1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

(Cancer Sci 2010; 101: 767–773)

Glutathione S-transferase μ (GSTM1) is mainly known as a detoxification enzyme but it has also been shown to be a negative regulator of apoptosis-related signaling cascades. Recently GSTM1 has been reported to be a significant risk factor for hematological relapse in childhood acute lymphoblastic leukemia, although the underlying mechanism remains largely unknown. Glucocorticoids play a crucial role in the treatment of childhood acute lymphoblastic leukemia, therefore we hypothesized that GSTM1 plays important roles in glucocorticoid-induced apoptotic pathways. To clarify the relationship between GSTM1 and drug resistance, GSTM1 was transfected into a T-acute lymphoblastic leukemia cell line, CCRF-CEM (CEM), and we established the GSTM1-expressing cell lines CEM/M1-4 and CEM/M1-9. Transduction of GSTM1 into CEM selectively decreased cellular sensitivity to dexamethasone in a manner that was independent of glutathione conjugation, but was due to apoptosis inhibition. Dexamethasone-induced p38-MAPK and Bim activation were concomitantly suppressed. Interestingly, nuclear factor kappa b (NF-κB) p50 activity was upregulated in GSTM1-expressing CEM. Inhibition of NF-κB by the pharmacological agent BAY11-7082 greatly enhanced the sensitivity of the GSTM1-expressing CEM to dexamethasone and was accompanied by an increase in Bim expression. Thus, we propose that GSTM1, a novel regulator of dexamethasone-induced apoptosis, causes dexamethasone resistance by suppression of Bim through dual mechanisms of both downregulation of p38-MAPK and upregulation of NF-κB p50.

Glutathione S-transferases (GST) are phase II detoxification enzymes that catalyze the conjugation of glutathione to xenobiotic compounds. Many anticancer drugs are substrates for GST and, therefore, overexpression of GST is responsible for resistance to anticancer drugs in tumor cell lines.(1,2) In addition to this catalytic activity, certain GST isozymes (GSTπ and GSTμ) can regulate mitogen-activated protein kinases (MAPK), which are involved in stress responses, apoptosis, and proliferation.(2,3) It has been reported that GSTM1 directly interacts with apoptosis signal-regulating kinase (ASK) 1 and MAPK kinase kinase (MEKK) 1, and modulates p38-MAPK signaling cascades.(3–6) GSTM1 may cause resistance to anticancer drugs, not only by catalyzing glutathione conjugation but also by downregulating the p38-MAPK pro-apoptotic pathway.

The GSTμ gene (GSTM1) is polymorphic in humans; 40–60% of the population has a homozygous deletion of this gene. This polymorphism, resulting in the absence of enzymatic activity, has been reported to be associated with the clinical outcome of various malignancies.(7–13) In childhood acute lymphoblastic leukemia (ALL), some studies have suggested that patients possessing GSTM1 have a greater risk of hematological relapse.(7,8) The precise underlying mechanism remains unknown.

Glucocorticoids play a crucial role in the treatment of childhood ALL, and patient response is an important determinant of clinical outcome.(14,15) Although the complete mechanism of glucocorticoid-induced apoptosis is still unclear, some of the critical pathways involved have recently been identified. The p38-MAPK pathway is an important signaling system for induction of apoptosis.(16) Glucocorticoids activate p38–MAPK and induce mRNA transcription and synthesis of Bim, a BH3-only pro-apoptotic Bcl-2 family member.(17,18) Bim activates Bax directly and the induction of Bim correlates with the onset of glucocorticoid-induced apoptosis.(19) On the other hand, glucocorticoids also interact with NF-κB through the glucocorticoid-receptor complex, thereby repressing NF-κB transcriptional activity.(20,21) As it has been reported that NF-κB is constitutively activated in childhood ALL,(22) the function of NF-κB may also be important for glucocorticoid responses.

In the present study, we hypothesized that GSTM1 plays important roles in glucocorticoid-induced apoptotic pathways, causing resistance, particularly in ALL. To verify this hypothesis, the T-lymphoblastic leukemia cell line CCRF-CEM (CEM) was stably transduced with GSTM1 and two GSTM1-expressing CEM clones (CEM/M1-4 and CEM/M1-9) were established. The effect of GSTM1 on cellular sensitivity to anticancer drugs was examined and GSTM1-expressing CEM clones were moderately resistant to dexamethasone.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Cell culture and reagents.  The human T-lymphoblastic leukemia cell lines CEM, Jurkat, and HSB2 were cultured in RPMI-1640 with 10% FCS at 37°C. Dexamethasone was a generous gift from Banyu Pharmaceutical Co. (Tokyo, Japan). G-418 was purchased from Invitrogen (Carlsbad, CA, USA), restriction enzymes were purchased from Takara Bio (Otsu, Japan), BAY11-7082 and the anti-GSTM1-1 antibody were from Merck (Darmstadt, Germany), antibodies to p38, phospho-p38 (Thr180/Tyr182), Bim, and phospho-Bim (Ser69) were from Cell Signaling Technology (Beverly, MA, USA), and β-actin and the other molecular biology reagents were from Sigma-Aldrich (St Louis, MO, USA).

