Novel Mechanism whereby Nuclear Factor κB Mediates DNA Damage Repair through Regulation of O6-Methylguanine-DNA-Methyltransferase (original) (raw)

Requests for reprints: Iris Lavon, Department of Neurology, Leslie and Michael Gaffin Center for Neuro-Oncology, Ein Karem, P.O. Box 12000, 91120 Jerusalem, Israel. Phone: 972-2-677-7712; Fax: 972-2-677-7712; E-mail: [email protected].

Received: October 16 2006

Revision Received: May 29 2007

Accepted: July 11 2007

Online ISSN: 1538-7445

Print ISSN: 0008-5472

2007

Cancer Res (2007) 67 (18): 8952–8959.

Split-Screen
Views Icon Views
Open the PDF for in another window
Tools Icon Tools
Search Site
Article Versions Icon Versions
- Version of Record September 17 2007

Abstract

_O_6-Methylguanine-DNA-methyltransferase (MGMT) and nuclear factor κB (NF-κB) are two key effectors associated with the development of resistance to alkylating agent–based chemotherapy. This prompted us to hypothesize that NF-κB might be involved in MGMT regulation. Consistent with this hypothesis, we have discovered two putative NF-κB binding sites within the MGMT promoter region and showed a specific and direct interaction of NF-κB at each of these sites. Forced expression of the NF-κB subunit p65 in HEK293 cells induced an increase in MGMT expression whereas addition of the NF-κB super repressor ΔNIκB completely abrogated the induction. We also found a significant correlation between the extent of NF-κB activation and MGMT expression in the glioma cell lines and the human glial tumors tested and showed that it was independent of MGMT promoter methylation. Our results are of potential clinical significance because we show that cell lines with ectopic p65 or high constitutive NF-κB activity are less sensitive to nitrosourea treatment and that suppression of MGMT activity with _O_6-benzylguanine completely abolishes the chemoresistance acquired by NF-κB. The findings of our study strongly suggest that NF-κB plays a major role in MGMT regulation and that MGMT is most probably the major player in NF-κB–mediated chemoresistance to alkylating agents. [Cancer Res 2007;67(18):8952–9]

Introduction

Nuclear factor κB (NF-κB) is a family of dimeric transcription factors. The inhibitor of κB (IκB) retains NF-κB in the cytoplasm, and following exposure to extracellular inducers, it undergoes phosphorylation, ubiquination, and subsequent degradation. This, in turn, allows NF-κB to translocate into the nucleus, where it binds a common sequence motif known as the NF-κB site (1) and stimulates gene expression (for review, see ref. 2).

DNA damage induced by the alkylating agent 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) results in a marked increase in NF-κB activity (3). Inhibition of NF-κB activity by a superrepressor strongly enhances the apoptotic potential of the alkylating agent. Therefore, it was suggested that in human tumors, the role of NF-κB in antiapoptotic mechanisms contributes to the high incidence of chemoresistance to alkylating agents (3).

Alkylating agents are highly reactive mutagens and carcinogens and their analogous compounds are used to treat human malignancies. The lethal and mutagenic effects of these compounds are inhibited by the cellular DNA repair enzyme _O_6-methylguanine-DNA-methyltransferase (MGMT), which transfers the alkyl/methyl adducts from the O6 atom of DNA guanine to its own cysteine residues. The guanine is then restored and the MGMT molecule is irreversibly inactivated (4). Hence, the repair capability of MGMT is dependent on de novo protein synthesis.

MGMT expression varies widely in tumor cells (5, 6). It has been suggested that hypermethylation of CpG islands within the promoter region is associated with epigenetic inactivation of the MGMT. Several studies showed that tumor cells with MGMT promoter methylation are more sensitive to alkylating agents such as BCNU and temozolomide (7–9). On the other hand, overexpression of MGMT in tumors has a protective effect against cell death induced by chlorethylating and methylating agents in both experimental and clinical settings (10–15). Inhibition of MGMT activity, using an artificial substrate such as _O_6-benzylguanine, sensitizes the tumor cells to the toxic effects of chemotherapeutic alkylating agents (16–19).

These data indicate that MGMT expression is a crucial player in tumor drug resistance and is an ideal target for modulation. Therefore, an understanding of the molecular mechanisms that control MGMT expression may have major clinical implications.

The promoter region of MGMT has been cloned and sequenced (20). The function of the transcription factors glucocorticoid-responsive element (21) and activator protein 1 (AP-1; ref. 22) in the regulation of MGMT have been described. We identified two putative NF-κB binding sites in the MGMT promoter region and analyzed their role in the regulation of MGMT expression. Using electrophoretic mobility shift assay (EMSA), we found that NF-κB binds specifically to the MGMT promoter and showed that transient transfection of HEK293 cells with the NF-κB subunit p65 induced a 55-fold increase in MGMT expression. The relationships between NF-κB and MGMT were further shown in glioma cell lines and human glial tumors, as we found a significant correlation between the extent of NF-κB activation and MGMT expression, which was independent of MGMT promoter methylation. We also showed that ectopic expression of p65 or high constitutive NF-κB activity has a protective effect against alkylating agents, which seems to be solely dependent on MGMT. In view of the fact that a large proportion of tumor cells display high constitutive activation of NF-κB (for review, see ref. 23) and that such tumors usually exhibit increased resistance to chemotherapy (24), clarification of the role of NF-κB in MGMT transcription regulation is of great importance.

