Bendamustine induces G2 cell cycle arrest and apoptosis in myeloma cells: the role of ATM-Chk2-Cdc25A and ATM-p53-p21-pathways
- First Online:
- Cite this article as:
- Gaul, L., Mandl-Weber, S., Baumann, P. et al. J Cancer Res Clin Oncol (2008) 134: 245. doi:10.1007/s00432-007-0278-x
- 473 Views
Multiple myeloma is a fatal hematological disease caused by malignant transformation of plasma cells. Bendamustine has been proven to be a potent alternative to melphalan in phase 3 studies, yet its molecular mode of action is still poorly understood.
The four-myeloma cell lines NCI-H929, OPM-2, RPMI-8226, and U266 were cultured in vitro. Apoptosis was measured by flow cytometry after annexin V FITC and propidium iodide staining. Cell cycle distribution of cells was determined by DNA staining with propidium iodide. Intracellular levels of (phosphorylated) proteins were determined by western blot.
We show that bendamustine induces apoptosis with an IC50 of 35–65 μg/ml and with cleavage of caspase 3. Incubation with 10–30 μg/ml results in G2 cell cycle arrest in all four-cell lines. The primary DNA-damage signaling kinases ATM and Chk2, but not ATR and Chk1, are activated. The Chk2 substrate Cdc25A phosphatase is degraded and Cdc2 is inhibited by inhibitory phosphorylation of Tyr15 accompanied by increased cyclin B levels. Additionally, p53 activation occurs as phosphorylation of Ser15, the phosphorylation site for ATM. p53 promotes Cdc2 inhibition by upregulation of p21. Targeting of p38 MAPK by the selective inhibitor SB202190 significantly increases bendamustine induced apoptosis. Additionally, SB202190 completely abrogates G2 cell cycle arrest.
Bendamustine induces ATM-Chk2-Cdc2-mediated G2 arrest and p53 mediated apoptosis. Inhibition of p38 MAPK augments apoptosis and abrogates G2 arrest and can be considered as a new therapeutic strategy in combination with bendamustine.
KeywordsMultiple myelomaBendamustineCell cycleAtaxia telangiectasia mutated proteinCheckpoint kinase 2
Multiple myeloma (MM) is an incurable hematological disease caused by malignant transformation of plasma cells. Over three decades neither a single agent nor any polychemotherapy was proven to be superior to melphalan and prednisone (Myeloma Trialist’s 1998). Only dose escalation of melphalan, enabled by the technology of autologous stem cell transplantation to overcome haematotoxicity, significantly improves overall survival (Child et al. 2003; Attal et al. 1996). However, since high dose melphalan has evolved to be gold standard for all eligible patients (Palumbo et al. 2004), highly active genotoxic compounds for the treatment of post-transplant relapse are rare.
Bendamustine is a bifunctional agent that consists of a benzimidazol nucleus, which is linked to a nitrogen mustard moiety and therefore combines the features of alkylating agents and purine analogs (Konstantinov et al. 2002). Due to this heterogeneity in structure and the fact that bendamustine induced double strand breaks have been shown to be more stable (Strumberg et al. 1996), its interaction with DNA has to be considered more complex than of other cytotoxic agents. Bendamustine has been proven to be active in lymphoma (Heider and Niederle 2001; Herold et al. 2006; Lissitchkov et al. 2006; Kath et al. 2001), multiple myeloma, and breast cancer (von Minckwitz et al. 2005; Zulkowski et al. 2002; Hoffken et al. 1998) and has been shown to be almost without cross-resistance to other alkylating agents like cyclophosphamide and melphalan (Leoni et al. 2003). According to a phase III study bendamustine plus prednison revealed to be superior to standard mephalan/prednisone in terms of time to treatment failure, complete remission rate, and quality of life (Ponisch et al. 2006). Furthermore, bendamustine has been proven to be save and effective as salvage treatment after high dose melphalan (Knop et al. 2005). This is in accordance with the preclinical data regarding lack of cross resistance (Leoni et al. 2003; Strumberg et al 1996). Although the clinical effectiveness of bendamustine in the treatment of myeloma is clearly demonstrated, we must resume that its molecular mechanism of action is still poorly understood.
The aim of our study was to examine the in vitro toxicity of bendamustine in myeloma cell lines and to detect the involved pathways as a prerequisite for development of molecular targeted combination therapies.
Materials and methods
NCI-H929, U266, RPMI-8226, OPM-2 cell lines were obtained from the American Type Culture Collection (Rockville, USA), grown in RPMI 1640 medium (Boehringer, Ingelheim, Germany) containing 10% heat-inactivated fetal calf serum (Boehringer) in a humidified atmosphere (37.0°C; 5% CO2), and seeded at a concentration of 1 × 105 cells/ml.
