O-GlcNAcylation inhibition redirects the response of colon cancer cells to chemotherapy from senescence to apoptosis

Abstract

The potential use of pro-senescence therapies, known as TIS (Therapy-Induced Senescence), for the treatment of colorectal cancer (CRC) generated significant interest since they require lower doses compared to those required for inducing apoptosis. However, the senescent cell cycle-arrested cancer cells are long-lived, and studies have revealed escape mechanisms contributing to tumor recurrence. To deepen our understanding of the survival pathways used by senescent cancer cells, we delved into the potential involvement of the hexosamine biosynthetic pathway (HBP). HBP provides UDP-GlcNAc, the substrate for O-GlcNAc transferase (OGT), which catalyzes O-GlcNAcylation, a post-translational modification implicated in regulating numerous cellular functions and aberrantly elevated in CRC. In this study, we demonstrated, in the p53-proficient colon cancer cell lines HCT116 and LS174T, that TIS induced by low-dose SN38 or etoposide treatment was accompanied with a decrease of GFAT (the rate limiting enzyme of the HBP), OGT and O-GlcNAcase (OGA) expression correlated with a slight reduction in O-GlcNAcylation levels. Further decreasing this level of O-GlcNAcylation by knocking-down GFAT or OGT redirected the cellular response to subtoxic chemotherapy doses from senescence to apoptosis, in correlation with an enhancement of DNA damages. Pharmacological inhibition of OGT with OSMI-4 in HCT116 and LS174T cells and in a patient-derived colon tumoroid model supported these findings. Taken together, these results suggest that combing O-GlcNAcylation inhibitors to low doses of conventional chemotherapeutic drugs could potentially reduce treatment side effects while preserving efficacy. Furthermore, this approach may increase treatment specificity, as CRC cells exhibit higher O-GlcNAcylation levels compared to normal tissues.

Fig. 1. SN38-induced senescence of HCT116 cells is accompanied by a decrease in GFAT2, OGT, OGA, and O-GlcNAcylation levels.

A Scheme depicting the Hexosamine Biosynthetic Pathway and O-GlcNAcylation processes. GFAT1/2: Glutamine Fructose-6-Phosphate Amido Transferase 1/2 (rate limiting enzyme of the HBP), OGT: O-GlcNAc Transferase, OGA: O-GlcNAcase. B HCT116 cells were treated with increasing concentrations of SN38 ranging from 1 to 25 nM for 96 hours. Cell apoptosis was investigated by Western Blot analyses of cleaved-caspase 7 and cleaved-PARP1. GAPDH was used as a loading control. Data shown are representative of three independent experiments. C–E HCT116 cells were treated with 1 nM SN38 for 96 or with DMSO as a negative control. C Top: Representative images of the microscopic analysis of cell morphology showing an increase in the size of senescent cells. Pictures were taken at the same magnification (x200). Bottom: SA-β-Galactosidase activity assay demonstrating blue staining of senescent cells. D Left: Western blot analysis of O-GlcNAcylation levels, OGT, OGA, and both isoforms of GFAT (GFAT1 and GFAT2), as well as the expression of p21, cyclin D1, and EZH2, three senescence markers. Right: Quantification of relative protein expression from four independent experiments (optical density measurement relative to GAPDH) (n = 4) (individual values and mean +/- SEM, ns: non-significant, *P < 0.05, **P < 0.01, multiple unpaired t-tests). E qRT-PCR analysis of p21, cyclin D1, EZH2, GFAT1, GFAT2, OGT, and OGA transcript expression. Results represent the individual values and mean +/- SEM of five independent experiments (n = 5) (ns: non-significant, **P < 0.01, ***P < 0.001, ****P < 0.0001, multiple unpaired t-tests).

Fig. 2. Preventing O-GlcNAcylation decrease through OGA silencing does not impact the entry of HCT116 cells into SN38-induced senescence.

HCT116 cells were transfected with a siRNA targeting OGA (si OGA) or a non-target control siRNA (siCTRL). 24 h later, the cells were treated with 1 nM SN38 for 72 hours. A Western blot analysis was performed to assess the expression of the three senescence markers p21, cyclin D1, and EZH2. The efficiency of siRNA was also confirmed by evaluating OGA los of expression and O-GlcNAcylation levels upregulation. GAPDH was used as a loading control. Results shown are representative of three independent experiments. B Quantitative fluorimetric determination of the SA-β-Galactosidase activity in the different experimental conditions. Results represent the individual values and mean +/- SEM of four independent experiments (n = 4) (ns: non-significant, *P < 0.05, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). Quantification of relative protein expression analyzed by Western Blot (optical density measurement relative to GAPDH) of p21 (C), cyclin D1 (D) and EZH2 (E) from three independent experiments (n = 3) (individual values and mean +/- SEM, ns: non-significant, *P < 0.05, ***P < 0.001, ****P < 0.0001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests).

Fig. 3. Silencing GFAT1 and GFAT2 prevents the entry of HCT116 cells into SN38-induced senescence and induces their apoptosis through enhancement of DNA damage.