Materials and Methods

Cell culture and transfection. Glioma cell lines U87MG, T98G, A172, RG2, and HEK293T (human embryonic kidney) were obtained from the American Type Culture Collection. U87MG and T98G were cultured in MEM supplemented with 2 mmol/L l-glutamine, 1.5 g/L sodium bicarbonate, Earle's balanced salt solution (Sigma), 0.1 mmol/L nonessential amino acids, 1.0 mmol/L sodium pyruvate, and 10% fetal bovine serum (FBS; Biological Industries). A172, RG2, and HEK293 cells were cultured in DMEM supplemented with 4 mmol/L l-glutamine, 1.5 g/L sodium bicarbonate (Sigma), and 10% FBS (Biological Industries). All cells were maintained in a humidified incubator at 37°C in 5% CO2.

Transient transfections were done in six-well plates using the FuGene6 transfection reagent (Roche Diagnostics GmbH) according to the manufacturer's instructions. U87MG, T98G, and A172 cells were transfected with pNF-κB-Luc reporter vector (Clontech) whereas RG2 and U87MG were transfected with _MGMT_-Luc (construct p-954/+24ML, 1 μg/well; kindly provided by Prof. Sankar Mitra, Department of Human Biological Chemistry and Genetics, The University of Texas Medical Branch, Galveston, TX; ref. 21). Cytomegalovirus-β-galactosidase expression vector (CMV-βgal; Clonthech) was included (0.1 μg/well) in each transfection to normalize transfection efficiency. HEK293 cells were transfected with CMV-βgal alone or along with one or more of the following plasmids: CMV-p65, CMV-ΔNIκB, CMV-p50, and CMV-cJun. CMV-p65 and CMV-ΔNIκB plasmids were kindly provided by Prof. Yinon Ben-Neriah (The Hebrew University, Jerusalem, Israel); CMV-p50 and CMV-cJun were kindly provided by Dr. Danielle Melloul (Hadassah Hebrew University Medical Center, Jerusalem, Israel).

Gel electrophoretic mobility shift assay. Nuclear extracts were prepared from HeLa cells treated with 200 units/mL of tumor necrosis factor-α (TNFα) for 15 min. Oligodeoxynucleotides spanning the NF-κB binding site within the MGMT promoter, with or without a C-G mutation or the consensus NF-κB site from HIV long terminal repeat (LTR; sequence shown in Fig. 1B), were end-labeled by a fill-in reaction using the Klenow fragment of DNA polymerase (New England Biolabs, Beverly, MA). DNA binding reactions were done by incubation on ice for 15 min of 10 μg of nuclear extracts with 0.3 ng of 32P-labeled synthetic double-stranded oligodeoxynucleotides in the presence of 10 mmol/L HEPES (pH 7.9), 10% glycerol, 50 mmol/L KCl, 5 mmol/L MgCl2, 5 mmol/L DTT, 2 μg of poly(deoxyinosinic-deoxycytidylic acid), and 0.1% NP40. Competitor oligonucleotides were incubated in 100-fold molar excess and preincubated in the reaction mixtures for 10 min before addition of the radiolabeled probe. For supershift experiments, 1 μL of p65 antibody (Chemicon International, Inc.) was added during the preincubation period.

Figure 1.

Specific binding of NF-κB to two putative NF-κB sites within the MGMT promoter. A, location of the two NF-κB putative binding sites within the MGMT promoter region. The sites were designated MGMT-κB1 and MGMT-κB2 according to their position within the MGMT promoter. B, sequence of the oligonucleotides used for EMSA: the consensus NF-κB site from HIV LTR (HIVκB) served as control. The oligonucleotides with mutated sites were designated MGMT-κB1-Mut and MGMT-κB2-Mut. The mutated nucleotides are in boldface and underlined. C, EMSA analysis. Nuclear extracts from HeLa cells exposed for 10 min to TNFα (200 units/mL) were incubated with 32P-labeled probes, as depicted. Incubation of consensus HIVκB (lane 2), MGMT-κB-1 (lane 5), and MGMT-κB-2 (lane 9) resulted in the formation of complexes of a similar size (black arrow). Specific binding was not observed when extracts were incubated with MGMT-κB1 Mut (lane 8) or MGMT-κB2 Mut (lane 12) or when the incubation was carried out in the presence of a 100-fold excess of unlabeled HIVκB probe (lanes 4, 7, and 11). Identification of the binding factor as NF-κB was confirmed by supershift analysis using mAbs against the p65 subunit (lanes 3, 6, and 10).

Figure 1.

Close modal

Real-time PCR. Total RNA was prepared using a SV total RNA kit (Promega). cDNA was prepared from 1 μg of total RNA using MuLV reverse transcriptase (Applied Biosystems) and random hexamers according to the manufacturer's instructions for first-strand cDNA synthesis. The reaction mixture included 1 μL of cDNA, 300 nmol/L concentrations of the appropriate forward and reverse primers (Syntteza), and 7.5 μL of the master mix buffer containing nucleotides, Taq polymerase, and SYBR green (SYBR Green Master Mix, Applied Biosystems), in a total volume of 15 μL. Gene amplification was carried out using the GeneAmp 7000 Sequence Detection System (Applied Biosystems). Amplification included one stage of 10 min at 95°C, followed by 40 cycles of a two-step loop: 20 s at 95°C and 1 min at 60°C. The gene expression results were normalized to the 18S rRNA gene.

