Evaluation of glucocorticoid sensitivity in 697 pre-B acute lymphoblastic leukemia cells after overexpression or silencing of MAP kinase phosphatase-1
- First Online:
- Cite this article as:
- Abrams, M.T., Robertson, N.M., Litwack, G. et al. J Cancer Res Clin Oncol (2005) 131: 347. doi:10.1007/s00432-004-0659-3
- 78 Views
To determine the effect of modulating MAP kinase phosphatase-1 (MKP-1) expression levels on cell death induced by glucocorticoid (GC) or hydroxyurea (HU) treatment in the human pre-B acute lymphoblastic leukemia cell line 697.
Stable MKP-1 overexpressing transformants of the 697 pre-B acute lymphoblastic leukemia cell line were created and tested for sensitivity to the GC triamcinolone acetonide (TA) and HU, and compared to a control 697 cell line containing normal MKP-1 expression levels. Small interfering RNAs (siRNAs) were designed to inhibit MKP-1 expression and evaluated for their effect on GC-mediated cell death.
MKP-1 overexpression caused a phenotype of partial resistance to HU-induced apoptosis but not to GC-induced apoptosis. Electroporation of siRNAs effectively silenced MKP-1 expression, and increased sensitivity to TA by 9.6±1.9%.
Because MKP-1 protects certain tumor cells from chemotherapy-induced apoptosis, its inhibition is being considered as a possible strategy for combination cancer therapy. However, this study suggests that while MKP-1 inhibition may improve the efficacy of DNA damaging agents, it may have only limited utility in combination with glucocorticoids.
Acute lymphoblastic leukemia
MAP kinase phosphatase-1
Reverse transcriptase-polymerase chain reaction
Acute lymphoblastic leukemia (ALL) is the most common childhood cancer, with approximately 5,000 new cases reported annually (Cortes and Kantarjian 1995). The standard course of therapy includes both classical chemotherapeutics such as vincristine as well as glucocorticoids (GC) such as dexamethasone, prednisone, or triamcinolone (Cortes and Kantarjian 1995). Although this regimen leads to a cure rate of >70%, poor prognosis often correlates with resistance to GCs (Tissing et al. 2003). Precursor B cell ALL is the most prevalent form of this disease, ~10–20% of which contain a t(1;19) chromosomal translocation (Raimondi et al. 1990). This translocation creates the E2A-Pbx chimeric transcription factor, which has oncogenic properties (Kamps et al. 1991) and causes aberrant expression of signaling proteins such as Wnt-16 (McWhirter et al. 1999) and N-Myc (Rutherford et al. 2001). The pre-B ALL cell line 697, a well-characterized GC-sensitive, t(1;19) positive cell line isolated from a 12-year-old patient, is a useful model for studying signal transduction pathways in this disease (Findley et al. 1982).
GCs induce apoptosis in numerous hematopoietic tissues of both lymphoid and myeloid lineage in a process that requires the glucocorticoid receptor (GR), a GC-activated transcription factor of the nuclear receptor superfamily (Planey and Litwack 2000). GR activation by GC induces dissociation of GC-GR from a large cytoplasmic complex and translocation to the nucleus, where GC-GR affects mRNA levels of target genes by interacting directly with promoter DNA or by binding to other transcription factors (McEwan et al. 1997). Apoptosis occurs downstream of these transcriptional effects, and involves the biochemical hallmarks of the intrinsic cell death pathway: release of cytochrome c from the mitochondria, cleavage of caspase-9, and DNA fragmentation (Planey et al. 2002). In the cell line 697, GC rapidly induces or represses >100 genes, including a number of interesting candidates for mediators of this pathway. These target genes include BCL2 family members, the FOXO3A transcription factor, a G-protein coupled receptor GPCR18, and MAP kinase phosphatase-1 (MKP1) (Planey et al. 2003).
MKP-1 protein is a dual-specificity phosphatase that exhibits anti-apoptotic effects by negatively regulating the stress-induced mitogen-activated protein kinases (MAPKs), c-Jun N-terminal kinase (JNK), and p38 (Slack et al. 2001). Hence, MKP-1 has emerged as an appealing GR target gene (Clark 2003). MKP1 mRNA and MKP-1 protein are both strongly and rapidly repressed by GC-GR in 697 cells before they apoptose, which led us to the hypothesis that alleviating negative regulation of pro-apoptotic MAPKs contributes to this mechanism (Planey et al. 2003). On the contrary, MKP-1 protein is induced by GC in numerous non-hematopoietic tissues including mammary epithelial cells, which initiates a survival response after exposure to GC (Wu et al. 2004). Interestingly, it has been shown recently that silencing MKP1 gene expression by RNA interference in MCF-7 breast cancer cells inhibits GC-mediated survival after paclitaxel treatment (Wu et al. 2004). Thus, it is possible that differential regulation of MKP-1 is a mechanism by which GC can cause opposite biological outcomes (survival versus apoptosis) in different tissues. Finally, given that MKP-1 is overexpressed in a variety of human cancers (Bang et al. 1998; Kurt et al. 1998), and that GCs are administered routinely to reduce the nausea and acute toxicity caused by chemotherapeutics, it is of considerable clinical interest to investigate the role of MKP-1 in GC-mediated effects.
