Analysis of senescence in glioblastoma cells exposed to temozolomide - a study for the identification of senescence maintaining pathways and novel senolytic agents
Date issued
Authors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
License
Abstract
The alkylating drug Temozolomide (TMZ) in combination with ionising radiation (IR) represents the first-line treatment in glioblastoma (GBM) therapy. While cancer treatment aims for triggering cell death to eliminate the tumour, cancer cells developed strategies to escape cell death: they activate cell survival pathways, including cellular senescence (CSEN) upon chemotherapeutic treatment. Over the past years in vitro and in vivo studies demonstrated that senescent cells could escape their senescent state and re-start proliferation, which may result in delayed tumour progression. Additionally, CSEN is accompanied by the senescence associated secretory phenotype (SASP), which includes pro-inflammatory cytokines, and drives senescence and tumorigenesis, rendering CSEN an unfavourable outcome of chemotherapy. A better understanding of the cellular responses to chemotherapy as well as strategies to eliminate senescent cells are therefore desirable.
Here, we focused on the effect of TMZ treatment regarding CSEN of the malignant brain tumour GBM. The following main questions were addressed: 1) What are the kinetics of TMZ-induced genotoxic, cytotoxic, and cytostatic effects and how are senescent cells characterised? 2) Which pathways are responsible for the maintenance of GBM cells in the TMZ-induced senescent state? 3) Which treatment options, including natural and synthetic compounds, can be used to reduce the load of TMZ-induced senescent cells in GBM?
To gain insight into the cellular processes following TMZ treatment, we analysed the formation of DSBs over time, and the induction of early and late apoptosis/necrosis (cell death) and CSEN. We found a significant level of DNA double-strand breaks (DSBs) 3 d after treatment which persisted up to 10 d following the onset of TMZ exposure. In line with the increase in DSBs, cell death and predominantly CSEN were provoked 3 d following TMZ treatment and increased up to day 5 and 8, respectively. We then analysed TMZ-induced senescent cells and identified pathways essentially involved in maintaining the senescent state. The DNA damage response (DDR) factors ATM, ATR, CHK1, CHK2, p53 and p21 were found to be active in TMZ-induced senescent cells. For senescence maintenance, the ATM/ATR-CHK1-p53-p21 axis as well was a pathway involving NF-kB were involved, as revealed from inhibitor experiments. Since the autophagy inhibitor chloroquine was senolytic, we conclude that activation of autophagy is required for keeping cells in the senescent state. To increase cell death levels following TMZ treatment, we co-treated LN229 and A172 cells with IR, N-(2-chloroethyl)-N'-cyclohexyl-N-nitrosourea (CCNU, lomustine) or methadone (MTD) and screened for natural compounds that specifically drive senescent cells into cell death, while proliferating cells are unharmed. No increase in cell death levels in the senescent cell population was observed upon co-treatment of GBM cells with TMZ and CCNU or MTD.
An additive effect on cell death was observed when GBM cells were first treated with TMZ and irradiated 3 h later, or first irradiated and treated with TMZ 6 h later.
Screening of natural compounds for their senolytic activity identified artesunate and fisetin, but not curcumin, as novel senolytic agents in GBM cells in vitro under the treatment conditions used. In a last step we identified senescent cells in patient derived tumour samples from matched primary and recurrent GBM via histochemistry staining. Increased DSBs combined with increased levels of trimethylated lysin on histone 3 (H3K27me3) served as senescence marker. Comparison of primary and recurrent tumour samples showed significantly higher DSBs and H3K27me3 but significantly lower cell death levels in the recurrent tumour samples. These results indicate that not only in vitro but also in vivo senescent cells accumulate following TMZ treatment of GBM.
Overall, the data presented in this study demonstrate that TMZ-induced CSEN is the main trait in GBM cells in vitro and very likely also in vivo. The identification of proteins involved in the TMZ-induced CSEN maintaining pathways in GBM cells might provide a reasonable support in the treatment of cancer. Also, the identification of novel senolytic agents and timing of the combination treatment with TMZ and IR in GBM cells in vitro can help to develop optimized treatment regiments for GBM.