Background
Glioblastoma is the most common WHO grade IV brain tumour in adults. The current standard therapy for patients with newly diagnosed glioblastoma includes primary resection, followed by radiotherapy (IR) and adjuvant temozolomide (TMZ) treatment. Despite of this, the median survival remains poor with about 14 months only [
1], asking urgently for novel therapy approaches.
Therefore, we tested a new strategy to overcome the high radio- and chemotherapy resistance of glioblastoma cells using the checkpoint kinase 1 (Chk1) inhibitor SAR-020106 (SAR) and the epigenetic modulator and cytotoxic agent decitabine (5-aza-2’-deoxycytidine, 5-aza-dC). The blockage of homologous recombination (HR) by SAR inhibits an important DNA repair mechanism of proliferating cells and should thereby lead to DNA damage accumulation and subsequently to cell death induction. Additionally, decitabine acts as an inhibitor of the de novo methyltransferase 1 (DNMT1) potentially releasing the expression of aberrant promotor-hypermethylated, silenced tumour suppressor genes. Moreover, DNA-incorporated decitabine induces further DNA damage and is therefore expected to synergize with SAR and standard glioblastoma treatment.
Immediately after DNA damage induction, proliferating cells undergo cell cycle arrest for DNA repair before cell division progresses or apoptotic cell death is induced. In mammalian non-cancerous cells usually a G1 phase arrest is activated in a p53-dependent manner and DNA double-strand breaks (DSB) are repaired by non-homologous end joining (NHEJ) [
2‐
5]. However, the
p53 gene is mutated in 30–50% (
p53-mut) of glioblastomas [
6‐
8]. In the absence of an intact p53 protein DSB are repaired p53-independently preferably by high-fidelity homologous recombination (HR) during S and G2 phases of the cell cycle [
9]. Most importantly, elevated HR was recently found to mediate acquired TMZ resistance in recurrent glioblastoma [
10]. A critical protein within the activation process of DNA HR is the serine/threonine kinase Chk1 by its interaction with RAD51 [
11]. When DNA damage activates ATR (ataxia telangiectasia and Rad3-related protein) Chk1 is activated by phosphorylation [
12] and becomes a central regulator of the intra-S and G2/M cell cycle checkpoints (reviewed in [
13,
14]). Additional roles of Chk1 in the modulation of cellular response to replication stress are the stabilization of stalled replication forks and the control of replication origin firing (reviewed in [
15]).
Inhibition of Chk1 in vitro and in vivo has been shown to have radio- and chemosensitizing effects on cell survival [
16] and several Chk1 inhibitors have already been applied in clinical trials (reviewed in [
17]). However, most of them failed to improve the therapy efficiency or induced normal tissue toxicity assumedly due to their lack of potency and specificity towards Chk1. SAR is a potent and highly selective ATP-competitive Chk1 inhibitor [
18]. It has been shown to enhance the efficacies of other DNA-damaging drugs like irinotecan and gemcitabine in human colon carcinoma xenograft models with minimal toxicities [
18,
19]. Furthermore, it sensitized cancer cells to IR-induced DNA damage in vitro and in vivo in head and neck cell carcinoma and colorectal cancer models [
18‐
21].
Decitabine got the FDA approval in 2006 for the treatment of the myelodysplastic syndrome (MDS) and chronic myelomonocytic leukemia (CMML) and EMA approval in 2012 for the treatment of acute myeloid leukemia (AML). Currently, clinical studies are ongoing also for treatment of solid tumours with decitabine in combination with other cytotoxic agents or therapies (
https://clinicaltrials.gov/).
Decitabine is an epigenetic DNA-hypomethylating and cytotoxic drug leading to proliferation inhibition and induction of apoptosis by reactivating cancer-related hypermethylation-induced gene silencing of tumour suppressor genes [
22,
23] or by induction of DNA damage [
24‐
26]. We have already shown that decitabine massively reduces clonogenic survival and has radioadditive effects in human medulloblastoma cell lines [
23]. Although in glioblastoma, the CpG island methylation phenotype (G-CIMP; about 9% of all GBMs) showed a better prognosis, G-CIMP-negative tumours also harbor about 1000 hypermethylated genes involved mainly in regulation of cell development, migration, cell-cell adhesion and transcription factors [
27,
28].