Retrovirus production and transduction.  Human GSTM1 cDNA was kindly provided by Dr C.S. Morrow (Wake Forest University School of Medicine, Winston-Salem, NC, USA). Recombinant GSTM1 retrovirus particles were obtained using the Retrovirus Packaging Ampho Kit (Takara Bio). CEM was seeded into RetroNectin (5 μg/cm2)-coated wells and preloaded with the retrovirus particles. After 48 h, cells resistant to 1.5 mg/mL G-418 were selected. CEM infected with an empty vector was used as a negative control.

Cell viability and apoptosis assay.  Cell viability was evaluated using the XTT assay kit (Roche, Basel, Switzerland) as previously described.(23) Apoptotic cells were detected by the Annexin V–FITC Apoptosis Detection kit (Roche) according to the manufacturer’s instructions. Fluorescence signals were detected using FACScan (Beckman Coulter, Marseille, France) by recording 10 000 events for each analysis. The incubation duration was determined as previously described.(24,25)

GST activity assay.  GST activity was measured using the GST-Tag assay kit (Takara Bio) as previously described.(26) GST concentration was determined from the standard curve.

Genotyping of GSTM1.  Genomic DNA was extracted from 2 × 106 cells using the DNeasy Blood and Tissue kit (Qiagen, Hilden, Germany). PCR amplification and detection were carried out as previously described.(27) DNA from subjects with positive GSTM1 alleles yielded a 215-bp fragment. Jurkat DNA was used as a GSTM1-positive control.

Quantitative RT-PCR analysis.  Total RNA was extracted from the cells using the RNeasy kit (Qiagen). First-strand cDNA was synthesized from 450 ng of total RNA using random nonamers. Quantitative RT-PCR was carried out using SYBR Green chemistry with the LightCycler system (Roche). Specific PCR primers were designed by Takara Bio and are listed in Table 1.

Table 1.   Sequence of oligonucleotides

Western blot analysis.  Western blotting was carried out as follows. Whole-cell lysates (20–35 μg) were separated by 7.5–12.5% SDS-PAGE and transferred to a polyvinylidene fluoride membrane. The membrane was blocked with 5% non-fat milk or 3% BSA for 1 h followed by incubation with the primary antibody: anti-GSTM1 (1:500), anti-phosho-p38 (1:750), anti-p38 (1:750), anti-phospho-Bim (1:750), anti-Bim (1:750), or anti-β-actin (1:2000). For visualization of specific proteins the blots were stained with horseradish peroxidase-conjugated secondary antibody (1:1000) and the signal developed with the ECL kit (GE-Amersham, Pittsburgh, PA, USA). Blots were stripped with the Restore Western Blot Stripping Buffer (Thermo Fisher Scientific, Waltham, MA, USA) and reprobed with different antibodies.

Electroporation of siRNA.  Synthetic sense and antisense oligoribonucleotides were synthesized by Takara Bio. The sense strand sequences were as follows: 5′-CCAUGGACAACCAUAUGCA (dTdT)-3′, for GSTM1 siRNA, and 5′-AUCCACCGGCAUAAUAGCA (dTdT)-3′, as scrambled siRNA.

Cells were washed twice with PBS and resuspended at 107 cells/mL in serum-free RPMI-1640 medium containing 25 mm HEPES. A 200-μL aliquot of cells was added to a 0.2-cm gap electroporation cuvette along with 10 μL of a 20 μm stock of siRNA and then incubated for 20 min at room temperature. Cells were then electroporated with a Gene Pulser Xcell (Bio-Rad, Hercules, CA, USA) using the pre-set mammalian Jurkat protocol. After electroporation, cells were incubated with an additional 2 mL RPMI-1640 medium (ATCC, Rockville, MD, USA). The final concentration of siRNA was 100 nm.

Assay of glucocorticoid receptor activity, NF-κB activity, and inhibitor-κBα phosphorylation.  Cells were harvested and whole-cell lysates were prepared according to the manufacturer’s instructions. The DNA-binding activity of glucocorticoid receptor (GR) and NF-κB was evaluated by the TransAM GR and NF-κB Kits (Active Motif, Carlsbad, CA, USA) by measurement of the binding of cell lysates to the consensus DNA binding sites, 5′-GGTACAnnnTGTTCT-3′ and 5′-GGGACTTTCC-3′, respectively. Phosphorylation of inhibitor-κBα (IκBα) on Ser32 was evaluated with a phospho-ELISA kit (BioSource, Camarillo, CA, USA).

Statistical analysis.  Statistical significance was determined by Student’s t-test, the Mann–Whitney U-test, or ANOVA when appropriate. Statistical significance was defined as < 0.05.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

GSTM1-expressing CEM exhibits decreased sensitivity to antileukemic drugs.  To investigate the role of GSTM1 in drug resistance, GSTM1-negative CEM (Fig. 1a) was transduced with a human GSTM1-expressing retrovirus and the effect of GSTM1 expression on cellular drug sensitivity was assayed. Two CEM clones that expressed the GSTM1 protein (CEM/M1-4 and CEM/M1-9) were used in this study (Fig. 1b). The GST activity of these clones was significantly higher than that of the CEM/mock control (Fig. 1c). CEM/M1-9 exhibited decreased sensitivity to chlorambucil, melphalan, carmustine, arsenic trioxide, and dexamethasone in a cell viability (XTT) assay (Table 2). The decreased sensitivity was most dramatic for dexamethasone, for which the relative resistance was increased 12-fold, whereas the relative resistance to chlorambucil, melphalan, carmustine, and arsenic trioxide was increased 1.5- to 2.1-fold. There was no change in cellular sensitivity to major antileukemic drugs including daunorubicin, doxorubicin, vincristine, etoposide, cytarabine, 4-hydroperoxycyclophosphamide (4-HC), and l-asparaginase. A similar response to the same drugs was observed in CEM/M1-4 (Table 2).


