Decoupling of DNA damage response signaling from DNA damages underlies temozolomide resistance in glioblastoma cells☆

Bo Cui, Stewrt P. Johnson, Nncy Bullock, Frncis Ali-Osmn, Drell D. Bigner, Henry S. Friedmn*

Departments of Surgerya and Pathologyb, Duke University Medical Center, Durham, NC 22710, USA Received 25 August 2010, Revised 4 November 2010, Accepted 15 November 2010

INTRODUCTION

Glioblastoma multiforme (GBM) is the most common, aggressive and difficult to treat primary brain tumour in adults with the majority of patients surving less than two years[1]. Current therapy includes surgery, radiation and adjuvant chemotherapy[2]. Temozolomide (TMZ), an SN1 type monofunctional DNA methylating agent., increases the survival of GBM patients in concurrent therapy with radiation[3]. However, resistance to TMZ emerges with prolonged treatment and posts a major therapeutic challenge. One of the major determinants of clinical response to TMZ is the methylation status of the promoter region of O6-methylguanine (O6-meG) DNA methyltransferase (MGMT). The hypermethylation of MGMT promoter is seen in approximately 40% to 60% of patients and correlates with a favorable response to TMZ[4-6]. The other known determinant of clinical response to TMZ is mismatch repair (MMR) status. Recurrence of GBM was associated with loss of MMR proteins in patients undergoing prolonged TMZ therapy[7,8].

Though MGMT promoter is methylated in up to 60% of GBM cases, for the first nine months of follow-up, addition of TMZ to radiotherapy does not change survival outcome[4], suggesting that MGMT methylation is the not the sole factor determining outcome. Furthermore, though MMR-proficient cells are approximately 100 fold more sensitive to SN1 type methylators than their MMR-deficient counterparts[9], MMR deficiency makes a minor contribution to clinical resistance in GBM patients[10], suggesting that there exist unrecognized mechanisms of resistance.

O6-meG accounts for approximately 9% of the DNA alkylation adducts by TMZ[11], but it provides the major signal for eliciting DNA damage response (DDR)[12,13]and is also the major initiator of cellular apoptotic response[14-16]. MGMT specifically demethylates O6-meG by transferring the methyl group to an active cysteine residue in itself. O6-meG adducts, if unrepaired, cause O6-meG:C to O6-meG:T transition mutations after DNA replication[17]. The MMR complexes MutSα and MutLα have an affinity for O6-meG:T mismatches[18]. These mispairs are initially detected and processed by MSHα and MLHα, which then signal to activate the ataxia-telangiectasiamutated kinase (ATM)-checkpoint 2 kinase (Chk2) and the Rad3-related kinase (ATR)-checkpoint 1 kinase (Chk1) pathway[12,13], leading to the activation of a transient checkpoint in the second S phase and an eventual cell cycle arrest at the following G2 phase[9].

The MRN complex, which includes Mre11, Rad50 and Nbs1, is also involved in DDR to TMZ. Mre11 forms discrete subnuclear foci in response to TMZ. Phosphorylated histone H2A variant protein H2AX (γ-H2AX), a reliable marker of DNA double-strand breaks (DSBs) and early apoptosis, also forms distinct foci in response to TMZ[19]. The DNA damage mediator protein MDC1 transduces DNA damage signals and controls the activation of intra-S phase checkpoint. It forms a complex with γ-H2AX and offers a platform to support retention of DDR proteins like ATM and Nbs1 at sites of DNA damages[20]. The involvement of MDC1 in DDR to temozolomide, however, remains unelucidated.

The functional availability of the DDR signaling network that responds to DNA damage likely directs how GBM responds to TMZ and decides the therapeutic outcome. We hypothesize that DDR signaling may be impaired in TMZ-resistant GBM cells, which could contribute to their phenotypic resistance. We have generated paired TMZ-resistant GBM cell lines whose resistance is due to neither high MGMT activities nor deficient MMR status. We report here that these resistant GBM cell lines exhibit methylationresistant DNA synthesis (MDS), defective cell cycle checkpoints, and impaired DDR signalling, suggesting that, in the TMZ-resistant GBM cell lines, DDR signaling is decoupled from ongoing DNA alkylation damages by TMZ, which may underlie the phenotypic resistance in the TMZ-resistant GBM cells.

MATERIALS AND METHODS

Drug and drug treatments

TMZ (Schering-Plough, Kenilworth, NJ) and the MGMT inactivator O6-benzylguanine (O6-BG) (National Cancer Institute, USA) were freshly prepared in dimethyl sulfoxide before use. For drug treatments, logarithmically growing cells were used and incubated with TMZ at a concentration equal to IC90for D54 and U251 for the time points indicated. Where indicated, U251 and U251 (OTR) were initially exposed to 100 μmol/L O6-BG, treated with TMZ for 4 h and plated in medium with 20 μmol/L O6-BG. For clonogenic assays, TMZ from 12.5 μmol/L up to 1 mmol/L was used.

