Malignant glioma is the most commonly occurring primary malignant brain tumor. It is difficult to completely remove the tumor surgically due to the unclear boundaries between tumor tissue and normal tissue. Comprehensive treatments of glioma are thus inefficient and the median survival for patients with high-grade glioma is <12 months (Grossman, 2003; Li et al., 2012). Cisplatin (cis-diamminedichloroplatinum) is a widely used anticancer drug in clinical and experimental studies for investigating the mechanisms of resistance of malignant tumors, including gliomas (Rosenberg, 1985; Madajewicz et al., 2000). According to the studies, there are several possible mechanisms involved in resistance to the cell damage by cisplatin, such as inhibition of the proapoptotic pathway and upregulation of the survival pathway (Peters et al., 2011; Skrzypski et al., 2011).
It is well known that deregulation of the Akt pathway plays an important role in the development of human cancers. Accumulated studies have indicated that Akt phosphorylation promotes cell survival when cells are treated with different apoptotic stimuli (Skeen et al., 2006). Recent studies have shown that Akt phosphorylation is involved in the resistance mechanisms of tumor cells to cisplatin (Zhang and Shen, 2007; Yang et al., 2008; Zhang et al., 2009b). The phosphatidylinositol-3,4,5-trisphosphate (PIP3) is generated by the activation of phosphatidylinositol 3 kinase (PI3K), which leads to the recruitment and phosphorylation of Akt. In malignant gliomas, the loss of PTEN or amplification of EGFR is very common (Shishodia et al., 2004).
Autophagy is an evolutionarily conserved process, which involves the cytosolic fraction wrapped in a double membrane structures called autophagosomes, and transported to lysosomes for degradation. Our previous studies and some other studies have shown that autophagy has a protective role for cisplatin, oxidative stress, or resveratrol-induced cytotoxicity in U251 cells (Harhaji-Trajkovic et al., 2009; Li et al., 2009; Zhang et al., 2009a). Under physiological conditions, autophagy plays a key role to maintain cellular homeostasis by degradation and recycling of long-lived proteins and organelles. Under pathological conditions or under stress, autophagy could play prosurvival role or promote cell death (Corcelle et al., 2009; Dalby et al., 2010; Dikic et al., 2010; Germain and Slack, 2011; Verma and Datta, 2012).
Recent studies have shown that chloride channel 3 (ClC-3) is highly expressed in glioma tissue and plays an important role in cellular proliferation, invasion, and migration of glioma cells (Olsen et al., 2003; Li et al., 2010; Yu et al., 2011). ClC-3 is a member of the CLC family of chloride channels, and it has been shown that the activity of Akt depends on the function of ClC-3 (Sun et al., 2003). However, the exact mechanism of how ClC-3 regulates Akt phosphorylation is unclear. Furthermore, ClC-3 is expressed predominantly in acidic intracellular compartments such as late-endosomes and lysosomes (Stobrawa et al., 2001; Weylandt et al., 2001) and participates in these vesicular acidification (Li et al., 2002; Hara-Chikuma et al., 2005; Weylandt et al., 2007; Zifarelli and Pusch, 2007). The inhibitors of v-ATPase such as bafilomycin A1 or lysosomotropic agents such as chloroquine can obstruct lysosome acidification and then suppresses autophagy through inhibiting autophagosome-lysosome fusion (Mizushima et al., 2010).
Therefore, we propose a hypothesis that suppression of ClC-3 expression could facilitate sensitivity of human glioma U251 cells to cisplatin through concomitant inhibition of Akt and autophagy. In this study, we utilized the PTEN-deficient human glioma cell line U251 that were relatively resistant to cisplatin. The results indicate that ClC-3 suppression can inhibit both Akt and autophagy to facilitate sensitivity of human glioma U251 cells to cisplatin.
MATERIALS AND METHODS
The human malignant glioma cell line U251 (glioblastoma multiforme, World Health Organization grade IV) was acquired from the American Type Culture Collection (Manassas, VA) (Ponten and Macintyre, 1968; Giard et al., 1973). Cells were grown at 37°C, in 5% CO2, in IMDM medium (Hyclone, Logan, UT) with 10% fetal calf serum (FCS) (Hyclone, Logan, UT), 50 U mL−1 penicillin, and 50 μg mL−1 streptomycin.
