Fig 2: Low pH triggers MakA-induced membrane damage and non-canonical autophagy—Supplemental immunofluorescence data. (A) Representative confocal images of MEF cells treated with 250 nM Alexa568-MakA for 16 h and co-stained for endogenous IST1, CHMP2A, or ALIX, as indicated. Nuclei were counterstained with DAPI. Scale bars, 10 µm. Alexa568-MakA refers to Alexa568 labeled monomeric MakA. (B) Representative confocal images of HeLa cells treated with 125 nM Alexa568-MakA for 16 h and co-stained for endogenous ALIX and Gal3 or Gal8, as indicated. Nuclei were counterstained with DAPI. Scale bars, 10 µm. Alexa568-MakA refers to Alexa568 labeled monomeric MakA. (C) Representative confocal images of HeLa cells transfected with mCherry or mCherry-SopF and treated with 250 nM FITC-MakA for 16 h. Cells were co-stained for endogenous LC3 (left) or ALIX (right). Scale bars, 10 µm. FITC-MakA refers to FITC labeled monomeric MakA.

Fig 3: TECPR1 recruitment to damaged lysosomes precedes galectin recruitment A, BRepresentative live‐cell fluorescent images of HeLa cells co‐transfected with EGFP‐TECPR1 and mCherry‐Gal3 and treated with 1 mM LLOMe for the indicated time. Arrowheads in (B) indicate the appearance of TECPR1 (green arrow) and Gal3 (magenta arrow). Scale bars = 10 μM for (A) and 1 μM for (B).CQuantification of TECPR1 and Gal3 recruitment to damaged lysosomes. Data are presented as mean ± SD from five independent experiments (each experiment represents a single cell with at least three individual lysosomes quantified).DRepresentative confocal image of a HeLa cell transfected with EGFP‐TECPR1, treated with LLOMe for 5 min, and immunostained for CHMP2A and LAMP1. Scale bars = 10 μm for whole image and 1 μm for insets. Fluorescence intensity profiles of the indicated channels across the dotted lines are shown in the lower subpanel.ERepresentative confocal images of a HeLa cells transfected with EGFP‐TECPR1 and TagBFP‐TMEM192 (a lysosomal/late endosomal protein), treated with 1 mM LLOMe for the indicated time, and immunostained for ALIX and Gal3. Scale bars = 10 μm. Source data are available online for this figure.

Fig 4: Multiomic interrogation reveals GAL3 as a cortical PODXL interactor.(A) Schema, multiomic (interactome, proteome, and transcriptome) profiling to identify high–surface PODXL interactors. (B) Network diagram of subgroup interactions among identified PODXL-binding partners. Networks, STRING analysis; layout, Cytoscape analysis. (C) Bubble plot, PODXL-binding partners sorted by subgroup then interaction score with high surface PODXL. For proteomics, color represents fold change (log2) interaction with PODXL compared to GFP alone; interactome, blue to red; proteome, grayscale. For RNA-seq, color represents fold change (log2) endogenous high– compared to low–surface PODXL expression. Bubble size, significance (−log10). (D) Cartoon, GFP-PODXL domains and known PODXL interactors. SP, signal peptide; TM, transmembrane domain; FBM, FERM-binding motif; PBM, PDZ-binding motif. (E) GFP-Trap immunoprecipitation of GFP or GFP-PODXL WT from PC3 cells. Western blotting of input and immunoprecipitates for GAL3, PODXL, GFP, and GAPDH (loading control for GFP). (F) Quantitation of relative GAL3 to precipitated GFP versus GFP-PODXL WT in PC3 cells. Floating bar graph: dots, experiments; midline, mean; n = 3 independent experimental culturing, lysate preparation, and protein isolation. P values (unpaired Student’s two-tailed t test), ***P ≤ 0.0005.

Fig 5: PODXL is a decoy receptor for GAL3 to release GAL3 repression of integrin-dependent invasion.(A to H) Receptor internalization by capture ELISA from GFP-PODXL WT–expressing PC3 cells coexpressing Scramble or GAL3 shRNA (A to D) or in cells with GFP-PODXL WT versus 4K>R mutant expression (E to H). Representative graphs, percentage of GFP-PODXL (A), a5β1-integrin (C), and GFP-PODXL WT versus 4K>R (E and G) internalization at indicated time points. Values, means ± SEM, n = 3 independent experiments. Graphs showing relative endocytosis at time point of 2 min (B and D) or 4 min (F and H) normalized to control; three independent experiments combined. (I to L) Western blotting and quantitation, PC3 expressing Scramble or ITGB1 shRNA for ITGB1, PODXL, GAL3, and GAPDH (loading control). Quantitation across replicates. Means ± SD, n = 5 independent experiments. (M and N) Co-IP, Western blotting, and quantitation of endogenous PODXL and GAL3 from PC3 in (I) for ITGB1, GAL3, PODXL, and GAPDH (loading control). Quantitation across replicates. Means ± SD, n = 5 independent experiments. (O to R) Phase images and quantitation from cells expressing (O and P) Scramble or GAL3 shRNAs or (Q and R) GFP-PODXL 4K>R mutant (IgG control) with β1-integrin blockade (AIIB2, blocking antibody). Cysts, yellow outlines. Outlines colored by time in single frames and overlaid. Heatmap, proportions of round, spread, and tunnel forming, z-score normalized. (P) n = 4 independent experiments, 3 technical replicates per condition, 444 to 702 cysts per condition per experiment. (R) n = 3 independent experiments, 2 technical replicates per condition, 137 to 354 cysts per condition per experiment. Bar graphs, P values, unpaired Student’s two-tailed t test; *P ≤ 0.05, **P ≤ 0.005, ***P ≤ 0.0005, and ****P ≤ 0.0001. Heatmap, P values, Cochran-Mantel-Haenszel test Bonferroni-adjusted, to compare the area and proportion of each classification to control, grayscale as indicated. Scale bars, 100 μm.