Oleanane triterpenoid CDDO-Me induces apoptosis in multidrug resistant osteosarcoma cells through inhibition of Stat3 pathway

Background The activation of signal transducer and activator of transcription 3 (Stat3) pathway correlates with tumor growth, survival, drug resistance and poor prognosis in osteosarcoma. To explore the potential therapeutic values of this pathway, we assessed both the expression and the activation of Stat3 pathway in several pairs of multidrug resistant (MDR) osteosarcoma cell lines, and tissues. To explore the potential therapeutic values of this pathway, we analyzed the ability of the synthetic oleanane triterpenoid, C-28 methyl ester of 2-cyano-3,12-dioxoolen-1,9-dien-28-oic acid (CDDO-Me), to inhibit Stat3 expression and activation as well as its effects on doxorubicin sensitivity in osteosarcoma cells. Methods Expression of Stat3, phosphorylated Stat3 (pStat3) and Stat3 targeted proteins, including Bcl-XL, Survivin and MCL-1 were determined in drug sensitive and MDR osteosarcoma cell lines and tissues by Western blot analysis. The effect of CDDO-Me on osteosarcoma cell growth was evaluated by MTT and apoptosis by PARP cleavage assay and caspase-3/7 activity. Results Stat3 pathway was activated in osteosarcoma tissues and in MDR cell lines. CDDO-Me inhibited growth and induced apoptosis in osteosarcoma cell lines. Treatment with CDDO-Me significantly decreased the level of nuclear translocation and phosphorylation of Stat3. The inhibition of Stat3 pathway correlated with the suppression of the anti-apoptotic Stat3 targeted genes Bcl-XL, survivin, and MCL-1. Furthermore, CDDO-Me increased the cytotoxic effects of doxorubicin in the MDR osteosarcoma cell lines. Conclusions Stat3 pathway is overexpressed in MDR osteosarcoma cells. CDDO-Me significantly inhibited Stat3 phosphorylation, Stat3 nuclear translocation and induced apoptosis in osteosarcoma. This study provides the framework for the clinical evaluation of CDDO-Me, either as monotherapy or perhaps even more effectively in combination with doxorubicin to treat osteosarcoma and overcome drug resistance.


Background
Osteosarcoma is the most common malignant tumor of bone, which mainly affects children and adolescents [1,2]. Current treatment of osteosarcoma consists of multi agent chemotherapy and surgical resection [3]. The advancement in intensive chemotherapy has significantly improved the 5-year survival rate from 10% with surgery alone to approximately 60-70% when combined with che-motherapy [1][2][3][4]. Even so, for 20 years, survival rate has not changed and nearly 30-40% of the patients still experience local recurrence or metastasis, possibly because of development of multidrug resistance (MDR). Overcoming drug resistance is one approach to improve the survival rate of osteosarcoma patients. The development of drug resistance is associated with many events, such as activation of transcription factors, overexpression of antiapoptotic proteins, and overexpression of multidrug resistance gene 1 (MDR1) [5][6][7][8]. Successful management of osteosarcoma might be greatly aided by the use of novel agents that could overcome drug resistance.
In this study, we investigated the molecular mechanism of CDDO-Me induced apoptosis and the effects of combinations of CDDO-Me with doxorubicin which is known to have established activity in the clinical therapy for osteosarcoma.

