α-Catenin homodimers are recruited to phosphoinositide-activated membranes to promote adhesion

Abstract

A unique feature of α-catenin localized outside the cadherin-catenin complex is its capacity to form homodimers, but the subcellular localization and functions of this form of α-catenin remain incompletely understood. We identified a cadherin-free form of α-catenin that is recruited to the leading edge of migrating cells in a phosphatidylinositol 3-kinase-dependent manner. Surface plasmon resonance analysis shows that α-catenin homodimers, but not monomers, selectively bind phosphatidylinositol-3,4,5-trisphosphate-containing lipid vesicles with high affinity, where three basic residues, K488, K493, and R496, contribute to binding. Chemical-induced dimerization of α-catenin containing a synthetic dimerization domain promotes its accumulation within lamellipodia and elaboration of protrusions with extended filopodia, which are attenuated in the α-catenin^KKR<3A mutant. Cells restored with a full-length, natively homodimerizing form of α-catenin^KKR<3A display reduced membrane recruitment, altered epithelial sheet migrations, and weaker cell-cell adhesion compared with WT α-catenin. These findings show that α-catenin homodimers are recruited to phosphoinositide-activated membranes to promote adhesion and migration, suggesting that phosphoinositide binding may be a defining feature of α-catenin function outside the cadherin-catenin complex.

Figure 1.

Forced dimerization of αCat is sufficient for its cortical recruitment. (A) Cadherin-independent recruitment of αCat to the leading edge. Fluorescent images of GFP–αCat localization in scratch-wounded A431D cells. αCat (green) colocalizes with F-actin (red) at wound front. (B) GFP–αCat localization in wounded R2/7 cells. (C) GFP–αCat does not colocalize with E-cadherin (red). Arrows show αCat enrichment at protrusions. Bars, 20 µm (A–C). (D) Schematic of iDimerize system. N terminus (aa 1–267) was replaced with the synthetic dimerization (DmrB) domain (brown), which is dimerized by the small molecule B/B (yellow). (E) BN-PAGE analysis of dimer formation (D) relative to monomer (M); B/B treatment = 3 h. (F) ΔNαCat is recruited to periphery within 5 s of B/B treatment. Bars, 10 µm. See also Video 1.

Figure 2.

Forced dimerization of αCat enhances filopodia on radiating protrusions at nascent contacts. (A) ΔNαCat dimerization by B/B promotes filopodia formation. Filopodia were counted every 1 s during a video of force dimerization (n = 6 FOVs from two BRs; data are mean ± SD). Bars, 10 µm. See also Video 1. (B) Actin ultrastructure of dimerized ΔNαCat (±B/B) by platinum replica electron microscopy. Bars, 500 nm. Arrows show αCat enrichment at protrusions. (C) Epifluorescence microscopy of radial protrusions (RPs; white arrows) induced by homodimerization. Bars, 20 µm. (D) Blinded quantification of RPs (n > 150 cells; FOV counts ratioed to total number of cells to account for variations in cell density; Materials and methods; data are mean ± SD). (E) Structured illumination microscopy (SIM) of RPs with filopodia. Bars, 5 µm. (F) Quantification of filopodia length (n > 13 FOVs; three BRs; data are mean ± SD). Significance in D and F by ANOVA. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. (G) Time-lapse analysis of B/B-treated ΔNαCat cells coinfected with GFP-LifeAct. Arrows indicate prolonged cell–cell contact upon homodimerization, which is reversed by washout ligand. Bars, 10 µm. See also Video 2.

Figure 3.

αCat homodimers prefer PIP₃ phosphoinositides. (A) Phosphoinositide (PtdInsP) selectivity of WT, full-length (FL) αCat dimer. SPR sensorgrams for seven different POPC/POPS/PtdInsP (77:20:3) vesicles are shown (from top to bottom): PIP₃ >> PIP₂ > phosphatidylinositol-3,4-bisphosphate (PI34P₂) ≈ phosphatidylinositol-5-monophosphate (PI5P) > phosphatidylinositol-4-monophosphate (PI4P) > phosphatidylinositol-3,5-bisphosphate (PI35P₂) ≈ phosphatidylinositol-3-monophosphate (PI3P). 200 nM αCat was used for all measurements. Data are representative of three experiments. (B) Binding of αCat homodimer versus monomer (200 nM each) to POPC/POPS/PIP₃ (77:20:3) vesicles. Data are representative of three experiments. (C) Chromatographic separation of monomer (M) and dimer (D) fractions of recombinant αCat proteins. (D) αCat schematic with KKR-basic patch overlaid on crystal structure of αCat dimer. Subunit A in gray; subunit B in green/blue; basic patch in red. (E) Circular dichroism (CD) spectra thermal denaturation analysis of purified recombinant protein; FLαCat (red trace) or FLαCat^KKR<3A (blue trace) proteins. No global differences between WT and mutant protein were detected. Wavelength scan: n = 4; temperature scan: n = 2. (F) SDS-PAGE (top) and BN-PAGE (bottom) analysis of purified proteins (10 µg). (G) Membrane binding isotherms for FL αCat dimer (open symbols) and FL^KKR<3A mutant dimer (closed symbols). n = 3; data are mean ± SD. Lines represent theoretical curves constructed from apparent dissociation constant (K_d) and the maximal R_eq value (R_max) determined from by nonlinear least squares analysis of the binding isotherm using the equation R_eq = R_max/(1 + K_d/P_o). K_d = 30 ± 3 nM for WT and 73 ± 15 nM for the mutant. (H) PIP₃ versus PIP₂ selectivity of WT FLαCat dimer (cyan) and FLαCat^KKR<3A mutant dimer (orange). Data are representative of three experiments. Notice that although the mutation decreased the overall membrane affinity of the αCat dimer, it did not affect the PtdInsP selectivity of the protein, as indicated by essentially the same (RU for PIP₃)/(RU for PIP₂) ratio for WT and the mutant. Each SPR measurement was performed in 50 mM Tris-HCl, pH 8.0, containing 0.1 M NaCl and 1 mM TCEP using L1 chip coated with POPC/POPS/PtdInsP (77:20:3) vesicles as the active surface. 200 nM αCat was used for both WT and mutant proteins. POPC vesicles were used to coat the control surface for most experiments. For A, POPC/POPS (80:20) vesicles were used for the control surface to eliminate the potential contribution of POPS to PtdInsP selectivity.

