The mode of subunit addition regulates the processive elongation of actin filaments by formin

Figures

• 

  Figure 1. FH2 stepping enables processive filament elongation.A, schematic representation of the two routes for actin subunits to bind a formin-bound barbed end. The actin filament is shown in gray and the FH2 dimer is shown in green. Top, an incoming actin monomer (yellow) or profilin–actin complex (not shown) can bind the barbed end directly. Bottom, a profilin–actin complex (dark blue and yellow) can bind a polyproline tract (pP tract; purple ovals) encoded in the FH1 domain for delivery to the barbed end. B, schematic representation of FH2 stepping. For clarity, the FH1 domains and profilin are not depicted. Left, the FH2 dimer binds the terminal subunits at the barbed end through engagement of three of its four actin-binding sites (3). Engagement of an actin-binding site is represented by “+”, and the free binding site is represented by “−”. Center, following binding of an incoming actin monomer, the FH2 dimer populates a dissociative transition state in which only two of its actin-binding sites engage the filament (3, 10). Right, the trailing FH2 domain (dark green) undergoes a translocation (i.e., FH2 step) in the direction of the barbed end to bind the newly added actin subunit.…

• 

  Figure 2. Bni1p mediates processive filament assembly.A, time series of inverted fluorescence micrographs depicting the polymerization of one Bni1p-bound filament (dim filament) and two filaments with free barbed ends (bright filaments) in the presence of 0.75 μM actin monomers (33% Oregon Green-labeled), 15 nM Bni1p FH1FH2, and 5 μM Saccharomyces cerevisiae (Sc) profilin. The blue arrowheads point to the barbed end of the filament as it is elongated by Bni1p, and the red arrowheads point to the location at which the fluorescence intensity of the filament increases abruptly. The scale bar represents 5 μm. B, fluorescence intensity along the lengths of three representative control (black) and formin-assembled filaments (blue) as a function of the distance from their pointed ends. C, average fluorescence intensity along the lengths of control filaments (white bar) and dim and bright segments of formin-assembled filaments (blue and red bars). Error bars are standard deviations of the average fluorescence intensity measured for at least 10 filaments or filament segments. ∗∗∗∗ corresponds to p < 0.0001 and “n.s.” indicates p > 0.05 as determined using a one-way ANOVA test (F(2,27) = 480, p <…

• 

  Figure 3. Bni1p′s processivity depends nonlinearly on the rate of subunit addition. The experimental conditions were as follows: 0.75 μM actin monomers (33% Oregon Green-labeled) in microscopy buffer with 4 to 35 nM Bni1p FH1FH2 and a range of concentrations of Sc profilin. The data were collected by TIRF microscopy. A, representative inverted fluorescence micrograph collected in the presence of 15 nM Bni1p and 5 μM profilin. The red arrowheads indicate positions at which formins have dissociated from filaments. The scale bar represents 10 μm. B, histogram of formin run lengths measured in the presence of 5 μM profilin. C, fraction of filaments bound by Bni1p as a function of length, obtained from four independent experiments performed in the presence of 5 μM profilin. A single-exponential fit was applied to the data to obtain the mean run length. D, dependence of the rate of filament elongation mediated by Bni1p on the concentrations of profilin-actin and profilin. Error bars are SDs of the mean elongation rate measured for at least 10 filaments. “n.s.” indicates p > 0.05 as determined using a one-way ANOVA test (F(5,54) = 62.7, p < 0.0001), followed by Tukey’s test (p = 0.9999). E,…

• 

  Figure 4. Profilin modulates Bni1p′s lifetime at the barbed end in a biphasic manner. The experimental conditions were as follows: 0.75 μM or 1 μM actin monomers (33% Oregon Green-labeled) in microscopy buffer with 2 to 35 nM Bni1p FH1FH2 and a range of concentrations of Sc profilin. The data were collected by TIRF microscopy. A, representative inverted micrographs collected 200 s and 600 s following the initiation of polymerization in the presence of 0.75 μM actin, 15 nM Bni1p, and 5 μM profilin. Blue arrowheads indicate the barbed ends of a representative set of formin-bound filaments. Red arrowheads indicate the positions at which formins have dissociated from the filaments. The scale bar represents 10 μm. B, fraction of filaments bound by Bni1p as a function of time, obtained from four independent experiments performed in the presence of 0.75 μM actin and 5 μM profilin. A single-exponential fit was applied to the data to obtain the dissociation rate (koff). C, dependence of the rate of Bni1p dissociation on the concentration of profilin-actin. Filled and open circles correspond to data collected in the presence of 0.75 and 1 μM actin. The error bars are the 95% confidence intervals…

• 

  Figure 5. A variable probability of dissociation dictates the processive behavior of Bni1p.A, dependence of the probability of Bni1p dissociation with each FH2 step (poff-step) on the concentration of profilin-actin. The probability values are calculated using survival curves depicting the fraction of filaments bound by Bni1p over time as depicted in Figure 4B, and the equation poff-step = 1 – exp(-koff-step∗ Δstep), where koff-step is the dissociation rate per FH2 step and Δstep is an increment of 1 step. Single-exponential fits applied to the survival curves yield dissociation rates (koff-time), which were divided by the experimental elongation rate measured in each condition to obtain a dissociation rate per FH2 step (koff-step). Errors are calculated using the 95% confidence intervals obtained from the exponential fits used to determine the values of koff-time. Data points are color-coded according to the concentration of profilin that was included in each reaction. B, dependence of the probability of Bni1p dissociation per unit time (poff-time) on the concentration of profilin-actin. The probability values are calculated using survival plots depicting the fraction of filaments bound…