According to the sequence for human ClC-3 (National Center for Biotechnology Information, accession numbers NM 173872.2), the ClC-3 siRNA corresponded to the following sequence: 5′- CTG CTT GAC CTA TGA TTA A -3′. Both short and long forms of ClC-3 can be eliminated by ClC-3 siRNA. The pSilencerTM3.0-H1-neo plasmid was purchased from Ambion (Ambion, TX). All the purified plasmids were transfected using LipofectamineTM 2000 transfection reagent (Invitrogen Corporation, Carlsbad, CA) according to the manufacturer's recommendations. Then cells were treated by cisplatin (Sigma-Aldrich, St. Louis, MO).
Total cellular RNA was extracted from U251 cells using the TRIzol reagent (Invitrogen Corporation, Carlsbad, CA). cDNA was synthesized using avian myeloblastosis virus (AMV) reverse transcriptase (Takara, Dalian, China). PCR cycling conditions were as follows: 94°C for 30 s, annealing for 30 s, and elongation at 72°C for 30 s. All PCR reactions were cycled 30 times. All primers were produced by Takara (Dalian, China) (Lu et al., 2002). The results were visualized by Tanon-1600 Figure Gel Image Processing System and analyzed by GIS 1D gel image system software (Tanon, Shanghai, China).
Western Blot Analysis
Cells were lysed in radioimmunoprecipitation assay buffer (Mizel, 1982) and 60 μg of protein per sample was loaded into SDS-PAGE gel by electrophoresis, and then transferred to polyvinylidene difluoride paper (Millipore, Bedford, MA). Bands were quantified using GIS 1D gel image system software. Primary antibodies used included anti-β-actin (1:400, sc-47778), anti-ClC-3 (1:100, sc-17572), anti-Akt (1:200, sc-8312), anti-p-Akt (1:200, sc-7985R), anti-LC3 (1:200, sc-28266), anti-Beclin1 (1:200, sc-48341), anti-Atg12 (1:200, sc-68884) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), anti-mTOR (1:200, 2972), anti-p-mTOR (1:200, 2974), anti-P70S6K (1:200, 9202), anti-p-P70S6K (1:200, 9205) were purchased from Cell Signaling Technology (MA, USA). All secondary HRP-conjugated antibodies were purchased from Thermo (Rockford, IL).
The cell survival rate was detected by MTT assay. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) was purchased from Sigma-Aldrich (St. Louis, MO). Experiments were performed in accordance with the existing methods (Mizel, 1982). Absorbance was measured at a wavelength of 570 nm using a microplate reader (Bio-Tek, Santa Barbara, CA). IC50 were calculated with SPSS 11.5 statistical software (SPSS, Chicago, IL).
The cells were harvested with 0.25% trypsin digestion and centrifuged (800 rpm) for 10 min at room temperature. Then pellets were fixed by adding 2.5% glutaraldehyde and 1% osmic acid. Fixed samples were proceeded according to published protocol (Kumar et al., 2003). Then the slices were observed and recorded by the transmission electron microscope.
Apoptosis Quantitative Detection
The experiments were performed according to the manufacture's instruction. The quantitative detection of apoptotic cells were determined by PI (propidium iodide)/Annexin V-FITC double staining apoptosis detection kit purchased from KeyGEN Biotech (Nanjing, China). Samples were detected by the BD flow cytometer. Early apoptosis cells were defined as being in the Q4 quadrant (Duan et al., 2001).
Monodansylcadaverin (MDC) (Sigma-Aldrich, St. Louis, MO) can be used to detect the formation of macrophage bubble autophagic vacuoles and acidic organelles. U251 cells were cultured on cover slips, incubated with MDC (1 μmol L−1, 37°C) for 10 min. Then the samples were analyzed by the laser scanning confocal microscope (excitation 380–420 nm, emission 450 nm) (Reddy and Reddanna, 2009).
