PMC

Active Vinculin Holds Integrin in an Active Conformation
(A and B) Expression of vinT12 or vin880, but not vinFL, in MEFvin^−/− cells increases the maximal detachment force and working energy when compared to control cells. (A schematic of the experimental setup is in A.)
(C) U2-OS cells were seeded on poly-D-lysine to hold them in a nonspread and tensionless shape. Note the increase in antibody staining for activated β1-integrins (12G10) upon expression of vin258, vin880, and vinT12 compared to control cells expressing vinFL (). Scale bar represents 5 μm.
(D and E) Quantification of the intensity of 12G10 labeling under control conditions (D) and after addition of Y-27632 (E).
(F) Quantification of total β1-integrin labeling under indicated conditions, using an antibody that binds to all conformations of β1-integrins (K20). All quantifications are expressed as ratios over the vinFL sample in the same condition (p < 0.01; ±SEM, n ≥ 15 cells from two independent experiments per condition).

Talin Binding of Active Vinculin Is Essential for Its Ability to Bypass the Requirement of Actomyosin Tension for FA Stabilization
(A) Graphical representation of vinculin constructs that were expressed as GFP-fusion constructs in this study. Proteins binding to the individual domains of vinculin are indicated below.
(B) MEFvin^−/− cells coexpressing LifeAct-mRFP together with indicated vinculin constructs were treated with Y-27632. Full-length vinculin, but not vin258, vin880, and vinT12 (), leaves FAs under these conditions. Vinculin forms carrying an A50I mutation that reduces their binding to talin (vinFL-A50I and vinT12-A50I) are released from FAs in the absence of intracellular tension. Similar results were obtained by treating the cells with cytochalasin D and blebbistatin ( and ). Scale bar represents 5 μm.
(C) Quantification of the normalized number and area fraction of vinculin-positive FAs over time. The vertical red dashed line indicates the beginning of the Y-27632 treatment (±SEM, n > 10 cells from two independent experiments).

Vinculin Regulates the Recruitment and the Release of Core Proteins of the FA Network
(A) Schematic of vinculin and its reported direct binding partners for the head (blue), neck (green), and tail (orange) regions (F1 generation). Indirectly associated FA core proteins that may bind through the F1 generation proteins are displayed as the F2 generation in red.
(B) Images from live recordings of U2-OS cells expressing vin880-CFP, talin-YFP, and LifeAct-mRFP (left panel) or vin880-CFP, α-actinin-YFP, and LifeAct-mRFP (right panel). Note that vin880 stabilizes talin, but not α-actinin, in FAs when intracellular tension is released during cytochalasin D treatment (). Scale bar represents 5 μm.
(C) Pearson’s correlation analysis after actin disruption using Y-27632 in MEFvin^−/− cells was used to quantify the extent of colocalization of vinculin constructs with indicated FA proteins. Note the high correlations of all FA stabilizing vinculin forms with talin, paxillin, zyxin, p130Cas, ILK, parvin, FAK, and tensin and the low correlation with α-actinin and VASP. High correlations of α-vinexin, β-vinexin, and ponsin are dependent on the presence of the proline-rich neck region in vinculin.

The Actin-Binding Site Present in Vinculin Tail Is Important for Stretch-Induced Reorganization of Cell Polarity
(A) Still images from time-lapse recordings of MEFvin^−/− cells undergoing cyclic stretching, expressing indicated vinculin constructs (see for higher resolution). The FAs are color coded according to the angle between their main axis and the stretching axis: red for FAs in an angle between 0° and 30° to the stretching axis, green between 30° and 60°, and blue between 60° and 90°. Note the reorientation of FAs in cells expressing vinFL and vinT12 and the static behavior of FAs expressing vin258. FAs of cells expressing vinT12 reorganized faster than in vinFL-expressing cells, having reached their final position within 30 min of stretching (). Scale bar represents 5 μm.
(B) Heat maps display the change in reorientation of the FAs (Δangle) relative to their initial angle (y axis) and during the time course of the stretching (x axis). The amplitude of the reorganization is color coded (see rainbow color bar). Most of the FAs with an initial angle between 0° and 60° to the stretching axis readily reoriented during the time of cyclic stretching. In contrast, FAs of cells without vinculin (MEFvin^−/− cells expressing paxillin-GFP as adhesion marker) or cells expressing vinculin constructs lacking the actin-binding site (vin258, vin880) were impaired in FA reorientation. The faster reorganization of FAs in cells expressing vinT12 highlights the importance of the vinculin-actin link in the transmission of stretch-induced forces.

Expression of FA-Stabilizing Vinculin Impairs Polarized Cell Migration
(A) Cell velocities of B16F10 cells expressing indicated constructs when plated on laminin.
(B) The plotted trajectories of 24 hr time-lapse experiments outline the decreased migration rate of cells expressing vin880 and vinT12 when compared to vinFL-expressing cells (experiment representative of three replicates; n = 69, 44, and 70 for vinFL, vin880, and vinT12 respectively; ±SEM; p < 0.01).
(C) High-magnification recordings of cells expressing indicated constructs were taken to outline potential differences in B16F10 polarity during migration. The color-coded shape outlines indicate representative protrusive activities recorded over a period of 2 hr (5 min intervals). Note the highly directional protrusion of vinFL-expressing cells leading to coordinated cell migration and the random unpolarized cell edge protrusions in cells expressing vin880 and T12 () (example representative of n > 10 cells from three independent experiments). Scale bar represents 5 μm.
(D) FRET-based Rac activation sensors (Raichu-Rac) coexpressed with mCherry-tagged vinculin constructs outline localized Rac1 activity in B16F10 plated on laminin. In cells coexpressing vinFL, Rac1 activity is visualized by a color shift toward red and usually found at a single protrusive edge at the front of a migrating cell. In cells expressing active vinculin constructs (vin258, vin880, and vinT12), all the numerous peripheral protrusions showed sites of intense Rac activity (examples representative of n > 10 cells from two independent experiments). Scale bar represents 5 μm.

Intracellular Tension Is Required to Maintain Full-Length Vinculin in FAs
(A) Expression of vin258 in U2-OS cells in the presence of Y-27632 prevents the release of endogenous talin, paxillin, and focal adhesion kinase (FAK), but not vinculin, in FAs. Note the overlap of the fluorescence intensity peaks for vin258 colocalization with talin, paxillin, and FAK and the absence of overlap of the profiles for GFP-vin258 with endogenous vinculin. Scale bar represents 5 μm.
(B) Panels of time-lapse recordings of a U2-OS cell expressing vinFL-mCherry and vinT12-GFP during Y-27632 treatment (refer to for the full time course). Intensity profiles of vinT12 (green) and vinFL (red) along the yellow line at indicated time points are displayed below the images. During the drug treatment, the profile corresponding to vinT12 remains stable, and the matching peaks in the profile of vinFL decreases at the same time as the overall background increases. Intensities were corrected for photobleaching using a histogram-matching method and normalized using the total fluorescence in each channel. Scale bar represents 5 μm.
(C and D) FRAP experiments were performed in MEFvin^−/− cells expressing either vinFL-GFP or vin880-GFP and treated with Y-27632. Note the slower fluorescence recovery in FAs of cells expressing vinFL when treated with Y-27632.
(E) Quantification of the mobile fraction of indicated GFP-fusion proteins in FAs. Note the significant decrease in mobile fraction of vinFL-GFP after treatment with Y-27632. No statistically significant difference was observed in the mobile fraction of vin880 upon drug treatment (±SEM; p < 0.01).