Therapy-induced senescence of glioblastoma cells is determined by the p21CIP1-CDK1/2 axis and does not require activation of DREAM

Abstract

Therapy-induced senescence (TIS) is a major challenge in cancer therapy as senescent cancer cells provoke local and systemic inflammation and might be the cause of recurrences. Elucidation of pathways leading to TIS is of utmost importance for establishing strategies to counteract this. Previously we have shown that temozolomide (TMZ), an alkylating drug used forefront in glioma therapy, causes majorly cellular senescence, which is triggered by the primary damage O⁶-methylguanine, activating the mismatch repair dependent ATR/ATM-CHK1/CHK2-p53 damage response pathway. The downstream pathways leading to TIS remained to be explored. Here, we show that TMZ-induced TIS in glioma cells does not require activation of the DREAM complex, but is bound on a G2-specific response. We show that the CDK inhibitor p21^CIP1 does not interact with CDK4, but with CDK1 and CDK2 causing abrogation of the B-Myb and FOXM1-signaling pathway and subsequently arrest of cells in the G2-phase. The induced G2-arrest is incomplete as DNA synthesis can be resumed leading to endoreduplications. This process, which is inhibited by the CDK4-blocking drug palbociclib, is preceded by reactivation of the G1/S-specific E2F1-signaling pathway due to lack of functional DREAM activation. These findings provide an explanation for the polyploidization and giant cell phenotype of anticancer drug-induced senescent cells. Incomplete DREAM activation may also explain the observation that downregulation of DNA repair is a transient phenomenon, which goes along with the entrance of cells into the senescent state.

Fig. 1. TMZ-induced cell death and senescence.

A–D LN229 cells were exposed to different concentrations of TMZ for 144 h (A, B) or to 50 µM TMZ for different time periods (C, D). Experiments were performed in triplicates. A, C Cell death was measured by flow cytometry using PI staining. B, D Frequency of senescent cells was detected microscopically by SA-β-Gal staining. E, F LN229 cells were transfected with p21^CIP1 specific siRNA or nonspecific siRNA and 24 h later exposed to 50 µM TMZ for 144 h. Cell death was measured by flow cytometry using PI staining (E) and frequency of senescent cells was detected by flow cytometry using C₁₂FDG staining (F). G–I LN229 cells were exposed to 50 µM TMZ for different time periods. G Proliferation was measured by cell counting. H DNA synthesis was measured by BrdU assay. I DNA content per cell was measured using the NanoDrop 1000 Spectrophotometer. J LN229 cells were exposed to 50 µM TMZ for 144 h, and cell cycle distribution was measured by flow cytometry using PI staining. A representative experiment is shown. B, D, I Differences between treatment and control were statistically analyzed using Student’s t test (p < 0.001). F Differences between treatment and control, as well as between ns-si/TMZ and p21-si/TMZ were statistically analyzed using Student’s t test (**p < 0.01***p < 0.001).

Fig. 2. Analysis of DREAM activation.

A LN229 cells were exposed to 50 µM TMZ for 96 or 144 h. Expression and phosphorylation of p130 (pp130), Rb (pRb), as well as expression of p21^CIP1, E2F1, E2F4, E2F5, B-Myb, FOXM1, LIN9, and p27 were measured by immunoblotting. β-Actin or HSP90 was used as internal loading control. Quantification of the immunoblot indicates x-fold induction in TMZ-exposed cells compared to untreated cells. B, C LN229 cells were exposed to 50 µM TMZ for 48 or 144 h. B Expression of FOXM1, MYBL2, CDK1, CDKN1A, CDKN1B, E2F1, E2F4 and E2F5 was measured by qPCR. C Expression of CCNA1, CCNA2, CCNB1, CCNB2, CCND1, CCND2, CCNE1 and CCNE2 was measured by qPCR. B, C Experiments were performed in triplicates, ACTB and GAPDH were used as internal loading control. Differences between the control and TMZ treatment were statistically analyzed using Student’s t test (non-labeled = not significant; *p < 0.1; **p < 0.01; ***p < 0.001).

Fig. 3. Transcriptional changes upon TMZ exposure.

LN229 cells were exposed to 50 µM TMZ for 48 or 144 h. RNA was isolated and subjected to RNA-Seq. A The genes commonly up-regulated (upper panel) or down-regulated (lower panel) after 48 and 144 h exposure are displayed as Venn diagrams. B Significantly regulated pathways were identified by the Reactome pathway browser (https://reactome.org/PathwayBrowser/). C The up-regulated genes and overlap with p53 targets are displayed as Venn diagrams. D The down-regulated genes and overlap with DREAM targets are displayed as Venn diagrams. E LN229 cells were exposed to 50 µM TMZ for 48 or 144 h. Expression of WEE1, PLK1, CENPA, MYT1, BIRC5, AURKA, AURKB, CDC20 and CDH1 was measured by qPCR. Differences between the control and TMZ treatment were statistically analyzed using Student's t test (non labelled = not significant; *p < 0.001 and log2 fold change > 1.5 are included.