Cell lines, tissues, antibodies and drugs
Human osteosarcoma cells KHOS, U-2OS, SaOS were obtained from the American Type Tissue Collection (Rockville, MD) and a normal human hipbone osteoblast cell line HOB-c was obtained from the Promo Cell GmbH (Heidelberg, Germany). The MDR cell line U-2OS TR was established as previously reported [1,2]. Briefly, the U-2OS cell lines were selected over a period of 8 months by continuous culturing in medium containing step-wise increases in drugs. Dr. Efstathios Gonos (Institute of Biological Research and Biotechnology, Athens, Greece) kindly provided the MDR KHOS R2 cell line [41]. In addi-tion, osteosarcoma tissue samples were obtained from the Massachusetts General Hospital Sarcoma Tissue Bank. Surgically treated patients diagnosed with osteosarcoma were identified and utilized for the study and were used in accordance with the policies of the Institutional Review Board of the institution. Doxorubicin was obtained through unused residual clinical material at the Massachusetts General Hospital. CDDO-Me was kindly provided by Dr. Jeff Supko (Massachusetts General Hospital). The stock solution of drugs were prepared according to the drug specifications and stored at -20°C. The rabbit polyclonal antibodies to Stat3, Bcl-X L , MCL-1, PARP, and the mouse monoclonal antibodies to phosphorylated Stat3 (pStat3), survivin were purchased from Cell Signaling Technologies (Cambridge, MA). The mouse monoclonal antibody to β-actin and MTT reagent were purchased from Sigma-Aldrich (St. Louis, MO). The Pgp1 monoclonal antibody C219 was purchased from Signet (Dedham, MA). Goat antimouse-HRP and goat antirabbit-HRP were purchased from BIO-Rad (Hercules, CA). SuperSignal ® West Pico Chemiluminescent Substrate was purchased from PIERCE (Rockford, IL).

Cell culture
Osteoblast cells HOB-c were cultured in Osteoblast Growth Medium (Promo Cell), and all other cell lines were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin (all obtained from Invitrogen, Carlsbad, CA). All cells were incubated at 37°C in 5% CO 2 -95% air atmosphere and passaged when near confluent monolayers were achieved using trypsin-EDTA solution. Resistant cell lines were continuously cultured in 0.1 μM doxorubicin. Cells were free of mycoplasma contamination as tested by MycoAlert ® Mycoplasma Detection Kit from Cambrex (Rockland, ME).

Cell viability analysis
The In vitro cell viability analysis was performed by MTT assay as described previously [42]. For analyzing effects of CDDO-Me on KHOS, KHOS R2 , U-2OS, U-2OS TR and HOB-c, 2 × 10 3 cells per well were plated in 96-well plates in culture medium containing increasing concentrations of CDDO-Me. For the effect of varying concentrations of CDDO-Me on U-2OS TR and KHOS R2 , 2 × 10 3 cells per well were plated in 96-well plates in culture medium containing increasing concentrations of doxorubicin and CDDO-Me at final concentrations of 0.1 and 0.3 μmol/L, respectively. After 7 days of culture at 37°C, 10 μl MTT (5 mg/ml in PBS, obtained from Sigma) was added to each well and the plates were incubated for 3 h. The resulting formazan product was dissolved with acid-iso-propanol and the absorbance at a wavelength of 490 nm (A 490 ) was read on a SPECTRAmax Microplate Spectrophotometer (Molecular Devices, Sunnyvale, CA). The absorbance values were normalized by assigning the value of the control line in the medium without drug to 1.0 and the value of the no cell control to 0. Experiments were performed in triplicate. The half inhibitory concentration (IC 50 ) was defined as the compound or chemo drug concentration required decreasing the A 490 to 50% of the control value.

Apoptosis assay
Lysates from whole-cell were immunoblotted with specific antibodies to PARP (Cell Signaling Technology) and its cleavage products. Positive immunoreactions were detected by using Super Signal West Pico Chemiluminescent Substrate. As a second parameter of apoptotic cell death, we measured caspase-3/7 activities in HOB-c, KHOS and KHOS R2 after treatment with CDDO-Me by using Apo-ONE Homogenous caspase-3/7 system according to the manufacturer's instructions (Promega, Madison, WI). The intensity of the emitted fluorescence was determined at a wavelength of 521 nm with the use of a SPECTRAmax ® Microplate Spectrofluorometer (Molecular Devices).

