This article is mentioned in:

Introduction

Cholangiocarcinoma is a malignant tumor originating from the bile duct epithelium (1,2). Complete surgical resection is currently the only curative treatment for patients with cholangiocarcinoma. However, due to difficulties in the early diagnosis and its high metastatic potential, most patients with cholangiocarcinoma have regional or distant metastasis at the time of clinical diagnosis (3–5). Few specific diagnostic or therapeutic tools are available because of the limited information on the molecular pathogenesis of cholangiocarcinoma. Therefore, new therapeutic targets are urgently needed.

Epigenetic gene silencing is an important mechanism that results in loss of gene expression and that mediates, along with genetic mutations, the initiation and progression of human cancer. Polycomb group proteins regulate and mediate epigenetic transcriptional silencing (6,7). They are involved in the maintenance of embryonic and adult stem cells and in repression of key tumor-suppressor pathways, which may contribute to their oncogenic function (8). The enhancer of zeste homolog 2 (EZH2) is the catalytic subunit of the polycomb repressive complex 2 (PRC2), which also includes the suppressor of zeste 12 (SUZ12) protein and embryonic ectoderm development (EED) protein. EZH2 acts as a histone lysine methyltransferase that mediates trimethylation of lysine 27 of histone H3 (3mH3K27) to silence expression of PRC2 target genes involved in lineage differentiation (9,10). EZH2 is overexpressed in a variety of malignancies including prostate cancer, breast cancer, melanoma, uterine cancer, gastric cancer and renal cell cancer. EZH2 expression levels are correlated with aggressiveness, metastasis and a poor patient prognosis in most of these cancers (11–15). EZH2 is barely expressed in normal tissues of various types (16).

More recently, we found that cholangiocarcinoma also overexpresses EZH2 and that high expression of EZH2 is correlated with poor patient prognosis (17). Furthermore, an activating mutation in EZH2 has been identified in a subset of B-cell lymphomas (18). Overexpression of EZH2 enhances aggressiveness in prostatic cancer cells (19) and produces a neoplastic phenotype characterized by independent growth and cell invasion in immortalized mammary epithelial cells and bronchial epithelial cells (13,20). Conversely, depletion of EZH2 results in reduced proliferation, increased apoptosis, and inhibition of tumorigenicity in cancer cells (12,19,21,22) including cholangiocarcinoma cells (17). These findings indicate that EZH2 may be an appropriate therapeutic target for various types of cancers.

A cyclopentenyl analog of 3-deazaadenosine, 3-deazaneplanocin A (DZNep), inhibits the activity of S-adenosyl-L-homocysteine (AdoHcy) hydrolase, the enzyme responsible for the reversible hydrolysis of AdoHcy to adenosine and homocysteine (23). This inhibition results in the intracellular accumulation of AdoHcy, which leads to inhibition of S-adenosyl-L-methionine-dependent lysine methyltransferase activity. Presently, DZNep was shown to reduce levels of the PRC2 complex, including EZH2, in breast cancer cells and cause concomitant loss of 3mH3K27 and repression of epigenetically silenced target genes (24). Moreover, DZNep was found to inhibit proliferation and promote apoptosis in several types of cancer cells (24–29), and the synergistic interaction of DZNep and gemcitabine enhanced the inhibition of cell proliferation and apoptosis in pancreatic cancer cells (30).

However, data on the activity of DZNep in cholangiocarcinoma cells are scarce. In the present study, we demonstrated the effects of DZNep alone and in combination with gemcitabine on cholangiocarcinoma cells.

Materials and methods

Cell lines and chemicals

The cholangiocarcinoma cell lines RBE and TFK-1 were purchased from RIKEN Bioresource Center (Ibaraki, Japan). RBE and TFK-1 cell lines were cultured in RPMI-1640 (Invitrogen, Tokyo, Japan) containing 10% fetal bovine serum (FBS). All cultures were maintained in a 5% CO2 air-humidified atmosphere at 37°C. DZNep and gemcitabine were provided by Cayman Chemical Japan and Eli Lilly Japan, respectively.