In this study, we determined the effect of modulating MKP-1 expression levels on GC-induced apoptosis in the cell line 697. Based on the hypothesis stated above, MKP-1 overexpression would be expected to cause GC resistance, compared with cells containing only endogenous levels of MKP-1. However, strong overexpression of MKP-1 cDNA in 697 cells did not affect GC sensitivity or caspase-3 activation, even though it caused partial resistance to another chemotherapeutic agent, hydroxyurea. We also designed and validated small RNA duplexes (reviewed by Caplen and Mousses 2003) that significantly reduced MKP-1 expression. Silencing of MKP1 mRNA by RNA interference had a modest effect on GC sensitivity. These data establish that MKP-1 is not likely to be a critical mediator of GC signaling in pre-B ALL and underscore the importance of investigating the tissue-specific mechanisms of GC signal transduction.
Materials and methods
Compounds and reagents
Triamcinolone acetonide (TA) was purchased from Sigma (St. Louis, Mo., USA) and dissolved in 100% ethanol. The final concentration of ethanol in all cell treatments was kept constant at 0.1%. Hydroxyurea (Sigma) was dissolved in phosphate buffered saline (PBS). All siRNAs were purchased from Dharmacon (Lafayette, Colo., USA). The target sequences used to generate the MKP-1 siRNAs are shown in Fig. 3. The GL2 firefly luciferase siRNA sequence (Elbashir et al. 2001) was used as the control for all siRNA experiments, and this sequence was also used for both the FITC and Cy3 labeled siRNAs. All tissue culture and electrophoresis reagents were purchased from Invitrogen (Carlsbad, Calif., USA) except where otherwise indicated.
Cell culture and cDNA overexpression
697 pre-B ALL cells were grown in RPMI 1640 medium containing 10% fetal bovine serum and penicillin/streptomycin mix. MKP-1 cDNAs were a generous gift from Steve Keyes and Chris Franklin. The BamH1/EcoR1 fragment from MKP-1/pSG5 (Groom et al. 1996) was subcloned into a pcDNA3.1 expression vector (Invitrogen). Sixty μg of this plasmid, as well as the empty vector, were transfected by electroporation into 697 cells as described (Planey et al. 2003). 48 h later, 1 μg/ml G418 was added to the growth medium. After 1 week, resistant cells were seeded in 96-well plates at limiting dilutions. Clonal isolates were expanded for screening after approximately 4 weeks. 697 cells overexpressing dominant-negative caspase-9 were as described in (Planey et al. 2003).
For all of the MKP-1 immunoblots, 40–50 μg of whole-cell lysate from each clone was electrophoresed on Tris-glycine 10% polyacrylamide gels. Samples were transferred to polyvinylidene fluoride membranes, blocked in PBS containing 5% powdered milk and 0.5% Tween-20, and immunoblotted with an anti-MKP-1 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, Calif., USA) diluted 1:100 in PBS containing 0.5% Tween-20, or an anti-GAPDH antibody (Research Diagnostics, Flanders, N.J., USA) diluted 1:20,000. After washing, detection was performed using an anti-rabbit secondary antibody conjugated to horseradish peroxidase (Santa Cruz Biotechnology) and a chemiluminescent substrate (SuperSignal, Pierce, Rockford, Ill., USA).
Real time RT-PCR
Total RNA was purified from 1×107 cells for each cell line using RNeasy columns (Qiagen, Valencia, Calif., USA). RT-PCR reactions were performed in duplicate using Taqman One Step reagents in an ABI 7900 Sequence Detection System (Applied Biosystems, Foster City, Calif., USA). Zero point one μg of template RNA was used per reaction. Primer/probe sequences and concentrations were the same as in Planey et al. (Planey et al. 2003).
For cell cycle analysis, cells were treated in 6-well plates, washed with PBS, fixed in 70% ethanol, and stained with 50 μg/ml propidium iodide containing 1 mg/ml RNase A and 1 mg/ml glucose (all from Sigma) for 30 min. Samples were processed on a FACSCalibur instrument (BD Biosciences, San Jose, Calif., USA) and data were analyzed using FACSComp software. Ten thousand cells were collected for each measurement. For viability analysis, samples were resuspended in 500 μl of 0.1 M HEPES-NaOH pH 7.4, 1.4 M NaCl, 25 mM CaCl2 containing 0.5 μg/ml 7-aminoactinomycin-D (BD Pharmingen, San Diego, Calif., USA). 7-AAD content was measured using a Beckman Coulter XL flow cytometer and data were analyzed using WinMDI software. Fifty thousand cells were counted for each measurement.
1 × 105 cells per well were seeded in 96-well plates in triplicate for each condition. After drug treatment, cells were subjected to assay using ApoAlert (Clontech, Palo Alto, Calif., USA) reagents including the DEVD-AFC substrate, and detected fluorimetrically according to the manufacturer’s protocol.