Real-time analysis of the effect of CDDO-Me on Stat3 nucleocytoplasmic translocation
To study the effects of CDDO-Me on Stat3 nuclear translocation in live cells, a real-time cell based assay was used as described below. Stable transfectants expressing EGFP-Stat3 were generated with the osteosarcoma cell line U-2OS using standard lipofectmine transfection techniques with G418 selection. EGFP-Stat3 expressing cells were seeded at a density of 4 × 10 3 cells per well in 96 well plates and incubated overnight at 37°C. The cells were then treated with either 1 μM CDDO-Me with 30 ng/ml human recombinant IL-6 (R&D Systems, Minneapolis, MN) or IL-6 alone. To counterstain the nuclei, the cells were incubated with 1 μg/ml Hoechst 33342 (Invitrogen, Carlsbad, CA) for 1 minute. IL-6 dependent nuclear translocation of EFGP-Stat3 was analyzed using an Olympus 1X71 fluorescence microscope and the pictures were captured as digital images using IPLab Software from Scanalytics (Rockville, MD).

Data analysis
Values shown are representative of triplicate determinations in two or more experiments. The IC 50 was calculated and the effects of treatment were evaluated using a two-sided Student's t test (GraphPad PRISM ® 4 software, GraphPad Software, San Diego, CA). Errors are a SD (standard deviation) of averaged results and with value p < 0.05 was considered as a significant difference between means.
The combination index (CI) for experimental treatment combinations was calculated to determine the synergistic, additive, or antagonistic effects using the Chou-Talalay method and GraphPad PRISM ® 4 software [43]. When CI = 1, the equation represents the conservation isobologram and indicates additive effects. CI values of <1.0 indicates synergistic effect, > 1.0 indicates antagonistic effect.

Stat3 pathway is activated in osteosarcoma cell lines and tissues
To evaluate the expression and activation of Stat3 pathway, we analyzed the protein expression in several pairs of both drug sensitive and MDR osteosarcoma cell lines and 8 samples of osteosarcoma tissues. Western blot analysis demonstrated that Stat3 and pStat3 were constitutively overexpressed in drug sensitive cell lines KHOS, U-2OS, SaOS and MDR cell lines KHOS R2 , U-2OS TR . Normal osteoblast cell line HOB-c expressed Stat3 as well as low levels of pStat3 (Fig. 1A). In osteosarcoma tissues, Stat3 was overexpressed in all of the samples and pStat3 was overexpressed in 7 out of 8 (88%) samples (Fig. 1B). Following this preliminary study, we analyzed the expression of Stat3-mediated antiapoptotic proteins Bcl-X L , survivin, and MCL-1. Western blot analysis demonstrated that Bcl-X L , survivin, and MCL-1 were constitutively overexpressed in drug sensitive cell lines and MDR osteosarcoma cell lines. Osteoblast cell lines HOB-c controls showed no expression of Stat3-mediated anti-apoptotic proteins. Pgp1 is also over expressed in MDR cell lines (Fig. 1A). Osteosarcoma tissues showed heterogeneous expression of Stat3 mediated antiapoptotic proteins and Pgp1 (Fig. 1B).

CDDO-Me inhibits growth and induces apoptosis in osteosarcoma cells
To confirm that CDDO-Me inhibits cell growth, KHOS, U-2OS, KHOS R2 , U-2OS TR , and HOB-c were evaluated using MTT assay. The results showed that the growth of all cell lines was inhibited after treatment with CDDO-Me ( Fig. 2A). CDDO-Me showed significantly higher antiproliferative activity in osteosarcoma cells than in osteoblast cells (p < 0.0001). The IC 50 of each cells were HOB-c: 0.  (Fig. 2B). In addition, caspase-3/7 activity was significantly increased when KHOS and KHOS R2 were treated with increasing concentration of CDDO-Me (Fig. 3).

CDDO-Me inhibits IL-6 induced nuclear translocation of Stat3
To identify the interruption of IL-6 dependent Stat3 nuclear translocation by CDDO-Me, a novel real-time cell-based method was developed to image the EGFP-Stat3 chimera in the nucleus and cytoplasm in human osteosarcoma cell line U-2OS. Resting cells demonstrated that the majority of EGFP-Stat3 was cytoplasmic (Fig. 4A,  A') until the addition of IL-6, which then promptly induced translocation of fluorescent Stat3 to the nucleus in U-2OS cells (Fig. 4B). Pretreatment of the cells with CDDO-Me (1 μM) blocked IL-6 dependent EGFP-Stat3 nuclear translocation (Fig. 4B'). Expression was assessed with total cellular protein isolated from the indicated cell lines and immunoblotted with specific antibodies as described in Materials and Methods. The blots were also probed with an antiactin monoclonal antibody to assess relative protein levels in the sample lanes.

