Salinomycin Inhibits Proliferation and Induces Apoptosis of Human Hepatocellular Carcinoma Cells In Vitro and In Vivo

The anti-tumor antibiotic salinomycin (Sal) was recently identified as a selective inhibitor of breast cancer stem cells; however, the effect of Sal on hepatocellular carcinoma (HCC) is not clear. This study aimed to determine the anti-tumor efficacy and mechanism of Sal on HCC. HCC cell lines (HepG2, SMMC-7721, and BEL-7402) were treated with Sal. Cell doubling time was determinated by drawing growth curve, cell viability was evaluated using the Cell Counting Kit 8. The fraction of CD133+ cell subpopulations was assessed by flow cytometry. We found that Sal inhibits proliferation and decreases PCNA levels as well as the proportion of HCC CD133+cell subpopulations in HCC cells. Cell cycle was analyzed using flow cytometry and showed that Sal caused cell cycle arrest of the various HCC cell lines in different phases. Cell apoptosis was evaluated using flow cytometry and Hoechst 33342 staining. Sal induced apoptosis as characterized by an increase in the Bax/Bcl-2 ratio. Several signaling pathways were selected for further mechanistic analyses using real time-PCR and Western blot assays. Compared to control, β-catenin expression is significantly down-regulated upon Sal addition. The Ca2+ concentration in HCC cells was examined by flow cytometry and higher Ca2+ concentrations were observed in Sal treatment groups. The anti-tumor effect of Sal was further verified in vivo using the hepatoma orthotopic tumor model and the data obtained showed that the size of liver tumors in Sal-treated groups decreased compared to controls. Immunohistochemistry and TUNEL staining also demonstrated that Sal inhibits proliferation and induces apoptosis in vivo. Finally, the role of Sal on in vivo Wnt/β-catenin signaling was evaluated by Western blot and immunohistochemistry. This study demonstrates Sal inhibits proliferation and induces apoptosis of HCC cells in vitro and in vivo and one potential mechanism is inhibition of Wnt/β-catenin signaling via increased intracellular Ca2+ levels.


Introduction
Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and the third most common cause of cancer deaths, resulting in more than 600,000 deaths annually [1]. Early HCC detection due to advances in diagnostic modalities and clinical screening has made it possible to perform curative or palliative treatment with surgical excision [2]. Unfortunately, HCC is often diagnosed at a late stage, when potentially curative therapies are least effective. For these patients, medical treatments including chemotherapy, chemoembolization, ablation, and proton beam therapy remain disappointing. Most patients show disease recurrence that rapidly progresses to advanced stages with vascular invasion and multiple intrahepatic metastases and the 5year relative survival rate is only 7%. Furthermore, the prognosis for HCC patients who have surgically resectable, localized tumors remains dismal, with a 5-year recurrence rate of 40-70%. Therefore, there is an urgent need for new therapies for this aggressive disease. Salinomycin (Sal) is a polyether organic anion used extensively in poultry as a coccidiostatic antibiotic and is commonly fed to ruminant animals to improve feed efficiency [3]. As an ionophore with strict selectivity for alkali ions, it acts in different biological membranes, including cytoplasmic and mitochondrial membranes [4]. Recently, a robotic high-throughput screening approach evaluated approximately 16,000 compounds from chemical libraries for activity against human breast cancer stem cells and found only Sal markedly and selectively reduced the viability of stem-like cells [5]. Additionally, other studies showed Sal could induce apoptosis in chronic lymphocytic leukemia cells [6] and human prostate cancer cells [7]. However, little is known about its impact on HCC cells. HCC is a complex and heterogeneous tumor with several genomic alterations and aberrant activation of several signaling cascades including Wnt, Hedgehog, transforming growth factor-beta (TGF-b), epidermal growth factor receptor (EGFR), vascular epidermal growth factor receptor (VEGFR), mitogen-activated protein kinase (MAPK), and AKT [8].In this study, we aimed to determine the effect of Sal on HCC cells and its underlying mechanisms of action.

Cell Lines and Culture
The HCC cell lines HepG2, SMMC-7721, and BEL-7402 were purchased from the Chinese Academy of Sciences Committee Type Culture Collection cell bank. The three cell lines were cultured in high glucose Dulbecco's modified Eagle's medium (DMEM-h; Thermo, China) supplemented with 10% fetal bovine serum (Hycione, South America), 100 U/ml penicillin, and 100 mg/ml streptomycin (Gibco,Canada) in a humidified incubator at 37uC in 5% CO 2 .