All experiments were repeated three to five times in triplicate and are presented as the mean ± SD. The primers used for real-time PCR were 18S, 5′-GGCCCTGTAATTGGAA-3′ (forward) and 5′-CCCTCCAATGGATCCTCGTT-3′ (reverse); MGMT, 5′-GCAATTAGCAGCCCTGGCA-3′ (forward) and 5′-CACTCTGTGGCACGGGAT-3′ (reverse).

Immunohistochemical staining. Five-micrometer-thick sections were deparaffinized in xylene, hydrated, and incubated with 10 mmol/L sodium citrate buffer (pH 6.5). These were then heated by microwave (500 W) for 30 min. The sections were left in the heated buffer for 10 min at room temperature. After 5-min treatment with 3% hydrogen peroxide, they were blocked by incubation with 3% bovine serum albumin for 30 min, followed by a 1-h incubation with mouse monoclonal antibodies (mAb) against human MGMT (1:50 dilution; Chemicon International) or mouse mAbs against human p65 (1:50 dilution; Chemicon International) at 37°C. The sections were then treated with secondary antibody (biotinylated antimouse; ABC Elite kit, Vector Laboratories) for 30 min at room temperature and then with avidin-peroxidase complex for 20 min, and finally developed with diaminobenzidine substrate (Sigma) according to the manufacturer's instructions. The nuclei were counterstained with hematoxylin (Sigma). In the negative controls, the primary antibody was omitted. The level of MGMT protein expression was defined semiquantitatively according to the fraction of positive nuclear staining and was scored as high (50–100% positive nuclear staining) or low (0–49% positive nuclear staining). The semiquantitative evaluation was done by a pathologist (Y.F.) who was blinded to all the patients' details.

_O_6-Benzylguanine and BCNU treatment and crystal violet viability test. Various glioma cells (A172, U87MG, and T98G) or HEK293 cells were transiently transfected with different plasmids as indicated above. For the _O_6-benzylguanine–treated groups, _O_6-benzylguanine was added to the cell medium 4 to 6 h after transfection at a final concentration of 80 μmol/L and incubated with the cells for the duration of the experiment. At 24 h after transfection, the cells were exposed for 2 h to increasing concentrations of BCNU (as indicated). The cells were fixed 48 h later with 4% paraformaldehyde for 20 min at room temperature and washed twice with PBS. A 400-μL volume of 0.5% crystal violet in dH2O was added and the cells were stained for 15 min on a vibrator at 300 rpm. The dye was aspirated, the wells rinsed with dH2O, and the plate was allowed to dry in the hood. For destaining, 400 μL of 10% acetic acid were added to each well and the cells were incubated for 15 min on a vibrator at 300 rpm. The absorbance (_A_590) of the destained solution in each well was read and recorded.

Luciferase assay. At 24 h after transfection, the cells were lysed for 15 min on ice with 200 μL of luciferase lysis buffer (Promega). Luciferase assays were carried out using a Promega assay kit and a luminometer (EG&G Berthold). The activity was normalized to β-galactosidase activity (Promega) and plotted as the mean ± SD of triplicates from a representative experiment.

Analysis of the methylation status of the MGMT promoter. Genomic DNA (500 ng) from each cell line was chemically modified with sodium bisulfite to convert unmethylated cytosine to uracil while leaving methylcytosine unaltered (EZ DNA methylation kit, Zymo Research). A 2-μL volume of the converted DNA was subjected to methylation-specific PCR using two primer sets designed for amplifying the methylated or unmethylated allele of the MGMT promoter. Primer sequences of MGMT were for unmethylated reaction, 5′-TTTGTGTTTTGATGTTTGTAGGTTTTTGT-3′ (forward) and 5′-AACTCCACACTCTTCCAAAAACAAAACA-3′ (reverse), and for methylated reaction, 5′-TTTCGACGTTCGTAGGTTTTCGC-3′ (forward) and 5′-GCACTCTTCCGAAAACGAAACG-3′ (reverse). PCR was done under the following conditions: an initial melting step of 10 min at 95°C; followed by 50 cycles of 20 s at 95°C, 20 s at 59°C, and 45 s at 72°C; and a final elongation step of 4 min at 72° in a GeneAmp 9700 thermocycler (Applied Biosystems) using AmpliTaq Gold DNA polymerase (Applied Biosystems). Amplified products were separated on a 3.5% MetaPhor gel and visualized under UV illumination.

Statistical analysis. The nonparametric Kruskal-Wallis ANOVA test was applied to examine the statistical differences between study groups subjected to quantitative reverse transcription-PCR (RT-PCR) analysis or to BCNU treatment. When significant results were obtained, multiple pairwise comparisons were carried out between pairs of groups using the Mann-Whitney test with the Bonferroni correction for the significance level.

Fisher's exact test was applied to test the association between NF-κB activation and the MGMT expression level in glial tumors or between MGMT expression and MGMT promoter methylation as categorical variables.

All tests applied were two tailed and P ≤ 0.05 was considered statistically significant.

Results

Detection of two putative NF-κB binding sites within the MGMT promoter. Inhibition of NF-κB sensitizes cancer cells to alkylating agents (3). Considering the fact that MGMT is a critical DNA repair enzyme involved in _O_6-alkylguanine–induced effects, we hypothesized that NF-κB might play a role in MGMT regulation. Using two different software tools (TFSEARCH, Genomatix) for computer analysis of the promoter region of MGMT, we detected two putative binding sites for NF-κB: one at −766 and the other overlapping a Sp1 binding site at −90. We named them MGMT-κB1 and MGMT-κB2, respectively (Fig. 1A).

