Fig. 1. Schematic presentation of actin used for the measurements. (a) Surface representation of actin molecules in the filament. Contrasting shades represent different molecules. The actin structure was drawn based on the rabbit skeletal muscle actin crystal structure (Kabsch et al., 1990) (PDB code 1ATN) and a theoretical model (Mendelson and Morris, 1997) (PDB code 1ALM). The donor (green) and the acceptor (red) fluorophores were positioned in the blue colored actin molecule. This figure was made with VMD (Humphrey et al., 1996). (b) The donor (green) and acceptor (red) probes are placed on the actin structure, shown in ribbon representation, based on the crystal structure of rabbit skeletal muscle actin (Kabsch et al., 1990) (PDB code 1ATN). This figure was made with MolMol (Koradi et al., 1996). (c) Actin filament adsorbed onto a quartz slide through an avidin–biotin system (Itakura et al., 1993). Actin (gray), biotin (blue), avidin (brown), and BSA (yellow) are depicted in the drawing. Single-molecule FRET was measured with TIRFM (Funatsu et al., 1995).

Fig. 2. Single-molecule FRET signals. (a) The donor and acceptor were excited alternatively by an Ar–Kr laser (λ = 514.5 nm) and a laser diode (λ = 635 nm), respectively, using optical shutters. (b) A pair of the donor (I_D, top left) and acceptor fluorescence images (I_A, top right) and the acceptor fluorescence images (I_R, bottom right) were taken by alternative excitation by the Ar–Kr laser for the excitation of the donor (top) and the laser diode for the excitation of the acceptor (bottom), respectively. (c) Time courses of the fluorescence I_D, I_A and I_R.

Fig. 3. Detection of dynamic FRET changes. (a) Time course of the donor (green line, I_D) and acceptor (red line, I_A) fluorescence of double-labeled actin molecules in the filament. Raw data (light lines) and the data after passing through a low-pass filter (dark lines) are superimposed. (b) Time course of florescence from single-labeled actin molecule. Left: fluorescence of TMR (green line) and right: IC5 (red line). Raw data (light lines) and the data after passing through a low-pass filter (dark lines) are superimposed. (c) Power spectrum densities of the fluctuation of FRET efficiency and the fluorescence intensities. Blue: sum of the donor (I_D) and acceptor fluorescence (I_A), I_D + I_A. Violet: FRET efficiency (I_A/(I_D + I_A)). Solid lines are best fits with least-squares approximation by splines. (d) Correlation factor as an index on how the signals are perturbed by noises. Computer simulation was performed to relate the correlation factor to the S/N (black dots). The S/N was defined as the ratio of the standard deviation of the artificial anti-correlated signal to given noise. The value (−0.31) for the correlation factor obtained experimentally for actin is shown as a red diamond.

Fig. 4. Single-molecule FRET from single filamentous actin molecules in the filament. (a) Typical FRET data from individual actin molecules. Each row represents different molecules. Left: time courses of donor (green line) and acceptor (red line) fluorescence. The data were passed through a low-pass filter (see text). Center: time courses of the FRET efficiencies calculated from donor and accepter fluorescence intensities shown on the left. Right: histograms of FRET efficiencies. The histograms were constructed from the data of the FRET efficiency time courses. (b) Histogram of the FRET efficiencies summed for 35 different actin molecules in 35 different actin filaments. Total areas of the histograms of individual molecules have been normalized for averaging. The histogram of the FRET efficiency of actin was fitted to a sum of two Gaussian distributions (green solid line) with means of 0.27 (S.D. = 0.12) and 0.54 (S.D. = 0.12).

Fig. 5. Kinetic analysis for the transition between two actin conformational states. (a) Spontaneous conformational transition of actin and the duration time for two major conformational states. The transition from the low to high FRET state is deemed to have occurred when the FRET efficiency increases through E_FRET = 0.42 (=the mean − the standard deviation of the Gaussian distribution of the high FRET state) (red broken line) and the transition from the high to low FRET state is deemed to have occurred at E_FRET = 0.39 (=the mean + the standard deviation of the Gaussian distribution of the low FRET state) (blue broken line). (b) Distribution of the duration time for the two actin conformational states. The red and blue histograms are the duration times in high FRET state, t_A and low FRET state, t_I, respectively. The solid lines are best fit to a single exponential decay with rate constants 0.34 ± 0.05 s⁻¹ for t_A and 0.24 ± 0.04 s⁻¹ for t_I.

Fig. 6. Conformational state of actin that activates and inactivates myosin motility. (a) The distribution of the FRET efficiency for actin molecules interacting with myosin V in the presence of ATP (red bars) is compared with the high FRET efficiency component (red line). (b) The distribution of the FRET efficiency for the cross-linked (blue bars) is compared with the low FRET efficiency component (blue line). (c) The conformational dynamics of actin molecules in the filament. Actin molecules undergo a transition between active (A) and inactive (I) states. The binding of myosin shifts the equilibrium toward the A state and cross-linking with glutaraldehyde shifts toward the I state.