Fig 2: a-Catenins are autophagy substrates.a Scatterplot of all proteins uniquely identified in both SILAC experiments (L light, M medium, H heavy amino acids). Fold change (x-axis and y-axis for experiment A and B, respectively) are shown as log2. Scatterplot of the proteins upregulated = 2 fold under bafilomycin A1 (BafA1) treatment is shown on the right. b Representative a-catenins (CTNNA3 and CTNNA1) immunoblotting in double ATG7/10 knockdown MCF10A cells. The experiment was repeated at least twice with similar results. c CTNNA3/GAPDH (left panel) and CTNNA1/GAPDH (right panel) densitometry in MCF10A cells exposed to either control or ATG7/ATG10 siRNAs. The graphs show the mean ± s.d. (CTNNA3/ GAPDH, n = 3 independent experiments) and mean ± s.e.m. (CTNNA1/ GAPDH, n = 5 independent experiments; ***P < 0.001, *P < 0.05; two-tailed one sample t-test). d Representative immunoblots of Ctnna3 levels in wild-type and Atg16L1 hypomorph (Atg16 hyp) primary cortical neurons. The graph shows the Ctnna3/Gapdh levels: mean ± s.d. (n = 3 independent experiments; *P < 0.5; two-tailed one sample t-test). e Representative a-catenins (CTNNA3 and CTNNA1) immunoblots in MCF10A cells under starvation with EBSS (6 h). The experiment was repeated at least twice with similar results. f CTNNA3/GAPDH (left panel) and CTNNA1/GAPDH (right panel) densitometry in MCF10A cells starved in EBSS for 6 h. The graphs show the mean ± s.d. (n = 3 independent experiments; **P < 0.01, *P < 0.05; two-tailed one sample t-test). g Representative a-catenins (CTNNA3 and CTNNA1) immunoblots in MCF10A cells treated with the autophagy inducer Tat-Beclin1 peptide (20 µM for 48 h). Tat-scrambled peptide was used as control. The experiment was repeated at least twice with similar results. h Representative immunoblot of mEm-CTNNA1 in MCF10A cells exposed to BafA1 (200 nM, 16 h). The MCF10A cells were transiently transfected with mEm-CTNNA1. The graph shows the mEm-CTNNA1/GAPDH levels: mean ± s.d. (n = 6 independent experiments; *P < 0.05; two-tailed one sample t-test). i Representative immunoblot of mEm-CTNNA1 in MCF10A cells exposed to autophagy inducers: SMER28 (20 µM) and EBSS for 6 h. The graph shows the mEm-CTNNA1/GAPDH levels: mean ± s.d. (n = 3 independent experiments; **P < 0.01, *P < 0.05; two-tailed one sample t-test). Exact P values for asterisks: c (from left to right) 0.0280, 0.0009; d 0.0461; f (from left to right) 0.0069, 0.0181; h 0.0301; and i (from left to right) 0.0340, 0.0036.

Fig 3: Autophagy regulates YAP/TAZ activity via a-catenins.a Representative confocal images of endogenous YAP/TAZ and F-actin immunostaining in MCF10A cells. MCF10A cells were initially exposed to ATG7/10 siRNAs for 48 h, and only after, followed by a-catenin depletion using CTNNA1/CTNNA3 siRNAs (together with ATG7/10 siRNAs) for other 48 h. Scale bars are 10 µm. The experiment was repeated with similar results. b YAP/TAZ localisation in MCF10A cells treated as in (a). CTNNA1/3 KD partially rescues the YAP/TAZ localisation phenotype in ATG7/10 KD cells. The experiment was repeated with similar results (n (from left to right) = 280/304/299/333 cells; **P < 0.01, ****P < 0.0001; chi-square test). c Quantification of percentages of MCF10A cells with F-actin stress fibres. MCF10A cells were initially exposed to ATG7/10 siRNAs for 48 h, followed by a-catenin depletion using CTNNA1/CTNNA3 siRNAs for other 48 h. The experiment was repeated with similar results (n (from left to right) = 202/ 200/ 206/ 233 cells; *P < 0.05, ****P < 0.0001; chi-square test). d Size of MCF10A cells initially exposed to ATG7/10 siRNAs for 48 h, followed by a-catenin depletion using CTNNA1/CTNNA3 siRNAs for other 48 h. Confocal images of 25 cells per condition were analysed for measuring cell area using ZEN software. Bars represent the mean ± s.e.m (****P < 0.0001, **P < 0.01; two-tailed t-test). e Mathematical modelling of YAP/TAZ activity when reducing autophagy levels. The graph shows the dependency of YAP/TAZ activity on the initial levels of a-catenins and the strength of the feedback loop involving YAP/TAZ control of autophagy vY21. f Representative immunoblot of a-catenin in various cell lines. Bars represent the mean ± s.d. (n = 3 independent experiments; ***P < 0.001, **P < 0.01, *P < 0.05; two-tailed one sample t-test). g Model for the autophagy-dependent YAP/TAZ activity in two initial conditions: low and high a-catenin levels. Autophagy inhibition in cells with high initial a-catenin levels promotes a-catenin accumulation which sequester YAP/TAZ into the cytosol (Hippo signalling is ON). However, autophagy inhibition in cells with low initial a-catenin levels causes only a small increase in these proteins, increase unable to sequester YAP/TAZ into the cytosol (as their levels also increase, being also autophagy substrates)—Hippo signalling is OFF. The feedback loop of YAP/TAZ controlling autophagy accentuates these effects. Exact P values for asterisks: b 0.0023; c 0.0105; d 0.0082; f (from left to right) <0.0001, 0.0027, <0.0001, 0.0027, 0.0276.