Interaction between NF-κB/p65 and the NF-κB sites within the MGMT promoter. To explore whether the NF-κB sites found within the MGMT promoter actually bind NF-κB, we carried out an EMSA analysis of nuclear extracts from HeLa cells treated with TNFα and with labeled DNA fragments spanning the region containing the NF-κB motif. This process generated a nuclear binding factor specific for these regions (Fig. 1C,, lanes 5 and 9) of the same size as the complex attached to the consensus HIV-κB site (Fig. 1C,, black arrows). The binding was abolished when a C-to-G substitution was introduced into the NF-κB-binding motif (Fig. 1B, and C, lanes 7 and 11), showing the specific requirement for this motif. Binding specificity was confirmed by competition with cold HIV-κB oligonucleotide. Addition of mAbs against the NF-κB/p65 resulted in a “supershift” (Fig. 1C,, lanes 6 and 10). This confirmed that p65 binds to both NF-κB putative sites. The upper band in lane 9 (Fig. 1C) is not the same size as the NF-κB complex attached to the canonical HIV-κB oligonucleotide (Fig. 1C,, lane 2). It was neither shifted when incubated with NF-κB/p65 antibodies (lane 10) nor abolished in the presence of unlabeled HIV-κB probe (Fig. 1C,, lane 11). Based on the finding that the MGMT-κB2 binding site overlaps the Sp1 site, we speculate that this complex may contain Sp1. This speculation is also supported by its size, which is the same as that of the complex attached to the Sp1 oligonucleotide (Fig. 1C , lane 1). The results of these experiments led us to conclude that there is a specific and direct interaction between NF-κB/p65 and the two NF-κB sites located within the MGMT promoter.

Induction of MGMT mRNA expression and MGMT promoter–dependent reporter gene by NF-κB. To determine whether NF-κB plays a functional role in MGMT transcription, we carried out a quantitative RT-PCR analysis. We found that the MGMT mRNA level was 55-fold higher in HEK293 cells transiently transfected with CMV-p65 versus the cells transfected with CMV-βgal (Fig. 2A and B). This elevation was almost completely abolished by the addition of the nondegradable IκBα mutant protein (CMV-ΔNIκB). Transfection with CMV-cJun increased MGMT by 8-fold, an observation that has been documented before (22). Cotransfection of CMV-p65 and CMV-AP-1/c-Jun did not result in augmented MGMT mRNA, whereas cotransfection of CMV-p65 and CMV-p50 reduced by 50% the activity driven by CMV-p65 alone (Fig. 2A and B). Induction of MGMT RNA following transfection with CMV-p65, CMV-cJun, CMV-p65 and CMV-AP-1/c-Jun, or CMV-p65 and CMV-p50 was significantly different as compared with that in cells transfected with CMV-βgal (P < 0.001). These data were further confirmed by using a luciferase reporter gene driven by an hMGMT promoter fragment containing the two NF-κB binding sites. Transfection of CMV-p65 into RG-2 and U87MG cell lines induced MGMT promoter–dependent luciferase activity by 6- and 24-fold, respectively, whereas CMV-ΔNIκB almost completely abolished the induction (Fig. 2C).

Figure 2.

Induction of MGMT mRNA expression and MGMT promoter–dependent reporter gene by NF-κB/p65. A, HEK293 cells were transiently transfected with CMVβGal alone or along with various expression vectors as indicated. At 24 h after transfection, changes in MGMT expression were analyzed by real-time RT-PCR using the SYBR Green assay. Fold-change (y axis) represents the relative expression of the MGMT mRNA versus that of the control group (cells transfected with CMV-βgal) normalized to 18S rRNA expression. Columns, mean; bars, SD. B, parallel patterns were observed when the real-time PCR end products from (A) were analyzed by agarose gel electrophoresis. C, U87MG (gray columns) and RG2 (white columns) cell lines were transiently transfected with the reporter _MGMT_-Luc construct alone or with other plasmids as indicated. CMV-βgal expression vector was included in each transfection to normalize transfection efficiency. The obtained promoter activity is relative to the basic pGL3-Luc reporter plasmid lacking the promoter sequence.

Figure 2.

Close modal

Significant correlation between NF-κB activation and MGMT expression level in glial tumors and cell lines. A high MGMT expression level is associated with tumor resistance to alkylating agents. To test whether the extent of NF-κB activation in tumor cells correlates with the protein expression level of MGMT, we immunohistochemically stained 29 human oligodendroglioma sections for both NF-κB and MGMT (Fig. 3A). To selectively detect the activated form of NF-κB, we used a mAb that recognizes an epitope overlapping the nuclear localization signal of the p65 subunit. The results of the semiquantitative assessment of the NF-κB and MGMT nuclear staining (Fig. 3B) revealed a significant correlation (P < 0.0001, Fisher's exact probability test) between NF-κB activation and MGMT expression (Fig. 3A). To confirm our results, we used three glioma cell lines (U87MG, A172, and T98G) that show differential RNA expression of MGMT. We evaluated NF-κB basal activity in each cell line by quantifying luciferase activity 24 h after transfection with the pNF-κB-Luc reporter vector (Fig. 3C). MGMT expression levels in the different cell lines were identified by real-time RT PCR and are presented as fold induction relative to that of human lymphocytes (Fig. 3D). There was a correlation between NF-κB activity, as shown by the luciferase assay, and MGMT mRNA expression level (Fig. 3C and D).