In this study, we investigated for the first time the effect of SAR combined with the current standard therapy (IR and TMZ) and decitabine in p53-mut and p53-wildtype (p53-wt) glioblastoma cell lines and primary cells. We analyzed DNA damage, cell cycle phase distribution, proliferation, apoptosis, long-term clonogenic survival, and the number of potential glioblastoma stem cells. Putative toxic effects of this novel treatment approach on non-cancerous normal brain tissue are evaluated on neural progenitor cells using a murine entorhinal-hippocampal slice culture model.
Material & Methods
Modulators
Temozolomide (TMZ, trade name Temodal®) and 5-aza-2’-deoxycytidine (5-aza-dC, decitabine, trade name Dacogen®) were purchased from Sigma-Aldrich, SAR-020106 (SAR) from SYNkinase, and CCT244747 from AdipoGen. Stock solutions were prepared as follows: 10 mM 5-aza-dC in PBS (phosphate buffered saline; Biozym; stored at − 20 °C); 100 mM TMZ in DMSO (stored at − 20 °C); 20 mM SAR in DMSO (stored at − 80 °C) or 1 mM in DMSO (stored at 4 °C for max. 1 week); 20 mM CCT244747 in DMSO (stored at − 20 °C). Further work solutions were made in cell culture medium immediately before use, appropriate DMSO controls were implemented.
Cell lines and cell culture
The human glioblastoma cell lines T98G, DBTRG, and A172 were purchased from the ATCC and LN405 was obtained from the DMSZ. Primary glioblastoma cells P0297 and P0306 were established as described before [
29] from primary glioblastoma (IDH wildtype). Patients provided written informed consent according to German laws and in accordance with the 1964 Helsinki declaration and its amendments, as confirmed by the local ethical committee (144/08-ek).
P53 mutation status of primary cells and cell lines was determined by sequencing (Ion AmpliSeq Cancer Hotspot Panel v2, Thermo Fisher Scientific) in the core unit for DNA technologies, Interdisciplinary Centre for Clinical Research Leipzig. The
MGMT promotor methylation status was determined by pyrosequencing of 5 CpG loci (74–78) (modified after [
30]) in the Division of Neuropathology, University of Leipzig. Results are summarized in Fig.
6a. T98G, LN405, and A172 were maintained in DMEM with 4.5 g/l glucose (Biozym) supplemented with 10% FCS (fetal calf serum; Biochrom). DBTRG cells were cultivated in Gibco™RPMI 1640 (Thermo Fisher Scientific) supplemented with 10% FCS, 25 mM HEPES buffer (Lonza), 2.5 g/l (D+) glucose, 0.11 g/l sodium pyruvate (AppliChem), 0.3 g/l L-glutamine, 30 mg/l L-proline, 35 mg/l L-cysteine, 15 mg/l hypoxanthine, 10 mg/l adenine, 1 mg/l thymidine, and 1 mg/l ATP (Sigma-Aldrich). All beforehand mentioned media were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin (Biochrom). Cells were passaged with trypsin/EDTA. Primary adherent cells were maintained in AmninoMAX-C100 basal medium (Gibco) with 10% AmninoMAX-C100 supplement (Gibco) and passaged using StemPro Accutase (Thermo Fisher Scientific). All cells were cultivated at 37 °C and 5% CO
2. Vital cells were counted by trypan blue exclusion assay. Tests to detect mycoplasma were performed in three-month intervals using PCR Mycoplasma test kit (AppliChem).
Animals
Nestin-CFPnuc C57BL/J6 mice [
31] were bred in the animal facility of the Faculty of Medicine, University of Leipzig according to European (Council Directive 86/609/EEC) and German guidelines (Tierschutzgesetz) for the welfare of experimental animals as previously described [
32]. All experiments had been approved in advance by the local authorities (Landesdirektion Sachsen T12/17).