CDDO-Me inhibits Stat3 pathway in a dose-dependent manner
After identifying CDDO-Me as an inhibitor of Stat3 nuclear translocation in osteosarcoma cells, the effect of CDDO-Me on Stat3 phosphorylation was examined in osteosarcoma drug sensitive and MDR cell lines. To evaluate the dose-dependent inhibition of Stat3 activation, the cell lines were treated with CDDO-Me alternatively with varying doses for 24 h. In a dose-dependent manner, concentrations as low as 0.5 μM CDDO-Me inhibited Stat3 phosphorylation (Fig. 5). Stat3 phosphorylation and nuclear translocation are required for Stat3 transcriptional activity. We hypothesized that inhibition of nuclear transport should suppress transcription and subsequent translation of Stat3-mediated proteins. Therefore, we next examined whether exposure of cell lines to CDDO-Me resulted in decreased expression of the antiapoptotic proteins Bcl-X L , survivin and MCL-1 by dose-dependent inhibition (Fig. 5). Results showed incubation of CDDO-Me for 24 h caused significant down-regulated Bcl-X L , survivin and MCL-1 expression in both drug sensitive

CDDO-Me inhibits Stat3 pathway and induces apoptosis in a time-dependent manner
To evaluate the time-dependent inhibition of Stat3 activation, the cell lines were treated with 1 μM of CDDO-Me for varying time periods for 24 h. In a time-dependent manner, the pStat3 level decreased as early as 4 hours after the addition of 1 μM of CDDO-Me treatment (Fig. 6). We next examined whether exposure of cell lines to CDDO-Me resulted in decreased expression of the antiapoptotic proteins Bcl-X L , survivin and MCL-1 by time-dependent inhibition (Fig. 6). Results showed incubation of CDDO-Me for 24 h caused significant downregulation of Bcl-X L , survivin and MCL-1 expression in both drug sensitive (Fig. 6A) and MDR (Fig. 6B) cell lines in a time dependent manner. The effect of CDDO-Me on the induction of apoptosis was assessed by evaluating PARP cleavage in a time-dependent manner. PARP cleavage was detected in all cells after 12 hours of treatment with 1 μM of CDDO-Me (Fig. 6).

CDDO-Me has no effect on Pgp-1 expression
While CDDO-Me significantly inhibits Stat3 phosphorylation in KHOS R2 , U-2OS TR cells, there was no significant change in Pgp1 expression as observed in the Pgp1-over expressing cell lines ( Fig. 5B and 6B).

CDDO-Me shows synergy with doxorubicin and reduces resistance in MDR osteosarcoma cell lines
To examine whether interruption of the Stat3 pathway shows synergistic effect when combined with chemotherapy, we analyzed the effect of CDDO-Me in the presence or absence of doxorubicin. Treatment of MDR cell lines with doxorubicin, CDDO-Me or combination of both drugs were analyzed using MTT. The combination of the two agents doxorubicin and CDDO-Me resulted in significantly greater cell death than either drug alone ( Fig  7A). To investigate the inhibition of Stat3 pathway's effect on drug sensitive osteosarcoma cell lines KHOS, U-2OS and MDR osteosarcoma cell lines KHOS R2 , U-2OS TR , they were exposed to varying concentrations of doxorubicin and CDDO-Me for 7 days. MTT demonstrated that CDDO-Me has a synergistic effect on doxorubicininduced antiproliferative effect in KHOS R2 and U-2OS TR (Fig 7B and 7C).

