Polymerization kinetics of ADP- and ADP-Pi-actin determined by fluorescence microscopy

Abstract

We used fluorescence microscopy to determine how polymerization of Mg-ADP-actin depends on the concentration of phosphate. From the dependence of the elongation rate on the actin concentration and direct observations of depolymerizing filaments, we measured the polymerization rate constants of ADP-actin and ADP-P(i)-actin. Saturating phosphate reduces the critical concentration for polymerization of Mg-ADP-actin from 1.8 to 0.06 microM almost entirely by reducing the dissociation rate constants at both ends. Saturating phosphate increases the barbed end association rate constant of Mg-ADP-actin 15%, but this value is still threefold less than that of ATP-actin. Thus, ATP hydrolysis without phosphate dissociation must change the conformation of polymerized actin. Analysis of depolymerization experiments in the presence of phosphate suggests that phosphate dissociation near the terminal subunits is much faster than in the interior. Remarkably, 10 times more phosphate is required to slow the depolymerization of the pointed end than the barbed end, suggesting a weak affinity of phosphate near the pointed end. Our observations of single actin filaments provide clues about the origins of the difference in the critical concentration at the two ends of actin filaments in the presence of ATP.

Fig. 1.

Polymerization of Mg-ADP-actin SI Movies 1 and 2. Conditions were as follows: 50 mM KCl/1 mM MgCl₂/1 mM EGTA/10 mM imidazole, pH 7.0/100 mM DTT/0.2 mM ADP/15 mM glucose/20 μg ml⁻¹ catalase/100 μg ml⁻¹ glucose oxidase/0.5% methylcellulose at room temperature. (A) Time series of fluorescence micrographs showing the elongation of filaments by 3 μM Mg-ADP-actin 30% labeled with Alexa green. (Scale bar, 10 μm.) (B) Time course of elongation of three sample filaments by 3 μM ADP-actin. Filled symbols, barbed end; open symbols, pointed ends. Average elongation rates from linear fitting of 17 actin filaments under these conditions were 5.2 ± 0.1 s⁻¹ at the barbed and 0.14 ± 0.04 s⁻¹ at the pointed end. (C) Time course of depolymerization of sample filaments polymerized from 5 μM ADP-actin. Circles, polymerized and then washed with polymerization buffer containing 50 mM KCl and no actin at 620 s; triangles, polymerized and then washed with polymerization buffer containing 20 mM P_i and no actin at 440 s. Filled symbols, barbed end; open symbols, pointed ends.

Fig. 2.

Effect of P_i on elongation of actin filaments. Conditions were as in Fig. 1. To maintain constant ionic strength, the concentration of sulfate was varied such that the total concentration of P_i and sulfate was 160 mM, except the samples with <20 mM P_i, which did not have sulfate. (A) Dependence of the rate of elongation of barbed ends on the concentration of P_i. Red circles, 1 μM Mg-ATP-actin monomers; blue circles, 1 μM Mg-ADP-actin monomers. The curves are drawn through the data points. (B) Four different concentrations of Mg-ADP-actin monomers. The curves are global fits to the three highest ADP-actin concentrations (see Discussion and SI Text). (C) Barbed end elongation rate as a function of the concentration of Mg-ADP-actin monomers over a range of P_i concentrations. Black line and circles, 0 mM; broken line and brown circles, 5 mM; red line and circles, 20 mM; other P_i concentrations are defined by symbols on the graph. Error bars are ±1 standard deviation of the mean rates. The lines intersect the x axis at the critical concentrations. (D) Dependence of the rate of pointed end elongation on the concentration of P_i. Red squares, 1 μM Mg-ATP-actin; blue squares, 1 μM Mg-ADP-actin. (E) Dependence of the rate of pointed end elongation by four concentrations of Mg-ADP-actin monomers on P_i concentration. The curves are global fits to the three highest ADP-actin concentrations (see Discussion). (F) As in C, for the pointed end.

Fig. 3.

Dependence of the dissociation of Mg-ADP-actin from the two ends of actin filaments on P_i. Conditions were as in Fig. 2 (the total ionic strength was standardized with sulfate). (A) Barbed end. The lines are fits to the data (see SI Text for all parameter values). Continuous line: rate of P_i dissociation from actin filaments is 0.003 s⁻¹ and equilibrium P_i dissociation constant K_d^B2,3 = 50 mM. Dashed line: same as continuous line but the rate of P_i dissociation from terminal subunit is 20 s⁻¹ and K_d^B2,3 = 1 mM. Dotted line: same as continuous line, but K_d^B2,3 = 1.5 mM. (B) Pointed end. Continuous line: uniform P_i dissociation rate 0.003 s⁻¹, dissociation equilibrium constant K_d^P2,3 = 60 mM. Dashed line: rate of P_i dissociation from terminal subunit is 20 s⁻¹, K_d^P2,3 = 6 mM. Gray line: same as dashed line, but the P_i dissociation equilibrium constant is 6 mM on the terminal subunit and 1.5 mM elsewhere. Dotted line: same as the continuous line, but K_d^P2,3 = 1.5 mM.

Fig. 4.

Barbed end reactions. (A) Kinetic and thermodynamic parameters of barbed end polymerization including reactions of Mg-actin with bound ATP, ADP-P_i, or ADP. Numbers next to light arrows are rate constants and those next to bold arrows are ratios of rate constants (see Discussion). We indicate the standard deviations of the least certain values. The values for P_i binding to filaments apply to interior subunits. (B) Polymerization rate vs. ATP-actin concentration. Dashed line: theoretical calculations, using the values in A. Continuous line: as dashed line, but the rate constant for P_i dissociation from the terminal subunit is 20 s⁻¹. (Inset) ATP-actin critical concentration vs. rate of P_i dissociation from the terminal subunit.

Fig. 5.

Pointed end reactions. (A) Kinetic and thermodynamic parameters. Values in brackets for ATP-actin are consistent with detailed balance (see Text). (B) Pointed end polymerization rate vs. ATP-actin concentration for a range of association rate constants for ATP-actin binding terminal ADP-actin subunits, assuming that P_i dissociates from terminal subunits at r_Pi = 20 s⁻¹, and using parameter values from A except for 0.16 s⁻¹ for the ATP-actin dissociation rate constant. The gray line is the extrapolated linear fit from ref. for ATP-actin.