Figure 4.

Filopodia promoted by force-dimerization are reduced in ΔNαCat^KKR<3A mutant. (A) αCat localization is sensitive to wortmannin and EGF. Schematic of PIP₃ synthesis and impact of drugs to right. n = 3, >14 FOVs. Data indicate mean ± SD. Significance by ANOVA. Arrows show αCat enrichment at protrusions. Bars, 20 µm. See also Video 3. (B) BN-PAGE analysis of dimer formation (D) relative to monomer (M); B/B treatment 3 h. (C) ΔNαCat^KKR<3A dimerization by B/B reduces filopodia formation compared with ΔNαCat. As in Fig. 2, filopodia were counted every 1 s during a video of force dimerization. Bars, 10 µm; n = 6 FOVs from two BR; data are mean ± SD). (D) Epifluorescence microscopy of radial protrusions (RPs; white arrows) reduced in ΔNαCat^KKR<3A mutants. Bars, 20 µm. (E) Blinded quantification of RP (n > 150 cells; FOV counts ratioed to total number of cells to account for variations in cell density; Materials and methods). (F) Length of filopodia decreased in ΔNαCat^KKR<3A, as imaged in structured illumination microscopy (SIM) of RP filopodia. Bars, 5 µm. (G) Quantification of filopodia length (n > 13 FOVs from three BRs), table of results below. Significance in E and G by ANOVA; data are mean ± SD. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. (H) Time-lapse analysis of B/B-treated ΔNαCat^KKR<3A cells coinfected with GFP-LifeAct. Prolonged cell–cell contact upon homodimerization was not observed in mutant construct. Bars, 10 µm. See corresponding Videos 1 and 2.

Figure 5.

Exogenous phosphoinositides are sufficient to recruit endogenous αCat. (A) Schematic of BODIPY-labeled phosphoinositide/histone H1 complex formation and integration into the apical membrane of polarized MDCK cells grown on filters for 2 wk. (B) Slice views of PI35P₂, PIP₂, or PIP₃ integrated via histone H1 complex into MDCK cells. Images were captured via confocal microscope using identical exposure time and laser intensity. Arrows indicate colocalization of αCat and phosphoinositide. Bar, 10 µm. (C) Colocalization was analyzed for each slice (0.2-µm steps) via Pearson’s correlation coefficient (chosen over Mander’s because it is independent of signal levels and background) in Nikon Elements (n = 3). (Left) Plotted value for each slice from apical (A) to basal (B) side. (Right) Plotted values represent all values calculated within a stack. Data are mean ± SD. Significance by one-way ANOVA with multiple comparisons: ****, P < 0.0001.

Figure 6.

αCat^KKR<3A mutant shows reduced recruitment to exogenous PIP₃. (A) PIP₃ apical membrane integration and recruitment assay; mCherry αCat (red); BODIPY–PIP₃ (green) with quantification of total colocalized surface. Arrows show colocalized surfaces as yellow (top) and purple (bottom) when reconstructed as a volume rendering. (B) BODIPY–PIP₃ did not alter expression of αCat (quantification not shown; n = 3). (C) Quantification of colocalization volumes between αCat constructs and PIP₃ (n = 45 from three BRs). Significance by unpaired t test: ***, P < 0.0005; ****, P < 0.0001. Error bars reflect the SD of the mean. (D) Surface renderings of colocalization volumes (purple) relative to BODIPY–PIP₃ integration (green). Slice shows en face view of Z-stack; arrows show colocalization (0.2-µm steps). (E) E-cadherin was not recruited to ectopic BODIPY–PIP₃ integrations, suggesting that this is a unique feature of extrajunctional αCat. Samples were fixed after live-cell imaging, so integration of BODIPY–PIP₃ was prolonged relative to Z-stacks in A and D. Bars, 20 µm.

Figure 7.

αCat^KKR<3A mutant alters cell–cell adhesion and sheet migration. (A) Live imaging of αCat-expressing R2/7 cells imaged for 12 h postwounding. Arrows indicate αCat recruitment in FLαCat but not FLαCat^KKR<3A. (B) Quantification of enrichments (n > 150 FOVs inclusive of all time points, two BRs). Data are mean ± SD. (C) Representative image of wound closure area. Pink overlay depicts area quantified. (D) Quantification of wound closure (n = 24 FOVs, two BRs, data are mean ± SD). See corresponding Video 4. (E) Phalloidin stain of F-actin at wound front. Red arrows indicate actin cables parallel to the front; black arrows indicate actin cables adjoining cells via junctions. High-resolution analysis of F-actin across adjacent cells by SIM (right). (F) Live imaging of GFP-LifeAct and mCherry FLαCat during wound migration. (Left) Fluorescent image; white arrows indicate protrusions with long filopodia; yellow arrowheads show areas of colocalization. (Right) Bright-field image. Bars, 50 µm. (G) Representative images from a mechanical disruption assay with quantification in H (Materials and methods; n = 5 BRs). Data are mean ± SD; significance in all panels by unpaired t test: ****, P < 0.0001.