Figure 1.  Viral transduction of glutathione S-transferase M1 (GSTM1) can influence dexamethasone sensitivity in an intracellular glutathione-independent manner. (a) Genomic PCR analysis of the GSTM1 gene in the indicated cell lines. β-actin was evaluated as an internal control. (b) Two CCRF-CEM (CEM) clones expressing GSTM1 (CEM/M1-4 and CEM/M1-9) were selected by western blot analysis. Similar results were obtained in repeat blotting experiments. (c) GST activity of GSTM1-expressing clones and control cells assayed by a colorimetric assay. Data are expressed as the mean ± SEM of three experiments. *< 0.01; **< 0.05 compared with CEM/mock. (d) Effect of the glutathione synthesis inhibitor (BSO) on cellular sensitivity to dexamethasone. The cells were incubated for 24 h with or without 50 μm BSO, exposed for a further 72 h to various concentrations of dexamethasone and then assayed in the XTT cell viability assay. Data are the mean ± SEM of at least three separate experiments. Treatment with BSO alone had no effect on cell viability. (e) The cells were incubated for 24 h with or without 50 μm BSO, exposed for a further 72 h to various concentrations of the dexamethasone, 4-HC, carmustine, or chlorambucil and then assayed in the XTT cell viability assay. The data represent the ratio of the IC50 in the absence of BSO to the IC50 in the presence of BSO.

Download figure to PowerPoint

Table 2.   Drug response-profile of glutathione S-transferase M1 (GSTM1)-expressing CCRF-CEM (CEM) cells
DrugIC50 [Relative resistance]
  1. GSTM1-transduced CEM/M1-9, CEM/M1-4, and empty vector-transduced CEM/mock cells were cultured for 72 h with various concentrations of drugs, and the IC50 for each drug was calculated using the XTT assay. Data are presented as the mean of at least three separate experiments. Relative resistance represents the ratio of the IC50 of CEM/M1-9 or CEM/M1-4 to the IC50 of CEM/mock. **< 0.01. N.D., not done.

Daunorubicin (nm)24.926.6 [1.07]24.5 [0.98]
Doxorubicin (nm)30.334.3 [1.13]32.3 [1.06]
Etoposide (nm)184192 [1.04]N.D.
Cytarabine (μm)17.917.6 [0.98]N.D.
Vincristine (nm)8.418.35 [0.99]8.38 [1.00]
4-HOO-cyclophosphamide (μm)3.814.4 [1.15]N.D.
l-Asparaginase (KU)0.110.10 [0.91]N.D.
Dexamethasone (μm)0.222.61 [11.9]**1.42 [6.45]**
Melphalan (μm)2.464.03 [1.64]**N.D.
Carmustine (μm)45.995.7 [2.08]**62.9 [1.37]**
Chlorambucil (μm)8.1212.2 [1.50]**N.D.
Arsenic trioxide (μm)1.282.03 [1.58]**N.D.

The response to dexamethasone was not altered by the addition of buthionine sulfoximine (BSO), an inhibitor of glutathione synthesis (Fig. 1d), whereas the sensitivity to carmustine, chlorambucil, and 4-HC was enhanced almost 3-fold in the presence of BSO (Fig. 1e). Treatment with BSO alone (50 μm) had no effect on cell viability. Forced expression of GSTM1 had no effect on the concentration of intracellular glutathione (data not shown). These results indicate that GSTM1-induced dexamethasone resistance is not due to glutathione-conjugating activity.

GSTM1 expression inhibits dexamethasone-induced apop-tosis.  The cellular response to glucocorticoid treatment is crucial for clinical outcome in childhood ALL. We therefore focused on elucidation of the mechanisms of dexamethasone resistance caused by GSTM1 expression and assayed if GSTM1 expression modulated cellular apoptosis. Treatment with 1 μm dexamethasone for 48 h increased the percentage of Annexin V-positive cells to 15 ± 4% (mean ± SEM) in CEM/mock, but not in GSTM1-expressing clones (Fig. 2a). After a 72-h treatment, the percentage of Annexin V-positive cells increased to 60 ± 3% in CEM/mock, but only to 27 ± 2% and 31 ± 4% in CEM/M1-4 and CEM/M1-9, respectively, indicating that dexamethasone resistance involves apoptosis inhibition. To verify that this resistance truly resulted from the expression of GSTM1, a GSTM1-RNA interference experiment was carried out. Transfection of GSTM1 siRNA into CEM/M1-4 and CEM/M1-9 decreased the expression of the GSTM1 protein to less than half of the control in 48 h, and increased dexamethasone-induced apoptosis (Fig. 2b). The contribution of GSTM1 to inhibition of dexamethasone-induced apoptosis was similarly demonstrated by GSTM1 RNA interference in other T-ALL cell lines, HSB2 and Jurkat, which, as shown in Figure 1(a), express endogenous GSTM1 (Fig. 2c) although cell sensitivity to the dexamethasone was different among the cell lines (IC50: 0.08 μm for HSB2 and 580 μm for Jurkat). These results indicate that GSTM1 is responsible for inhibition of dexamethasone-induced apoptosis.