Fig 4: a-Catenins interact with LC3 via two LIR regions.a Representative images of endogenous CTNNA3–LC3 colocalization in MCF10A cells treated with BafA1. The control cells were treated with DMSO. Scale bars are 10 µm. The graphs on the right show the image intensities for all channels (CTNNA3, LC3 and DAPI). Scale bars are 10 µm. The experiment was repeated twice with similar results. b Quantification of colocalization for the experiment in (a). The Pearson’s and Mander’s (fraction of CTNNA3 overlapping LC3) coefficients were quantified for n = 45 cells (DMSO) and n = 38 cells (BafA1). Bars represent the mean ± s.e.m. (****P < 0.0001; two-tailed t-test). c Representative confocal images of endogenous CTNNA3–LC3 colocalisation in MCF10A cells exposed to either Tat-scrambled or Tat-beclin1 peptide (20 µM, 48 h). Scale bars are 10 µm. The graph on the right shows the fluorescent intensity for each channel (red-CTNNA3, green-LC3 and blue-DAPI) along the indicated distances. d Co-immunoprecipitation of endogenous LC3 with CTNNA1 in BafA1-treated MCF10A cells. The experiment was repeated three times with similar results. e Co-immunoprecipitation of endogenous LC3 with wild-type and Flag-CTNNA1 mutants in BafA1-treated MCF10A cells. MCF10A cells were initially transfected with empty Flag (control vector), wild-type Flag-CTNNA1 or the indicated Flag-CTNNA1 mutants. The Flag-tagged proteins were pulled down using the Flag-Trap technology. The experiment was repeated three times with similar results. f Co-immunoprecipitation of endogenous LC3 with wild-type and mEm-CTNNA1 mutants. MCF10A cells were transfected with empty mEm (control vector), wild-type mEm-CTNNA1 or the indicated mEm-CTNNA1 mutants. The mEm-tagged proteins were pulled down using the GFP-Trap technology. The experiment was repeated three times with similar results. g 3D structure of the LIR region (417 KEYAQV 422) and its conservation among various species of a-catenins. h Representative images of mEm-CTNNA1—endogenous LC3 colocalization in MCF10A cells transiently transfected with either wild-type or Y419A-V422A mEm-CTNNA1. Scale bars are 10 µm. i The Pearson’s (PC), Overlap (OC) and Mander’s (M1—fraction of LC3 overlapping mEm-CTNNA1, and M2—fraction of mEm-CTNNA1 overlapping LC3) coefficients for MCF10A cells treated as in (h). Thirty-one wild-type cells and 21 Y419A-V422A mutant cells were quantified over three independent experiments. Bars represent the mean ± s.e.m. (****P < 0.0001, ***P < 0.001; two-tailed t-test). Exact P values for asterisks: i 0.0009.

Fig 5: LIR-defective CTNNA1 mutants are not autophagy substrates.a Representative immunoblots of MCF10A cells transiently expressing wild type, L897A-F900A, Y419A-V422A or Y419A-V422A-L897A-F900A mEm-CTNNA1 constructs and exposed to BafA1 treatment (400 nM, 6 h). GFP/GAPDH densitometry is shown on the right: mean ± s.d. (n = 5 independent experiments; *P < 0.05, ns not significant; two-tailed one sample t-test). b Representative immunoblot of MCF10A cells expressing Flag-wild-type and the indicated CTNNA1 mutants upon BafA1 treatment (400 nM, 6 h). c Representative immunoblots of MCF10A cells transiently expressing wild-type or Y419A-V422A mEm-CTNNNA1 and exposed to autophagy inducers: SMER28 (20 µM) and EBSS for 6 h. The graphs show the GFP/GAPDH densitometry of one representative experiment: mean ± s.d. (n = 3 independent experiments; **P < 0.01, *P < 0.05, ns not significant; two-tailed one sample t-test). The experiment was repeated with similar results. d Representative immunoblot of MCF10A cells expressing Flag-wild-type and the indicated CTNNA1 mutants upon exposure to different concentrations of SMER28 (20 and 50 µM) for 24 h. e Representative immunoblot of MCF10A cells expressing Flag-wild-type and the indicated CTNNA1 mutants upon exposure to EBSS for 4 h. Bars represent the mean ± s.e.m. (n = 4 independent experiments; *P < 0.05, ns not significant; two-tailed one sample t-test). Exact P values for asterisks: a (from left to right) 0.0317, 0.1743, 0.4750, 0.3386; c (from left to right) 0.0496, 0.0096, 0.5166, 0.1364; e (from left to right) 0.0336, 0.9923, 0.3824, 0.4799.