Cell lines, generation of drug-resistant cells and clonogenic assay

The GBM cell lines D54, D54 (OTR), U251, U251 (TR) and U251 (OTR) were maintained in Improved MEM Zinc Option (Richter's modification) (Invitrogen, USA) supplemented with 20% fetal bovine serum (FBS) (Invitrogen). For generation of drugresistant TR cell lines, we began treating cells with 10 μmol/L TMZ for 4 h. The cells were maintained at this drug concentration until they showed markedly increased survival to treatment. After every three passages, the drug concentration was increased by 5 to 10 μmol/L. The OTR cell lines were generated the same way, except that: (1) the cells were exposed to 100 μmol/L O6-BG for 10 min before TMZ treatment; (2) cells were maintained in medium with 20 μM O6-BG. The cytotoxicity of TMZ against the GBM cells was determined by clonogenic assays as previously described[21].

Measurement of MGMT activity and mismatch repair assays

The activity of MGMT was measured as described previously[22]and the enzyme activity was defined as the fmol of O6-[3H] meG removed from3H-methylated DNA per mg of extracted protein. Mismatch correction in U251 and U251 (OTR) or D54 and D54 (OTR) nuclear extracts was determined using heteroduplex DNA containing a GT mismatch as previously described[23].

[3H]-thymidine ([3H] TdR) incorporation assay

GBM cells were pre-labeled for 24 h with 0.05 μCi [14C] TdR/mL (Perkin-Elmer, Wellesley, MA). [14C] TdR incorporation was stopped by washing the cells twice with cold PBS (Sigma). The cells were then pulse-labeled for 2 h with 2.0 μCi [3H] TdR/mL (Perkin-Elmer) and, after [3H] TdR incorporation was stopped by washing the cells twice with cold PBS, the cells were incubated for 30 min with unlabeled medium to chase the labeled precursor pool. Thereafter, the cells were lysed with 0.25 mol/L NaOH and samples were counted by dual-label liquid scintillation, and the amount of [3H] TdR was normalized to the amount of [14C] TdR in each sample. DNA synthesis was measured as a ratio of [3H] to [14C] in the treated samples divided by the ratio of [3H] to [14C] in the untreated control samples.

Comet assays

The alkaline and neutral comet assays were carried out according to the manufacturer's recommendations (Trevigen, Gaithersburg, MD). Gel electrophoresis was carried out in alkaline electrophoresis solution (pH ＞13) at 1 V/cm for 10 min or in 1×Tris-borate-EDTA buffer at 4.5 V/cm for 5 min at room temperature. Slides were stained with SYBER Green and analyzed by fluorescence microscopy (Leica DMS 4000B). The percentage of DNA in the comet tail or tail length was determined using the software package "Comet Assay II" (Perceptive Instruments, UK). A minimum of 50 cells per experiment was analyzed. All the experiments were done in triplicate.

Antibodies, immunoblotting analysis and immunofluorescence microscopy

The antibodies for the following proteins were used for immunoblotting studies: phospho-Chk1 (serine 345) and phospho-Chk2 (threonine 68) (Cell Signaling Technology, USA), phospho-ATM (serine 1981) (Rockland, USA) and α-tubulin (Santa Cruz Biotechnology, USA). The following antibodies were used for immunofluorescence microscopy: mouse anti-phospho-ATM antibody (serine 1981), rabbit anti-phospho-Nbs1 antibody (serine 343), rabbit anti-MDC1 antibody, mouse and rabbit anti-γ-H2AX antibody (serine 139; Abcam, USA), and rabbit anti-phospho-Chk2 antibody (threonine 68; Santa Cruz Biotechnology). Secondary antibodies Alex Fluor 488 goat antirabbit IgG and 594 goat anti-mouse IgG (Invitrogen) were used. Immunoblotting and immunofluorescence microscopy were done as previously described[24,25]. Nuclear staining was done using 4', 6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, USA).

Cell-cycle analysis

Treated GBM cells were washed once with PBS, trypsinized, and washed again in PBS with 2% FBS and fixed in ice-cold ethanol for at least 1 h at -20°C, washed, and stained with propidium iodide (30 μg/mL) and treated with RNase (0.6 mg/mL) in PBS plus 0.5% (v/v) Tween 20 and 2% FBS. Stained cells were analyzed on a FACSCalibur flow cytometer (BD Bioscience, USA) using Cellquest software, and the ModFit program (Verity Software House Inc., USA) was used to analyze the cell-cycle profiles.