Figure 2.  Glutathione S-transferase M1 (GSTM1) mediates inhibition of dexamethasone-induced apoptosis. (a) Cells were cultured with or without dexamethasone (DEX; 1 μm) for 48 or 72 h. The percentage of apoptotic cells was determined by Annexin V-staining. Data from at least three independent experiments were expressed as the mean ± SEM. *< 0.01 compared with CCRF-CEM (CEM)/mock at the indicated times. (b) GSTM1-specific siRNA (100 nm) or scrambled siRNA (100 nm) was transfected into CEM/M1-4 or CEM/M1-9. After 48 h, whole-cell lysates were analyzed by immunoblotting with anti-GSTM1 or the loading control anti-β-actin antibody (left). Comparable results were obtained in three experiments. Twenty-four hours after electroporation, cells were exposed to DEX (1 μm) and apoptotic cells were measured at the indicated time points (right). Data are the mean ± SEM of at least three separate experiments. *< 0.01 compared with respective controls. (c) HSB2 and Jurkat cells were transfected with GSTM1 siRNA and repression of GSTM1 protein expression was confirmed as described in (b) (left). Twenty-four hours after electroporation, cells were treated with DEX (0.05 μm for HSB2 or 420 μm for Jurkat) and stained with Annexin V at the indicated time points (right). Data are the mean ± SEM of at least three separate experiments. *< 0.01; **< 0.05 compared with respective controls.

Download figure to PowerPoint

We therefore investigated GSTM1-mediated anti-apoptotic mechanisms in the GSTM1-overexpressing clones.

GSTM1 expression does not appear to alter GR func-tion.  Glucocorticoid resistance in ALL cell lines almost invariably involves defects in GR function.(24,28) However, there was no detectable difference in glucocorticoid response element (GRE)-binding activity between CEM/M1-9, CEM/mock, and parental CEM (Fig. 3a). As a further assay of GR function, we next evaluated GR-mediated transactivation by measurement of the expression of mRNA for GR and thioredoxin-interacting protein that are primary upregulated via different GRE.(24,29) Again, no difference in the kinetics of expression of these mRNA was apparent between CEM/mock and CEM/M1-9 (Fig. 3b). These data indicate that the possibility of GR dysfunction can be excluded as a cause of resistance in GSTM1-expressing CEM.


Figure 3.  Glutathione S-transferase M1 (GSTM1) transduction did not alter glucocorticoid receptor (GR) activity but suppressed the dexamethasone-induced p38-MAPK/Bim pathway. (a) Binding activity of GR to GRE was assessed by ELISA. Wholecell lysates were subjected to the assay, and activities were expressed as a percentage of parental CCRF-CEM (CEM). Mean ± SEM was calculated from three independent experiments. There was no statistical significance (> 0.05). (b) Cells were treated with 1 μm dexamethasone (DEX) for 24 or 48 h and the expression of GR and thioredoxin-interacting protein (TXNIP) mRNA was then measured by quantitative RT-PCR. Relative expression is shown as the mean ± SEM of three separate experiments. N.S., not significant compared with CEM/mock. (c) CEM/mock, CEM/M1-9, and CEM/M1-4 cells were left untreated (cont) or were treated with 1 μm DEX for 8 or 12 h following which p-p38, p38, and β-actin were detected by western blot analysis. Similar results were obtained in repeat experiments. The expression of dual-specificity phosphatase 1 (DUSP1) mRNA was measured by quantitative RT-PCR during 1 μm DEX for 24 or 48 h. Relative expression is shown as the mean ± SEM of three separate experiments. (d) GSTM1-specific siRNA (100 nm) or scrambled siRNA (100 nm) was transfected into CEM/M1-9. After 48 h, cells were left untreated (cont) or were treated with 1 μm DEX for 8 or 12 h following which p-p38, p38, and β-actin were detected by western blot analysis. Similar results were obtained in repeat experiments. (e) CEM/mock and CEM/M1-9 were treated with 1 μm DEX for 12 or 24 h and the level of phospho-Bim or Bim expression was determined by immunoblotting. β-actin was evaluated as a loading control. Similar results were obtained in repeat blotting experiments.

Download figure to PowerPoint

GSTM1 downregulates dexamethasone-induced p38-MAPK and Bim activation.  We next determined if GSTM1 might modulate signaling pathways known to be regulated by dexamethasone. Dexamethasone-induced p38-MAPK activation is well documented in CEM,(16) and western blot analysis revealed a time-dependent phosphorylation of p38-MAPK in CEM/mock following dexamethasone treatment for 8 and 12 h (Fig. 3b). However, in GSTM1-expressing CEM, the level of p38-MAPK phosphorylation was not increased even after 12 h of treatment. The low level of p38-MAPK phosphorylation in GSTM1-expressing CEM was not due to enhanced negative regulation of p38-MAPK phosphorylation by the dual-specificity phosphatase 1 (DUSP1)(30) because dexamethasone-induced DUSP1 expression in GSTM1-expressing CEM was not higher than that of CEM/mock (Fig. 3c). Moreover, knockdown of GSTM1 expression in CEM/M1-9 increased phosphorylation of p38-MAPK (Fig. 3d), indicating that GSTM1 is involved in downregulation of dexamethasone-induced p38-MAPK activation.