Cell growth assay

Cells were seeded in a 96-well plate at a density of 5000 cells per well and treated with or without DZNep (0.5 μM) or gemcitabine (30 nM) or the combination. Viable cholangiocarcinoma cell numbers were measured with the Cell Counting Kit-8 (CCK-8) containing 2-(2-methoxy-4- nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (WST-8) (Dojin Laboratories, Kumamoto, Japan) for 24, 48 and 72 h, according to the manufacturer’s instructions. OD 450 was measured using an automatic microplate reader (Molecular Devices, Osaka, Japan). Each experiment was performed in triplicate.

Flow cytometric analysis of cell cycle distribution and detection of apoptosis

RBE and TFK-1 cells were seeded in a 12-well plate. After 72 h of treatment with or without DZNep (0.5 nM) or gemcitabine (30 nM) or the combination, cells were trypsinized, adjusted to 5×105/ml, and fixed in 70% ethanol overnight at −20°C. Subsequently, cells were pelleted and resuspended in phosphate-buffered saline containing 1 μg/ml RNase A (Sigma, Tokyo, Japan) and 100 μg/ml propidium iodide (PI) (Sigma). Cell cycle analyses were performed using a FACS Aria II (BD Biosciences, Tokyo Japan). The distribution of cells in different cell cycle phases was calculated using FlowJo software (Tomy Digital Biology Co., Ltd., Tokyo, Japan). Phosphatidylserine externalization was detected by Annexin V staining (Millipore, Tokyo, Japan) following the manufacturer’s instructions. Annexin V-positive and PI-negative cells were detected as early apoptotic cells.

Statistical analysis

Statistical analysis was performed using the JMP program (SAS Institute, Cary, NC, USA). Quantitative data are expressed as means ± SD, unless otherwise stated. The χ2 test was used to analyze the relationship between categorical variables. For all statistical analyses, the level of significance was set at 0.05.

Results

DZNep reduces EZH2 expression and demethylated H3K27 in cholangiocarcinoma cells

The expression of EZH2 was studied at the protein level in untreated RBE and TFK-1 cells and in cells treated with 0.1–5 μM DZNep. As shown in Fig. 1, DZNep reduced the expression of EZH2 in a dose-dependent manner. In addition, we investigated the expression of the 3mH3K27 protein, which was also reduced by DZNep in a dose-dependent manner in both cell lines.

Figure 1 Western blot analysis of the RBE and TFK-1 cholangiocarcinoma cell lines. Cell lysates were collected 72 h after treatment with DZNep. Lysates were subjected to western blot analysis. Representative western blots of EZH2 and 3mH3K27; β-actin acted as a loading control.

DZNep inhibits the growth of cholangiocarcinoma cells and interacts synergistically with gemcitabine

We assessed the effect of DZNep-mediated EZH2 repression on the growth of RBE and TFK-1 cholangiocarcinoma cells (Fig. 2). The cell growth assay demonstrated that 0.5 μM DZNep caused significantly inhibited proliferation of cholangiocarcinoma cells. As compared with control cells on day 3, DZNep reduced the growth of RBE and TFK-1 cells to 67.0 and 67.6%, respectively. Similarly, 30 nM gemcitabine also significantly inhibited cell proliferation to 58.4 and 69.5% (in RBE and TFK-1 cells, respectively) as compared with the control cells on day 3. The combination of 0.5 μM DZNep and 30 nM gemcitabine enhanced the antiproliferative activity of gemcitabine. The combination of DZNep and gemcitabine reduced the proliferation of RBE and TFK-1 cells to 61.2 and 78.6% compared with gemcitabine alone, and to 35.7 and 53.1% compared with the control cells (Fig. 2).

Figure 2 Inhibition of cell proliferation of the RBE and TFK-1 cholangiocarcinoma cell lines and pharmacological interaction of DZNep and gemcitabine. Experiments were performed in triplicate. Standard deviations are indicated. *p<0.05, **p<0.01.