Measurements of Intracellular Emission Spectra of Acridine Orange
Acridine orange (AO) staining was used to detect intracellular acidic vesicles. AO is a fluorescent tertiary amine that becomes concentrated in the acidic vesicles by protonation. AO concentration increased from monomer to dimer, and the fluorescence changed from green to red (Suzuki et al., 2000). The cells were cultured on cover slips, incubated with AO (1 μmol L−1, 37°C) for 15 min. The AO-induced granular red fluorescence in each group was measured using laser confocal microspectrofluorometry (FluoViewTM FV300, Olympus, Japan) (excitation 528–553 nm, emission 600–660 nm) (Millot et al., 1997; Al-Mohammad and Mant, 2011).
The 2,7-dichloro dihydro fluorescein diacetate (DCFH-DA) fluorescent probe can be used to detect intracellular reactive oxygen species content. The cells were cultured on cover slips, incubated with DCFH-DA (Sigma-Aldrich, St. Louis, MO) (50 μmol L−1, 37°C) for 30 min. Then the samples were analyzed by the laser scanning confocal microscope (excitation 488 nm, emission 525 nm).
Statistical analysis was performed by SPSS 11.0 software. The experimental data were expressed as mean±standard deviation. ANOVA was used to compare two groups (P<0.05 was considered statistically significant).
Cisplatin-Induced Autophagy and Phosphorylation of Akt in U251 Cells
After 24 and 48 h of cisplatin treatment, the IC50 values for U251 cells were 13.40 and 5.23 μg mL−1, respectively. However, the peak plasma concentration of cisplatin is only 2.5 μg mL−1in vivo. Therefore, U251 cells were relatively resistant to cisplatin. In this study, cisplatin concentration of 5.23 μg mL−1 was used in the following experiments. U251 cells were treated with 5.23 μg mL−1 (IC50 of cisplatin treatment for 48 h) cisplatin for 6, 12, and 24 h (Fig. 1). Bright green dots generated in MDC staining and red fluorescence dots generated in AO staining revealed increased acidic vacuoles especially late-lysosome or mature autophagic vacuoles. Compared to untreated cells, the results of both MDC staining and AO staining indicated that acidic vacuoles were enhanced in U251 cells after 5.23 μg mL−1 cisplatin exposure for 6, 12, and 24 h (Fig. 1A). LC3-I conversion to LC3-II is a landmark feature of autophagic activity. Compared to untreated cells, both LC3-I and LC3-II expressions were enhanced in U251 cells after 5.23 μg mL−1 cisplatin exposure for 6, 12, and 24 h (Fig. 1B,C). Both acidic vacuoles and conversion of LC3-I to LC3-II revealed that autophagy occurred in response to cisplatin and increased to a peak at 12 h. In addition, subcellular structures detection by transmission electron microscope showed a large number of double membrane vesicles in the cytoplasm of U251 cells treated with cisplatin for 12 h, compared to control cells (Fig. 1D). The above results demonstrated that autophagy occurred in U251 cells with cisplatin treatment.
In this study, we utilized PTEN-deficient human glioma cell line U251 that has high levels of Akt. The results of Western blot assay showed that the phosphorylated form of Akt enhanced at 6 h of cisplatin treatment and maintained a high level until 12 h, then decreased at 24 h slightly (Fig. 1E,F). So, in this study, both autophagy and Akt phosphorylation were induced by cisplatin in U251 cells.
ClC-3 Knock Down by siRNA Facilitates U251 Cells Sensitivity to Cisplatin
To evaluate the function of ClC-3 in cisplatin resistance, pSilencer 3.1-ClC-3 siRNA plasmids were transfected into U251 cells. RT-PCR and Western blot analysis showed that ClC-3 mRNA and protein expression levels were significantly decreased in ClC-3 siRNA transfected cells (inhibition ratio>80%) (Fig. 2A–D). ClC-3 knock down by siRNA increased the cell inhibition ratio of U251 cells with cisplatin treatment for 6, 12, and 24 h (Fig. 2E–G). The drug sensitivity of ClC-3 siRNA transfected U251 cells to cisplatin was determined by IC50 (Fig. 2H).
