36 degrees step size of proton-driven c-ring rotation in FoF1-ATP synthase

Abstract

Synthesis of adenosine triphosphate ATP, the 'biological energy currency', is accomplished by F(o)F(1)-ATP synthase. In the plasma membrane of Escherichia coli, proton-driven rotation of a ring of 10 c subunits in the F(o) motor powers catalysis in the F(1) motor. Although F(1) uses 120 degrees stepping during ATP synthesis, models of F(o) predict either an incremental rotation of c subunits in 36 degrees steps or larger step sizes comprising several fast substeps. Using single-molecule fluorescence resonance energy transfer, we provide the first experimental determination of a 36 degrees sequential stepping mode of the c-ring during ATP synthesis.

Figure 1

Single-molecule FRET approach to detect the 36° step size of the rotary c subunits in F_oF₁-ATP synthase during ATP synthesis. (A) Model of FRET-labelled E. coli F_oF₁-ATP synthase with EGFP (green; fused to the C terminus of subunit a, orange) and Alexa568 (red, at residue E2C) at one of the c subunits (blue). Rotation of c results in stepwise distance changes to EGFP. (B) The FRET-labelled E. coli F_oF₁-ATP synthase is reconstituted and diffuses freely through the dual laser foci of the duty cycle-optimized alternating laser excitation scheme. (C) Photon burst of a single FRET-labelled F_oF₁-ATP synthase. In the lower panel, fluorescence intensities of FRET donor EGFP (green trace) and acceptor Alexa568 (red trace) are shown excited with 488 nm. In the upper panel, the fluorescence intensity of the directly excited Alexa568 (pulsed interleaved excitation with 561 nm, see text) of the same F_oF₁-ATP synthase is shown (black trace) confirming the existence of both markers on the enzyme.

Figure 2

(A–D) Photon bursts of single FRET-labelled F_oF₁-ATP synthases during ATP synthesis. Lower panels show fluorescence time trajectories for EGFP (green trace) and Alexa568 (red trace), upper panels show the corresponding intramolecular FRET distances (most-likely distance in red and deviations as blue band). Black arrows mark small steps, red arrows large steps. (E) Dwell time distributions of FRET levels in F_oF₁-ATP synthases during ATP synthesis (red bars), and (F), with 20 μm aurovertin B (gray bars, monoexponential fits in black). Adding an additional rising component apparently improved the fitting at shorter dwells. This is due to the fact that the lower limit to determining dwell times is 2 ms, and thus the histograms lack those data points.

Figure 3

Model for the geometrical constraints for the FRET measurements within a single F_oF₁-ATP synthase and FRET transition density plot. (A) Model for FRET distances between the EGFP chromophore (green dot) and three of the 10 stopping positions of Alexa568 (red dots) during c-ring rotation. Distance changes for a 36° step correspond to arrows blue and cyan; a 144° step corresponds to arrows blue and black. (B) Five zones for the 10 c positions as defined from symmetry reasons. 36° stepping of c results in changing between adjacent zones, and, correspondingly, incremental FRET distance changes. (C) Larger step sizes of 108° require switching between non-neighbouring zones. (D) Transitions between FRET distances according to 36° steps and geometries as shown in panel B. (E) Transitions between FRET distances according to 120° steps. (F) FRET transition density plot for proton-driven c subunit rotation with constraint curves for 36° steps in white and for 120° steps in black.

Figure 4

Monte Carlo simulations of FRET distance changes during stepwise c-ring rotation and comparison with experimental data during ATP synthesis. (A) Experimental FRET distance changes. Simulations with 36° (B), 72° (C), 108° (D), 144° (E), 120° (F), 40° plus 80° (G); the weighted sum of 36°, 72°, 108° and 144° step sizes (H) reproduced the FRET distance difference plot and the experimental FRET transition density plot (see Figure 3F) fairly well. I, FRET transition density plot from the Monte Carlo simulation shown in (H).