Preparation of murine entorhinal–hippocampal slice cultures
Murine organotypic entorhinal–hippocampal slice cultures (OEHSC) were generated from nestin-CFPnuc C57BL/J6 mice on postnatal day (p) 3 to p 6 as initially described by Gahwiler et al. [
33] and verified as detailed [
32]. Briefly, after decapitation of the mice and preparation of the brains, 350 μm-thick horizontal slices were cut on a vibratome (Leica VT 1000) under sterile conditions. Up to four slices (eight hippocampi) per mouse were collected, the entorhinal–hippocampal formation resected, and transferred onto porous membrane inserts (Millicell PICMORG50, Millipore) in six-well culture plates. The cultivation medium consisted of MEM (Invitrogen) with 25% Hank’s balanced salt solution (Invitrogen), 25% horse serum (Invitrogen), 1% L-glutamine (Sigma), 1% penicillin/streptomycin (Lonza), and 1% glucose (stock solution 45%, AppliChem). Slices were cultivated at 37 °C and 5% CO
2.
Irradiation
A 150 kV X-ray machine (DARPAC 150-MC, RayTech) with dose rates of 0.69 Gy/min (75-cm2 cell culture flasks), 0.86 Gy/min (6-well plates), or 1.394 Gy/min (96-well plates) was used for IR.
Apoptosis/cell cycle distribution
Cell death induced by apoptosis was detected by AnnexinV Apoptosis Detection Kit II (BD Pharmingen™) or Annexin-V-FLUOS Staining Kit (Sigma-Aldrich) according to manufacturers’ instructions. Cells were seeded in 6-well plates and allowed to attach for 24 h. One hour after treatment with SAR (0.125 μM, 0.25 μM), cells were irradiated with 8 Gy single dose and treated with 5-aza-dC (0.1 μM, 0.5 μM) or TMZ (50 μM, 100 μM). Cells were harvested by trypsinization 4, 24, and 96 h after IR, washed twice with PBS and stained with Annexin V FITC antibody and propidium iodide. Cell staining was measured by flow cytometry (Beckman Coulter, EPICS XL). Additionally, apoptosis-induced DNA fragmentation was determined in sub-G1 fraction of cell cycle analysis (Nicoletti assay) after propidium iodide staining. Cells were treated and harvested as mentioned above, washed twice with PBS, fixed with 70% ethanol, and stored at − 20 °C overnight. After two washes with PBS cells were incubated with 0.1 mg/ml RNAseA solution (Sigma-Aldrich) at 37 °C for 20 min. Then, 50 μg/ml propidium iodide (Sigma-Aldrich) was added and the solution was incubated at 4 °C for 10 min before DNA content was measured by flow cytometry. Sub-G1 fraction was determined using EXPO32 software (Beckman Coulter).
Alkaline comet assay
To determine DNA damage in p53-mut cells, we used the alkaline comet assay. Cells were seeded in 6-well plates and allowed to attach for 24 h. Then, cells were treated with 0.25 μM SAR and irradiated with 8 Gy single dose 1 h later. Immediately and 24 h after IR, cells were detached using HBSS (Hanks’ balanced salt solution; Gibco) supplemented with 20 mM Na2-EDTA and 10% DMSO, washed with PBS, and resuspended in 1% low-gelling temperature agarose (LMPA; Sigma-Aldrich). Then, cells were rapidly spread onto microscope slides pre-coated with a thin layer of 1% normal melting point agarose (NMPA; SeaKem® LE agarose; Biozym) and coated with a thin layer 1% LMPA. Slides were immersed in lysis buffer (2.5 M NaCl, 100 mM Na2-EDTA, 10 mM Tris, pH 10.0) at 4 °C overnight. Alkaline denaturation was carried out in pre-chilled electrophorese buffer (300 mM NaOH, 1 mM Na2-EDTA, pH > 13) for 60 min, followed by electrophoresis (1.2 V/cm; 370 mA) at 4 °C for 30 min. Then, slides were neutralized by incubation with 0.4 M Tris, pH 7.5 three times for 5 min and stained with 1 μg/ml DAPI. DNA content/tail size of 50 randomly selected nuclei per treatment were measured by image analyses (Comet Assay IV software, Perceptive Instruments ltd.) using a Zeiss AxioLab microscope.