Drugs and Antibodies
Sal was purchased from Sigma Aldrich (St. Louis, MO, USA). A 50 mM Sal stock solution was made in dimethyl sulfoxide (DMSO; Gibco, Canada) and stored in the dark at 220uC. The final Sal concentrations used for different experiments were prepared by diluting the stock solution with high-glucose DMEM. The antibodies used for Western blotting and immunohistochemistry staining were as follows: rabbit anti-PCNA, anti-Bcl-2, anti-Bax, anti-p-GSK-3b, anti-GSK-3b, anti-b-catenin, and mouse anti-b-actin. All antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Cell Growth Curve and Determination of Doubling Time
Three HCC cell lines HepG2, SMMC-7721 and BEL-7402 in logarithmic growth phase were re-suspended, seeded into 24-well plates at a density of 2610 4 cells/well. One day after seeding, 3 wells were taken to calculate the number of cells as the starting cell number, then Sal was added in three replicates of 0 mM, 2.5 mM, 5 mM, 10 mM for each cell population. Trypan blue staining was performed once daily since the 1st day of culture, and the viable cells in 3 wells of each group were counted for five consecutive days. The doubling time (Td) of each cell line was calculated according to the Patterson formula as follow: Td = T6lg2/ (lgN22lgN1), where N1 is cell number on the 1st day, and N2 is cell number at T h after culture; T (h) is the time from N1 to N2.

Cell Viability
HepG2, SMMC-7721, and BEL-7402 cells were plated in 96well plates (100 ml media per well). One day after seeding, Sal was added in five replicates of 0 mM, 5 mM, 10 mM, 15 mM, 20 mM, 25 mM, and 30 mM for each cell population. Cell viability was measured after 24 h, 48 h, and 72 h using the Cell Counting Kit 8 (Peptide Institute Inc., Osaka, Japan) and a microplate reader at a wavelength of 450 nm. A calibration curve was prepared using the data obtained from wells that contained a known number of viable cells. CD133 + Cell Subpopulation Analyses using Flow Cytometry HepG2, SMMC-7721, and BEL-7402 cells were plated in 6well plates. After 48 h, cells from the control group and the Sal [half maximal inhibitory concentration (IC 50 ) and 80% inhibitory concentration (IC 80 )] group were collected, washed twice in cold phosphate buffered saline (PBS).Dissociated cells were stained with phycoerythrin-conjugated CD133 antibody (Miltenyi Biotec, Auburn, CA) and incubated for 30 min at 4uC. Mouse IgG1phycoerythrin was used as isotype control antibody. Dead cells were eliminated with 7-aminoactinomycin D. The labeled cells were analyzed by the BD FACS Vantage Systems in accordance with the manufacturer's protocols. Gating was implemented on the basis of negative-control staining profiles.

Cell Cycle Analyses using Flow Cytometry
HepG2, SMMC-7721 and BEL-7402 cells were plated in 6-well plates and synchronized with DMEM containing 1% fetal bovine serum. After 48 h, control cells, DMSO-treated cells, and Saltreated cells (IC 50 ) were collected, washed twice in cold PBS,

Intracellular Calcium Analyses using Flow Cytometry
HepG2, SMMC-7721 and BEL-7402 cells were plated in 6-well plates. After 48 h, control cells, DMSO-treated cells, and Saltreated cells (IC 50 ) were collected, washed twice in cold PBS, mixed in 300 ml 16 binding buffer, and incubated at room temperature for 30 min with intracellular calcium fluorescent probes (fluo-3) (Sigma Aldrich) staining solution. The intracellular calcium levels were analyzed by flow cytometry and the mean fluorescence intensity was calculated using ModFit LT software (Verity Software House).

Animal Experiments
Animal experiments were performed on 6-week-old male nude mice (athymic, BALB/C nu/nu). Mice were housed in a standard animal laboratory with free access to water and food. They were kept under constant environmental conditions with a 12-hour light-dark cycle. All operations were performed under aseptic conditions. All procedures were approved by the Animal Care and Use Committee of Shanghai Tongji University.

