Molecular basis and cellular functions of vinculin-actin directional catch bonding

Figures

• 

  Fig. 1. Evaluation of energetics and hydrogen bond (H-bond) properties between F-actin and Vt. a Schematic representation of pulling forces imposed on Vt in different directions during MD simulations. b Quantification of Vt:F-actin binding strength through potential of mean force (PMF) calculations. The X-axis represents the reaction coordinate (ξ), and the Y-axis represents the PMF. The graph displays PMF profiles for three pulling directions: along the Fpointed, Fbarbed, and Fnormal directions, as Vt is pulled away from the actin filament. c A circular plot depicting Vt and F-actin interactions in the Fpointed direction to probe Vt:F-actin engagement. The occupancy percentages of inter-molecular H-bonds between F-actin and Vt as quantified from constant-force pulling DMD simulations were used to create the circular plot. A scatter plot was used to illustrate the occupancies of H-bonds, allowing for a detailed analysis of the H-bond occupancies at specific forces. Each dot on the plot represents the existence of H-bonds at different pulling forces, color-coded according to force magnitude. The size of each dot indicates the occupancy percentage of the corresponding H-bond during the…

• 

  Fig. 2. Characterization of directionally asymmetric force-strengthening (DAFS) H-bond interactions between F-actin and Vt. Close-up view of the Vt/F-actin interface (PDB ID: 3JBI) highlighting a, H-bonds between Vt and F-actin and b, DAFS interactions for the force-loaded state of Vt:F-actin complex in the Fpointed direction. Newly induced DAFS interactions from Fpointed trajectories are enclosed in dotted circles. Color scheme: Vt is shown in red and F-actin in gray. The residue pairs involved in H-bond interactions are shown in ball and stick representation.

• 

  Fig. 3. DAFS residue variants retain Vt structure and stability. a Ribbon diagram of Vt with residues predicted to form DAFS interactions highlighted in blue and a key actin binding residue (I997) in grey licorice. b Far-UV CD spectral profiles of DAFS variants. c Near-UV spectral overlay of DAFS variants. d Melting temperature (TM) of DAFS variants calculated from CD thermal melt curves. Representative far-UV and near-UV CD spectral profiles shown are from N = 3 independent experiments. The CD thermal melt profiles are presented as an average of 3 scans from N = 3 independent experiments. Error in TM is shown with standard errors.

• 

  Fig. 4. DAFS variants retain Vt interactions with F-actin and PIP2. a High speed F-actin co-sedimentation assays comparing actin binding of WT Vt, DAFS and the actin-deficient I997A Vt variant. Vt I997A is an actin-binding deficient control. b Low-speed F-actin co-sedimentation assays with WT Vt, DAFS and Vt I997A variants comparing the fraction of F-actin present in bundles or in higher-order assemblies. Box-whisker plots with n = 12 independent measurements pooled from N = 3 preparations are presented for F-actin a binding and b bundling co-sedimentation assays, respectively. c Box-whisker plot with n = 7 independent measurements from N = 2 preparations comparing WT Vt and DAFS Vt variant association with PIP2-containing LUVs. Box depicts the median as the center value, the 25th percentile as the lower bound, and the 75th percentile as the upper bound. Whiskers extend 1.5 times the interquartile region (IQR) from the bottom and top of box, or to the minimum and maximum of the data if the data does not extend to the whiskers. Values outside the whiskers are plotted as individual points. One-way analysis of variance (ANOVA) and Tukey’s HSD test were used for statistical analysis.…

• 

  Fig. 5. Vcn DAFS variants do not display altered FA characteristics. a–d Representative images of Vcn−/− MEFs expressing WT VcnVenus or VcnVenus containing DAFS variants. Scale bar: 10 μm. WT VcnVenus is shown in gray and VcnVenus DAFS variants are shown in shades of blue. Quantification of FA morphometric characteristics are shown including e, FA number, f, FA area, g FA axis ratio, and h, standard deviation of FA orientation. Plots shown for WT VcnVenus and VcnVenus DAFS variants (n = 45, 52, 59, 51 cells, respectively, collected over N = 3 independent experiments). Box depicts the median as the center value, the 25th percentile as the lower bound, and the 75th percentile as the upper bound. Whiskers extend 1.5 times the IQR from the bottom and top of box, or to the minimum and maximum of the data if the data does not extend to the whiskers. Values outside the whiskers are plotted as individual points. One-way ANOVA paired with a Steel-Dwass test (e, f, g) or a Tukey’s HSD test (h) was conducted across all VcnVenus, VcnCS, and VcnTS constructs used for statistical analyses. Differing letters denote significant difference at p < 0.05. For more than two comparisons, see Supplementary…