Immunofluorescence and western blot analyses of DNA damage proteins
To determine the influence of Chk1 inhibition by SAR on proteins involved in DNA damage repair, we used immunofluorescence microscopy and western blot (method see next section) of gH2A.X and phosphorylated replication protein A 32/2 (pRPA). Cells were seeded on 8-well chamber slides and allowed to attach for 24 h. Then, cells were treated with 1 μM SAR-020106 and irradiated with 8 Gy single dose 1 h later. After 1.5, 24 and 72 h, cells were fixed with 2% formaldehyde in PBS for 15 min and immunofluorescence staining was performed as described before [
34]. The following primary antibodies were used: mouse anti-phospho-histone H2A.X (Ser139) clone JBW301 (1:100; Millipore); rabbit phospho-RPA32/RPA2 (Ser8) clone E5A2F (1:500; Cell Signaling Technology). Secondary antibodies were as follows: Alexa568 goat anti-mouse IgG F(ab’)2 (1:1000, Invitrogen); Alexa488 goat anti-rabbit IgG F(ab’)2 (1:1000, Invitrogen).
GH2A.X foci were quantified in p53-wt primary cells (P0306) in at least 50 cells for each treatment 1 and 24 h after IR. A maximum of 30 gH2AX-foci per nucleus were counted by fluorescence microscopy.
To determine DNA damage-related proteins in T98G cells, we used western blot analysis as already described by Oppermann et al. [
35]. In brief, cells were seeded in 100-mm cell culture dishes and allowed to attach for 24 h. Then, cells were treated with 1 μM SAR and 8 Gy single dose irradiation 1 h later. At 0.5, 1.5, 24, and 72 h after irradiation, cells were washed twice with ice-cold PBS, harvested by scraping in 1 ml ice-cold PBS, and the cell suspension was transferred into a 1.5-ml reaction vial. After resuspension in ice-cold RIPA buffer containing phosphatase and protease inhibitors, cell pellets were lysed by sonification. After centrifugation (5500×g; 5 min; 4 °C), the supernatant was transferred into fresh reaction vials. Proteins were immediately frozen at − 80 °C until western blot was performed.
Protein concentration was determined using Pierce™ 660 nm Protein Assay Reagent (Thermo Fisher Scientific) and a BSA (bovine serum albumin) reference standard curve.
SDS-PAGE was performed with a 15% acrylamide gel, 20 μg of protein per lane, and Chameleon Duo protein ladder (Li-COR Biosciences) using a Mini-PROTEAN System (Bio-Rad). After electrophoresis, proteins were transferred to PVDF membranes (Low-Fluorescence Membrane 0.2 μm pore size, Biozym) using a Mini Trans-Blot Cell (Bio-Rad). Then, the membranes were blocked with TBST (Tris-buffered saline with polysorbate 20: 20 mM Tris, 134 mM NaCl, 0.1% Tween 20; pH 7.6) + 2% BSA for 1 h, washed once with TBST for 5 min, and incubated with primary antibodies (mouse anti-phospho-histone H2A.X (Ser139) clone JBW301, Millipore; rabbit phospho-RPA32/RPA2 (Ser8) clone E5A2F, Cell Signaling Technology; rabbit anti-histone H3 clone D1H2, XP® ChIP formulated, Cell Signaling Technology) diluted 1:1000 in TBST for 1 h. Then, membranes were washed three times with TBST for 5 min and incubated with secondary antibodies (red fluorescent IRDye 680RD goat anti-mouse and green fluorescent IRDye 800CW goat anti-Rabbit; both diluted 1:10,000 in TBST; LI-COR Biosciences) for 1 h in the dark. The membranes were washed three times with TBST and once with double-distilled water. All blocking, antibody incubation, and washing steps were performed at room temperature on an orbit shaker. Membranes were scanned using an Odyssey Imaging System (Li-COR, Bad Homburg, Germany).
Proliferation
Cell proliferation was measured using the colorimetric BrdU cell proliferation ELISA (Sigma-Aldrich) according to the manufacturers’ instructions. For single dose treatment, cells were seeded into 96-well plates and allowed to attach for 24 h. Then, cells were treated with SAR (0.125 μM, 0.25 μM) 1 h before irradiation with 2 or 8 Gy single dose and immediate treatment with 5-aza-dC (0.1 μM, 0.5 μM) or TMZ (50 μM, 100 μM). BrdU solution was added 72 h or 7 d after treatment, 24 h before measurement. For fractionated dose treatment, cells were seeded into cell culture flasks and allowed to attach for 24 h. Then, cells were treated at 7 consecutive days with SAR (0.25, 0.5, 1 μM) and IR (2.2/3.4 Gy per fraction) 1 h later (see treatment schedule Fig.
6c). Five days after treatment, cells were seeded in BrdU-containing medium into 96-well plates and measured 24 h later.