We next compared the activation of Bim, a pro-apoptotic member of the Bcl-2 family of the intrinsic apoptosis pathway whose synthesis and phosphorylation are known to be regulated through the p38-MAPK pathway,(17–19,31) in CEM/mock and CEM/M1-9. Phosphorylation of Bim was increased in CEM/mock but not in CEM/M1-9 following treatment with dexamethasone (Fig. 3e), which correlates with phospho-p38-MAPK levels in these cells (Fig. 3c). The expression of Bim was also increased by dexamethasone treatment in CEM/mock, but not in CEM/M1-9 (Fig. 3e).

These results suggest that the expression of GSTM1 inhibits the p38-MAPK-Bim pathway during dexamethasone-induced apoptosis.

NF-κB p50 activity and Bcl-3 expression are increased in GSTM1-expressing CEM.  The transcription factor NF-κB network is also an important pathway that interacts with, and affects, glucocorticoid signaling.(20,21,32) As shown in Figure 4(a), the DNA-binding activity of the NF-κB p50 subunit was higher in both CEM/M1-4 and CEM/M1-9 than in CEM/mock, whereas the activity of the NF-κB p65 subunit, which is known to interfere with glucocorticoid signaling,(20) was similar in these cells. The high activity of p50 in CEM/M1-4 and CEM/M1-9 was not altered even after dexamethasone treatment (data not shown). We could not detect the activity of the p52 subunit in these cells using ELISA (data not shown).


Figure 4.  NF-κB p50 is responsible for glutathione S-transferase M1 (GSTM1)-induced resistance to dexamethasone. (a) The DNA binding activity of NF-κB p50 and p65 subunits in the indicated CCRF-CEM (CEM) cells was measured by ELISA. Relative activities are shown as the mean ± SEM of three separate experiments. **< 0.05, and N.S., not significant, compared with CEM/mock. (b) Bcl-3 mRNA expression was measured by quantitative RT-PCR. Data are the mean ± SEM of relative expression levels in four independent experiments. *< 0.01; **< 0.05 compared with CEM/mock. (c) Cells were treated with or without 3 μm BAY11-7082 (BAY) for 3 h, and the activity of p50 and p65 was measured by ELISA. Relative activities are shown as the mean ± SEM of three separate experiments. (d) Cells were treated for 12 h with or without BAY (3 μm), exposed for a further 36 h to 1 μm dexamethasone or medium and the amount of Bim protein was then determined by western blot analysis. Similar results were obtained in repeat experiments. (e) Cell viability was measured by the XTT assay and the mean IC50 was calculated from at least three separate experiments. *< 0.01 compared with respective controls.

Download figure to PowerPoint

In many tumor cells, the constitutively activated IκB kinase complex induces degradation of IκB proteins, resulting in NF-κB activation.(22,33,34) However, no statistical difference in the phosphorylation of IκBα was observed between CEM/mock and CEM/M1-9 (data not shown). Bcl-3 has been implicated in NF-κB activation by its ability to stabilize DNA binding of the p50–p50 homodimer ([p50]2) in the absence of IκB phosphorylation.(35,36) This observation suggested that Bcl-3 might be causally related to p50 activation in GSTM1-expressing CEM. We therefore compared the expression of Bcl-3 mRNA between CEM/mock and GST-expressing CEM. As expected, the level of Bcl-3 mRNA was higher in CEM/M1-4 and CEM/M1-9 than in CEM/mock (Fig. 4b).

Suppression of NF-κB restores dexamethasone sensitivity by upregulation of Bim expression.  We further investigated the involvement of highly activated NF-κB p50 in dexamethasone resistance mediated by GSTM1 with the NF-κB-inhibitor, BAY11-7082. Treatment with 3 μm of BAY11-7082 decreased p50 activity in both CEM/M1-4 and CEM/M1-9 down to a level comparable to that in CEM/mock (Fig 4c). In contrast, the inhibitor-induced decrease in p65 activity was modest and was comparable in these cells. In the presence of BAY11-7082, dexamethasone treatment increased the expression of Bim in association with an increase in apoptosis. The increase in Bim was more prominent in CEM/M1-4 and CEM/M1-9 than in CEM/mock in the setting of the NF-κB-inhibition (Fig 4d).

Notably, treatment with BAY11-7082 decreased the IC50 value of CEM/mock, CEM/M1-4, and CEM/M1-9 for dexamethasone (from 1.42 to 0.06 μm in CEM/M1-4, from 2.61 to 0.06 μm in CEM/M1-9, and from 0.22 to 0.03 μm in CEM/mock), thereby decreasing the relative resistance up to 2-fold (Fig 4e). The IC50 value of both CEM/M1-4 and CEM/M1-9 treated with BAY11-7082 were lower than that of CEM/mock without inhibitor treatment.

These results demonstrate that, in GSTM1-expressing CEM, the highly activated NF-κB p50 is critically involved in resistance to dexamethasone in association with a decrease in Bim expression.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

In the present study, we demonstrate new properties of the GST isozyme GSTM1, indicating an inhibitory action on dexamethasone-induced apoptosis by dual mechanisms. GSTM1 is responsible for regulation not only of the p38-MAPK and Bim pathways but also of the NF-κB p50 pathway.