Figure 3.

Correlation between NF-κB activation, MGMT mRNA, and protein expression. Summary of semiquantitative immunohistochemical analysis of NFκB activation (nuclear translocation) and MGMT expression in 29 paraffin-embedded sections of high-grade and low-grade human gliomas. A, significantly more tumors with high MGMT expression (≥50% nuclear staining) are seen in the group of tumors with high nuclear NF-κB staining than in the group of tumors with low NF-κB staining (<50% nuclear staining). B, representative microscopy of MGMT and NF-κB nuclear staining assessed by immunohistochemistry. Top, a patient with negative staining of both NFκB and MGMT; bottom, a patient with positive staining. C and D, extent of NFκB activation and MGMT mRNA expression assessed in different glioma cell lines. C, NF-κB activation level determined by transfection with an NF-κB-luciferase reporter construct. Columns, mean reporter activity of the relative luciferase activity (RLU) after normalization to β-galactosidase expression; bars, SD. D, MGMT mRNA expression evaluated by real-time RT-PCR. Fold change (y axis) represents the relative expression of MGMT mRNA versus that of the control group (human lymphocytes) normalized to 18S rRNA expression. Columns, mean; bars, SD.

Figure 3.

Close modal

MGMT expression correlates with NF-κB activation regardless of the methylation status of MGMT promoter. The relationship between MGMT promoter hypermethylation and loss of MGMT expression is controversial ever since it was first reported by Esteller et al. (16, 25, 26). To determine whether there is a correlation between NF-κB activation, MGMT expression, and MGMT promoter methylation, we analyzed the methylation status of the MGMT promoter by methylation-specific PCR assay. In contrast to the differential pattern of NF-κB activation and MGMT expression in the various cell lines (Fig. 3C and D), the MGMT promoter was uniformly methylated in the A172, U87MG, and T98G cell lines (Fig. 4A). This discrepancy was also found in human oligodendrogliomas. Assessment of the MGMT promoter status of 16 oligodendrogliomas of the 29 used earlier did not reveal any correlation between NF-κB activation, MGMT expression, and MGMT methylation status (P = 1.00, Fisher's exact probability test; Fig. 4B). Furthermore, forced overexpression of p65 in HEK293 increased the MGMT expression level by 55-fold (Fig. 2) despite the fact that the MGMT promoter is methylated in those cells (Fig. 4A).

Figure 4.

MGMT promoter methylation status in glioma cell lines and oligodendroglioma tumors. A, methylation status of the MGMT promoter region in various cell lines, as determined by methylation-specific PCR assay. PC, positive control for methylated DNA; NC, normal control (DNA from a normal blood sample) with an unmethylated MGMT. C, control without DNA. A 100-bp marker ladder was loaded to estimate molecular size (right). B, methylation status of 16 of the 29 tumors presented in Fig. 3. Of these tumors, 12 exhibited both low MGMT and low p65 nuclear staining (<50%), 4 tumors exhibited both high MGMT and high p65 nuclear staining (≥50%), and 1 tumor showed high MGMT and low p65. Despite the significant correlation between NF-κB activation and MGMT expression, methylation of the MGMT promoter varied.

Figure 4.

Close modal

NF-κB activation induces chemoresistance. As shown above, high constitutive activation of NF-κB or ectopic p65 stimulated MGMT expression. To determine whether this also results in induction of cellular resistance to alkylating agents, we compared the sensitivity of the three glioma cell lines to increasing quantities of BCNU (5–80 μg/mL). Viability tests showed that the glioma cell line T98G, which exhibits high NF-κB activation and high MGMT expression, was significantly (P < 0.001) more resistant to toxic doses of BCNU than the other two cell lines (Fig. 5A). To investigate whether this effect would also be obtained following forced expression of p65, HEK293 cells were transfected with CMV-βgal alone or along with CMV-p65 or CMV-p65 and CMV-ΔNIκB. The cells were treated 24 h after transfection with BCNU (20–80 μg/mL) for 2 h. Viability tests done 48 h later showed that cells transfected with CMV-p65 had acquired chemoresistance as compared with cells transfected with CMV-βgal alone (at BCNU concentrations of 40 and 80 μg/mL; _P_ < 0.001). Addition of CMV-ΔNIκB abrogated this resistance (_P_ > 0.4; Fig. 5B).

Figure 5.

NF-κB–related chemoresistance to alkylating agent (BCNU). A, glioma cell lines A172, U87MG, and T98G, were exposed for 2 h to increasing concentrations of BCNU, as indicated. B, HEK293 cells transiently transfected with CMV-βgal alone or along with CMV-p65 or with CMV-p65 and CMV-ΔNIκB. In the _O_6-benzylguanine (BG)–treated groups, _O_6-benzylguanine was added to the cell medium 4 to 6 h after transfection at a final concentration of 80 μmol/L, and the cells were incubated with the agent for the duration of the experiment. BCNU was added 24 h after transfection. A viability test was done 48 h after the BCNU treatment. Columns, mean percent survival of three different experiments; bars, SD. Cells with endogenous high NF-κB activity (A) or with forced p65 (B) display lower sensitivity to nitrosourea treatment than cells with low NF-κB activity. Depletion of MGMT by _O_6-benzylguanine restores chemosensitivity.