Statistical analysis

Data were expressed as mean±SD and analyzed using the SPSS10.0 sofeware. P ＜ 0.05 was considered statistically significant.

RESULTS

The resistant GBM cell lines exhibited novel mechanisms of chemoresistance

We treated U251 and U251 (TR) cells with TMZ and clonogenic assays indicated that, compared with U251 cells, U251 (TR) cells were markedly resistant to TMZ (Fig. 1A). With the addition of 25 μmol/L O6-BG, U251 (TR) cells showed a clonogenic survival profile similar to that of U251 cells, suggesting that resistance in U251 (TR) cells to TMZ was due to high MGMT activities.TMZ markedly inhibited the growth of U251 with an IC50and IC90of 50 and 100 μmol/L, respectively, whereas TMZ exerted no noticeable growth-inhibitory effect on the TMZ-resistant U251 (OTR) (Fig. 1B). D54 was sensitive to growth inhibition by TMZ with an IC50and IC90of 20 and 40 μmol/L, respectively, while D54 (OTR) was resistant to the growth-inhibitory effects of the drug (Fig. 1C). Additionally, 20 μmol/L O6-BG failed to sensitize U251 (OTR) or D54 (OTR) to TMZ.

Furthermore, U251 exhibited virtually no MGMT activity (＜10 fmol/mg protein) and U251 (OTR) expressed moderate amounts of MGMT activities (419±37 fmol/mg protein) while TMZ-resistant U251 (TR) expressed high levels of MGMT activity (715±70 fmol/mg protein). Co-administration of 20 μmol/L O6-BG with TMZ reduced the MGMT activities essentially to zero in U251 (OTR) and U251 (TR). D54 and D54 (OTR) also expressed virtually no MGMT activity (＜10 fmol/mg protein). In vitro MMR assays using nuclear extracts from these GBM cell lines and DNA substrates containing a G:T mismatch[23]indicated that there was also no difference in MMR capacity between D54 and D54 (OTR) or between U251 and U251 (OTR) (Fig. 1D).

Fig. 1 The temozolomide (TMZ)-resistant GBM cell lines exhibit unknown mechanisms of resistance.

The resistant GBM cells exhibited MDS in response to TMZ

DNA alkylation damages were shown to inhibit DNA synthesis[26,27]. We analyzed DNA synthesis in U251 and U251 (OTR) by the [3H] TdR incorporation method, which revealed that these cells responded acutely to TMZ by inhibiting the overall rate of DNA replication within minutes after drug exposure, producing a rapid and steep decline in DNA synthesis (Fig. 2A). This sharp depression was followed by a dramatic recovery from the nadir of DNA synthesis. Thereafter, these cells underwent a second prolonged decline in DNA synthesis, which started at 36 h and recovered to control levels by 96 h after treatment. During this period, TMZ caused a greater inhibition of DNA synthesis in U251 than U251 (OTR) (64.51±3.64% of controls vs 85.35±6.25% of controls, 48 h post treatment), suggesting that U251 (OTR) exhibited MDS in response to TMZ.

Similar findings were also observed in D54 and D54 (OTR) (Fig. 2B). A marked inhibition of DNA synthesis was noticed in D54 48 h post treatment (81.2±0.48% of controls). D54 (OTR), on the other hand, exhibited a similar rate of DNA synthesis to that in control cells (119.84±8.81% of controls), indicating ongoing DNA synthesis in these cells despite the presence of DNA damages.

Fig. 2 TMZ causes a biphasic inhibition of DNA synthesis. TMZ causes an acute decline in DNA synthesis followed by a second prolonged phase of inhibition of DNA synthesis. U251 (OTR) (A) and D54 (OTR) (B) exhibit methylation-resistant DNA synthesis in this second phase. Each data point represents the average of three independent experiments; bars: mean±SD. TMZ: temozolomide

O6-meG contributed to the inhibition of DNA synthesis in GBM cells

As TMZ generates several different methyl adducts like N7-methylguanine, N3-methylguanine and O6-meG, we investigated whether O6-meG contributed to TMZ-induced inhibition of DNA synthesis. Analysis of [3H] TdR incorporation in U251 and U251 (TR) revealed a marked inhibition of DNA synthesis in U251 (40.74±14.11% of controls) while DNA synthesis remained virtually unchanged in U251 (TR) (94.13±7.12% of controls) at 4 h after treatment, suggesting that O6-meG contributed to the acute inhibition of DNA synthesis. At 48 h after treatment, DNA synthesis remained virtually unchanged in U251 (TR) (106.79±7.63% of controls) while that in U251 stood at 67.94±7.47% of controls, suggesting that O6-meG also contributed to the second prolonged phase of DNA synthesis inhibition.