DZNep induces cell cycle arrest and the combination of DZNep and gemcitabine enhances cell cycle arrest

We used flow cytometry to determine whether the reduction in proliferation was due to cell cycle arrest or apoptosis in the two cholangiocarcinoma cell lines. DZNep alone and in combination with gemcitabine affected the cell cycle of cholangiocarcinoma cells (Fig. 3). In particular, DZNep increased the percentage of RBE and TFK-1 cells in the G0/G1 phase from 46.5 to 50.2% and from 49.1 to 59.2%, respectively (p<0.05). Gemcitabine alone reduced the number of cells in the G0/G1 phase to 28.3 and 36.1%, respectively (p<0.05). The combination of DZNep and gemcitabine significantly increased the percentage of RBE and TFK-1 cells in the G0/G1 phase compared with gemcitabine alone (from 28.3 to 60.2% and from 36.1 to 45.4%, respectively). Additionally, DZNep reduced the percentage of cells in the S and G2/M phases, and the combination of DZNep and gemcitabine significantly reduced the percentage of cells in the S and G2/M phases compared with gemcitabine alone.

Figure 3 Effect of DZNep and gemcitabine on cell cycle distribution in the RBE and TFK-1 cholangiocarcinoma cell lines. Cells were treated with 0.5 μM DZNep and 30 nM gemcitabine and in combination. After 72 h, the percentage of cells in each cell cycle phase was measured using a FACS flow cytometer. Experiments were performed in triplicate. Standard deviations are indicated. *p<0.05, **p<0.01.

DZNep induces cell apoptosis and the combination of DZNep and gemcitabine enhances apoptosis

Flow cytometry using Annexin V and PI staining, and analysis of the sub-G1 region demonstrated that DZNep or gemcitabine alone significantly enhanced apoptosis when compared with the control cells (Fig. 4). Treatment with 0.5 μM DZNep increased apoptosis 2-fold, and treatment with 30 nM gemcitabine resulted in ~5-fold increase in apoptosis. Furthermore, the combination of DZNep and gemcitabine significantly increased the number of apoptotic cells when compared with either agent alone (~5-fold and 2-fold increases in the RBE and TFK-1 cells, respectively). We also investigated expression of cleaved caspase-3 and cleaved polyADP ribose polymerase (PARP) as markers of apoptosis to confirm the apoptotic effect of the combination of DZNep and gemcitabine. A slight increase in these proteins was noted followed treatment with DZNep alone, but the combination of DZNep and gemcitabine resulted in a significant increase in apoptotic proteins in the RBE and TFK-1 cells (Fig. 5). These results confirmed the flow cytometric analysis.

Figure 4 Analysis of the apoptosis of the RBE and TFK-1 cholangiocarcinoma cell lines using flow cytometry with Annexin V and PI staining, and determination of sub-G1 cells. Cells were treated with 0.5 μM DZNep and 30 nM gemcitabine or the combination. After 72 h, the percentage of cells in the apoptotic fraction was measured using a FACS flow cytometer. Experiments were performed in triplicate. Standard deviations are indicated. *p<0.05, **p<0.01.

Figure 5 Western blot analysis of the RBE and TFK-1 cholangiocarcinoma cell lines. Cell lysates were collected 72 h after treatment with 0.5 μM DZNep and 30 nM gemcitabine or the combination. Lysates were then subjected to western blot analysis. Representative western blots of EZH2, 3mH3K27, p16INK4a, p27KIP1, cleaved PARP and cleaved caspase-3; β-actin acted as a loading control.

DZNep depletes EZH2 and 3mH3K27 and increases p16INK4a and p27KIP1 in cholangiocarcinoma cells and the combination of DZNep and gemcitabine enhances these effects

DZNep reduced the expression of EZH2 protein and 3mH3K27 in cholangiocarcinoma cell lines (Fig. 5). The combination of DZNep and gemcitabine also reduced the expression of EZH2 and 3mH3K27 when compared with gemcitabine alone. We determined the effect of DZNep on EZH2 target genes such as p16INK4a and p27KIP1. Expression of p16INK4a and p27KIP1 was increased by DZNep. The combination of DZNep and gemcitabine enhanced the increase in expression of p16INK4a and p27KIP1.