Fig. 4. Binding of p21^CIP1 to CDKs and activation of the DDR.

A LN229 cells were exposed to 50 µM TMZ for 48 or 144 h. Interaction between p21^CIP1 and CDK1, CDK2, and CDK4 was measured by co-immunoprecipitation. B LN229 cells were exposed to 50 µM TMZ for 48, 96, or 144 h. Expression of CDK1, CDK2, and CDK4 was measured by immunoblotting. β-Actin was used as internal loading control. C LN229 cells were exposed to 50 µM TMZ for 24, 48, or 72 h. D LN229 cells were exposed to 50 µM TMZ for 96 or 144 h. C, D Expression of p21^CIP1, CDC25c and PLK1, as well as expression and phosphorylation of CDK1 (pCDK1), CHK1 (pCHK1), WEE1 (pWEE1) and B-Myb (pB-Myb) was measured by immunoblotting. β-Actin or HSP90 were used as internal loading control. B, C Quantification of the immunoblot indicates x-fold induction in TMZ exposed cells compared to untreated cells.

Fig. 5. TMZ-induced cell death, senescence and transcriptional regulation of cell cycle factorsin A172 and U87MG cells.

A, B LN229, U87MG, and A172 cells were exposed to 50 µM TMZ for 144 h. A Cell death and cell cycle distribution were measured by flow cytometry using PI staining. B Frequency of senescent cells was measured by flow cytometry using C₁₂FDG staining. C A172 and U87MG cells were exposed to 50 µM TMZ for 144 h. Expression of FOXM1, E2F1, MYBL2, PLK1, WEE1, AURKA, AURKB, MYT1, CDKN1A, BIRC5 and CENPA was measured by qPCR. A–C Experiments were performed in triplicates. B–C Differences between treatment and control were statistically analyzed using Student’s t test (not-labeled = not significant; *p < 0.1; **p < 0.01; ***p < 0.001).

Fig. 6. Binding of p21^CIP1 to CDKs and activation of the DDR in A172 and U87MG cells.

A U87MG and A172 cells were exposed to 50 µM TMZ for 96 or 144 h. Expression of p21^CIP1, CDC25c, and PLK1, as well as expression and phosphorylation of CDK1 (pCDK1), CHK1 (pCHK1), WEE1 (pWEE1) and B-Myb (pB-Myb) were measured by immunoblotting. β-Actin or HSP90 was used as an internal loading control. Quantification of the immunoblot indicates x-fold induction in TMZ-exposed cells compared to untreated cells. B U87MG and A172 cells were exposed to 50 µM TMZ for 48 or 144 h. Interaction between p21^CIP1 and CDK1, CDK2, and CDK4 was measured by co-immunoprecipitation. C LN229, U87MG, and A172 cells were exposed to 50 µM TMZ for 48 or 144 h. In addition, the cells were exposed to 1 µM Palbociclib for 144 h or the cells were exposed to 50 µM TMZ, and 48 h later 1 µM Palbociclib was added for additional 96 h (TMZ/Palbo). Cell death and cell cycle distribution were measured by flow cytometry using PI staining; experiments were performed in triplicates. Differences between the frequency of cells in the G2-phase or showing a DNA content >2n after TMZ and TMZ/Palbo treatment were statistically analyzed using Student’s t test (***p < 0.001).

Fig. 7. Expression of DNA repair factors upon TMZ exposure.

A LN229 cells were exposed to 50 µM TMZ for 144 h. RNA was isolated and subjected to RNA-Seq. Number of downregulated DNA repair genes is displayed as Venn diagram. B LN229 cells were exposed to 50 µM TMZ for different time periods, and the expression of MSH2, MSH6, and EXO1 was analyzed by qPCR. C A172 and U87MG cells were exposed to 50 µM TMZ for 144 h, and the expression of MSH2, MSH6, and EXO1 was analyzed by qPCR. B–D Differences between the control and TMZ treatment were statistically analyzed using Student’s t test (not-labeled = not significant; *p < 0.1; **p < 0.01; ***p < 0.001. D Expression of MSH2, MSH6, and MGMT was analyzed in LN229 cells and four LN229-derived clones escaping/evading TMZ-induced senescence by qPCR. B–D Experiments were performed in triplicates, ACTB and GAPDH were used as internal loading control, and the control was set to 1. E Expression of MSH2, MSH6, and MGMT was analyzed in LN229 cells and four LN229-derived clones escaping/evading TMZ-induced senescence by immunodetection. HSP90 was used as an internal loading control.