To investigate the functional role of GSTM1 in leukemia cells, we chose a human lymphoblastic leukemia cell line, CEM, because CEM is glucocorticoid-sensitive and exhibits a GSTM1 null genotype (Fig. 1a). Viral transduction of GSTM1 into CEM dramatically decreased cellular sensitivity to dexamethasone in two established cell lines (Table 2), and knockdown of the transduced GSTM1 restored its sensitivity (Fig. 2b). This effect was ascribed to GSTM1-derived anti-apoptotic activity, but not to the general property of glutathione conjugation by GST (Fig. 1d).

Additionally, GSTM1-derived anti-apoptosis was also observed in the other T-ALL cell lines, HSB2 and Jurkat, which endogenously express GSTM1 (Fig. 2c). These data suggest the possibility that GSTM1 is involved in cellular resistance to dexamethasone-induced apoptosis in T-ALL.

Based on the data obtained from the experiment with enforced expression of GSTM1, we propose a model of GSTM1-regulated dexamethasone-induced apoptosis that is outlined in Fig. 5, in which GSTM1 inhibits signals downstream of GR that converge on the activation of p38-MAPK and Bim. Dexamethasone treatment activates p38-MAPK and Bim, and consequently causes apoptosis in several leukemia and lymphoma cell lines.(17–19,37) In contrast, GSTM1 is known to suppress stress-induced p38-MAPK activation via inhibition of the MAP3 kinases ASK1 and MEKK1(3–6) and we have shown that dexamethasone-induced phosphorylation of p38-MAPK and Bim were also suppressed in GSTM1-expressing CEM (Fig. 3c–e). Therefore, under conditions of GSTM1 overexpression, p38-MAPK/Bim-mediated apoptosis would be inhibited.


Figure 5.  Hypothetical model of glutathione S-transferase M1 (GSTM1)-induced dexamethasone resistance. In this model, GSTM1 inhibits the p38-MAPK/Bim pro-apoptotic pathway, which is activated by dexamethasone, and upregulates NF-κB p50, which inhibits Bim induction directly and/or indirectly. As a consequence, GSTM1-expressing cells escape dexamethasone-induced apoptosis. The expression of anti-apoptotic genes might be also stimulated. DEX, dexamethasone; p38, p38-MAPK; p-p38, phosphorylated p38.

Download figure to PowerPoint

The NF-κB pathway plays a central role as a regulator of apoptosis. In GSTM1-expressing CEM, upregulation of NF-κB activity could be restricted to the p50 subunit (Fig. 4a). The activated p50 might reflect the enhanced formation of (p50)2-Bcl-3 because the level of Bcl-3 was high in GSTM1-expressing CEM. In lymphoid cells, a glucocorticoid response results in the induction of apoptotic genes and the repression of proliferation and growth genes.(38) The expression of at least one of these genes, Bim, may be repressed by GSTM1-activated p50, as the promoter of Bim contains a binding site for NF-κB but not for GR,(39–42) and furthermore, the expression of Bim was increased by treatment with BAY11-7082 (Fig. 4d). Indeed, in the setting of NF-κB inhibition, the inhibited Bim expression was more strongly expressed. The function of p50 depends on cellular conditions and genes.(35,36,43–45) Therefore the possibility that other genes also contribute to anti-apoptosis cannot be excluded. We compared the expression of other proposed anti-apoptotic genes, such as Bcl-2, Bcl-xL, Mcl1, and CyclinD1, in CEM/mock and CEM/M1-9. However, no difference in the expression of these genes was observed during dexamethasone treatment (data not shown). The model of possible mechanisms underlying GSTM1-mediated apoptosis shown in Fig. 5 is based on a combination of the above concepts and our data.

It is noteworthy that, in GSTM1-expressing CEM, suppression of NF-κB activity by BAY11-7082 partially overcame dexamethasone resistance in association with an increase in Bim expression. BAY11-7082 treatment decreased the activities of both p50 and p65 subunits and increased the expression of Bim in both CEM/mock and CEM/M1-9. These data imply that NF-κB may be intrinsically involved in the control of Bim expression in CEM. Our data suggest that even selective activation of p50 by GSTM1 may be sufficient to confer to CEM the resistance to Bim induction that is stimulated by dexamethasone treatment. Although GSTM1 upregulated the expression of Bcl-3 (Fig. 4b), the mechanism by which GSTM1 activates p50/Bcl-3 is unclear and is currently being investigated in our laboratory.

Our study shows that overexpression of GSTM1 inhibits dexamethasone-induced apoptosis in the ALL cell line CEM, by suppression of Bim, through the dual mechanisms of downregulation of p38-MAPK and upregulation of NF-κB p50. These mechanisms appear to cooperate in induction of dexamethasone resistance. We propose GSTM1 as a novel regulator by dual mechanisms of dexamethasone-induced apoptosis.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

We thank Dr C.S. Morrow for his kind gift of the GSTM1 plasmid, and Dr Hitoshi Inoue for helpful discussions. This study was supported by grants from the Japan Society for the Promotion of Science, Japan (C19591102), a Grant-in-Aid for Young Scientists (Start-up 20890089), and the Japan Research Foundation for Clinical Pharmacology and Research on Publicly Essential Drugs and Medical Devices of JHSF (KHC1021).


