Fig 2: KDM1A silencing reduces CRC cells clonogenic potential and anoikis resistance.Cells were transduced with control shNT and shRNAs specific for KDM1A, sh68, and sh71. Representative pictures of the methylcellulose colony formation assay (CRC-SC#1 and CRC-SC#2) stained with MTT or acquired with a phase-contrast microscope (scale bar 250 µm) CRC-SC#1 (A) and CRC-SC#2 (B). Quantification of the number of colonies revealing a significant reduction in the clonogenic potential in CRC-SC#1 (C) and CRC-SC#2 (D). Data are presented as mean ± SD from three independent experiments. Representative pictures of the colony formation assay of HCT116 and SW480 cells stained with crystal violet (E). Quantification of the number of colonies revealing a significant reduction in the clonogenic potential (F). Data are presented as mean ± SD from three independent experiments. Representative pictures of transduced HCT116 (G), and SW480 (H) cells seeded in soft agar and stained with MTT. Quantification graphs showing a remarkable decrease in the number of colonies and their area in HCT116 (I), and SW480 (J) KDM1A-silenced cells. Data are presented as mean ± SD from at least three independent experiments. The number and area of the colonies were assessed by ImageJ software. Student’s t-test was conducted (sh68 or sh71 vs shNT): *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig 3: KDM1A knockdown induces apoptosis, cell cycle arrest, and defective DDR.Western blots showing KDM1A protein silencing upon shRNA-mediated transduction of HCT116 cells and CRC-SCs with shNT and the shKDM1A (sh68 and sh71) on day 3 after transduction (protein quantification normalized on tubulin, relative to shNT) (A). Cell viability assay using CellTiter-Glo to measure ATP content in CRC cells on days 5, 7, and 9 after transduction. Data represent the mean ± SD of at least three independent experiments (B). Viability of CRC-SCs (n = 4) 11 days after transduction with shNT, sh68, and sh71 shRNA expressing lentiviruses. Data represent the mean ± SD of at least three independent experiments (C). Representative dots plots showing HCT116 cells distribution after AxV/PI staining (D). Graphs showing the analysis of apoptotic cells (AxV pos/PI neg) after KDM1A silencing in HCT116 and CRC-SC#1 cells on day 4 and day 7 after transduction. Data represent the mean ± SD of at least three independent experiments (E). Cell cycle distribution quantification (F) of transduced HCT116 cells by flow cytometry. The percentage of cells in sub G0, G0/G1, S, and G2 phases of the cell cycle was assessed 48 h after serum addition to starved transduced cells (day 2 after transduction). Data represent the mean ± SD of three independent experiments. For DDR investigation, cells (day 2 after transduction) were pretreated with oxaliplatin 10 µmol/L for 8 h to induce DSBs formation. After oxaliplatin removal cells were stained for nuclei (blue), 53BP1 (red), or pH2AX (red), and mounted on microscope slides 8- and 24 h following treatment removal. Pictures were acquired by confocal microscopy (63X magnification, scale bar 25 µm). Representative pictures (G, I). Graphs showing 53BP1 or pH2AX foci average number per cell (H, J). Data show the mean ± SEM of three independent experiments. RT-qPCR analysis of BUB1, PLK1, and AURKA in CRC-SC#1 and CRC-SC#2 KDM1A-silenced cells (K). Data represent the mean ± SD of at least three independent experiments. Student’s t-test was conducted (sh68 or sh71 versus shNT): *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig 4: KDM1A expression foresees a worse prognosis in CRC.Immunohistochemistry staining of KDM1A in normal and CRC tissues, based on nuclear staining intensity score multiplied by the percentage of KDM1A-positive cells (<80 negative; >80 positive) (A). Chi-square analysis displaying the correlation between KDM1A and tumor stage in CRC patients (B). Kaplan–Meier curves with log-rank survival analyses showing that KDM1A expression is associated with reduced progression-free survival (PFS) and overall survival (OS) (C, E). Univariable and multivariable Cox proportional hazards regression analysis showing hazard ratio (HR) and 95% confidence interval (CI) for each covariate (D, F), p-values < 0.05 were considered statistically significant.

Fig 5: KDM1A targeting impairs CRC cells migration and invasion.Timeline of the investigation of the effect of KDM1A targeting on invasion and migration (A). Representative phase-contrast pictures of wound healing assays of HCT116 and SW480 cells (B). Data analysis of the migration potential of adherent KDM1A-silenced cells vs control cells (shNT). Data represent the mean ± SD of at three independent experiments (C). Representative phase-contrast pictures of wound healing assays of CRC-SC#1 and CRC-SC#2 (D). Data analysis of the migration potential of KDM1A-silenced CRC-SCs vs controls (shNT). Data represent the mean ± SD of at least three independent experiments (E). Representative phase-contrast pictures of transwell invasion assay in adherent CRC cells stained with DiffQuick (F). Data analysis of invading cells at 48 h. Data represent the mean ± SD of at least three independent experiments (G). Wound healing assay of HCT116 cells treated with the KDM1A inhibitors ORY-1001 (10µmol/L) and GSK2879552 (40 µmol/L); representative pictures (H) and data analysis (I) of the migration potential of adherent cells treated with KDM1A inhibitors vs vehicle. Data represent the mean ± SD of at least three independent experiments. The percentage of wound area was assessed by ImageJ software. Student’s t-test was conducted (sh68 or sh71 versus shNT): *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.