Clonogenic survival
To examine long-term survival of clonogenic cells after single drug treatment, cells were seeded in 6-well plates at three different cell densities in duplicates, allowed to attach overnight, and treated with different drug concentrations at day 1–3 and 6–9 (7 fractions, see Fig.
6c). Irradiation and immediate TMZ or 5-aza-dC treatment was performed 1 h after SAR administration. On day 14, fixation and staining of colonies was performed as detailed below.
For clonogenic assays after multimodal treatment, the setting had to be adapted to the higher cell death rate compared to single treatments: Cells were seeded in 75-cm
2 cell culture flasks and allowed to attach overnight. Medium was changed daily and fractionated treatment was executed (see Fig.
6c). At day 14, vital cells were counted using trypan blue exclusion test and seeded for clonogenic assay at three different cell densities in duplicates in 6-well cell culture plates. Ten to 17 days later (dependent on cell line), colonies were washed with PBS, fixed with ice-cold ethanol/acetone (1/1, V/V) for 10 min, stained with Giemsa (Dr. K. Hollborn & Söhne GmbH & Co. KG) solution (1/1, V/V with distilled water) for 5 min, and washed with distilled water. Colonies with > 50 cells were counted indicating the plating efficiency (PE). The ratio between PE of treated cells and PE of untreated cells represented the surviving fraction (SF) of clonogenic cells. The overall clonogenic survival (OSF) was calculated from the relative number of vital cells at day 14 multiplied with the SF (only multimodal treatment).
Immunofluorescence microscopy of potential tumour stem cells
For nestin and GFAP staining, cells were seeded on 8-well chamber slides 5 d after fractionated treatment (7 fractions, see Fig.
6c) and allowed to attach for 24 h. Then, cells were washed in PBS, fixed in ethanol/acetone (1/1, V/V) for 10 min, permeabilized with 0.5% Triton-X100 for 5 min, and immersed in PBS with 10% normal goat serum and 0.25% Triton-X-100 for 30 min to block unspecific binding. Slides were then incubated with the primary antibody: mouse IgG1 anti-human nestin clone10C2 (Millipore, Cat# MAB5326) 1: 200; rabbit Ig anti-human GFAP (DAKO, Cat# Z0334) 1: 500 in PBS with 2% normal goat serum and 0.25% Triton-X100 at 4 °C overnight. After three washes with PBS, the slides were incubated with the secondary antibodies: goat anti-mouse IgG F(ab’)2-Alexa488 or goat anti-rabbit IgG F(ab’)2-Alexa568 (Invitrogen) 1: 1000 in PBS with 2% normal goat serum and 0.25% Triton-X100 at IR for 1 h. Nuclei were counterstained with DAPI (4′6-diamidino-2-phenylindole-dilactate 10 mg, 1: 10,000; Invitrogen) for 5 min and slides were mounted in Mowiol 4–88/DABCO (Roth, Sigma-Aldrich). For all stainings, specific IgG isotype controls (nestin: mouse IgG1, 1: 100, Millipore; GFAP: rabbit Ig, 1: 2500; DAKO) were applied.
Live imaging analyses of neural progenitor cells
To assess the neurotoxic potential of the multimodal treatment, murine entorhinal–hippocampal slice cultures were fractionated treated (7 fractions, see Fig.
6c) and the nestin expression, characteristic for neural progenitor cells, was visualized after 9 and 16 days using an Olympus BX51 confocal fluorescence microscope at 458 nm excitation and quantified using ImageJ and the Plugin Cell Counter (
http://imagej.nih.gov/ij/) as previously described [
32].
Statistics
Statistical data analyses were, if not otherwise noted, performed using the parametric, two-way, and paired Student’s t-test with Microsoft Excel 2003 software.
Statistical analyses of clonogenic survival data were performed using non-parametric Mann-Whitney test with SPSS statistic software version 24.
P-values ≤0.05 (*;#) and ≤ 0.01 (**;##) were considered as statistically significant and p-values ≤0.001 (***;###) as highly statistically significant.
