Dose-response curves of the cytotoxicity of ProA and digoxin in U87-MG, U251-MG, GBM6, GBM9 and human astrocytes (a). At least three independent experiments were performed. Three-dimensional (3D) tumor spheroids of U87-MG (b–g) and GBM6 cells (h–m) were treated with indicated concentrations of ProA (b–d and h–j) and digoxin (e–g and k–m) for 7 days. Bright-field photomicrographs (b, e, h, k), dose-dependance and time-dependance on spheroid area (c, f, i, l) and viability at day 7 of exposure, assessed by resazurin reduction test (d, g, j, m). Bar = 500 μm. At least three independent experiments were performed. Bar ± SEM. *p < 0.05; **p < 0.005; ***p < 0.001

Representative images and quantification of U87-MG (a, d), U251-MG (b, e) migratory cells treated with vehicle (control), ProA and digoxin (at EC0 and EC50) for 5 h using the transwell migration assay. Representative images and quantification of migratory cancer stem-like cells GBM6 and GBM9 (c, f) treated with vehicle (control) or ProA (at EC0 and EC50) for 5 h using the transwell migration assay. Bar = 200 μm. Quantification was expressed as percentage of migrating cells relative to 100% of control cells. At least three independent experiments were performed for each condition. Bar ± SEM. *p < 0.05; **p < 0.005

Immunofluorescence staining of EB1 (red) in U87-MG (a) and GBM6 (c) treated or not (control) with ProA and digoxin at EC0 and EC50 for 5 h. Bar = 10 μm. Quantification of EB1 comet length in U87-MG (b) and GBM6 (d) treated with vehicle (control), ProA and digoxin at EC0 and EC50 for 5 h. Measurement of MT dynamics parameters by live microscopy of EB3 at MT plus-ends in EB3-GFP transfected U87-MG cells exposed to vehicle (control), ProA and digoxin at EC50 for 5 h. EB1 comets were tracked over time to measure MT distance-based and time-based catastrophe frequency (e–g). Bar ± SEM. *p < 0.05; **p < 0.005; ***p < 0.001; n.s., p > 0.05

ProA (a), digoxin (b) or vehicle (control) was added at different concentrations (10, 20, and 100 µM) to unpolymerized tubulin, and the percentage of MT assembly was quantified by turbidimetry after 30 min. Vinblastine (20 µM) was used as positive control (c). When ProA (20 µM) was added to fully polymerized MTs (arrow), no depolymerization occurred within 30 min (d). At least three independent experiments were performed. Bar ± SEM

Western blot analysis of ATP1-A1 levels in U87-MG or GBM6 cells transfected with a siRNA against ATP1-A1 (siATP1-A1) or control empty vector (si0). Ratios ATP1A1/GAPDH, relative to control si0, from at least three independent experiments are presented under the blots (a). Dose-response curves of the cytotoxicity of ProA in U87-MG or GBM6 cells transfected with a siRNA against ATP1-A1 (siATP1-A1) or control empty vector (si0) (b). Quantification of EB1 comet length in U87-MG and GBM6 cells transfected with a siRNA against ATP1-A1 (siATP1-A1) or control empty vector (si0) and treated with vehicle (control), ProA or digoxin at EC50 for 5 h (c). Quantification of migratory U87-MG and GBM6 cells transfected with a siRNA against ATP1-A1 (siATP1-A1) or control empty vector (si0) and treated with vehicle (control), ProA or digoxin at EC50, for 5 h by transwell migration assay. Quantification was expressed as percentage of migrating cells relative to 100% of control cells (d). Percentage of intracellular sodium change by analysis of SBFI fluorescence (340/380 nm), in response to treatment with ProA, digoxin (at EC0, EC50, and EC100) or vehicle (control) in U87-MG, U251-MG, and GBM6 cells (e). At least three independent experiments were performed for each condition. Bar ± SEM. *p < 0.05; **p < 0.005; n.s., p > 0.05

Western blot analysis of EB1 expression in wtEB1-GFP, EB1 S155A-GFP, and EB1 T166A-GFP transfected U87-MG cells. U87-MG wild type (wt) was used as a negative control. EB1 expression analysis showed two protein bands for endogenous (31 kDa) and exogenous EB1-GFP (58 kDa) in the stably transfected cells. GAPDH was used as loading control (a). Quantification of EB1 comet length in wtEB1-GFP, EB1 S155A-GFP, and EB1 T166A-GFP transfected U87-MG cells treated with vehicle (control) or ProA at EC50 for 5 h (b). Quantification of migratory wtEB1-GFP, EB1 S155A-GFP, and EB1 T166A-GFP transfected U87-MG cells treated with vehicle (control) or ProA at EC50 for 5 h (c). Survival of wtEB1-GFP, EB1 S155A-GFP and EB1 T166A-GFP transfected U87-MG cells exposed to ProA, measured by the MTT test (d). Quantification of EB1 comet length in U87-MG (e) and GBM6 cells (f) treated or not with SB216763 and concomitantly exposed or not (control) to ProA or digoxin at EC50 for 5 h. Quantification of migratory U87-MG (g) and GBM6 cells (h) treated or not with SB216763 and concomitantly exposed or not (control) to ProA or digoxin at EC50 for 5 h, by transwell migration assay. Quantification was expressed as percentage of migrating cells relative to 100% of control cells. Western blot analysis of expression and activity of GSK3β under 5 h-treatment with ProA at EC0 or EC50 in U87-MG or GBM6 cells transfected with a siRNA against ATP1-A1 (siATP1-A1) or control empty vector (si0). Quantification of western blot bands was expressed as phospho/total GSK3β ratio. GAPDH was used as loading control (i). At least three independent experiments were performed for each condition. Bar ± SEM. *p < 0.05; **p < 0.005; ***p < 0.001; n.s., p > 0.05