Hepatoma Orthotopic Tumor Model and Treatment Regimen
Nude mice (nu/nu; 4-6 weeks of age) were purchased from Shanghai Slac Laboratory Animal Co., Ltd. HepG2 cells were suspended in 100 ml1:1 serum-free DMEM and Matrigel (BD Biosciences). Mice were anesthetized with ketamine/xylazine and after surgically opening the abdomen, HepG2 cells were inoculated into the liver parenchyma and mice were monitored every 3 days for 35 days. Finally, 18 nude mice were divided into three groups that were intraperitoneally injected daily for 6 weeks: two Sal-treated groups (4 mg/kg Sal group, 8 mg/kg Sal group) and the control group (saline water group)

Immunohistochemistry
Tumor tissue from control group (saline water group) and Saltreated groups are used for immunohistochemistry. Sections (4 mm thick) from paraffin-embedded tumors were deparaffinized and rehydrated using xylene and ethanol, respectively, and immersed in 3% hydrogen peroxide solution for 10 min to block endogenous peroxidases. Sections were boiled for 30 min in 10 mM citrate buffer solution (pH 6.0) for antigen retrieval. Slides were incubated for 45 min with 5% bovine serum albumin and incubated overnight at 4uC with anti-PCNA, anti-Bcl-2, anti-Bax, or anti-b-catenin. These specimens were incubated for 45 min at 37uC with the appropriate peroxidase-conjugated secondary antibody and visualized using the Real Envision Detection Kit (Gene Tech Shanghai Company Limited, China) following the manufacturer's instructions. All slides were counterstained with hematoxylin and eosin (HE).

Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling (TUNEL) assays
Tumor tissue from control group (saline water group) and Saltreated groups are used for TUNEL assays. Sections (4 mm) from formalin fixed, paraffin-embedded tumors were deparaffinized and rehydrated using xylene and ethanol, respectively. The slides were rinsed twice with PBS and treated for 15 min at 37uC with proteinase K (15 mg/ml in 10 mM Tris/HCl, pH 7.4-8.0). Endogenous peroxidases were blocked using 3% hydrogen peroxide in methanol at room temperature for 10 min. The tissue sections were then analyzed with an in situ Cell Death Detection Kit-POD (Roche, Germany) in accordance with the manufacturer's instructions. The reaction was visualized using fluorescence microscopy.  Table 1. PCR  reactions were amplified for 40 cycles. Each cycle consisted of denaturation for 1 min at 94uC, annealing for 1 min at 60uC, and polymerization for 2 min at 72uC. PCR products were quantified using the Molecular Analyst software (Bio-Rad). The ratio of each gene against b-actin was calculated by standardizing the ratio of each control to the unit value.

Western Blot Assays
HepG2, SMMC-7721 and BEL-7402 cells were treated with Sal (IC50) for two days. Cells were washed twice with PBS solution, then cells were lysed with RIPA Lysis Buffer (Beyotime Institune of Biotechnology, Shanghai, China) and protease inhibitor (Thermo Scientific). Protein concentrations were determined with Pierce BCA Protein Assay Kit (Thermo Scientific), Equivalent amounts of total protein (60 ug) were boiled and electrophoretically separated on a 10% polyacrylamide gel at 80 volts. The proteins were transferred to a polyvinylidene difluoride membrane. Membranes were blocked for 60 min with a 5% milk solution prepared in PBS. The membranes were incubated overnight at 4uC with 1:500 dilutions of the primary antibodies (PCNA, Bcl-2, Bax, p-GSK-3b, GSK-3b, b-catenin, and b-actin). Membranes were washed three times for 5 min each with Tween 20(1:1000 dilution)-PBS and incubated for 45 min with the appropriate peroxidase-conjugated secondary antibody (1:1000 dilution). Membranes were washed with Tween 20-PBS three times for 10 min each and were developed using the Odyssey two-color infrared laser imaging system. The signal generated by b-actin was used as an internal control.

Statistical Analyses
The SPSS17.0 software was used for statistical analyses. Experiments were repeated at least three times with consistent results. Unless otherwise stated, data are expressed as the mean 6 standard deviation. The data of Real-time PCR were 2-DDCt transformed before analysis and were analyzed using ANOVA, the results of CCK8 assay were analyzed by two-Way Analysis of Variance, The results of Western blots, CD133, cell cycle, annexin-V/PI, and fluo-3 staining were analyzed using analysis of ANOVA. Tumor mean diameter, and mean volume were analyzed for statistical significance using paired Student's t tests. In all cases, p,0.05 was considered significantly statistic difference.