ClC-3 siRNA Decreases ROS Production and Activation of Akt Induced by Cisplatin
Cisplatin induces reactive oxygen species (ROS) production in various normal and cancer cells with different mechanisms. To determine whether the activation of Akt/mTOR pathway requires cisplatin-induced ROS generation, we examined ROS levels in U251 cells by DCFH-DA fluorescence staining. The results showed that the green fluorescence was increased in the cytoplasm with cisplatin treatment. Both ClC-3 knock down and NADPH oxidase inhibitor DPI pretreatment for 60 min decreased ROS production induced by cisplatin (Fig. 3A). The results indicate that cisplatin induces production of ROS, at least in part, through NADPH oxidase. Akt, mTOR, p70S6K, and their phosphorylated forms were detected by Western blot assay. A significant enhancement of Akt, mTOR, and p70S6K phosphorylation was observed in 5.23 μg mL−1 cisplatin-treated U251 cells. Both ClC-3 knock down and pretreatment with DPI for 60 min decreased Akt, mTOR, and p70S6K phosphorylation as well as cisplatin-induced ROS generation (Fig. 3B).
ClC-3 siRNA Decreases Autophagy and Increases Apoptosis by Cisplatin in U251 Cells
Both ClC-3 knock down and the autophagy inhibitor 3-MA pretreatment for 30 min significantly decreased cisplatin-induced autophagy (Fig. 4A,B). The results showed that levels of beclin 1 were enhanced after treatment with cisplatin, but decreased dramatically by ClC-3 knock down or pretreatment with 3-MA for 30 min (Fig. 4B). The Akt/mTOR pathway negatively regulates autophagy. So, DPI pretreatment increased cisplatin-induced autophagy through inhibition of the Akt/mTOR pathway. We examined apoptosis by PI and Annexin V-FITC double staining. The results showed that DPI or 3-MA pretreatment increased apoptotic cells through the Akt/mTOR pathway inhibition or autophagy inhibition, respectively. ClC-3 knock down significantly increased apoptotic cells through concomitant inhibition of autophagy and Akt/mTOR pathway.
The platinum-based drug, Cisplatin, is a potent DNA-damaging anticancer agent. Discovered >30 years ago, it remains one of the most widely used anticancer agents in the world. The traditional theory is that cisplatin exerts its cytotoxic effects through the formation of DNA adducts, which eventually culminates in irreversible apoptosis (Ding et al., 2012). But the mechanism of cisplatin resistance is not entirely clear (Borst et al., 2008).
Upregulation of survival pathways such as the Akt/mTOR pathway, caused by the progressive acquisition of multiple genetic changes, at least partly, induce chemoresistance of cancer cells. In this study, we utilized PTEN-deficient human glioma cell line U251 that has high levels of Akt. According to the results of MTT assay, the IC50 values for cisplatin after 24 and 48 h were 13.40 and 5.23 μg mL−1, respectively. U251 cells showed relatively high resistance to cisplatin (Kondo et al., 1998). However, the peak plasma concentration of cisplatin is only ∼2.50 μg mL−1in vivo (Daley-Yates and McBrien, 1984). So, in our study, cisplatin concentration of 5.23 μg mL−1 was used in all experiments. The results showed that enhancement of Akt phosphorylation and autophagy was induced by 5.23 μg mL−1 cisplatin for 6, 12, and 24 h. The Akt phosphorylation enhanced at 6 h with cisplatin treatment and maintained a high level until 12 h, then decreased at 24 h slightly. Here we encountered a puzzling question. It is well known that the Akt/mTOR signaling pathway inhibits autophagy. But in our results both Akt phosphorylation and autophagy were induced by cisplatin. So we think that cisplatin may induce autophagy through other pathways, such as ubiquitinated proteins, as shown in our previous study of cisplatin-resistant human ovarian cancer cells (Yu et al., 2012, 2011). Both Akt phosphorylation and autophagy play survival roles in resistance of U251 cells to cisplatin.