Figure 5.

Close modal

To test whether the chemoresistance acquired by p65 is associated with MGMT and not with other NF-κB target genes, MGMT was depleted from the HEK293 transfected cells by administration of _O_6-benzylguanine at 4 to 6 h after cell transfection. The results show that _O_6-benzylguanine treatment augmented cell sensitivity to BCNU (Fig. 5B), bringing the control cells and the p65-transfected cells to display the same sensitivity. Therefore, it can be assumed that MGMT depletion restored the cell sensitivity to BCNU (at BCNU concentrations of 40 and 80 μg/mL; P < 0.001) in the p65-induced resistant cells.

Discussion

MGMT is the only gene known to be critical to direct reversal repair of the biological effects of _O_6-methylguanine on DNA. In this study, we show that NF-κB plays a major role in the regulation of MGMT, a function not recognized before. It is known that NF-κB is activated as part of the DNA damage response. The role of Rel/NF-κB factors in the signaling cascade that is initiated with double-stranded DNA breaks has become more clear in the last few years (27–30). It was recently shown that in response to _O_6-alkylating agents, TNFα-induced protein 3 is involved in a putative cytoplasmic signaling cascade that mediates NF-κB activation (31); however, the NF-κB–mediated pathway in response to stimuli induced by alkylating agents remains largely obscure.

The results of our transient expression experiments suggest that the NF-κB/p65 homodimer is a significant factor involved in MGMT regulation because transfection of p65 into HEK293 resulted in a 55-fold increase in the induction of MGMT expression, compared with an 8-fold increase in induction using AP-1/c-Jun. Cotransfection of p65 with AP-1/c-Jun or p50 did not exert a synergistic effect on p65-derived MGMT expression (Fig. 2A and B). The indication that the p65 homodimer is the main player in NF-κB–induced MGMT expression is also supported by EMSA analysis, showing that both NF-κB sequences in the MGMT promoter can bind a complex of similar size to the complex that binds to the canonical NF-κB site. Furthermore, the addition of mAbs against the active form of p65 to the probe-extract mixture resulted in a complete supershift. The EMSA results imply that the −90 binding site, which overlaps a Sp1 site, probably binds both NF-κB and Sp1 (Fig. 1C). Therefore, the relationship between NF-κB and Sp1 might affect MGMT induction. However, additional studies are required to determine whether the p65 homodimer alone is sufficient to regulate MGMT expression or whether its association with other factors, such as Sp1, AP-1, or additional Rel/NF-κB proteins, is of importance.

MGMT promoter methylation has been associated with prolonged survival in patients with various cancers, especially malignant gliomas (7–9). A recent study analyzed time to tumor progression in relation to MGMT promoter methylation in patients with glioblastoma moltiforme and found significantly improved results in patients with methylated MGMT promoter (7–9). It is conceivable that methylated regions of the MGMT promoter are located in closed nucleosome structures (32), impeding transcription factor access to the promoter (33, 34). DNA methylation may affect transcription through this mechanism. Thus, we posed the question: Does MGMT promoter methylation impair NF-κB–induced MGMT expression in case of ectopic or constitutive NF-κB activation? We found that transient expression of p65 in HEK293 cells induces a 55-fold increase in MGMT mRNA (Fig. 2A) although the cells bear a methylated MGMT promoter (Fig. 4A). We also found a significant concordant relationship between the nuclear pattern of activated NF-κB/p65 molecules and MGMT expression in human oligodendrogliomas (Fig. 3A and B). However, in these tumors, there was no correlation between the methylation status of the MGMT promoter and MGMT expression (Fig. 4B). This was confirmed also in the glioma cell lines (Figs. 3C and D and 4A), indicating that MGMT promoter methylation is not an overruling factor in relation to MGMT expression.

Our findings imply that tumors displaying high constitutive NF-κB activity (23) should also exhibit high MGMT expression. Furthermore, the NF-κB–derived MGMT expression should not be dependent on MGMT methylation status as shown in the current study. This could account for conflicting observations on the correlation between MGMT promoter methylation and gene expression (25, 26).

MGMT is overexpressed in many types of human tumors (for review, see ref. 35). The expression level can serve as a major predictor of chemosensitivity to alkylating agents such as temozolomide and BCNU (36, 37). We found that either high constitutive NF-κB activity (Fig. 5A) or ectopic p65 (Fig. 5B) stimulated significant cellular resistance to BCNU treatment, most probably through induction of MGMT expression. On the other hand, inhibition of NF-κB activity by the dominant ΔNIκB sensitized the cells to BCNU (Fig. 5B). These results are confirmed by other studies showing that insertion of a mutant NF-κB transgene inhibitor increases the efficacy of BCNU in human gliomas (3, 24, 38).

It was previously shown that NF-κB is activated in response to alkylating agents and that high NF-κB activation is associated with chemoresistance (3). Our results suggest that MGMT is most likely the major player in NF-κB–induced chemoresistance mediated by alkylating agents because we found that inhibition of MGMT activity by _O_6-benzylguanine completely abrogated the p65-induced chemoresistance. Thus, we propose a model for a novel DNA damage repair molecular mechanism induced by NF-κB in response to exposure to alkylating agents. According to this simplified model (Fig. 6), cell exposure to alkylating agent (A) induces NF-κB activation (B, C), which is followed by augmented MGMT expression (D). MGMT then removes the alkyl/methyl adducts from the O6 atom of DNA guanine (E) and the guanine is restored (F).