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  • 1
    Tew KD, Monks A, Barone L et al. Glutathione-associated enzymes in the human cell lines of the National Cancer Institute Drug Screening Program. Mol Pharmacol 1996; 50: 14959.
  • 2
    Tew KD, Dutta S, Schultz M. Inhibitors of glutathione S-transferases as therapeutic agents. Adv Drug Deliv Rev 1997; 26: 91104.
  • 3
    McIlwain CC, Townsend DM, Tew KD. Glutathione S-transferase polymorphisms: cancer incidence and therapy. Oncogene 2006; 25: 163948.
  • 4
    Dorion S, Lambert H, Landry J. Activation of the p38 signaling pathway by heat shock involves the dissociation of glutathione S-transferase Mu from Ask1. J Biol Chem 2002; 277: 307927.
  • 5
    Cho SG, Lee YH, Park HS et al. Glutathione S-transferase mu modulates the stress-activated signals by suppressing apoptosis signal-regulating kinase 1. J Biol Chem 2001; 276: 1274955.
  • 6
    Ryoo K, Huh SH, Lee YH, Yoon KW, Cho SG, Choi EJ. Negative regulation of MEKK1-induced signaling by glutathione S-transferase Mu. J Biol Chem 2004; 279: 4358994.
  • 7
    Rocha JC, Cheng C, Liu W et al. Pharmacogenetics of outcome in children with acute lymphoblastic leukemia. Blood 2005; 105: 47528.
  • 8
    Hall AG, Autzen P, Cattan AR et al. Expression of mu class glutathione S-transferase correlates with event-free survival in childhood acute lymphoblastic leukemia. Cancer Res 1994; 54: 52514.
  • 9
    Voso MT, D’Alo F, Putzulu R et al. Negative prognostic value of glutathione S-transferase (GSTM1 and GSTT1) deletions in adult acute myeloid leukemia. Blood 2002; 100: 27037.
  • 10
    Khedhaier A, Remadi S, Corbex M et al. Glutathione S-transferases (GSTT1 and GSTM1) gene deletions in Tunisians: susceptibility and prognostic implications in breast carcinoma. Br J Cancer 2003; 89: 15027.
  • 11
    Howells RE, Holland T, Dhar KK et al. Glutathione S-transferase GSTM1 and GSTT1 genotypes in ovarian cancer: association with p53 expression and survival. Int J Gynecol Cancer 2001; 11: 10712.
  • 12
    Medeiros R, Pereira D, Afonso N et al. Platinum/paclitaxel-based chemotherapy in advanced ovarian carcinoma: glutathione S-transferase genetic polymorphisms as predictive biomarkers of disease outcome. Int J Clin Oncol 2003; 8: 15661.
  • 13
    Kishi S, Cheng C, French D et al. Ancestry and pharmacogenetics of antileukemic drug toxicity. Blood 2007; 109: 41517.
  • 14
    Reiter A, Schrappe M, Ludwig WD et al. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86. Blood 1994; 84: 312233.
  • 15
    Pieters R, Huismans DR, Loonen AH et al. Relation of cellular drug resistance to long-term clinical outcome in childhood acute lymphoblastic leukaemia. Lancet 1991; 338: 399403.
  • 16
    Miller AL, Webb MS, Copik AJ et al. p38 Mitogen-activated protein kinase (MAPK) is a key mediator in glucocorticoid-induced apoptosis of lymphoid cells: correlation between p38 MAPK activation and site-specific phosphorylation of the human glucocorticoid receptor at serine 211. Mol Endocrinol 2005; 19: 156983.
  • 17
    Lu J, Quearry B, Harada H. p38-MAP kinase activation followed by BIM induction is essential for glucocorticoid-induced apoptosis in lymphoblastic leukemia cells. FEBS Lett 2006; 580: 353944.
  • 18
    Wang Z, Malone MH, He H, McColl KS, Distelhorst CW. Microarray analysis uncovers the induction of the proapoptotic BH3-only protein Bim in multiple models of glucocorticoid-induced apoptosis. J Biol Chem 2003; 278: 238617.
  • 19
    Willis SN, Adams JM. Life in the balance: how BH3-only proteins induce apoptosis. Curr Opin Cell Biol 2005; 17: 61725.
  • 20
    Smoak KA, Cidlowski JA. Mechanisms of glucocorticoid receptor signaling during inflammation. Mech Ageing Dev 2004; 125: 697706.
  • 21
    Schoneveld OJ, Gaemers IC, Lamers WH. Mechanisms of glucocorticoid signalling. Biochim Biophys Acta 2004; 1680: 11428.
  • 22
    Kordes U, Krappmann D, Heissmeyer V, Ludwig WD, Scheidereit C. Transcription factor NF-kappaB is constitutively activated in acute lymphoblastic leukemia cells. Leukemia 2000; 14: 399402.
  • 23
    Yamamoto S, Yamauchi T, Kawai Y et al. Fludarabine-mediated circumvention of cytarabine resistance is associated with fludarabine triphosphate accumulation in cytarabine-resistant leukemic cells. Int J Hematol 2007; 85: 10815.
  • 24
    Schmidt S, Irving JA, Minto L et al. Glucocorticoid resistance in two key models of acute lymphoblastic leukemia occurs at the level of the glucocorticoid receptor. FASEB J 2006; 20: 26002.