Discussion
Stat3 pathway plays an important role in tumor cell growth and proliferation, and is constantly activated in many human cancer cell lines and tumor tissues [9][10][11][12][13][14]17,20]. Constitutive activation of Stat3 pathway could be an early indicator of drug resistance [15]. Activation of Stat3 pathway was also reported to be present in osteo-sarcoma cells and tissues [14,20]. In this study, we observed that elevated levels of Stat3 and pStat3 are detected in osteosarcoma drug sensitive cell lines KHOS, U-2OS, SaOS and osteosarcoma MDR cell lines KHOS R2 , U-2OS TR . The data is consistent with previous studies that Stat3 pathways play an important role in not only drug sensitive but also drug resistant osteosarcoma cells [14,20]. The downstream effect of Stat3 activation is the Stat3-dependent regulation of several antiapoptotic genes including Bcl-X L , survivin, and MCL-1. Overexpressions of these survival-promoting genes have been shown to be highly expressed and prevent apoptosis in human cancer cells, especially in high-grade tumors [13,[44][45][46]. In our study, these antiapoptotic genes were highly expressed in both drug sensitive and resistant osteosarcoma cell lines, but not in normal human osteoblast cells.
Nuclear translocation of pStat3 is a crucial event for its transcriptional function. Blocking the phosphorylation and translocation of Stat3 is a rational approach for the inhibition of Stat3 activation. Recently, Stat3 has been implicated as a promising target for cancer therapy [12,13,20,[28][29][30][31]. Stat3 inhibitors, SD-1029 and SD-1008, greatly induce apoptosis in drug resistant ovarian cancer cells by blocking Stat3 nuclear translocation [29,30]. Stat3, pStat3 and Stat3 targeted antiapoptotic proteins are over expressed in drug resistance osteosarcoma cells. Inhibition of the Stat3 pathway and interruption of antiapoptotic response may also play an important role for treatment of these cells. Our result is consistent with recent reports showing that the novel Stat3 inhibitor, Indirubin, significantly induces apoptosis in human breast cancer cells [13]. In addition, LLL3, Stat3 DNA binding and transcription activities inhibitor, and Stat3 siRNA significantly decrease cell proliferation and viability, ultimately inducing apoptosis in osteosarcoma cells [20].
CDDO is widely used in Asian herbal medicine and was originally identified as an active compound for antiinflammatory and anti-carcinogenic treatments [34,39].
The novel compound CDDO-Me has been shown to be highly effective in vivo models for the prevention and treatment of cancer [33][34][35]. CDDO-Me is able to induce the differentiation of tumor cells, suppress the growth of tumor cells, and induce apoptosis in cancer cells that are resistant to conventional chemotherapeutic agents [32]. Recent report has shown that CDDO-Me inhibits activation of the JAK/Stat3 pathway by forming adducts with both JAK1 and Stat3 in human cervical and breast cancer cells [34]. Previously, we also demonstrated that CDDO-Me inhibits Stat3 pathway in ovarian cancer cells by down-regulation of antiapoptotic genetic expression, and resulted in a dramatic induction of apoptosis [17]. Furthermore, recent study showed that the novel Stat3 target gene Bcl-X L inhibitor, ABT-737, greatly enhanced the activities of paclitaxel in lung cancer cells [47]. In this study, CDDO-Me inhibits Stat3 phosphorylation and down-regulates Stat3-mediated antiapoptotic proteins Bcl-X L , survivin and, MCL-1, in both dose and time dependent manner. This was through apoptotic cell death in both drug sensitive and resistant osteosarcoma cells. Furthermore, the apoptotic threshold also increases the sensitivity of these cells to the cytotoxic effects of doxorubicin. The results are consistent with a series of studies showing that CDDO-related compounds are not monofunctional drugs that strictly target single steps in signal transduction pathways [32,40,48]. Biochemical and cellular assays support the hypothesis that the molecule CDDO-Me inhibits Stat3 activity with resultant inhibition of Stat3 phosphorylation, nuclear translocation, and decrease in Stat3-dependent transcription, leading to apoptosis and enhanced chemosensitivity.

Conclusion
Our results demonstrated that Stat3 pathways were constantly activated in osteosarcoma and MDR osteosarcoma cells. CDDO-Me significantly inhibits Stat3 phosphorylation and consequently Stat3 nuclear translocation, and eventually resulted in inducing apoptosis in these cells. In addition, this compound is able to overcome drug resistance in these cells.