Mechanoregulated inhibition of formin facilitates contractile actomyosin ring assembly

Figures

• 

  Fig. 1. Formin Cdc12-mediated actin assembly is inhibited by myosin Capture-Pull. a Schematic of reconstituted Search-Capture-Pull using node mimics, 1 µm microspheres coated with either formin Cdc12 or myosin Myo2. b–d Dual-color TIRFM micrograph time-series (min:s) (b, scale bar = 5 µm), Kymograph (c, scale bars = 200 s and 5 µm in x- and y-direction, respectively) and filament elongation rates d, of an actin filament elongating (searching, blue) from a Cdc12-associated bead (red circle) that is captured and pulled (red) by a Myo2-associated bead (white circle) before dissociating (green). For the time series, initial bead positions are marked with opaque filled circles. For the kymograph, dotted yellow lines mark filament elongation rates and white arrowheads mark the Myo2 bead position at the beginning and end of Capture-Pull (Supplementary Movie 1). Shaded lines represent the regression line fits for the respective filament elongation rate trace during the Search, Capture-Pull and Dissociation phase. e Normalized Cdc12-mediated actin filament elongation rates during Search-Capture-Pull and Dissociation (n = 11 independent Search-Capture-Pull events). One-way ANOVA with the Tukey’s…

• 

  Fig. 2. Active myosin Myo2 pulling is required to inhibit formin Cdc12. a Schematic of reconstituted Search-Capture-Pull with individual formin dimers fixed to the glass surface via the formin homology FH1 domain. b–g TIRFM kymographs b, e, corresponding elongation rates c, f, and normalized elongation rates (to pre-capture elongation rates) before (blue), during (red) and after (green) Search-Capture events d, g. Events are either surface-adhered Cdc12 and Myo2-associated beads b–d (n = 7 independent Search-Capture-Pull events), or surface-adhered Cdc12 and NEM-modified Myo2-associated beads e–g (n = 11 independent Search-Capture-Pull events) (Supplementary Movie 3). The Capture-Pull rate of 2.2 subs s−1 in c derives from interpolating the regression across three Capture-Pull intervals that are separated by very brief (~ 10–15 s) myosin-bead dissociation events. For kymographs, dotted yellow lines mark filament elongation rates and white arrowheads mark the Myo2 bead position at the beginning and end of Capture-Pull. Scale bars = 200 s b, e and 5 µm b or 10 µm e in x- and y-direction, respectively. Shaded lines c, f represent the regression line fits for the respective filament…

• 

  Fig. 3. Formin mDia2 remains active during Capture-Pull. a–c TIRFM kymograph (a, scale bars = 100 s and 5 µm in x- and y-direction, respectively) and corresponding elongation rates (b), and normalized elongation rates (to pre-capture elongation rates) for multiple events before (blue), during (red) and after (green) Search-Capture-Pull c between surface-adhered formin mDia2 and Myo2-associated beads (n = 15 independent Search-Capture-Pull events) (Supplementary Movie 4). Shaded lines b represent the regression line fits for the respective filament elongation rate trace during the Search and Capture-Pull phase. One-way ANOVA with the Tukey’s multiple comparisons test was performed on the entire data set (c, n.s., not significant when p ≥ 0.05). For kymograph, dotted yellow lines mark filament elongation rates and white arrowheads mark the Myo2 bead position at the beginning and end of Capture-Pull. Average formin elongation rates are listed in Supplementary Table 1

• 

  Fig. 4. Mechanosensitive inhibition maps to formin Cdc12’s FH1 domain. a Wild-type and chimeric formin constructs generated by exchanging the FH1 domains of Cdc12 and mDia2. b, c Normalized mDia2(FH1)-Cdc12(FH2)- b and Cdc12(FH1)-mDia2(FH2)-mediated c actin filament elongation rates before (blue), during (red), and after Capture-Pull (green) (n ≥ 10 independent Search-Capture-Pull events) of reconstituted Search-Capture-Pull between chimeric formin- and Myo2-associated beads. One-way ANOVA with the Tukey’s multiple comparisons test was performed on the entire data set (b, c, n.s., not significant when p ≥ 0.05, ***highly significant with p < 0.0001). Average formin elongation rates are listed in Supplementary Table 1

• 

  Fig. 5. Formin Cdc12 remains active when its FH2 domain experiences pulling force. a Schematic of reconstituted Search-Capture-Pull with individual formin dimers fixed to the glass surface via the formin homology FH2 domain. b, c Normalized Cdc12(FH1FH2)-SNAP- b and mDia2(FH1FH2-C)-SNAP-mediated c actin filament elongation rates before (blue), during (red), and after Capture-Pull (green) (n ≥ 8 independent Search-Capture-Pull events) of reconstituted Search-Capture-Pull between FH2-surface attached formin and Myo2-associated beads. One-way ANOVA with the Tukey’s multiple comparisons test was performed on the entire data set (b, c, n.s., not significant when p ≥ 0.05). Average formin elongation rates are listed in Supplementary Table 1

• 

  Fig. 6. Formin Cdc12’s force response does not require filament tension. a Illustration of filament tension T, the ratio of node-to-node contour filament length and formin node-to-myosin node distance. b Dependence of normalized formin elongation rates on filament tension. c Data from (b) summarized in a Box–Whisker plot. Elongation rates were normalized to the rate measured prior to myosin Capture-Pull. nCdc12 = 19 independent Search-Capture-Pull events, nmDia2 = 15 independent Search-Capture-Pull events. Average formin elongation rates are listed in Supplementary Table 1

• 

  Fig. 7. Myosin-mediated inhibition of formin Cdc12 facilitates contractile ring assembly in vivo. a Representative fields and time-series of dividing control (cdc12) and chimera mutant mDia2 (Cdc12N-mDia2FH1FH2-Cdc12C) fission yeast cells expressing Rlc1-3GFP (Supplementary Movies 6 and 7). Cytokinesis node clumping is marked with blue arrowheads. Scale bars = 5 µm. b Average node band width (3.0 ± 0.6 µm vs. 2.2 ± 0.6 µm for cdc12 and mDia2, Mean ± S.D., n ≥ 21 cells; p = 0.00008, and c 3D surface plots of representative node distributions, 8 min before ring maturation. d BODIPY-phallicidin staining of F-actin organization in cdc12 (left) and mDia2 (right) cells. Cell midzone used for quantification in e is boxed with dashed lines. Scale bar = 5 µm. e Plot of median F-actin fluorescence in the contractile ring for cdc12 and mDia2 cells. Median relative fluorescence, 5.6 ± 0.5 vs. 12.9 ± 0.7 for cdc12 and mDia2, Median ± s.e.m., n ≥ 20 cells; p = 0.02. One-sided (unpaired) Student’s t-test was used to compare the two groups of data sets (b, e). f Cartoon model showing uninhibited actin assembly by mDia2 leads to node clumping by disorganized connections between multiple nodes. g…

• 

  Fig. 8. Cartoon model of how mechano-inhibition of formin Cdc12 is critical for facilitating proper ring assembly from cytokinesis nodes. By slowing down in vivo F-actin polymerization rates to a level where Myo2 pulling net velocity is sufficiently larger (Fig. 7i), mechano-inhibition of formin Cdc12 facilitates effective node coalescence (Fig. 7h). In addition, mechano-inhibition of formin Cdc12 prevents excessive F-actin generation, which would impede broad band formation once neighboring nodes are inter-linked upon capture (Fig. 7a–f)