Fig 2: Knock-down of Willin/FRMD6 alters vinculin expression, focal adhesion assembly, and actin cytoskeleton dynamics. (A) Representative phase-contrast images (upper row), immunostainings against F-actin (red), vinculin (green) and DNA (blue; middle row) and TAZ (lower row) of control (shScr) and Willin/FRMD6 knock-down (shWillin) SH-SY5Y cells. Scale bars: 30 μm. Comparison of (B) the number of focal adhesions per cell, (C) the mean area of a single focal adhesion, and (E) the nuclear/cytosolic TAZ ratio in shScr and shWillin cells. Each data point in (B), (C) and (E) represents the measurement value for one cell taken from two independent experiments. Lines indicate means, error bars SEM. (D) Quantitative Western blot analysis of vinculin expression in shScr and shWillin cells. Means were calculated from three independent experiments. Error bars represent SEM. GAPDH serves as a loading control of the lysates. Correlation between the ratio of nuclear to cytoplasmic TAZ concentration, (F) the number of focal adhesions, and (G) the volume by which cells indent into the ERISM substrate for shScr (green dots) and shWillin (yellow dots) SH-SY5Y cells. Groups were compared using Student’s t-test; **p ≤ 0.01; ***p ≤ 0.001.

Fig 3: Knockdown of Willin/FRMD6 in SH-SY5Y cells activates the ERK1/2 pathway and elicits differentiation into a neuronal phenotype by activation of the transcriptional factor NeuroD1. Quantitative Western blot analysis of the levels of pERK1/2 relative to ERK1/2 (A) in control (shScr) and Willin/FRMD6 knock-down (shWillin) SH-SY5Y cells, (B) in control (Vector) and Willin/FRMD6 overexpressing (Willin) SH-SY5Y cells, (C) in SH-SY5Y wild type cells treated with either DMSO or retinoic acid (RA) and (D) in SH-SY5Y wild type cells treated with either DMSO or U0126. Means in (A) to (D) were calculated from three technical repeats. Error bars represent SEM. Comparison of (E) percentage of fine extension-bearing cells and (F) volume by which cells deform the ERISM substrate (each data point represents the value measured for one cell) for shScr cells treated with DMSO and shWillin cells treated with either DMSO or U0126 for 24 h. In (E) means were calculated from two independent experiments conducted in triplicate. In (F) each data point represents the value measured for one cell. Lines indicate means, error bars SEM. (G) Percentage of differentiating shScr and shWillin cells quantified from bright-field images of cells treated with 10 μM RA along with either DMSO or 10 μM U0126 for 4 days. Means and SEM (error bars) were calculated from two independent experiments, each of which was conducted in triplicates. Quantification of the levels of NeuroD1 in shScr and shWillin cells from (H) Western blot analysis and (I) qPCR analysis. Means in (H) and (I) were calculated from two independent experiments conducted in triplicate. Error bars represent SEM. (J) Immunofluorescent imaging of shScr (upper row) and shWillin (lower row) cells for NeuroD1 (left column) and DNA (right column). Scale bars: 40 μm. (K) Quantification of extension bearing cells from bright-field images for shScr cells treated with 10 μM RA for 1, 4, and 7 days. Means and SEM (error bars) were calculated from two independent experiments, each of which was conducted in triplicates. (L) Development of the volume by which cells indent into the ERISM substrate over 1 week during RA-initiated (10 μM) neuronal differentiation of shScr cells. Data points represent means measured for ni individual cells; shaded areas represent SEM (Day 1: nshScr = 16, nshScrRA = 31, nshWillin = 22; Day 3: nshScr = 21, nshScrRA = 17, nshWillin = 14; Day 6: nshScr = 12, nshScrRA = 16, nshWillin = 19; Day 8: nshScr = 18, nshScrRA = 12, nshWillin = 11). Groups were compared using Student’s t-test; ns: p > 0.05; *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001.

Fig 4: CRB3 is an upstream regulator of the Hippo pathway and affects FRMD6 transcription. (a) Protein levels of Hippo pathway components were examined in cells grown under sparse and confluent conditions. The numbers below the images indicate the relative expression levels of p-Lats1/Lats1 and p-YAP/YAP. (b) YAP localization was assessed in cells grown under sparse and confluent conditions. Scale bar, 25 μm. (c) Protein expression of the core regulators of the Hippo pathway was examined by immunoblotting. The numbers below the images represent relative expression levels of p-Lats1/Lats1 and p-YAP/YAP. (d-f) The expression levels of the FRMD6 and Kibra proteins and mRNAs were analysed by immunoblotting (d) and real-time PCR (e and f). ***P<0.001. Data are mean±S.D. (g) Cellular fractionation experiments were performed to examine the levels of the CRB3 protein in the nucleus and cytoplasm. GAPDH and Lamin A were used as loading controls. The numbers below the images indicate the relative expression level of YAP. (h) YAP localization was assessed in MCF10A cells grown under sparse and confluent conditions. Scale bar, 25 μm. (i, j) The expression levels of the CTGF and CYR61 mRNAs in MCF10A and T47D cells were examined using real-time PCR. shCRB3 cells were transfected with siRNAs to silence the expression of YAP. Cells were collected 48 h post-transfection. **P<0.01, ***P<0.001. Data are mean±S.D.

Fig 5: CRB3 stabilizes the Kibra protein by inhibiting its degradation, and Kibra rescues the changes in cell proliferation induced by CRB3 deregulation. (a) MCF10A cells were treated with cycloheximide (100 μg/ml) and MG132 (5 μM) for the indicated times. Cell lysates were examined for the levels of the FRMD6 and Kibra proteins by immunoblotting. (b) T47D cells were treated with cycloheximide (100 μg/ml) and MG132 (5 μM) for the indicated times. The levels of the FRMD6 and Kibra proteins were determined by immunoblotting. (c, d) MCF10A cell lysates were subjected to immunoprecipitation experiments with an anti-CRB3 or anti-Kibra antibody. Immunoblot analyses were conducted for the indicated proteins. (e, f) Immunoprecipitation experiments were performed in T47D cells. (g, h) The protein levels were detected by immunoblotting. (i, j) DNA synthesis was examined in MCF10A cells using BrdU incorporation. ***P<0.001. Data are mean±S.D. (k and l) Aberrant formation of acini was detected in 3D cultures of MCF10A cells. Scale bar, 50 μm.***P<0.001. Data are mean±S.D. (m, n) BrdU incorporation was assessed to examine DNA synthesis in T47D cells. *P<0.05, ***P<0.001. Data are mean±S.D.