Exponentially growing 697 cells were washed twice with PBS and resuspended in serum-free RPMI-1640 media containing 25 mM HEPES-NaOH pH 7.4 (without phenol red or antibiotics) at 107 cells/ml. Two hundred μL aliquots of cells were added to 0.2 cm gap electroporation cuvets (Bio-Rad, Hercules, Calif., USA). siRNA (10 μl of a 20 μM stock solution for a final concentration of 1 μM) was added and incubated at room temperature for 20 min. Cells were then electroporated with a Bio-Rad Gene Pulser using voltage and capacitance settings of 220 V and 500 μF, respectively. Cells were incubated for an additional 20 min., and added to 3 ml of the above medium supplemented with 10% FBS (final siRNA concentration after dilution was 66 nM) and allowed to recover at 37 °C for the indicated times.
Cells were transfected as above except that FITC-siRNA was added to 0.25 μM in each cuvet. Three electroporations were pooled together for each condition to obtain enough cells to sort efficiently. After 18 h, cells were collected and resuspended in medium containing 5% FBS. Cells were sorted based on FITC content using a EPICS Elite cytometer (Beckman Coulter, Fullerton, Calif., USA) into 50% growth medium/50% FBS and allowed to recover for 30 min. Cells were then re-seeded and either harvested for immunoblotting or treated with compounds for viability assay after an additional 24 h.
MKP-1 overexpression confers resistance to ultraviolet radiation, cisplatin, paclitaxel, doxorubicin, and Fas ligand in well-characterized in vitro models (Franklin et al. 1998; Sanchez-Perez et al. 2000; Srikanth et al. 1999; Wu et al. 2004). In this study, we also observed that MKP-1 overexpression renders 697 pre-B ALL cells resistant to hydroxyurea. Furthermore, siRNA, antisense or small molecule inhibitors of MKP-1 all exhibited drug sensitizing or antiproliferative effects in tumor cells (Guo et al. 1998; Liao et al. 2003; Wu et al. 2004). These data suggest that MKP-1 functions as a generalized resistance gene, affecting cell death induced by agents that initiate both intrinsic (caspase-9 mediated) apoptosis, such as DNA damage inducers, and extrinsic (caspase-8 mediated) apoptosis, such as Fas ligand. In this study, we investigated the role of MKP-1 in GC-induced apoptosis in pre-B ALL cells. We observed that MKP-1 overexpression did not confer GC resistance and that MKP-1 knockdown only slightly increased GC sensitivity (9.6±1.9%). These conclusions underscore the complexity and unique mechanism of glucocorticoid-induced apoptosis.
Inhibition of MKP-1 might contribute to therapy of solid tumors in the future. One recent study (Small et al. 2004) effectively demonstrated that antagonizing MKP-1 by either siRNA or anthracyclines increases the extent of apoptosis induced by proteasome inhibitors in breast carcinoma cells. In another, downregulation of MKP-1 by antisense DNA decreased growth of pancreatic tumors in a nude mouse model (Liao et al. 2003). Finally, in addition to its potential therapeutic value, MKP-1 expression may also have prognostic value, as it was shown to be a predictor of survival in non-small cell lung cancer. Based on our study, however, we predict that MKP-1 inhibition will have limited utility in sensitizing leukemias and lymphomas to GCs.
There are numerous examples of other GR target genes that, when overexpressed, confer GC resistance in leukemic cells. These include BCL2 (Planey et al. 2003), catalase (Tome et al. 2003), and dig120 (Wang et al. 2003b). In addition, we have recently reported that siRNA silencing of the pro-apoptotic BCL2 relative BIM, a target for GC-mediated repression, partially inhibits GC-induced apoptosis (Abrams et al. 2004). Therefore, the mechanism of GC-induced apoptosis is likely to involve the combinatorial effect of activation or repression of numerous specific genes by the GR. MKP-1 is likely to be a key mediator of GC-induced survival in breast cancer cells (Wu et al. 2004), but not GC-induced apoptosis in leukemia as previously hypothesized (Planey et al. 2003).
One of the most important unresolved issues in GC biology is that GR activation causes apoptosis in some tissues and survival in others. These observations are directly related to the fact that the transcriptional program activated by GC is highly dependent on the cellular context. MKP-1 can either be induced or repressed by GC depending on the tissue (see Introduction). Another example of this phenomenon is that while Bcl-2 protein and its proapoptotic family member Bim are oppositely regulated by GC in 697 cells (Planey et al. 2003), CEM T-lymphocytes (Wang et al. 2003a) and S49A murine lymphoma cells (Wang et al. 2003a), neither appear to be affected in MCF-7 breast cancer cells (Wu et al. 2004) or U20S osteosarcoma cells (Rogatsky et al. 2003). Understanding the mechanism of tissue specificity of GR activation outcomes is critical to developing novel therapeutic strategies for GC resistant cancers.
The authors would like to thank Paul Hallberg for providing flow cytometry expertise, and Dr. Chris Franklin and Dr. Steve Keyes for providing MKP1 cDNA constructs