Panel A: Fluorescence micrographs showing the overlap of Galectin-9 (red) with cET-ASOKras (green) in vesicles within Epha2 wild-type and Epha2 knockout KPC cells. Panel B: Bar graph quantifying the size and overlap of Galectin-9/ASO-positive vesicles in Epha2 wild-type (blue bars) and Epha2 knockout (magenta bars) KPC cells after 16 hours of treatment with cET-ASOKras. Panel C: Fluorescence micrographs of Epha2 knockout KPC cells expressing either Epha2 wild-type or Epha2NLS (green) together with Galectin-9 (red) following addition of cET-ASOKras (16 hours, white). Arrowheads indicate ASO accumulation. Fluorescence profiles corresponding to the yellow lines in the micrographs are shown on the right. Panel D: Violin plots quantifying the colocalization (Pearson correlation coefficient) between EPHA2 and ASOs or Galectin-9 and ASO in intracellular vesicles. Panel E: Density plots showing EPHA2 fluorescence intensity z-scores per vesicle binned into nuclear proximal (0–1 micrometer) or distal (greater than 1 micrometer) categories. Panel F: Density plots showing Galectin-9 fluorescence intensity z-scores per vesicle binned into nuclear proximal (0–1 micrometer) or distal (greater than 1 micrometer) categories. Panel G: Three-dimensional reconstruction images showing the overlap of Galectin-9 (red) with cET-ASOKras (magenta) and G3BP1-GFP (green) in Epha2 wild-type or Epha2 knockout KPC cells. Panel H: Bar graph showing the percentage of cells containing at least one G3BP1-positive condensate in either Epha2 wild-type or Epha2 knockout KPC cells after treatment with cET-ASOKras. Panel I: Scatter plot showing the average G3BP1-positive foci per cell in either Epha2 wild-type or Epha2 knockout KPC cells after cET-ASOKras treatment. Panel J: Western blot images showing phospho-Ser51 eIF2α following cET-ASOKras treatment (4 hours) in either Epha2 wild-type or Epha2 knockout KPC cells. Vinculin is used as a loading control. Panel K: Fluorescence micrographs showing the distribution of poly-d(T) (magenta) and G3BP1-GFP (green) after 16 hours of treatment with cET-ASOKras (red) in KPC cells. Panel L: Fluorescence micrographs showing the overlap between G3BP1-GFP, cET-ASOKras, and CD44 in Epha2 wild-type, Epha2 knockout, or ISRIB (1 micromolar)-treated Epha2 wild-type KPC cells. Panel M: Bar graph quantifying the average total area of intracellular G3BP1-GFP condensates per cell in cET-ASOKras-positive vesicle-containing cells. Panel N: Bar graph quantifying the average total area of intracellular ASO vesicles per cell in cET-ASOKras-positive vesicle-containing cells. Panel O: Bar plot showing the effect of ISRIB on cET-ASOKras-induced reduction of Kras messenger RNA expression in KPC cells. Panel P: Bar plot showing the effect of G3bp1/G3bp2 CRISPR double knockout on cET-ASOKras-induced reduction of Kras messenger RNA expression in KPC cells.
EPHA2 is required for trafficking of cET-ASOs to leaky endosomes. (A and B) Galectin-9 (red) overlaps with cET-ASOKras (green) in vesicles, and (B) quantification in either Epha2+/+ (blue bars) or Epha2−/− (magenta bars) KPC cells, after 16-h treatment with the indicated concentrations of cET-ASOKras (16 h); data are mean ± SEM, n = 7 fields. TWA, Tukey. (C) Fluorescence micrographs (Airyscan) of Epha2−/− KPC cells expressing either Epha2WT or Epha2NLS (green) in combination with galectin-9 (red) following addition of cET-ASOKras (16 h, white). Arrowheads show instances of ASO accumulation. Fluorescence profiles (right panels) corresponding to the yellow lines in the micrographs. (D) Quantification of the colocalization (Pearson’s correlation coefficient) between EPHA2 and ASOs or galectin-9 and ASO in intracellular vesicles corresponding to the groups in C, n = 3 independent experiments, unpaired t test. (E and F) Quantification and binning of cET-ASOKras vesicle-nucleus distance into proximal (0–1 µm) or distal (>1 µm) categories. Density plots show EPHA2 (E) and galectin-9 (F) normalized fluorescence intensity z-scores per vesicle, n = 3 independent experiments, ANOVA, Tukey post hoc test. (G) 3D-reconstruction of galectin-9 (red) overlap with cET-ASOKras (magenta) and G3BP1-GFP (green) in Epha2+/+ or Epha2−/− KPC cells. (H) Percentage of cells that contain at least one G3BP1-positive condensate in either Epha2+/+ or Epha2−/− KPC cells after treatment with cET-ASOKras. (I) Average G3BP1-positive foci per cell in either Epha2+/+ or Epha2−/− KPC cells after cET-ASOKras treatment, n = 3 independent experiments, OWA, Tukey. (J) Effect on phosphoSer51 eIF2α following cET-ASOKras (4 h) addition in either Epha2+/+ or Epha2−/− KPC cells. Vinculin is a loading control. (K) Distribution of poly-d(T) (determined by in situ hybridization; magenta) and G3BP1-GFP (green) after 16-h treatment with cET-ASOKras (red) in KPC cells. (L) Overlap between G3BP1-GFP, cET-ASOKras, and CD44 in either Epha2+/+, Epha2−/−, or ISRIB (1 μM)-treated Epha2+/+ KPC cells, n = 3 independent experiments. (M and N) Quantification of the average sum of the area of intracellular G3BP1-GFP condensates (K) and ASO vesicles (L) per cell, in cET-ASOKras–positive vesicles, n = 3 independent experiments, Tukey. (O and P) Effect of ISRIB and G3bp1/G3bp2 CRISPR knockout (G3BPdKO) on cET-ASOKras–induced reduction of Kras mRNA expression in KPC cells, n = 6 and 4 independent experiments, respectively, TWA, Sidak. Source data are available for this figure: SourceData F5. OWA, one-way ANOVA; TWA, two-way ANOVA.