The resistant GBM cells incurred extents of DNA methylation damages similar to their parental counterparts

With the exclusion of high MGMT activities and differential MMR capacity as the mechanisms of resistance to TMZ in U251 (OTR) and D54 (OTR), we investigated whether the enhanced clonogenic survival in the resistant GBM cells was attributed to enhanced repair of DNA methyl adducts. We monitored the formation and resolution of TMZ-induced DNA adducts in the GBM cells over time by using the alkaline comet assay, which detects single-strand breaks and alkali-labile sites. We treated U251 and U251 (OTR) with 100 μM TMZ in the presence of O6-BG and D54 and D54 (OTR) with 40 μM TMZ. The results indicated that U251 and U251 (OTR) all incurred the greatest amount of DNA damages at 24 h post treatment [P ＜ 0.05, U251 vs U251 (OTR)], which, though at reduced levels, persisted well into 72 h post treatment (Fig. 3A and 3B). The extent of DNA damages as reflected in the percentage of DNA in the comet tail was not significantly different in U251 and U251 (OTR) at all time points examined (P ＞ 0.05). Though D54 incurred DNA damages to a greater extent than D54 (OTR) at 24 h after treatment [P ＜ 0.05, U251 vs U251 (OTR)], both D54 and D54 (OTR) exhibited similar extent of DNA damages at later time points (Fig. 3A), suggesting that TMZ-induced DNA methylation damages were not preferentially repaired in the resistant GBM cells.

γ-H2AX failed to localize to sites of DNA damages in the resistant GBM cells

DNA DSBs arise from processing of O6-meG:C and O6-meG:T mismatches by the DNA repair apparatus[14,28]. We carried out neutral comet assays to detect the presence of DNA DSBs in the GBM cells 72 h following treatment. Similar amounts of DNA DSBs were found in U251 and U251 (OTR) while DNA DSBs were present to a greater extent in D54 than D54 (OTR) (Fig. 3C). The presence of DNA DSBs was confirmed by immunofluorescence microscopy using anti-γ-H2AX antibody. γ-H2AX formed distinct foci in response to TMZ in U251 and D54 (Fig. 3D). However, γ-H2AX failed to localize to sites of DNA damages in the resistant GBM cells despite the presence of DNA DSBs as revealed by the neutral comet assay.

The resistant GBM cells exhibited impaired DDR to TMZ

O6-MeG adducts signal to activate the ATR-Chk1 and the ATM-Chk2 pathway[12,13]. We examined the activation of the two pathways by immunoblotting analysis using phospho-specific antibodies against ATM, Chk1 and Chk2. We found that the ATR-Chk1 pathway was activated from 24 to 72 h post TMZ exposure in both U251 and U251 (OTR) (Fig. 4A). However, the ATR-Chk2 pathway was activated from 72 to 96 h post treatment only in U251, suggesting that DNA damages in U251 (OTR) failed to signal through the ATM-Chk2 pathway. Examination by indirect immunofluorescence microscopy further revealed that phosphorylated ATM formed discrete foci in U251, but not in U251 (OTR) (Fig. 4B). Phosphorylated Chk2 also formed distinct foci in U251 and D54, but not in their resistant counterparts (Fig. 4C and Fig. 4D).

In addition, phosphorylated Nbs1 and the DNA damage mediator protein MDC1 formed discrete foci in U251 and D54, but failed to do so in the resistant GBM cells (Fig. 5A to Fig. 5D). These findings suggest that these DDR proteins were activated in response to TMZ in D54 and U251. However, they failed to respond to DNA damages induced by TMZ in the resistant GBM cells. Furthermore, Chk2, Nbs1 and MDC1 co-localized with γ-H2AX (data not shown), suggesting that these DDR proteins were intimately associated with one another in response to TMZ.

Fig. 3 DNA methylation damages persist in the GBM cells,

Fig. 4 The resistant GBM cells exhibit defective signaling through the ATM-Chk2 pathway.

The resistant GBM cells showed defective cell cycle checkpoints

We investigated if impaired DDR signaling through the ATM-Chk2 pathway in the resistant GBM cells was associated with any defect in the activation of cell cycle checkpoints. Flow cytometric analysis revealed a marked rise in the percentage of U251 and D54 in the S phase and G2 phase 72 h post treatment (Fig. 6A and 6B). The resistant GBM cells, on the other hand, exhibited virtually no change in cell cycle distributions compared with controls. These findings suggested that DNA lesions in the resistant GBM cells failed to trigger the activation of cell cycle checkpoints that was normally seen in cellular response to TMZ.

DISCUSSION

TMZ has become part of standard care in combination with surgery and radiotherapy for GBM patients. However, resistance emerges with the prolonged use of the drug and the tumor often recurs in a more aggressive nature. Moreover, a significant subset of GBM patients shows clinical resistance to TMZ with operant mechanisms other than high MGMT activities or MMR deficiency.