We obtained annexin V/FITC-conjugated from BD Bioscience (San Jose, CA, USA), propidium iodide from Calbiochem (Darmstadt, Germany). Bendamustine from Ribosepharm (Gräfelfing, Germany), caffeine from Sigma–Aldrich (Taufkirchen, Germany), olomoucine, roscovitine, Chk2-inhibitor, p38-inhibitor SB202190 from Calbiochem (Darmstadt, Germany). P-ATM, P-ATR, P-cdc2, P-Chk1, P-Chk2, P-Cdc25C and P-p53-antibodies were obtained from Cell-signaling technology (Frankfurt am Main, Germany). Bad, bax, bcl-2, bcl-XL, Cyclin B and D2, p21, p53, XIAP, and actin-antibodies were requested from Santa Cruz (Santa Cruz, CA, USA)
Analysis of cell death and apoptosis by flow cytometry
Myeloma cells were seeded in 6-well plates at a concentration of 0.5 × 105 cells/ml. After 48 h myeloma cells were detached by pipetting vigorously and by using a cell scraper. Cells were stained with FITC-conjugated annexin V and propidium iodide. Briefly, after two washes with washing buffer (8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4, 0.24 g KH2PO4, 1 l H2O, pH7.2), cells were resuspended in binding buffer (10 mM HEPES/NaOH, pH7.4, 140 mM NaCl, 2.5 mM CaCl2). A total of 100 μl of this cell suspension was incubated with 5 μl annexin V-FITC and 10 μl of 50 μg/ml PI for 15 minutes at room temperature in the dark. Cells were analyzed on a Coulter EPICS XL-MCL flow cytometer (Beckman Coulter, Krefeld, Germany) within 30 minutes.
Analysis of cell cycle
Cells were fixed overnight with 70% (w/v) ice-cold ethanol. After two washes with ice-cold phosphate-buffered saline (PBS), the fixed cells were resuspended in 1 ml of PBS containing 40 μg/ml propidium iodide (PI) and 500 U/ml RNase A. Following incubation for 30 min in the dark at room temperature, the cells were analyzed by flow cytometer using the System II software. The PI fluorescence signal peak versus the integral was used to discriminate G2/M cells from G0/G1 doublets.
Cells were washed three times in ice-cold PBS and lysed in a buffer containing 10 mM Tris–HCl (pH 7.6), 137 mM NaCl, 1 mM Na3VO4, 10 mM NaF, 10 mM EDTA, 1% (v/v) Igepal CA-630 (NP-40) with the addition of 10 μg/ml aprotinin, 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonylfluoride (PMSF). Protein concentration was adjusted using a colorimetric assay (Bio-Rad Protein Assay, Bio-Rad, Munich, Germany). Proteins were subjected to SDS-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Germany). The transfer buffer contained 25 mM Tris–HCl, 192 mM glycine, 0.037 (w/v) SDS, and 20% (v/v) methanol. The membranes were blocked with PBS containing 5% dried milk and 0.05% Tween-20 (PBS/Tween). After washing four times with PBS/Tween, the membranes were incubated with appropriate primary and secondary antibody and visualized by autoradiography using the ECL western blotting detection system (Amersham Pharmacia, Freiburg, Germany).
Cells were cultured in microplates after addition of the agent in a final volume of 100 μl/well. 10 μl/well WST-agent (Roche, Penzberg, Germany) was added for 1 h. After shaking gently absorbance at 450 nm was measured using a microplate enzyme-linked immunosorbent assay reader to detect metabolically intact cells.
Kruskal–Wallis one-way analysis of variance on Ranks was used to determine the statistical significance of treatment results. The pair wise multiple comparison procedure was performed according to Dunn’s method. Values of P < 0.05 were considered statistically significant.
Bendamustine induces apoptosis in myeloma cell lines
Bendamustine induces G2/M cell cycle arrest
Flow cytometry images of the two cell lines NCI-H929 and U-266 are presented in Fig. 2b. To exclude the so-called “mitotic catastrophe” as the pathogenetic mechanism for the effect, we counted cells and mitotic figures by microscopy after Giemsa staining (not shown). The number of apoptotic bodies but not of mitotic figures was increased with dose escalation. Furthermore, we analyzed cell proliferation after incubation with WST-1 over 72 h and measurement of the extinction at wavelength 450 nm on days 0, 1, 2, and 3. Figure 2c shows a time and concentration dependent inhibition of proliferation without any sign for a “proliferation burst” in lower concentrations. Primary MM cells from patients arrest in vitro in G1 phase, even in the presence of bone marrow stromal cells. By supplementing growth factors und cytokines ex vivo survival can be improved, but a sufficiently high proliferation rate is not achievable. Therefore the effects obtained with MM cell lines could not be proved in patient cells.