Figure 6.

A proposed simplified model for a DNA damage repair mechanism induced by NF-κB in response to alkylating agents. Cell exposure to alkylating agent (A) induces NF-κB activation and subsequent degradation of IκB (B), allowing the translocation of NF-κB into the nucleus (C) and stimulation of MGMT expression (D). MGMT removes the alkyl/methyl adducts from the O6 atom of DNA guanine (E) and the guanine is restored (F).

Figure 6.

Close modal

Several therapeutic inhibitors that block NF-κB activation are under development (for review, see ref. 39). Our findings indicate that such inhibitors would be of great clinical value for sensitizing MGMT-positive expressing cells to alkylating chemotherapeutic treatment and may assist in overcoming treatment-induced chemoresistance.

In conclusion, our findings provide the first evidence that MGMT is a target gene for NF-κB. It is possible that MGMT is only the first example of the role played by NF-κB in the regulation of DNA repair mechanisms. NF-κB involvement in DNA damage repair may include additional DNA repair genes.

Acknowledgments

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Y. Ben-Neriah and D. Melloul for providing various plasmids, K. Baum for technical assistance, T. Bdolah-Abram for help in the analysis of the statistical data, Prof. B. Kaina and Drs. O. Gerlitz and A. Mahler for reviewing the manuscript and useful comments, and Prof. S. Mitra for providing the MGMT-Luc plasmid and for the support and encouragement given so kindly.

References

Ghosh S, May MJ, Kopp EB. NF-κB and Rel proteins: evolutionarily conserved mediators of immune responses.

Annu Rev Immunol

1998

;

225

–60.

Senftleben U, Karin M. The IKK/NF-κB pathway.

Crit Care Med

2002

;

S18

–26.

Weaver KD, Yeyeodu S, Cusack JC, Jr., et al. Potentiation of chemotherapeutic agents following antagonism of nuclear factor κB in human gliomas.

J Neurooncol

2003

;

187

–96.

Srivenugopal KS, Yuan XH, Friedman HS, Ali-Osman F. Ubiquitination-dependent proteolysis of _O_6-methylguanine-DNA methyltransferase in human and murine tumor cells following inactivation with _O_6-benzylguanine or 1,3-bis(2-chloroethyl)-1-nitrosourea.

Biochemistry

1996

;

1328

–34.

Citron M, Decker R, Chen S, et al. _O_6-Methylguanine-DNA methyltransferase in human normal and tumor tissue from brain, lung, and ovary.

Cancer Res

1991

;

4131

–4.

Washington WJ, Foote RS, Dunn WC, et al. Age-dependent modulation of tissue-specific repair activity for 3-methyladenine and _O_6-methylguanine in DNA in inbred mice.

Mech Ageing Dev

1989

;

–52.

Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma.

N Engl J Med

2005

;

352

997

–1003.

Danam RP, Howell SR, Remack JS, Brent TP. Heterogeneous methylation of the O(6)-methylguanine-DNA methyltransferase promoter in immortalized IMR90 cell lines.

Int J Oncol

2001

;

1187

–93.

Silber JR, Blank A, Bobola MS, et al. Lack of the DNA repair protein _O_6-methylguanine-DNA methyltransferase in histologically normal brain adjacent to primary human brain tumors.

Proc Natl Acad Sci U S A

1996

;

6941

–6.

Dumenco LL, Warman B, Hatzoglou M, et al. Increase in nitrosourea resistance in mammalian cells by retrovirally mediated gene transfer of bacterial _O_6-alkylguanine-DNA alkyltransferase.

Cancer Res

1989

;

6044

–51.

Schold SC, Jr., Brent TP, von Hofe E, et al. _O_6-Alkylguanine-DNA alkyltransferase and sensitivity to procarbazine in human brain-tumor xenografts.

J Neurosurg

1989

;

573

–7.

Wu ZN, Chan CL, Eastman A, Bresnick E. Expression of human _O_6-methylguanine-DNA methyltransferase in a DNA excision repair-deficient Chinese hamster ovary cell line and its response to certain alkylating agents.

Cancer Res

1992

;

–5.

Citron M, White A, Decker R, et al. _O_6-Methylguanine-DNA methyltransferase in human brain tumors detected by activity assay and monoclonal antibodies.

Oncol Res

1995

;

–55.

Kokkinakis DM, Ahmed MM, Delgado R, et al. Role of _O_6-methylguanine-DNA methyltransferase in the resistance of pancreatic tumors to DNA alkylating agents.

Cancer Res

1997

;

5360

–8.

Jaeckle KA, Eyre HJ, Townsend JJ, et al. Correlation of tumor _O_6-methylguanine-DNA methyltransferase levels with survival of malignant astrocytoma patients treated with _bis_-chloroethylnitrosourea: a Southwest Oncology Group study.

J Clin Oncol

1998

;

3310

–5.

Esteller M, Garcia-Foncillas J, Andion E, 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.

Friedman HS, Kokkinakis DM, Pluda J, et al. Phase I trial of _O_6-benzylguanine for patients undergoing surgery for malignant glioma.

J Clin Oncol

1998

;

3570

–5.