TMZ causes alkylation damages and the DDR, which these lesions initiate, likely determines the fate of a tumor cell in survival or death. DNA alkylation damages inhibit DNA synthesis in the second S phase following treatment[26,27]. We show that TMZ exhibits a biphasic inhibition of DNA synthesis: an initial acute but temporary inhibition followed by a more prolonged inhibition of DNA synthesis in the second S phase. In the acute phase, D54 and U251 exhibit a profile of inhibition of DNA synthesis similar to their resistant counterparts. In the prolonged phase, however, TMZ causes a greater depression of DNA synthesis in D54 and U251 compared with their TMZ-resistant counterparts, which exhibit MDS.

Fig. 6 The resistant GBM cells are defective in cell cycle checkpoints.

Though the decrease in the rate of DNA synthesis is small in this second phase, given the sharp increase in the proportion of S-phase cells in TMZ-treated D54 and U251, the degree of inhibition of DNA synthesis in these cells may be significantly greater than appreciated. Earlier studies indicated that this second S-phase inhibition of DNA replication was related to cytotoxicity of O6-meG adducts[17,29]. In a GBM patient who suffered relapses, the recurrent tumors were less aggressive and highly sensitive to TMZ[30]. Intriguingly, TMZ caused a greater inhibition of DNA synthesis in the recurrent tumor samples than the primary tumor sample. These insights indicate that, in response to DNA alkylation damages, the GBM cells slow down ongoing DNA synthesis and allow the processing and repair of these DNA lesions, which, if unsuccessful, leads to the onset of cellular death. The lethality of this type of replication-arresting DMA damages may be due to futile, reiterative processing of O6-meG:C and O6-meG:T mispairs or due to attempted repair of DNA DSBs generated as a result of processing or repairing the primary DNA lesions[31]. This futile reiterative process leads to extensive lethalities of treated cells, and the more marked inhibition of DNA synthesis in D54 and U251 may, therefore, explain the greater sensitivity to TMZ in these cells than their resistant counterparts.

DNA lesions that are quickly repaired do not trigger the global DDR. However, we have observed similar extents of DNA alkylation damages between U251 and U251 (OTR) and between D54 and D54 (OTR), suggesting that DNA alkylation damages in the resistant GBM cells fail to signal to slow down ongoing DNA synthesis or that processing or repair of these DNA lesions is decoupled from DNA replication. O6-meG:T mismatches can only arise during the S phase. Alkylated DNA bases are incapable of pairing with any other natural bases, resulting in gaps or breaks on the newly synthesized DNA strand. TMZ also induces formation of DNA DSBs, which precede O6-meG-induced apoptosis in glioma cells, and replication is required for O6-meG-induced apoptosis[14,16]. Though the resistant GBM cells display extensive alkylation damages, the failure to slow down ongoing DNA synthesis may be due to a failure to convert these DNA adducts into DNA strand breaks. However, this appears not to be the case as there exist similar amounts of DNA DSBs in U251 and U251 (OTR). DNA DSBs are also present in D54 (OTR) though to a smaller extent than those in D54. The presence of ongoing DNA alkylation damages and DNA strand breaks in the resistant GBM cells and their failure to trigger an inhibition of ongoing DNA synthesis unmask the presence of MDS in these cells.

Cancer-prone patients suffering from ataxiatelangiectasia exhibit radioresistant DNA synthesis (RDS)[32]and disruption of the ATM-Chk2 pathway results in defective S phase checkpoint and RDS[33]. The MRN complex is required for the activation of ATM and also functions downstream of ATM[34]and its deficiency was associated with defective Chk2 signaling in response to camptothecin-induced DNA DSBs[35]. The complex is also implicated in DDR to TMZ[19]. The DNA damage mediator protein MDC1 binds directly to Nbs1 and targets the latter to sites of DNA DSBs[20]. We show here that ATM, Chk2, MDC1, and Nbs1 form distinct foci in response to TMZ in U251 and D54 and these proteins also co-localize with γ-H2AX, suggesting that these DDR proteins are intimately associated with one another in response to TMZ. Though DNA alkylation damages and DSBs are present in the resistant GBM cells, γ-H2AX fails to localize to the sites of these DNA lesions. Additionally, these DNA damages fail to signal through the ATM-Chk2 pathway. MDC1 and Nbs1 also fail to form discrete foci in the resistant GBM cells.