ClC-3 is expressed predominantly in acidic intracellular compartments such as late-endosome, lysosome, and mature autophagyosome (Jentsch, 2007). A growing body of research shows that ClC-3 plays an important role in cell proliferation, invasion, migration, and drug-resistance (Mao et al., 2008; Tang et al., 2008; Lui et al., 2010; Xu et al., 2010). In this study, ClC-3 was efficiently suppressed by transfection with pSilencer 3.1-ClC-3 siRNA plasmids to investigate the function of ClC-3 in chemoresistance. ClC-3 knock down by siRNA enhances the drug sensitivity of U251 cells to cisplatin. Consistent with other previous studies, the results indicated that ClC-3 plays cytoprotective role in chemotherapy (Weylandt et al., 2007). But one question we faced now is whether suppression of ClC-3 increased the drug sensitivity of U251 cells to cisplatin and is simultaneously related to cisplatin-induced Akt phosphorylation. Recent studies show that NADPH oxidase (Nox) generates cisplatin-induced ROS in the inner ear and human gliomas (Shono et al., 2008; Kim et al., 2010). The ROS generated by Nox is closely related to cancer development, inhibition of apoptosis, chemotherapy, and radiation resistance (Mates et al., 2008; Reuter et al., 2010). Cisplatin activates the PI3K/Akt signaling pathway that in part protects cells from drug-induced apoptosis (Belyanskaya et al., 2005). In this study, cisplatin-induced ROS was suppressed by ClC-3 knock down or pretreatment with Nox inhibitor DPI. Both ClC-3 knock down and DPI pretreatment decreased Akt, mTOR, and p70S6K phosphorylation as well as ROS generation induced by cisplatin. The results indicated that cisplatin-induced ROS, at least in part, was generated through Nox and activated the Akt/mTOR signaling pathway, and ClC-3 is required for the process involved. As a chloride/proton exchanger, ClC-3 can control ROS generation by Nox. The processing of generating ROS by Nox is electrogenic and needs charge neutralization. ClC-3 provides charge neutralization for ROS generation by Nox (Yamaguchi and Wang, 2004; Hyoda et al., 2006; Bromati et al., 2011). In this study, ROS was increased with cisplatin treatment. Both DPI and ClC-3 siRNA could decrease ROS levels and Akt phosphorylation induced by cisplatin. Therefore, phosphorylation of Akt is activated by cisplatin through Nox and requires ClC-3.
Our previous studies and results from other studies show that autophagy has a protective role for cisplatin (Yu et al., 2011; Xu et al., 2012). Autophagy could play prosurvival role or promote cell death. The 3-MA is an autophagy inhibitor and was used in this study to investigate the role of autophagy on cisplatin-induced cell death. Our results showed that autophagy-related gene beclin 1 levels were increased with cisplatin treatment, but decreased dramatically by ClC-3 knock down or 3-MA pretreatment. The Akt/mTOR pathway negatively regulates autophagy and DPI pretreatment increased cisplatin-induced autophagy through inhibition of the Akt/mTOR signaling pathway. DPI or 3-MA pretreatment increases cell apoptosis through the Akt/mTOR pathway inhibition or autophagy inhibition, respectively.
The above results suggest that ClC-3 plays double roles in cisplatin resistant mechanism in U251 cells. On one hand, ClC-3 promotes the Akt/mTOR pathway through generating ROS by Nox. On the other hand, since ClC-3 has indispensable role in acidification of acidic intracellular compartments such as late-endosome, lysosome and mature autophagosome, the deficiency of ClC-3 probably induce autophagy. Although the two possible roles of ClC-3 in cisplatin resistance have been discussed in this study, the exact mechanism still needs further investigation. So we think ClC-3 knock down significantly increased apoptotic cells, at least in part, through concomitant inhibition of autophagy and Akt.
In summary, we describe the role of autophagy and Akt pathway involved in the mechanism of resistance to cisplatin in U251 cells. Here, we showed that cisplatin elicited autophagy and activation of Akt/mTOR pathway in human malignant glioma U251 cells. We also showed that ClC-3 knock down resulted in concomitant inhibition of autophagy and Akt/mTOR pathway and significantly enhanced apoptotic cells induced by cisplatin. ClC-3 suppression causes the inhibition of Akt and autophagy, which can enhance the therapeutic benefit of cisplatin in U251 cells.
The authors thank Dr. William Orr (University of Manitoba, Winnipeg, Manitoba, Canada) for assistance with revising the manuscript.