Discussion

The present study demonstrated that the combination of the EZH2 inhibitor DZNep and the cytotoxic compound gemcitabine exhibited strong synergistic effects on the cholangiocarcinoma cell lines. EZH2 has a master regulatory role in the fate of native embryogenic cells (31), as well as in cancer development via methylation-mediated repression of the transcription of several genes (32,33). Overexpression of EZH2 is a marker of advanced and metastatic disease in many solid tumors, including cholangiocarcinoma (12,13,34). EZH2 is crucial for regulating the cell cycle by repressing several tumor-suppressor genes such as p16INK4a, p27KIP1 and RUNX3, and by mediating the pRb pathway. In addition, EZH2 regulates E2F1-dependent apoptosis directly in cancer cells, through modulation of Bim expression.

Our previous study on RBE and TFK-1 cells showed that suppression of EZH2 expression by RNA interference with siRNA targeting EZH2 markedly inhibited cellular proliferation in vitro, and drastically induced both cell cycle arrest and apoptosis (17,35). Knockdown of EZH2 by siRNA increased expression of p16INK4a and p27KIP1, which are considered to be the targets of EZH2 (17). According to other research, transfection of shEZH2 construct cells sensitized MIA-PaCa-2 and Pac04.02 cells to doxorubicin and gemcitabine in pancreatic cancer (21), suggesting that the combination of EZH2 inhibitors and gemcitabine might overcome the intrinsic chemoresistance of cholangiocarcinoma. To the best of our knowledge, this is the first study evaluating the pharmacological interaction of the small-molecule EZH2 inhibitor DZNep with gemcitabine in cholangiocarcinoma cells.

In the cholangiocarcinoma cell lines, we observed a significant reduction of both EZH2 and 3mH3K27 expression in a dose-dependent manner. Analyses of the cell cycle distribution and apoptosis indicated that the growth suppression by DZNep was associated with G1 cell cycle arrest and increased apoptosis in cholangiocarcinoma cells. This conclusion is consistent with the findings from acute myeloid leukemia (AML) cells treated with DZNep (27). Knockdown of EZH2 by siRNA has also been shown to induce G1 cell cycle arrest and to increase apoptosis in cholangiocarcinoma cells and other types of carcinoma cells (36). DZNep induced accumulation of p16INK4a and p27KIP1 in RBE and TFK-1 cells in the present study, and this may have resulted in G1 arrest and increased apoptosis. An association between either p16INK4a or p27KIP1 accumulation and G1 cell cycle arrest and apoptosis has been shown in various types of cells (37,38). Fiskus et al reported that DZNep treatment induced p16INK4a, p21CIP1, p27KIP1 and FBXO32 while reducing cyclin E and HOXA9 levels in human AML cells (27). Taken together, these results indicate that growth suppression by DZNep is associated with G1 phase arrest and apoptosis in cholangiocarcinoma cells, partly via p16INK4a and p27KIP1 accumulation.

The Advanced Biliary Cancer (ABC)-02 study showed that systemic combination chemotherapy of gemcitabine and cisplatin prolonged the survival of patients with inoperable cholangiocarcinoma, making it a treatment standard, and gemcitabine is still the key drug used in chemotherapy of advanced cholangiocarcinoma (39,40). In the present study, we demonstrated that the combination of DZNep and gemcitabine has a synergistic effect on two representative cholangiocarcinoma cell lines, RBE and TFK-1. This synergistic interaction against cell proliferation was associated with a significant increase in apoptosis induction and cell cycle arrest. The combination of gemcitabine and DZNep increased the expression of p16INK4a and p27KIP1 when compared with gemcitabine alone, so that the synergistic effect might be partly explained by the significant upregulation of p16INK4a and p27KIP1.

In conclusion, EZH2 inhibitor DZNep appears to be a promising anticancer agent, by attacking key mechanisms involved in the proliferation, cell cycle control and apoptosis of cholangiocarcinoma cells. Moreover, the favorable modulation of p16INK4a and p27KIP1 makes DZNep an optimal candidate for combination therapy with gemcitabine. The synergistic effects observed in the present study may have critical implications for the rational development of innovative regimens consisting of DZNep and gemcitabine to improve the effectiveness of the treatment of cholangiocarcinoma.

References