Chk2 is implicated in irradiation-induced apoptosis[36,37]and MDC1 interacts with Chk2 and is critical for Chk2-mediated apoptosis[38]. Chk2 also activates E2F-1 target genes which may determine the balance between cell survival and death[39,40]. Furthermore, the suppression of TMZ-induced activation of Chk2 due to Akt overexpression protects GBM cells from TMZ-induced cytotoxicity[41]. Nbs1 is required for the induction of apoptosis by the MRN complex[42]. It also regulates apoptosis by disrupting the Ku70-Bax complex, leading to the activation of Bax and caspase-3[43]. The MRN complex has also been shown to contribute to TMZ-induced cytotoxicity[19]. MDS and the failed signaling through the ATM-Chk2 pathway and the lack of involvement of the MRN complex in DDR to TMZ may all contribute to the increased clonogenic survival that we have observed in the TMZ-resistant GBM cells.

Cell cycle checkpoints monitor the state of DNA and arrest cell cycle progression if DNA damage is detected and until a DNA lesion is repaired. If repair cannot be effected, the cell is committed to die. ATM, Chk2 and Nbs1 control the intra-S phase checkpoint in response to irradiation-induced DNA DSBs[44]. MDC1 interacts with Nbs1, which is required for irradiationinduced intra-S phase checkpoint[20]. The defective signaling through the ATM-Chk2 pathway and the failure of MDC1 and Nbs1 to form discrete foci in response to TMZ in the resistant GBM cells suggest that these cells exhibit defective cell cycle checkpoints. Glioma cells exhibited a robust cell cycle arrest in the S phase following treatment with 100 μmol/L TMZ[45]. We also observed a vigorous activation of an intra-S phase checkpoint following TMZ exposure in D54 and U251. However, we failed to observe any noticeable cell cycle distribution alterations in response to TMZ in the resistant GBM cells. These findings suggest that processing or repair of DNA alkylation damages and strand breaks is decoupled from cell cycle checkpoints or that these DNA lesions have escaped from the monitoring of cell cycle checkpoints.

Signaling of O6-meG lesions through the ATRChk1 pathway depends upon a functional MMR apparatus[12]. Our immunoblotting analysis indicates that the ATR-Chk1 is activated in all the GBM cell lines, suggesting that a functional MMR apparatus in these cells. We further showed that these GBM cells exhibited similar MMR capacity. The failed DDR signaling in the resistant GBM cells is, therefore, unlikely due to defective MMR. On the other hand, the failed DDR signaling through the ATM-Chk2 pathway in the resistant GBM cells may arise from a disconnection between the MMR complex and the MRN complex during DDR to TMZ. MMR has been implicated in S-phase checkpoint activation following ionizing irradiation and the MMR protein Msh2 was found to interact with Chk2 and Mlh1 was found to interact with ATM[46]. Additionally, Msh2 was required for the relocalization of Mre11 following irradiation and Msh2 defect led to defective Chk2 phosphorylation[47]. The MRN complex was also found to interact with Mlh1 in response to TMZ[19]. As the MRN complex is required for the activation of ATM[34], failure of the MMR apparatus to interact with the MRN complex may lead to defective phosphorylation by ATM of downstream effectors such as Chk2 or H2AX. There is also another distinct possibility that processing and repair of DNA lesions in the resistant GBM cells are decoupled from DDR to avoid triggering the onset of cellular death programs. The defective DDR signaling through the ATM-Chk2 pathway may also be due to a defect in an upstream regulator of ATM such as FOXO3a[48]. Silencing FOXO3a by siRNA was shown to impair ATM-regulated DDR signaling.

In conclusion, we have generated TMZ-resistant GBM cell lines whose resistance cannot be explained by high MGMT activities or MMR deficiency, the currently two known determinants of clinical response to TMZ in patients with GBM. Additionally, we have unmasked the presence of MDS in the TMZ resistant GBM cells. Furthermore, these resistant cells exhibit defective cell cycle checkpoints and impaired DDR signaling through the ATM-Chk2 pathway. As the functional availability of the DDR signaling network that responds to DNA damage likely directs how GBM responds to TMZ and decides the therapeutic outcome, the defective DDR signaling that we have observed likely drives the resistance in the temzolomide-resistant GBM cells.

[1] Wen PY, Kesari S. Malignant gliomas in adults. N Engl J Med 2008;359:492-507.

[2] Mason WP, Cairncross JG. Drug Insight: temozolomide as a treatment for malignant glioma--impact of a recent trial. Nat Clin Pract Neurol 2005;1:88-95.

[3] Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-96.

[4] Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352:997-1003.

[5] Esteller M, Garcia-Foncillas J, Andion E, Goodman SN, Hidalgo OF, Vanaclocha V, et al. Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 2000; 343:1350-4.

[6] Maxwell JA, Johnson SP, Quinn JA, Mclendon RE, Ali-Osman F, Friedman AH, et al. Quantitative analysis of O6-alkylguanine-DNA alkyltransferase in malignant glioma. Mol Cancer Ther 2006;5:2531-9.