• 

  Fig. 6. DAFS variant VcnTSs experience reduced, but not negligible load. a Schematic of TSMod and VcnTS constructs in loaded and unloaded states. b Schematic of VcnTS in the FA, loaded by actomyosin contractility. c–g FA masked acceptor intensity images are shown for single Vcn−/− MEFs expressing WT VcnTS, DAFS variants of VcnTS and a previously validated variant with reduced actin affinity, VcnTS I997A. h Representative cytoplasm acceptor intensity shown for a single Vcn−/− MEF expressing unloaded FRET control, TSMod. i–m, Corresponding FA masked FRET efficiency shown for (c–g). n Corresponding cytoplasm FRET efficiency shown in h. o Box-whisker plots are shown for cell-averaged FRET efficiency of VcnTS and VcnTS DAFS variants compared to expected unloaded FRET efficiency (dotted line) controls (n = 232, 85, 79, 97, 58, and 69) biologically independent cells respectively, examined from over N = 3 independent experimental days. Box depicts the median as the center value, the 25th percentile as the lower bound, and the 75th percentile as the upper bound. Whiskers extend 1.5 times the IQR from the bottom and top of box, or to the minimum and maximum of the data if the data does not extend…

• 

  Fig. 7. DAFS variant VcnTSs exhibit loss of Vcn force-stabilized dynamics. a, b Representative Vcn−/− VcnTS and Vcn−/− VcnTS E1015A-E1021A FAs undergoing FRAP time lapse. Scale bar: 2 μm c FRAP recovery curves are shown for cells expressing VcnTS and VcnTS E1015A-E1021A (n = 25, 22 from N = 11 independent experimental days). Measure of center for error bars is average of n normalized intensities at respective timepoint for VcnTS or VcnTS E1015A-E1021A expressing cells. Exponential fit curves are fitted to averages at respective timepoints. d Box-whisker plots of mobile fraction of FAs analyzed in c, using two-tailed Welch’s t-test for unequal variances (p = 0.0000054) e Box whisker plots of FRAP recovery half-time of FAs analyzed in c, using one-way ANOVA followed by F-test (p = 0.25). Box depicts the median as the center value, the 25th percentile as the lower bound, and the 75th percentile as the upper bound. Whiskers extend 1.5 times the IQR from the bottom and top of box, or to the minimum and maximum of the data if the data does not extend to the whiskers. Values outside the whiskers are plotted as individual points. f Correlation between FRAP recovery half-time and FRET efficiency…

• 

  Fig. 8. Local FA-SF alignment and directed migration ability differ between WT VcnTS and DAFS double variant VcnTS. a VcnTS or b, DAFS double variant VcnTS E1015A-E1021A stained for F-actin (purple). Scale bar (and zoom-in height and width): 10 μm. Box-whisker plots shown for stably expressed VcnTS or VcnTS E1015A-E1021A in Vcn−/− MEFs (n = 36, 45 respectively, from N = 3 independent experimental days) quantifying c spread of actin SF orientations (p = 0.34, one-way ANOVA followed by F-test) d SF density per FA-defined local region (p = 0.45, two-tailed Student’s t-test paired with post-hoc Tukey Test), and e angle between FA and average angle of actin filaments within respective Voronoi regions of cells (p = 0.0173, one-way ANOVA followed by F-test). f Schematic depicting transwell setup for directed migration assay. g Box-whisker plots are shown of migrated cells per field of view (FOV) in a Boyden chamber haptotaxis migration assay (n = 19,18,16,18 FOVs per given group from N = 3 independent experiments). Box depicts the median as the center value, the 25th percentile as the lower bound, and the 75th percentile as the upper bound. Whiskers extend 1.5 times the IQR from the bottom and…