• 

  Figure 6. Bni1p′s lifetime at the barbed end is insensitive to the concentration of free profilin. The experimental conditions were as follows: 0.75 μM actin monomers (33% Oregon Green-labeled) in microscopy buffer with 4 to 60 nM Bni1p FH1FH2 and a range of concentrations of Sc or Homo sapiens (Hs) profilin. Data points are color-coded according to the concentration of profilin that was included in each reaction. A, time series of inverted fluorescence micrographs depicting the polymerization of one Bni1p-bound filament (dim filament) and two filaments with free barbed ends (bright filaments) in the presence of 40 nM Bni1p FH1FH2 and 1 μM Hs profilin. The blue arrowheads point to the barbed end of the filament as it is elongated by Bni1p, and the red arrowheads point to the location at which the fluorescence intensity of the filament increases abruptly. The scale bar represents 10 μm. B, dependence of the rate of filament elongation mediated by Bni1p on the concentrations of Sc (circles) and Hs (squares) profilin-bound actin. Rates obtained from reactions performed with Sc profilin-actin are identical to those depicted in Figure 3D and are included for comparison purposes. Error bars are…

• 

  Figure 7. FH1-mediated delivery regulates the dissociation of Bni1p from barbed ends. The experimental conditions were as follows: 0.75 or 1 μM actin monomers (33% Oregon Green-labeled) in microscopy buffer with 4 to 35 nM Bni1p FH1FH2 or 7.5 to 35 nM Bni1p PCPD-FH2 and a range of concentrations of Sc profilin. Data points are color-coded according to the concentration of profilin that was included in each reaction. A, domain architecture of the Bni1p FH1FH2 and PCPD-FH2 constructs. Polyproline tracts are depicted as purple ovals and labeled profilin–actin, PB, etc. In the PCPD-FH2 construct, the profilin–actin and PB polyproline tracts are replaced with (Gly-Gly-Gly-Ser)n sequences matching the length of each tract. B and C, dependence of the run lengths of Bni1p FH1FH2 (circles) and Bni1p PCPD-FH2 (diamonds) on the concentration of profilin-actin measured in the presence of (B) 0.75 μM and (C) 1 μM actin. Run lengths obtained from reactions performed with Bni1p FH1FH2 and 0.75 μM actin are identical to those depicted in 3E and are included for comparison purposes. Error bars are 95% confidence intervals obtained from single-exponential fits of survival plots generated from measurements…

• 

  Figure 8. Bni1p′s tail domain speeds dissociation from barbed ends. The experimental conditions were as follows: 0.75 μM actin monomers (33% Oregon Green-labeled) in microscopy buffer with 4 to 35 nM Bni1p FH1FH2 or 5 to 85 nM Bni1p FH1FH2-C and a range of concentrations of Sc profilin. Data points are color-coded according to the concentration of profilin that was included in each reaction. A, domain architecture of the Bni1p FH1FH2-C construct. B, time series of inverted fluorescence micrographs depicting the polymerization of a Bni1p FH1FH2-C–bound filament (dim filament) in the presence of 5 μM Sc profilin. The blue arrowheads point to the barbed end of the filament as it is elongated by Bni1p, and the red arrowheads indicate the position at which the formin has dissociated from the filament. The scale bar is 10 μm. C, dependence of the rate of filament elongation mediated by Bni1p FH1FH2 (circles) and Bni1p FH1FH2-C (triangles) on the concentration of profilin-actin. Rates obtained from reactions performed with Sc profilin-actin are identical to those depicted in Figure 3D and are included for comparison purposes. The error bars are SDs of the mean elongation rate measured for at…

• 

  Figure 9. Model for processive filament elongation by Bni1p.A, schematic representation of actin filament elongation mediated by the FH1 (purple), FH2 (green), and C-terminal tail (black) domains of Bni1p. Top row, when actin monomers (not shown) and profilin–actin complexes (blue and yellow) bind the FH2-bound barbed end directly, dissociation of the formin from the barbed end occurs at a basal rate of ∼0.0039 s−1. Bottom row, when the fraction of monomers polymerized via the FH1 domain is below two-thirds (i.e., fraction FH1 < 0.67), the formin dissociates at the basal rate of 0.0039 s−1. When the fraction of monomers polymerized via the FH1 domain exceeds two-thirds (i.e., fraction FH1 > 0.67), the dissociation rate increases by 0.02 multiplied by the fraction of monomers polymerized via the FH1 domain that exceeds two-thirds (i.e., 0.0039 + 0.02∗(fraction FH1 – 0.67) s−1). B, dependence of mean simulated run lengths on the fraction of actin subunits that are polymerized via FH1-mediated delivery. Simulations were performed using filament elongation rates of 5 sub/s (dark blue data), 15 sub/s (light blue data), 25 sub/s (yellow data), and 35 sub/s (red data). SDs of the mean simulated…