HCT116 cells were transfected with a siRNA targeting GFAT1 and GFAT2 (si GFAT1/2) or a non-target control siRNA (siCTRL). 24 h later, the cells were treated with 1 nM SN38 for 72 h. A Western blot analysis was performed to assess the expression of the senescence markers p21 and cyclin D1, the DNA damage marker γH2AX and the apoptosis marker cleaved-caspase 7. The efficiency of siRNA was also confirmed by evaluating GFAT1 and GFAT2 loss of expression. GAPDH was used as a loading control. Results shown are representative of three independent experiments. B Quantitative fluorimetric determination of the SA-β-Galactosidase activity in the different experimental conditions. Results represent the individual values and mean +/- SEM of four independent experiments (n = 4) (ns: non-significant, **P < 0.01, ***P < 0.001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). Quantification of relative protein expression analyzed by Western Blot (optical density measurement relative to GAPDH) of γH2AX (C) and cleaved-caspase 7 (D) from three independent experiments (n = 3) (individual values and mean +/- SEM, ns: non-significant, *P < 0.05, **P < 0.01, ***P < 0.001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). E Quantification of the ratio of viable cells through MTT assay from three independent experiments (n = 3) (individual values and mean +/- SEM, ****P < 0.0001, ***P < 0.001, *P < 0.05, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests).

Fig. 4. Silencing OGT prevents the entry of HCT116 cells into SN38-induced senescence and induces their apoptosis through enhancement of DNA damage.

HCT116 cells were transfected with a siRNA targeting OGT (si OGT) or a non-target control siRNA (siCTRL). 24 h later, the cells were treated with 1 nM SN38 for 72 h. A Western blot analysis was performed to assess the expression of the senescence markers p21 and cyclin D1, the DNA damage marker γH2AX and the apoptosis marker cleaved-caspase 7. The efficiency of siRNA was also confirmed by evaluating GFAT1 and GFAT2 loss of expression. GAPDH was used as a loading control. Results shown are representative of four independent experiments. B Quantitative fluorimetric determination of the SA-β-Galactosidase activity in the different experimental conditions. Results represent the individual values and mean +/- SEM of four independent experiments (n = 4) (ns: non-significant, **P < 0.01, ***P < 0.001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). Quantification of relative protein expression analyzed by Western Blot (optical density measurement relative to GAPDH) of γH2AX (C) and cleaved-caspase 7 (D) from three independent experiments (n = 3) (individual values and mean +/- SEM, ns: non-significant, *P < 0.05, **P < 0.01, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). E Quantification of the ratio of viable cells through MTT assay from four independent experiments (n = 4) (mean +/- SEM, **P < 0.01, ****P < 0.0001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests).

Fig. 5. Pharmacological inhibition of OGT combined with low-dose SN38 treatment causes HCT116 cells to switch from senescence to apoptosis in a dose dependent-manner.

HCT116 cells were treated with increasing concentrations of the OGT inhibitor OSMI-4 ranging from 5 to 20 µM. 24 h later, the cells were treated with 1 nM SN38 for 72 h. A Western blot analysis was performed to assess the expression of the senescence markers p21 and cyclin D1, the DNA damage marker γH2AX and the apoptosis markers cleaved-caspase 7 and cleaved-PARP1. The treatment efficiency was also confirmed by evaluating O-GlcNAcylation levels. GAPDH was used as a loading control. Results shown are representative of three independent experiments. B Quantitative fluorimetric determination of the SA-β-Galactosidase activity in the different experimental conditions. Results represent the individual values and mean +/- SEM of three independent experiments (n = 3) (ns: non-significant, *P < 0.05, **P < 0.01, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). Quantification of relative protein expression analyzed by Western Blot (optical density measurement relative to GAPDH) of γH2AX (C) cleaved-caspase 7 (D) and cleaved-PARP1 (E) from three independent experiments (n = 3) (individual values and mean +/- SEM, ns: non-significant, *P < 0.05, **P < 0.01, ****P < 0.0001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests). F Quantification of the ratio of viable cells through MTT assay from three independent experiments (n = 3) (mean +/- SEM, ***P < 0.001, ****P < 0.0001, ordinary one-way ANOVA with Fisher’s LSD multiple comparison tests).

Fig. 6. The OGT inhibitor OSMI-4 enhances the cytotoxicity of SN38 in patient-derived colon tumor organoids.

A PDTOs derived from a human left colon tumor were allowed to form before being treated with increasing concentrations of SN38 ranging from 5 to 100 nM for 72 h. B Cell viability was evaluated through cell-titer glo® assay. Values represent the mean +/- SD of two independent experiments (n = 2) (ns: non-significant, *P < 0.05, ****P < 0.0001, ordinary one-way ANOVA with Dunnett’s multiple comparisons tests). C,D PDTOs were allowed to form before being treated either with 10 nM or 25 nM of SN38 alone or in combination with 20 µM OSMI-4 for five days. C Representative brightfield images of the tumoroïds in the different experimental conditions. Scale bar = 200 µM. D Box-plots depicting the diameters (in µM) of more than 30 tumoroïds obtained from three independent images acquired in one experiment and measured with the image J software. Statistical significance was evaluated using ordinary one-Way ANOVA with Tukey’s multiple comparisons tests (ns: non-significant, *P < 0.05, **P < 0.01). E Evaluation of apoptosis through caspase-glo® 3/7 activity assay. Values represent the individual values and mean +/- SEM of two replicates (ns: non-significant, *P < 0.05, ****P < 0.0001, ordinary one-way ANOVA with Dunnett’s multiple comparisons tests).