[7] Hunter C, Smith R, Cahill DP, Stephens P, Stevens C, Teague J, et al. A hypermutation phenotype and somatic MSH6 mutations in recurrent human malignant gliomas after alkylator chemotherapy. Cancer Res 2006;66:3987-91.

[8] Cahill DP, Levine KK, Betensky RA, Codd PJ, Romany CA, Reavie LB, et al. Loss of the mismatch repair protein MSH6 in human glioblastomas is associated with tumor progression during temozolomide treatment. Clin Cancer Res 2007;13:2038-45.

[9] O'Brien V, Brown R. Signalling cell cycle arrest and cell death through the MMR System. Carcinogenesis, 2006;27:682-92.

[10] Maxwell JA, Johnson SP, McLendon RE, Lishter DW, Home KS, Rasheed A, et al. Mismatch repair deficiency does not mediate clinical resistance to temozolomide in malignant glioma. Clin Cancer Res 2008;14:4859-68.

[11] Marchesi F, Turriziani M, Tortorelli G, Avvisati G, Torino F, De Vecchis L. Triazene compounds: Mechanism of action and related DNA repair systems. Pharmacol Res 2007;56:275-87.

[12] Yoshioka K, Yoshioka Y, Hsieh P. ATR kinase activation mediated by MutSalpha and MutLalpha in response to cytotoxic O6-methylguanine adducts. Mol Cell 2006;22:501-10.

[13] Caporali S, Falcinelli S, Starace G, Russo MT, Bonmassar E, Jiricny J, et al. DNA damage induced by temozolomide signals to both ATM and ATR: role of the mismatch repair system. Mol Pharmacol 2004;66:478-91.

[14] Roos WP, Batista LF, Naumann SC, Wick W, Weller M, Menck CF, et al. Apoptosis in malignant glioma cells triggered by the temozolomide-induced DNA lesion O(6)-methylguanine. Oncogene 2007;26:186-97.

[15] Hickman MJ, Samson LD. Apoptotic signaling in response to a single type of DNA lesion, O(6)-methylguanine. Mol Cell 2004;14:105-16.

[16] Roos W, Baumgartner M, Kaina B. Apoptosis triggered by DNA damage O6-methylguanine in human lymphocytes requires DNA replication and is mediated by p53 and Fas/CD95/Apo-1. Oncogene 2004;23:359-67.

[17] Karran P, Bignami M. Self-destruction and tolerance in resistance of mammalian cells to alkylation damage. Nucleic Acids Res 1992;20:2933-40.

[18] Stojic L, Brun R, Jiricny J. Mismatch repair and DNA damage signalling. DNA Repair (Amst) 2004;3:1091-101.

[19] Mirzoeva OK, Kawaguchi T, Pieper RO. The Mre11/Rad50/Nbs1 complex interacts with the mismatch repair system and contributes to temozolomide-induced G2 arrest and cytotoxicity. Mol Cancer Ther 2006;5: 2757-66.

[20] Wu L, Luo K, Lou Z, Chen J. MDC1 regulates intra-Sphase checkpoint by targeting NBS1 to DNA doublestrand breaks. Proc Natl Acad Sci USA 2008; 105:11200-5.

[21] Friedman HS, Colvin OM, Kaufmann SH, Ludeman SM, Bullock N, Bigner DD, et al. Cyclophosphamide resistance in medulloblastoma. Cancer Res 1992;52:5373-8.

[22] Bacolod MD, Johnson SP, Ali-Osman F, Modrich P, Bullock NS, Colvin DM, et al. Mechanisms of resistance to 1,3-bis(2-chloroethyl)-1-nitrosourea in human medulloblastoma and rhabdomyosarcoma. Mol Cancer Ther 2002;1:727-36.

[23] Holmes J, Jr., Clark S, Modrich, P. Strand-specific mismatch correction in nuclear extracts of human and Drosophila melanogaster cell lines. Proc Natl Acad Sci U S A 1990;87:5837-41.

[24] Schild-Poulter C, Pope L, Giffin W, Kochan JC, Ngsee JK, Traykova-Andonova M, et al. The binding of Ku antigen to homeodomain proteins promotes their phosphorylation by DNA-dependent protein kinase. J Biol Chem 2001;276:16848-56.

[25] Goldberg M, Stucki M, Falck J, D'Amours D, Rahman D, Pappin D, et al. MDC1 is required for the intra-S-phase DNA damage checkpoint. Nature 2003;421:952-6.

[26] Plant JE, Roberts JJ. A novel mechanism for the inhibition of DNA synthesis following methylation: the effect of N-methyl-N-nitrosourea on HeLa cells. Chem Biol Interact 1971;3:337-42.

[27] Plant JE, Roberts JJ. Extension of the pre-DNA synthetic phase of the cell cycle as a consequence of DNA alkylation in Chinese hamster cells: a possible mechanism of DNA repair. Chem Biol Interact 1971;3:343-51.