Bendamustine activates the ATM-Chk2-Cdc25A pathway
The key controller of the progression from G2 to mitosis is the Cdc2/CyclinB complex, the mitosis-promoting factor (MPF). It accumulates in an inactive form in late G2 and is regulated by nuclear import and export. This physiological cell cycle progression can be interrupted at several levels by apoptotic and antiproliferative stimuli.
P53 and p21
As p53 is known to be involved in apoptosis and cell cycle arrest at several levels, we incubated the blots with p53 and P-p53 antibodies. Figure 3d illustrates that p53 protein level is increased in NCI-H929 but not in RPMI-8226 cells. However, both cell lines exhibit strong activation which appears as phosphorylation at Ser15. The linkage of p53 activation to Cdc2 inhibition is mediated by p21, which is known to directly inhibit the MPF. P21 protein levels were increased in both cell lines, suggesting that p53 activation by bendamustine is as well involved in the observed G2-arrest. The p53-target 14-3-3σ, known to sequester MPF in the cytoplasm and by this to prevent initiation of mitosis, was not altered (data not shown). Levels of members of the pro- and antiapoptotic bcl-family like bax, bad, bcl2, bcl-x/l and XIAP remained unchanged as well (Fig. 3e).
Inhibition of p38α-MAPK potentiates bendamustine induced apoptosis
P38 activation is known to arrest cells in G2 (Garner et al. 2002; Pedraza-Alva et al. 2006). Therefore we concluded that inhibition should be able to overcome G2 arrest. Cell cycle was analyzed as demonstrated in Fig. 4b. As we expected, the inhibitor SB202190 totally abrogated bendamustine-induced G2/M arrest (P = 0.015). To verify the involvement of p38 in intracellular bendamustine effects, we detected p38 level and activation status by immunoblotting. Figure 4b shows that p38 is expressed in the cells and activated by phosphorylation when treated with bendamustine.
Anti-cancer agents like purine analogs and alkylating agents target the DNA and induce cross-linking of the double strands. This mechanism makes the cell vulnerable in the two cell cycle checkpoints G1 and G2. Many cytotoxic drugs have been shown to cause changes in checkpoint regulation resulting in proliferation arrest (Damia and Broggini 2004; Bozko et al. 2005). This enables the cell to DNA repair in order to prevent cell death and in this line proliferation arrest may mean drug resistance. However, above a certain threshold of pathway activation the cells commit programmed suicide, termed apoptosis. We show that bendamustine induces apoptosis in MM cells with an IC50 of about 35–65 μg/ml, and causes G2 arrest. To understand the importance of G2 arrest it is essential to illuminate the complex regulation mechanisms of this checkpoint. The progression from G2 to mitosis is mainly regulated by the mitosis-promoting factor (MPF), a complex of Cyclin B and Cdc2 (Nurse 1990). Its activity is a result of numerous pathways reflecting the complexity of stimuli that occur during cell cycle progression and its inhibition a hallmark for G2-arrest. In our experiments Cdc2 is inhibited by phosphorylation at Tyr15 accompanied by an increased level of Cyclin B at low concentrations. The majority of cells arresting in late G2 maintain high levels of Cyclin B that is synthesized equally in G2 (Maity et al. 1996). The activity of the complex is reflected by the activity of Cdc2, which is decreased with rising concentrations. But which pathway is responsible for apoptosis and the inhibition of CyclinB/Cdc2? The two protein kinases ATM (Ataxia-telangiectasia mutated kinase) and ATR (Ataxia-telangiectasia and Rad3-related kinase), both exhibiting homology to the PI3 kinase family, have been shown to play a major role in the primary signaling of DNA-damage (Abraham 2001). ATM phosphorylates and activates Chk2 (Matsuoka et al. 1998; Chaturvedi et al. 1999; Falk et al. 2001), whereas ATR phosphorylates Chk1 (Liu et al. 2000). Furthermore, ATM is able to activate the tumor suppressor p53 on several serine residues (Canman et al. 1998; Turenne et al. 2001; Miyakoda et al. 2002). The grade of activation has been shown to correlate with the amount of DNA strand breaks (Buscemi et al. 2004). Activation of ATR and Chk1 is caused by UV-damage (Tibbetts et al. 1999, O’Driscoll et al. 2003) and has proven to be the mechanism the purine analogon thioguanine mediates G2 arrest (Yan et al. 2004; Yamane et al. 2004), while ATM activation has frequently been associated with double strand breaks due to ionising radiation (IR) or radiomimetic agents but not to UV or alkylating agents (Canman et al. 1998). Chk1 and Chk2 can inactivate Cdc25C by inhibitory phosphorylation of Tyr162 (Chaturvedi et al. 1999; Sanchez et al. 1997) and directly activate p53 by phosphorylation at Ser20 (Hirao et al. 2000). In contrast, activation of p53 by ATM and ATR is mediated by phosphorylation at Ser15. Inactivated Cdc25C is no longer able to promote cell cycle progression by activating the MPF. Activated p53 itself can inhibit MPF by regulation of p21 and 14-3-3σ and force cells to undergo apoptosis due to its complex function as a transcription factor. Our experiments demonstrate that ATM but not ATR is activated and this result is underlined by the further activation of Chk2 but not of Chk1. The fact that these results are in contrast with studies denying a role for ATM in the action of alkylating agents points towards a more complex interaction with the DNA maybe due to the homology of the substance with purine analogs. As Cdc25c is not altered we conclude that this kinase is not essential for bendamustine induced G2 arrest, but the opposite can be demonstrated for Cdc25A. Cdc25A as an activator of cdk1(Cdc2) and cdk2 has been proven to regulate G1/S and G2/M transition and to play a role in UV, IR and chemotherapeutic activated checkpoint pathways (Busino et al. 2004; Agner et al. 2005; Mailand et al. 2000). Our experiments show that Cdc25A is degraded which correlates with inhibition of MPF. As Chk1 is not activated and ATR even inhibited, we conclude that only Chk2 is able to activate Cdc25A thus causing its proteolytic degradation.
A second checkpoint pathway stimulated by bendamustine is based on the tumor suppressor p53. Phosphorylation of p53 at Ser15 points towards a direct activation by ATM. On the one hand, p53 additionally blocks cell cycle progression via Cdc2/CyclinB inhibitor p21. On the other hand it initiates the apoptotic program, which explains the cytotoxic potency of bendamustine. We inhibited the main kinase for G2 arrest Cdc2, with selective inhibitors to test whether this inhibition alters apoptosis. Interestingly none of them showed any effect, which means that inhibition of the MPF alone is not sufficient to further enhance the apoptotic potential of bendamustine. This could be due to the relatively low potential of the inhibitors in comparison to the concurrence of stimuli.
Nonetheless we decided targeting G2 interacting pathways as a promising option to further optimize and understand bendamustine impact on checkpoint activation. One of these is the p38 MAPK pathway. P38 has been shown to interfere with G2 checkpoint by inducing G2 arrest (Bulavin et al. 2001; Garner et al. 2002; Pedraza-Alva et al. 2006). Concerning multiple myeloma, p38 inhibition increased the cytotoxicity of proteasome inhibitors (Hideshima et al. 2004; Navas et al. 2006) and there is evidence that targeting the microenvironment, which is of major importance in the pathophysiology of the disease, by specific inhibitors may become an interesting therapeutic option in the future (Hideshima et al. 2003; Nguyen et al. 2006; Wang et al. 2006). We investigated the ability of the p38 MAPK inhibitor SB202190 to act in synergism with bendamustine and found that, if added in a non-toxic dose, it was able to increase apoptosis by twofold. This effect has been investigated in three cell lines with similar results. In addition, we observed high levels of p38 in myeloma cells, which were further activated under treatment with bendamustine suggesting the potential of p38 to initiate G2 arrest. This may be considered a regulatory activation upon treatment with DNA alkylating drugs in order to ensure cell viability during cell cycle arrest and DNA repair. We further demonstrated that the inhibitor was able to abrogate bendamustine induced G2 arrest. The consequence is that cells which were arrested due to a certain DNA damage in order to allow repair mechanisms are forced to break the checkpoint with defective DNA and undergo apoptosis.
In summary bendamustine is a highly active anti-myeloma agent in vitro and in vivo. Myeloma cells respond to bendamustine induced DNA damage with apoptosis and ATM/Chk2/Cdc25A and ATM/p53/p21 mediated cell cycle arrest and p38 activation to enable DNA repair and cell survival. Targeting of p38 by SB202190 concomitantly with bendamustine treatment synergistically induces apotosis in myeloma cells and herewith provides a good rationale for new therapeutic strategies in multiple myeloma.