Fig 2: PP1a interacts with the HR2 domain of Cavin3 and Cavin1. a Schematic representation of the helical regions (HR)1 and HR2 regions of Cavin1 and Cavin3. b AlphaLISA screen of Cavin3 HR1 domain, Cavin3 HR2 domain, and Cavin3 HR1+HR2 domain and PP1a, FKBP, and PP1a (negative control), **p-value < 0.01, ****p-value < 0.0001 (one-way ANOVA). NS = no significance. Scatter dots in each group represent the values from three independent experiments (labeled by three colors). c AlphaLISA screen of Cavin1 HR1 domain and PP1a, Cavin1 HR2 domain and PP1a, Cavin1 HR1+HR2 domain and PP1a, Cavin1 FL and PP1a, and FKBP and PP1a (negative control), *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, ****p-value < 0.0001 (one-way ANOVA). Mean ± SD are presented for three sets of experiments. d–e GFP-tagged full-length Cavin3 (lane 2) and truncation mutants (HR1+HR2) domain, HR1 domain or HR2 domain were expressed in MCF-7 cells followed by lyzed GFP Trap pulldown assays. Transfection and pulldown efficiency were examined by western blot analysis using anti-GFP antibodies (d). Endogenous PP1a (e) was detected by western blotting using PP1a specific antibodies. Results are representative of three independent experiments

Fig 3: Hypo-osmotic interactions of endogenous PP1a and cavin proteins. a, b Representative immunofluorescence images of endogenous Cavin1 (green) and PP1a (red) (a) or Cavin3 (green) and PP1a (red) (b) in untreated or hypo-osmotic treated A431 cells. Yellow or white circles (in merged images) indicate the outline of the nucleus. Scale bar, 10 µm. Images are representative of three independent experiments. c Representative images of PLA signals (red)/cell representing the interactions of endogenous PP1a/Cavin1 or PP1a/Cavin3 in A431 cells from three independent experiments. Cavin1/Cavin3 interaction was used as a positive control. DAPI was used to visualize nuclear DNA (blue). Scale bar, 10 µm. d, e PLA signal/cell of Cavin1/PP1a and Cavin3/PP1a interactions in A431 cells with or without hypo-osmotic treatment. The PLA signals between cells were distinguished by the matched DIC microscopy and are represented as the mean ± SD. Scale bar, 10 µm. f, h Quantitation of PLA signals for the Cavin1/PP1a (f) and Cavin3/PP1a (h) interactions upon hypotonic treatment, n = 150 cells for three independent experiments. ****p-value (Student’s t-test, two tail) < 0.0001. g, i The number and ratio (%) of nuclear (Nuc.) and cytoplasmic (Cyto.) localized PLA signal for Cavin1/PP1a (g) and Cavin3/PP1a (i) interactions was quantified using CellProfiler, where DAPI staining was utilized for nuclear identification. Histogram represents the mean ± SD values of PLA signal number in each cell from three independent experiments (n = 50 cells for each experiment). Cytoplasmic and nuclear signals were labeled as blue and gray. The values of the ratio are presented in in the table as mean ± SD. j MCF-7 cells transfected with GFP-cavins and GFP-vector (upper two panels), or with additional mCherry-CAV1 and Flag-Cavin1 (not expressed in Cavin1-GFP cells) (lower two panels). PLA signals for the interaction between GFP and PP1a were detected and visualized as magenta puncta in merged images. PLA signal images alone were inverted to gray scale. Enlarged images showing the colocalization between mCherry-CAV1 and GFP-tagged Cavin1 and Cavin3 are presented. Scale bar, 10 µm. Images are representative of three independent experiments

Fig 4: ALPHAScreen analysis of interacting proteins. a–c Cavin1 (a), Cavin2 (b), and Cavin3 (c) following co-expression of 47 potential interacting proteins in the eukaryotic Leishmania tarentolae-based cell-free-expression system. Each protein pair was tested in triplicate and values above 2000 counts were considered positive interactions. Red = positive interactions, blue = no interactions. *PPP1CA serine/threonine-protein phosphatase PP1alpha catalytic subunit. d Interaction heatmap of the mean values of 47 tested potential interacting proteins with each of Cavin1, Cavin2, and Cavin3 from three independent experiments