[28] Stojic L, Cejka P, Jiricny J. High doses of SN1 type methylating agents activate DNA damage signaling cascades that are largely independent of mismatch repair. Cell Cycle 2005;4:473-7.

[29] Black KA, McFarland RD, Grisham JW, Smith GJ. Cell cycle perturbation and cell death after exposure of a human lymphoblastoid cell strain to N-methyl-N'-nitro-Nnitrosoguanidine. Am J Pathol 1989;134:53-61.

[30] Spiegl-Kreinecker S, Pirker C, Marosi C, Buchroithner J, Pichler J, Silyer R, et al. Dynamics of chemosensitivity and chromosomal instability in recurrent glioblastoma. Br J Cancer 2007;96:960-969.

[31] Karran P. Mechanisms of tolerance to DNA damaging therapeutic drugs. Carcinogenesis 2001;22:1931-7.

[32] Painter RB, Young BR. Radiosensitivity in ataxia-telangiectasia: a new explanation. Proc Natl Acad Sci U S A 1980;77:7315-7.

[33] Falck J, Mailand N, Syljuasen RG, Bartek J, Lukas J. The ATM-Chk2-Cdc25A checkpoint pathway guards against radioresistant DNA synthesis. Nature 2001;410:842-7.

[34] Dupre A, Boyer-Chatenet L, Gautier J. Two-step activation of ATM by DNA and the Mre11-Rad50-Nbs1 complex. Nat Struct Mol Biol 2006;13:451-7.

[35] Takemura H, Rao VA, Sordet O, Furnta T, Mao ZH, Meng L, et al. Defective Mre11-dependent activation of Chk2 by ataxia telangiectasia mutated in colorectal carcinoma cells in response to replication-dependent DNA double strand breaks. J Biol Chem 2006;281:30814-23.

[36] Hirao A, Kong YY, Matsuoka S, Wakeham A, Ruland J, Yoshida H, et al. DNA damage-induced activation of p53 by the checkpoint kinase Chk2. Science 2000; 287:1824-7.

[37] Takai H, Naka K, Okada Y. Chk2-deficient mice exhibit radioresistance and defective p53-mediated transcription. Embo J 2002;21:5195-205.

[38] Lou Z, Minter-Dykhouse K, Wu X, Chen J. MDC1 is coupled to activated CHK2 in mammalian DNA damage response pathways. Nature 2003;421:957-61.

[39] Stevens C, Smith L, La Thangue NB. Chk2 activates E2F-1 in response to DNA damage. Nat Cell Biol 2003;5:401-9.

[40] Hallstrom TC, Mori S, Nevins JR. An E2F1-dependent gene expression program that determines the balance between proliferation and cell death. Cancer Cell 2008;13:11-22.

[41] Hirose Y, Katayama M, Mirzoeva OK, Berger MS, Pieper RO. Akt activation suppresses Chk2-mediated, methylating agent-induced G2 arrest and protects from temozolomide-induced mitotic catastrophe and cellular senescence. Cancer Res 2005;65:4861-9.

[42] Stracker TH, Morales M, Couto SS, Hussein H, Petrini JH. The carboxy terminus of NBS1 is required for induction of apoptosis by the MRE11 complex. Nature 2007;447:218-21.

[43] Iijima K, Muranaka C, Kobayashi J, Sakamoto S, Komatsu K, Matsuura S, et al. NBS1 regulates a novel apoptotic pathway through Bax activation. DNA Repair (Amst) 2008;7:1705-16.

[44] Falck J, Petrini JH, Williams BR, Lukas J, Bartek J. The DNA damage-dependent intra-S phase checkpoint is regulated by parallel pathways. Nat Genet 2002;30:290-4.

[45] Bocangel DB, Finkelstein S, Schold SC, Bhakat KK, Mitra S, Kokkinakis DM. Multifaceted resistance of gliomas to temozolomide. Clin Cancer Res 2002;8: 2725-34.

[46] Brown KD, Rathi A, Kamath R, Beardsley DI, Zhan Q, Mannino JL, et al. The mismatch repair system is required for S-phase checkpoint activation. Nat Genet 2003;33:80-4.

[47] Franchitto A, Pichierri P, Piergentili R, Crescenzi M, Bignami M, Palitti F. The mammalian mismatch repair protein MSH2 is required for correct MRE11 and RAD51 relocalization and for efficient cell cycle arrest induced by ionizing radiation in G2 phase. Oncogene 2003;22:2110-20.

[48] Tsai WB, Chung YM, Takahashi Y, Xu Z, Hu MC. Functional interaction between FOXO3a and ATM regulates DNA damage response. Nat Cell Biol 2008; 10:460-7.