Dolan ME, Mitchell RB, Mummert C, et al. Effect of _O_6-benzylguanine analogues on sensitivity of human tumor cells to the cytotoxic effects of alkylating agents.

Cancer Res

1991

;

3367

–72.

Dolan ME, Moschel RC, Pegg AE. Depletion of mammalian _O_6-alkylguanine-DNA alkyltransferase activity by _O_6-benzylguanine provides a means to evaluate the role of this protein in protection against carcinogenic and therapeutic alkylating agents.

Proc Natl Acad Sci U S A

1990

;

5368

–72.

Harris LC, Potter PM, Tano K, et al. Characterization of the promoter region of the human _O_6-methylguanine-DNA methyltransferase gene.

Nucleic Acids Res

1991

;

6163

–7.

Biswas T, Ramana CV, Srinivasan G, et al. Activation of human _O_6-methylguanine-DNA methyltransferase gene by glucocorticoid hormone.

Oncogene

1999

;

525

–32.

Boldogh I, Ramana CV, Chen Z, et al. Regulation of expression of the DNA repair gene _O_6-methylguanine-DNA methyltransferase via protein kinase C-mediated signaling.

Cancer Res

1998

;

3950

–6.

Amit S, Ben-Neriah Y. NF-κB activation in cancer: a challenge for ubiquitination- and proteasome-based therapeutic approach.

Semin Cancer Biol

2003

;

–28.

Baldwin AS. Control of oncogenesis and cancer therapy resistance by the transcription factor NF-κB.

J Clin Invest

2001

;

107

241

–6.

Lin Z, Gao M, Zhang X, et al. The hypermethylation and protein expression of p16 INK4A and DNA repair gene _O_6-methylguanine-DNA methyltransferase in various uterine cervical lesions.

J Cancer Res Clin Oncol

2005

;

131

364

–70.

Brell M, Tortosa A, Verger E, et al. Prognostic significance of _O_6-methylguanine-DNA methyltransferase determined by promoter hypermethylation and immunohistochemical expression in anaplastic gliomas.

Clin Cancer Res

2005

;

5167

–74.

Janssens S, Tschopp J. Signals from within: the DNA-damage-induced NF-κB response.

Cell Death Differ

2006

;

773

–84.

Wu ZH, Mabb A, Miyamoto S. PIDD: a switch hitter.

Cell

2005

;

123

980

–2.

Bartek J, Lukas J. Cell biology. The stress of finding NEMO.

Science

2006

;

311

1110

–1.

Wu ZH, Shi Y, Tibbetts RS, Miyamoto S. Molecular linkage between the kinase ATM and NF-κB signaling in response to genotoxic stimuli.

Science

2006

;

311

1141

–6.

Bredel M, Bredel C, Juric D, et al. Tumor necrosis factor-α-induced protein 3 as a putative regulator of nuclear factor-κB-mediated resistance to _O_6-alkylating agents in human glioblastomas.

J Clin Oncol

2006

;

274

–87.

Watts GS, Pieper RO, Costello JF, et al. Methylation of discrete regions of the _O_6-methylguanine DNA methyltransferase (MGMT) CpG island is associated with heterochromatinization of the MGMT transcription start site and silencing of the gene.

Mol Cell Biol

1997

;

5612

–9.

Patel SA, Graunke DM, Pieper RO. Aberrant silencing of the CpG island-containing human _O_6-methylguanine DNA methyltransferase gene is associated with the loss of nucleosome-like positioning.

Mol Cell Biol

1997

;

5813

–22.

Costello JF, Futscher BW, Kroes RA, Pieper RO. Methylation-related chromatin structure is associated with exclusion of transcription factors from and suppressed expression of the _O_6-methylguanine DNA methyltransferase gene in human glioma cell lines.

Mol Cell Biol

1994

;

6515

–21.

Margison GP, Povey AC, Kaina B, Santibanez Koref MF. Variability and regulation of _O_6-alkylguanine-DNA alkyltransferase.

Carcinogenesis

2003

;

625

–35.

Hermisson M, Klumpp A, Wick W, et al. _O_6-Methylguanine DNA methyltransferase and p53 status predict temozolomide sensitivity in human malignant glioma cells.

J Neurochem

2006

;

766

–76.

Pollack IF, Hamilton RL, Sobol RW, et al. _O_6-Methylguanine-DNA methyltransferase expression strongly correlates with outcome in childhood malignant gliomas: results from the CCG-945 Cohort.

J Clin Oncol

2006

;

3431

–7.

Orlowski RZ, Baldwin AS, Jr. NF-κB as a therapeutic target in cancer.

Trends Mol Med

2002

;

385

–9.

Karin M, Yamamoto Y, Wang QM. The IKK NF-κB system: a treasure trove for drug development.

Nat Rev Drug Discov

2004

;

–26.

2007

956 Views

115 Web of Science

Novel Mechanism whereby Nuclear Factor κB Mediates DNA Damage Repair through Regulation of O6-Methylguanine-DNA-Methyltransferase (original) (raw)

Abstract

Introduction

Materials and Methods

Results

Discussion

Acknowledgments

References

Citing articles via

Email alerts

Novel Mechanism whereby Nuclear Factor κB Mediates DNA Damage Repair through Regulation of O6-Methylguanine-DNA-Methyltransferase (original) (raw)

Abstract

Introduction

Materials and Methods

Results

Discussion

Acknowledgments

References

Citing articles via

Email alerts

Related Content