Discussion
Anticancer treatment by irradiation- or drug-induced DNA damage is limited by the DNA repair capacity of tumour cells. The DNA damage-induced G1/S arrest is accompanied by the non-homologous end joining (NHEJ) repair pathway which is inactive in about 50% of glioblastomas due to aberrant p53 signalling. The inhibition of the alternative p53-independent G2/M DNA repair checkpoint by Chk1 inhibitors may lead to the accumulation of DNA damage resulting in a specific enhancement of tumour cell kill. To exploit this promising strategy, we investigated in this study for the first time antitumour effects after combined application of the Chk1 inhibitor SAR-020106 together with irradiation, temozolomide and decitabine in p53-wildtype and -mutated human glioblastoma cells in a clinical relevant, fractionated setting.
Analysis of cell death induction in two
p53-mut glioblastoma cell lines (LN405, T98G) revealed that IR induces apoptotic and non-apoptotic cell death. This is going along with the notion that IR-induced DNA damage mainly results in cell death by mitotic catastrophe or replicative senescence (reviewed in [
13,
37]). Addition of the S and G2/M checkpoint inhibitor SAR led to a significant enhancement even of apoptotic cell death (AnnexinV
pos/PI
neg; Fig.
1). This is in line with findings of Borst et al. [
21] for SAR and also for other Chk1 inhibitors [
17] and may indicate the induction of p53-independent apoptotic pathways e.g. via activation of p73 through Chk2 [
38,
39]. Also, given as single mediator, SAR strongly induced cell death in these
p53-mut cells (Fig.
1). This might be explained by the high baseline amount of DSB observed in gH2A.X assays, presumably caused by inefficient NHEJ and further accumulation of DSB after SAR administration through its abrogation of G2/M arrest (Fig.
2) and homologue recombination (HR) DNA repair [
18,
21]. Additionally, the prolongation of the S cell cycle phase by SAR (Fig.
2) indicates SAR-induced DNA synthesis problems, in line with its known inhibition of Chk1, and the role of Chk1 during S phase activities (reviewed in [
15]). In comet assays, we revealed that only in SAR-treated
p53-mut cells IR-induced DSB remained for at least 24 h (Fig.
3a-d) going along with its function as DNA repair inhibitor [
11]. Syljuasen et al. [
40] suggested that the inhibition of Chk1 in S phase cells increased the binding of pRPA to single-stranded DNA leading to genomic instability and DSB. Additionally, the hyperphosporylation of RPA indicated by phosphorylation at S4/S8 (Fig.
3f) was shown to go along with DSB generated from the collapse of replication forks after treatment with DNA-damaging agents, e.g. Chk1 inhibitors, stalling DNA replication [
41]. The enhanced accumulation of the DNA damage/repair proteins pRPA and gH2A.X observed here by western blot and immunofluorescence supports the notion that DNA damage is induced, especially DSB, and that DNA repair in response of combined IR and SAR treatment is delayed. At high concentrations, also SAR alone seems to inhibit the repair most probably of spontaneous DSB in
p53-mut cells (Fig.
3f,g), which is in line with our cell death (Fig.
1), cell cycle (Fig.
2), and proliferation results (Fig.
4a).
In the
p53-mut glioblastoma cell line LN405 we observed a significant reduction of BrdU incorporation by SAR and by IR after 72 h which was pronounced after combination of both (Fig.
4a), probably as a result of cell death as shown in the apoptosis assay (Fig.
1). This result was verified in primary glioblastoma cells which showed a lower proliferation rate, resulting in a postponed response. Here, SAR moderately reduced the BrdU incorporation after 72 h and partly reversed the IR-induced reduction of proliferation in
p53-mut cells (presumably by overwriting the IR-induced G2/M arrest shown in Fig.
2). In contrast, SAR enhanced the IR effect on BrdU incorporation in
p53-wt cells at 72 h, where the unrepaired DSB may lead immediately to apoptosis, explaining the lack of effects after 7 days. In the
p53-mut cells however, only at the later time point (7 days) SAR pronounced the IR-induced reduction of BrdU incorporation, going along with the hypothesis, that accumulation of DSB after combination of SAR and IR may lead to delayed cell death by mitotic catastrophe (Fig.
4b) (reviewed in [
13]). These results were also confirmed in a fractionated and therefore more clinical relevant setting in
p53-mut primary GBM cells (Fig.
4c).