  • 25
    Bornhauser BC, Bonapace L, Lindholm D et al. Low-dose arsenic trioxide sensitizes glucocorticoid-resistant acute lymphoblastic leukemia cells to dexamethasone via an Akt-dependent pathway. Blood 2007; 110: 208491.
  • 26
    Takemura H, Urasaki Y, Yoshida A, Fukushima T, Ueda T. Simultaneous treatment with 1-beta-d-arabinofuranosylcytosine and daunorubicin induces cross-resistance to both drugs due to a combination-specific mechanism in HL60 cells. Cancer Res 2001; 61: 1727.
  • 27
    Chen CL, Liu Q, Relling MV. Simultaneous characterization of glutathione S-transferase M1 and T1 polymorphisms by polymerase chain reaction in American whites and blacks. Pharmacogenetics 1996; 6: 18791.
  • 28
    Ploner C, Schmidt S, Presul E et al. Glucocorticoid-induced apoptosis and glucocorticoid resistance in acute lymphoblastic leukemia. J Steroid Biochem Mol Biol 2005; 93: 15360.
  • 29
    Wang Z, Rong YP, Malone MH, Davis MC, Zhong F, Distelhorst CW. Thioredoxin-interacting protein (txnip) is a glucocorticoid-regulated primary response gene involved in mediating glucocorticoid-induced apoptosis. Oncogene 2006; 25: 190313.
  • 30
    Slack DN, Seternes OM, Gabrielsen M, Keyse SM. Distinct binding determinants for ERK2/p38alpha and JNK map kinases mediate catalytic activation and substrate selectivity of MAP kinase phosphatase-1. J Biol Chem 2001; 276: 16491500.
  • 31
    Cai B, Chang SH, Becker EB, Bonni A, Xia Z. p38 MAP kinase mediates apoptosis through phosphorylation of BimEL at Ser-65. J Biol Chem 2006; 281: 2521522.
  • 32
    Tissing WJ, Meijerink JP, Den Boer ML, Pieters R. Molecular determinants of glucocorticoid sensitivity and resistance in acute lymphoblastic leukemia. Leukemia 2003; 17: 1725.
  • 33
    Adli M, Baldwin AS. IKK-i/IKKepsilon controls constitutive, cancer cell-associated NF-kappaB activity via regulation of Ser-536 p65/RelA phosphorylation. J Biol Chem 2006; 281: 2697684.
  • 34
    Traenckner EB, Pahl HL, Henkel T, Schmidt KN, Wilk S, Baeuerle PA. Phosphorylation of human I kappa B-alpha on serines 32 and 36 controls I kappa B-alpha proteolysis and NF-kappa B activation in response to diverse stimuli. EMBO J 1995; 14: 287683.
  • 35
    Fujita T, Nolan GP, Liou HC, Scott ML, Baltimore D. The candidate proto-oncogene bcl-3 encodes a transcriptional coactivator that activates through NF-kappa B p50 homodimers. Genes Dev 1993; 7: 135463.
  • 36
    Watanabe N, Iwamura T, Shinoda T, Fujita T. Regulation of NFKB1 proteins by the candidate oncoprotein BCL-3: generation of NF-kappaB homodimers from the cytoplasmic pool of p50-p105 and nuclear translocation. EMBO J 1997; 16: 360920.
  • 37
    Bachmann PS, Gorman R, Mackenzie KL, Lutze-Mann L, Lock RB. Dexamethasone resistance in B-cell precursor childhood acute lymphoblastic leukemia occurs downstream of ligand-induced nuclear translocation of the glucocorticoid receptor. Blood 2005; 105: 251926.
  • 38
    Helmberg A, Auphan N, Caelles C, Karin M. Glucocorticoid-induced apoptosis of human leukemic cells is caused by the repressive function of the glucocorticoid receptor. EMBO J 1995; 14: 45260.
  • 39
    Stahl M, Dijkers PF, Kops GJ et al. The forkhead transcription factor FoxO regulates transcription of p27Kip1 and Bim in response to IL-2. J Immunol 2002; 168: 502431.
  • 40
    Zhao Y, Tan J, Zhuang L, Jiang X, Liu ET, Yu Q. Inhibitors of histone deacetylases target the Rb-E2F1 pathway for apoptosis induction through activation of proapoptotic protein Bim. Proc Natl Acad Sci U S A 2005; 102: 160905.
  • 41
    Inta I, Paxian S, Maegele I et al. Bim and Noxa are candidates to mediate the deleterious effect of the NF-kappa B subunit RelA in cerebral ischemia. J Neurosci 2006; 26: 12896903.
  • 42
    Biswas SC, Shi Y, Sproul A, Greene LA. Pro-apoptotic Bim induction in response to nerve growth factor deprivation requires simultaneous activation of three different death signaling pathways. J Biol Chem 2007; 282: 2936874.
  • 43
    Zhong H, May MJ, Jimi E, Ghosh S. The phosphorylation status of nuclear NF-kappa B determines its association with CBP/p300 or HDAC-1. Mol Cell 2002; 9: 62536.
  • 44
    Pan J, McEver RP. Regulation of the human P-selectin promoter by Bcl-3 and specific homodimeric members of the NF-kappa B/Rel family. J Biol Chem 1995; 270: 2307783.
  • 45
    Bours V, Franzoso G, Azarenko V et al. The oncoprotein Bcl-3 directly transactivates through kappa B motifs via association with DNA-binding p50B homodimers. Cell 1993; 72: 72939.