Force-dependent binding of vinculin to α-catenin regulates cell-cell contact stability and collective cell behavior

Abstract

The shaping of a multicellular body and repair of adult tissues require fine--tuning of cell adhesion, cell mechanics, and intercellular transmission of mechanical load. Adherens junctions (AJs) are the major intercellular junctions by which cells sense and exert mechanical force on each other. However, how AJs adapt to mechanical stress and how this adaptation contributes to cell-cell cohesion and eventually to tissue-scale dynamics and mechanics remains largely unknown. Here, by analyzing the tension-dependent recruitment of vinculin, α-catenin, and F-actin as a function of stiffness, as well as the dynamics of GFP-tagged wild-type and mutated α-catenins, altered for their binding capability to vinculin, we demonstrate that the force-dependent binding of vinculin stabilizes α-catenin and is responsible for AJ adaptation to force. Challenging cadherin complexes mechanical coupling with magnetic tweezers, and cell-cell cohesion during collective cell movements, further highlight that tension-dependent adaptation of AJs regulates cell-cell contact dynamics and coordinated collective cell migration. Altogether, these data demonstrate that the force-dependent α-catenin/vinculin interaction, manipulated here by mutagenesis and mechanical control, is a core regulator of AJ mechanics and long-range cell-cell interactions.

FIGURE 1:

Substrate stiffness-dependent recruitment of α-catenin, vinculin and F-actin at cell–cell contacts. (A) MDCK cells were cultured for 24 h on PAA gels of the indicated stiffness (4.5, 9, or 35 kPa), on which 100-µm-diameter disks of FN had been patterned. Preparations were then fixed and immunostained for α-catenin, α18 epitope, vinculin, and F-actin and then imaged by confocal microscopy (panels show 0.5-µm-thick z-projections taken at the level of the apical complexes). Scale bar: 50 µm. (B) The histograms represent the mean fluorescence intensities measured for αE-catenin, α18 epitope, vinculin, and F-actin stainings as indicated in Materials and Methods (mean ± SEM, n = 640–1260 junctions in total per condition, out of three independent experiments; ****, p < 0.0001; ns, not significant; one-way analysis of variance (ANOVA) test. (C) Western blot analysis of α-catenin and vinculin from protein extracts of cells grown for 24 h on FN-coated PAA gels of 4.5, 9, and 35 kPa rigidity, respectively. α-Tubulin was used as a loading control.

FIGURE 2:

E-cadherin–dependent cell–cell contacts form independently of α-catenin/vinculin interactions. (A) Schematics of GFP-tagged wild-type (α-cat-wt-GFP), L344P (α-cat-L334P-GFP), and Δmod (α-cat-Δmod-GFP) αE-catenin constructs. (B) Confocal analysis of apical vinculin and F-actin distribution in αE-catenin KD MDCK cells expressing α-cat-wt, α-cat-L334P, and α-cat-Δmod grown on glass surfaces. The expression of mutant proteins restored cell–cell contacts, as did the expression of wt α-catenin. Scale bar: 5 µm. (C) FRAP experiments were performed on cell–cell contacts of α-cat-wt (green), α-cat-L334P (blue), or α-cat-Δmod (red)-expressing cells grown on glass substrates. Mean intensity recoveries over time (±SEM) fitted with a one-term exponential equation (n = 50 regions of interest out of three independent experiments for each condition). (D) Mobile fractions extracted from the fits of individual recovery curves (scatter dot plot, mean values ± SD). ****, p < 0.0001; ***, p < 0.001; ns, nonsignificant; one-way ANOVA test.

FIGURE 3:

Binding of α-catenin to vinculin is required for its tension-dependent stabilization at cell–cell contacts. (A, B) MDCK α-catenin–KD cells expressing GFP-tagged α-cat-wt (green), α-cat-L334P (blue), or α-cat-Δmod (red) were cultured for 24 h on 4.5 (light colors) or 35 kPa (dark colors) PPA gels before FRAP experiments were performed. Graphs represent mean GFP fluorescence recovery over time (±SEM, n = 50 out of three independent experiments for each condition) fitted with a one-term exponential equation. (C) Mobile fraction values (scatter dot plot, mean values ± SD) extracted from the fits of individual recovery curves considered in panels A and B. **, p < 0.01; ns, nonsignificant; two-way ANOVA test. Notice the nonsignificant differences in mobile fraction values observed for the mutant proteins on soft and stiff substrates, contrasting with the significant decrease in mobile fraction observed for the wt protein as a function of increasing substrate compliance. (D) Magnetocytometry applied on Ecad-Fc–coated bead doublets bound to the surface of MDCK cells expressing α-cat-wt, α-cat-L334P, and α-cat-Δmod mutants. The histogram reports the mean values of the SD of the bead fluctuation angles.

FIGURE 4:

Vinculin/α-catenin association controls collective cell behavior and cell–cell contact lifetime. (A) MDCK cells silenced for α-catenin (α-cat KD), as well as cells expressing α-cat-L334P, α-cat-Δmod, or wt α-catenin (α-cat-wt), were seeded on 500-µm Ø FN patterns and phase contrast imaged for 24–36 h (still images of Supplemental Videos 1–4). The collective behavior of cell monolayers was analyzed by PIV over 6 h providing heat maps of instantaneous local velocities (B). Mean velocities (C) and correlation lengths (D) characteristic of each cell type were then extracted from these instantaneous velocity maps (mean values ± SD) out of three independent experiments; n = 360 frames analyzed per condition, derived from 10 patterns per condition coming from three independent experiments. ****, p < 0.0001; *, p < 0.1; ns, nonsignificant; one-way ANOVA test. (E) Mean lifetime of individual cell–cell contacts measured for each cell type (scatter dot plot: mean ± SD, n = 30 cell doublets for α-cat KD and α-cat-wt, and n = 51 for α-cat-L334P, α-cat-Δmod, out of ≥4 patterns derived from ≥2 independent experiments for each condition; ****, p < 0.0001; **, p < 0.01; one-way ANOVA test).