Sal Inhibits HCC Cell Proliferation in vitro
The HCC cell lines HepG2, SMMC-7721 and BEL-7402 were treated with increasing concentrations of Sal for five days and we drew the growth curve (Fig. 1A). The data showed that the growth of Sal-treated cells was relatively slower than that of DMSOtreated cells in a dose-dependent manner, the Td of Sal-treated HepG2, SMMC-7721, and BEL-7402 cells were significantly longer than that of DMSO-treated cells (Table 2,* p,0.05). Sal cytotoxicity was determined using a CCK8 assay. The data indicate that Sal reduced the viability of HepG2, SMMC-7721, and BEL-7402 cells in a dose-and time-dependent manner ( Fig. 1B; p,0.05). The Sal concentrations that reduced growth by  50% and 80% (IC 50 and IC 80 ) are summarized in Table 3. We evaluated PCNA expression by real-time PCR and Western blot. The results showed that Sal could down-regulate the expression of PCNA mRNA and protein ( Fig. 1C; * p,0.05)..

Sal Reduces the Proportion of HCC CD133 + Cell Subpopulations in vitro
A small population of cancer cells, termed cancer stem cells, have unlimited proliferation, self-renewal, and multi-lineage differentiation potentials. The CD133+ protein, a pentaspan cell surface receptor, is a putative HCC stem cell marker. Flow cytometry was performed to analyze the effect of Sal on the proportion of HCC cells with a CD133+ antigenic phenotype. Sal reduced the proportion of CD133+ cells in HepG2 cells (Blank

Sal Causes Cell Cycle Arrest of HCC Cells in vitro
In this study, we found that Sal could inhibit proliferation of the HCC cell lines. Therefore, we investigated the mechanism of this inhibition. The effect of Sal on cell cycle was tested using flow cytometry analyses. As shown in Figure 3A and Table 4

Sal Induces HCC cell Apoptosis in vitro
To examine if Sal could induce apoptosis, we used flow cytometry analyses (Fig. 3B and Table 5, * p,0.05). An increase in the percentage of early apoptosis (quadrant 2) plus late apoptosis (quadrant 3) was identified in HepG2, SMMC-7721, and BEL-7402 cells after Sal treatment. Cell morphology changes were also observed using Hoechst 33342 staining (Fig. 3C). Using a laserscanning microscope, cell nuclei were round and homogeneous in the absence of Sal, whereas Sal treatment resulted in nuclear condensation, a hallmark feature of apoptotic cells. Moreover, real-time PCR and Western blotting revealed that Sal treatment increased the Bax/Bcl-2 ratio in HepG2, SMMC-7721, and BEL-7402 cells (Fig. 3D, * p,0.05).

Sal Increases Intracellular Calcium Levels in vitro
To assess whether Sal elevates intracellular calcium levels, Fluo-3 FACS analyses were used. As shown in Figure 3C, the mean fluorescence intensity of intracellular calcium was remarkably increased in HepG2, SMMC-7721, and BEL-7402 cells after treatment with Sal (IC 50 ; Fig. 5).

Anti-tumor Activity of Sal in vivo
In order to investigate the effect of Sal on anti-tumor activities in vivo, we established a HepG2 orthotopic tumor model in nude mice. After administration of 4 mg/kg Sal, 8 mg/kg Sal and 10 ul/g saline water for 6 weeks, the mice were sacrificed. The size of the liver tumors in the Sal treatment groups diminished compared with the control group (Fig. 6A). The mean diameter of the tumors decreased from 12.17 mm to 3.67 mm (Fig. 4B, * p,0.05) and the mean volume (V = length6width260.5) [9] of the tumors decreased from 819 mm 3 to 25.25 mm 3 (Fig. 6B, * p,0.05). Next, the tumors were harvested, followed by HE staining, immunohistochemistry, and TUNEL assays, to assess the anti-tumor activity of Sal in vivo. HE staining (Fig. 7A) showed that the structure of the liver cancer tissue:nuclei of different sizes, hepatic cord structure was destroyed. Immunohistochemistry showed that PCNA expression (Fig. 7B) was lower after Sal treatment. HE staining (Fig. 7C) and TUNEL (Fig. 7C) assays indicated the Sal-treated groups had higher apoptosis rates than control. Furthermore, immunohistochemistry showed an increased Bax/Bcl-2 ratio after Sal treatment (Fig. 7C). Finally, we examined in vivo expression of p-GSK-3b, total GSK-3b, and b-catenin using Western blotting. Expression of p-GSK-3b decreased but total GSK-3b did not change, indicating increased GSK3b activity. The protein expression of b-catenin decreased in the Sal treatment groups compared with control (Fig. 7D). Finally, we evaluated expression of b-catenin using immunohistochemistry (Fig. 7D) and verified Sal could down-regulate the b-catenin expression in HCC.

