin (d).

Figure 8

Comparison of the level of relaxation enhancement of the FeS cluster in cytochrome bc₁ in the presence and absence of inhibitors. The level of the enhancement in native cytochrome bc₁ (a) and the FeS motion knockout (b) is expressed as the integrated area under the curves obtained from subtraction of temperature dependences of 1/T_M for dithionite-reduced samples from those for ascorbate-reduced samples. Abbreviations: No inh, Stg, Myx, Ant, and Myx + Ant denote the samples without inhibitor and with stigmatellin, myxothiazol, antimycin, and myxothiazol plus antimycin, respectively. Gray bar illustrates a correction accounting for the apparent weakening of the enhancement caused by stigmatellin (see Results for details).

Figure 9

Pure dipolar decay curves of the FeS cluster obtained from Q-band measurements at 12 K in the presence and absence of inhibitors. (a) shows dipolar decays in native cytochrome bc₁. The gray scale illustrates a progressive weakening of the phase relaxation which follows the order stigmatellin (black), noninhibited (dark gray), myxothiazol-treated (intermediate gray), antimycin-treated (gray), and myxothiazol plus antimycin-treated (light gray) cytochrome bc₁. (b) shows dipolar decays in the FeS motion knockout. Color code is as in (a).

Figure 10

Temperature dependence of the offset in the dipolar decay curves of native cytochrome bc₁ determined for noninhibited (squares) and stigmatellin- (circles) or myxothiazol plus antimycin-treated (triangles) samples.

Figure 11

Comparison of the simulated and experimental dipolar traces of the FeS cluster in native cytochrome bc₁. Experimental dipolar decay curve for stigmatellin-treated cytochrome bc₁ at 19.5 K (Stg) is best fitted with the spin−lattice relaxation rate of heme b_L of 4 × 10⁶ s⁻¹, when r = 2.64 nm and ∑Δ_A(θ) is calculated from the crystal structure of cytochrome bc₁ (see Figure S2 in Supporting Information) with the FeS in position A. If the same spin−lattice relaxation rate of heme b_L is taken, a simulation using r = 3.55 nm and ∑Δ_D(θ) calculated from the crystal structure of position D (see Figure S2 in Supporting Information) yields the dipolar curve for the FeS cluster in the single c₁ position (dotted line). Experimental dipolar decay curves for noninhibited (No inh) and antimycin plus myxothiazol-treated (Myx + Ant) cytochrome bc₁ are also shown. Gray area denotes the spectrometer dead time (0 on the t scale denotes the beginning of the relaxation).