Fig 5: Cavin3 regulates BRCA1 protein expression and localization.(A) Representative image of MCF7 cells stably expressing GFP alone, cavin1-GFP, and cavin3-GFP fixed and stained with a BRCA1 antibody. (B) Percentage of MCF7 cells showing strictly nuclear, nuclear-cytoplasmic, or cytoplasmic localization of BRCA1 was counted for 50 cells from 4 to 5 independent experiments as mean ± SD using a one-way ANOVA and Bonferroni’s multiple comparisons test. Each biological replicate was color-coded. NS: not significant, *p<0.05, **p<0.01. (C) Lysates from stably expressing MCF7 cells western blotted for GFP, BRCA1, and Tubulin as a load control. (D) MCF7-GFP and MCF7-cavin3-GFP cells, untreated (-) or treated with MG-132 for 6 hr. Lysates were western blotted with GFP, BRCA1, and Tubulin antibodies as a loading control. (E) A431 cells treated with control siRNAs (Con) or two siRNAs specific to cavin3. Lysates were western blotted using cavin3, BRCA1, CAV1 antibodies, and Tubulin as the loading control. (F) A431 cells treated with control siRNAs or two siRNAs specific to BRCA1. Lysates were western blotted using cavin3, BRCA1, CAV1 antibodies, and Tubulin as the loading control. (G) A431 cells treated with control (Con) or siRNAs specific to cavin3, untreated or treated with MG132 for 6 hr. Lysates were western blotted using cavin3, BRCA1, and Tubulin as a loading control. Quantitation of all blots in Figure 4 is provided in Figure 4—figure supplement 1A–E. Figure 4—source data 1.Raw western data for MCF7 cells with molecular weight markers for Figure 4C.(A) Western blot analysis of anti-rabbit BRCA1, (B) anti-mouse Tubulin, and (C) anti-mouse GFP antibodies in (1) GFP lysates, (2) cavin1-GFP lysates, and (3) cavin3-GFP lysates. Figure 4—source data 2.Raw western data for MCF7 cells with molecular weight markers for Figure 4D.(A) Western blot analysis of anti-mouse GFP, (B) anti-rabbit BRCA1, and (C) anti-mouse Tubulin antibodies in (1) MCF7/GFP untreated, (2) MCF7/GFP + MG132-treated lysates, and (3) MCF7/cavin3-GFP untreated lysates. Figure 4—source data 3.Raw western data for A431 cells with molecular weight markers for Figure 4E.(A) Western blot analysis of anti-rabbit cavin3, (B) anti-rabbit CAV1, (C) anti-mouse Tubulin, and (D) anti-rabbit BRCA1 antibodies in (1) A431 cells treated with control siRNA oligos, (2) A431 cells treated with cavin3-specific siRNA oligo 1, and (3) A431 cells treated with cavin3-specific siRNA oligo 2. Figure 4—source data 4.Raw western data for A431 cells with molecular weight markers for Figure 4F.(A) Western blot analysis of anti-rabbit BRCA1, (B) anti-mouse Tubulin, (C) anti-rabbit cavin3, and (D) anti-rabbit CAV1 antibodies in A431 cells treated with (1) control siRNA oligos, (2) A431 cells treated with BRCA1-specific siRNA oligo 1, and (3) A431 cells treated with BRCA1-specific siRNA oligo 2. Figure 4—source data 5.Raw western data for A431 cells with molecular weight markers for Figure 4G.(A) Western blot analysis of anti-rabbit BRCA1, (B) anti-rabbit cavin3, and (C) anti-mouse Tubulin antibodies in (1) A431 cells treated with control siRNA (control KD) oligos no treatment, (2) A431 cells treated with control siRNA oligos (control KD) and MG132 for 6 hr, (3) cavin3-specific siRNA (cavin3 KD) oligo 1 no treatment, (4) cavin3-specific siRNA (cavin3 KD) oligo 1 and MG132 for 6 hr, (5) cavin3-specific siRNA (cavin3 KD) oligo 2 no treatment, and (6) cavin3-specific siRNA (cavin3 KD) oligo 2 and MG132 for 6 hr.