Discussion
It has been reported that Sal can decrease the malignant traits of several types of cancer cells [10,7]. In this study, we found that Sal inhibited proliferation of HCC cell lines and reduced the proportion of HCC CD133+ cell subpopulations. Recent studies support the theory that tumors can be initiated and maintained by a small population of cells that have stem-like features, and these highly tumorigenic cells within the tumor bulk are considered cancer stem cells [11]. Studies have identified CD133 as an adequate marker of HCC cancer stem cells [12][13][14]. CD133expressing cancer cells with a stem-cell-like phenotype are believed to account for chemotherapy resistance, tumor recurrence, and disease progression. Compared with paclitaxel [5] and oxaliplatin [15] that kill general cancer cells, Sal selectively kills cancer stem cells, providing a new strategy for cancer therapy.
In this study, we show that Sal repressed HCC cell proliferation. Cell-cycle dysregulation and apoptosis resistance are hallmarks of tumor cells. Sal caused different cell phase arrests in different HCC cells; however, our data showed the induction of apoptosis in Sal-treated HCC cells was similar. Apoptosis resistance is one of the main causes of tumorigenesis and tumor drug resistance [16]. The Bcl-2 family plays a critical role in apoptosis regulation. Proapoptotic Bax promotes intrinsic apoptosis by forming oligomers in the mitochondrial outer membrane, facilitating the release of the apoptogenic molecules, whereas anti-apoptotic Bcl-2 blocks mitochondrial apoptosis by blocking the release and oligomerization of Bax [17]. Sal can induce apoptosis and overcome apoptosis resistance of hepatoma cells in vitro and in vivo, which might make it an effective chemotherapy agent for treating HCC.
b-catenin, encoded by the CTNNB1 gene, has multiple functions, including mediation of cell adhesion and signal transduction. It combines with a variety of proteins to regulate cell proliferation and differentiation, which is critical for embryonic development and tumorigenesis. In agreement with another report [18], this study demonstrates that Sal inhibited proliferation and promoted hepatoma apoptosis in vitro and in vivo through down-regulating b-catenin expression. The b-catenin gene mutation rate is approximately 20% in HCC [19,20]. However, bcatenin accumulated within 67% of HCC tissues and is closely related to the clinicopathological features of HCC [21][22][23]. It is likely that some upstream molecules play an important role in b-catenin activation in HCC. In canonical Wnt signaling pathways, Gsk-3b is the upstream adjustment factor of b-catenin and can compose a complex with APC and axin to phosphorylation bcatenin, leading to b-catenin degradation. Gsk-3b is inactivated by phosphorylation as well because in this study, we showed that Sal suppressed GSK-3b phosphorylation, suggesting active GSK-3b was up-regulated, followed by b-catenin degradation.
Sal functions on different biological membranes, including cytoplasmic and mitochondrial membranes, as an ionophore with strict selectivity for alkali ions and a strong preference for potassium and it can disrupt Na + /Ca 2+ exchange and cause an increase in intracellular calcium levels [24]. Previous studies have demonstrated that non-canonical Wnt family members repress canonical Wnt signaling by inducing calcium influx and the increase in intracellular calcium inhibits the canonical Wnt pathway [25][26][27][28][29]. Here, we show that Sal can increase intracellular calcium levels and inhibit the Wnt pathway. Moreover, increased intracellular Ca 2+ is an initiating factor of apoptosis because nucleosome DNA degradation is calcium-dependent, providing a reasonable explanation of Sal-induced apoptosis [30].
In this study, we demonstrate the anti-tumor antibiotic Sal can inhibit proliferation, induce apoptosis, and repress canonical Wnt/ b-catenin signaling involved in HCC growth via increasing intracellular calcium levels. Although the exact mechanism remains unclear, this research provides a starting point for exploring Sal-related HCC cancer therapy.