Clonogenic survival experiments in the fractionated setting underlined the above findings, showing accelerated reproductive cell death by SAR in all four glioblastoma cell lines after IR, 5-aza-dC, or TMZ treatment with slightly less response in
p53-wt cell lines. Interestingly, Bao et al. reported an enhanced Chk1/2 activity especially in glioma stem cells which are thought to promote radioresistance, underlining the relevance and selectivity of DNA checkpoints as therapeutic targets [
42]. We could also confirm here, that decitabine can sensitize glioblastoma cells towards TMZ and IR. This might be induced by the observed gene body hypomethylation and reexpression of
MGMT, a gene coding for a DNA mismatch repair protein essential for TMZ-induced DNA damage repair, and by the enhancement of residual DNA damage after IR [
43,
44]. Treatments including TMZ induced stronger cell death in
p53-wt compared to
p53-mut cell lines with similar
MGMT promoter methylation status (55.6–75%) which is in accordance with the reported role of p53 in cancer drug resistance [
45]. However, strongest anti-clonogenic effects were seen after triple combination of SAR with IR, 5-aza-dC, and TMZ again in both,
p53-mut and
p53-wt glioblastoma cell lines (Fig.
6) supporting the multimodal treatment approach.
The significance of the
p53 mutation status regarding the sensitivity of tumour cells to Chk1 inhibitors like SAR varies in the literature (overview in [
13]). Especially in
p53-wt cells aberrations of proteins downstream of p53 may also lead to abnormal G1/S checkpoint control resulting in similar effects as seen in
p53-mut cells. However, in our case the low number of unrepaired spontaneous DSB (Fig.
3b, e; gH2A.X assay) indicates a functional NHEJ in the
p53-wt primary cells (P0306). Also crossreactivity at the kinase level, which is usually seen with nonselective Chk1/2 inhibitors such as AZD7762 [
46], is rather unlikely to account for this effect, as SAR inhibits Chk1 with high specificity at the concentrations used here (Chk1 K
i = 13.3 nM, Chk2 K
i > 10 μM) [
36]. Nevertheless, Chk1-dependent DNA repair of replication-induced DSB during S phase and of drug-induced DSB in the S and G2 phase takes place also in
p53-wt cells, which together with some functional overlaps [
47] most likely explains the inhibitory effects of SAR on both,
p53-wt and -mut cells.
The efficiency of such treatments also in
p53-wt glioblastoma patients is of high clinical relevance as, although about 30% of patients with primary and about 60% of patients with secondary glioblastoma have mutant
p53 [
48,
49], intratumoural heterogeneity of
p53 mutation status has been reported and is thought to trigger tumour recurrence after p53-dependent treatment [
50,
51]. However, it has to be kept in mind that enhanced adverse effects of Chk1 inhibitors on
p53-wt normal tissue cells may occur if systemic DNA-damaging therapeutics are used.
It is therefore encouraging that no toxic effects of SAR-including single or multimodal treatment on neural progenitors occurred in our murine hippocampal slice model (Fig.
7d, e).
The specificity of the anti-clonogenic effects shown above was verified by concentration dependencies for 5-aza-dC, TMZ, SAR, and CCT244747, with and without single dose irradiation, at concentrations known to be reached or even exceeded in vivo (5-aza-dC: [
52]; TMZ: [
53]; SAR: [
21]; CCT: [
54]). Interestingly, radiosensitization by SAR was similar to that of the closely related [
36] and orally available Chk1 inhibitor CCT244747 (Fig.
5a-d). Our findings go along with recent reports of radiosensitization by these specific Chk1 inhibitors in human lung, colon, and head and neck cancer cell lines and xenograft models (SAR: [
18,
21]; CCT244747: [
55]). Human primary glioblastoma cells showed a similar dose-dependent radiosensitization by SAR confirming the relevance of the results found in glioblastoma cell lines (Fig.
5e).
The lack of nestin/GFAP expression changes in human primary glioblastoma cells implicates no induction of cell differentiation by the treatments (Fig.
7a-c), although such responses have been described for 5-aza-dC in hypermethylated
IDH1-mutant secondary gliomas [
56].
In the future, the here documented effects of the ChK1 inhibitor SAR might be further enhanced by addition of other mediators. For example, ATR inhibitors are already entering clinical trials and could inhibit ATR-mediated phosphorylation and activation of Chk1, thereby lowering the threshold for induction of cell death by Chk1 inhibitors [
57].
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.