INTRODUCTION
Gastric cancer (GC), a malignant tumor that develops in the gastrointestinal tract, is the fifth most commonly diagnosed cancer and the fourth leading cause of cancer-related deaths worldwide [1]. GC is caused by environmental factors, such as Helicobacter pylori infection and poor dietary habits, and genetic factors [2]. Surgical resection and chemotherapy remain the standard treatment modality for GC; however, the prognosis of advanced or metastatic GC remains poor.
The persistent presence of cancer stem cells (CSCs) is associated with the poor prognosis of patients [3]. CSCs are a subpopulation of tumors that are characterized by self-renewal, multilineage differentiation, and tumor initiation abilities [2]. Several stem cell regulators, including CD44, CD133, integrin α6, aldehyde dehydrogenase 1A1 (ALDH1A1), Oct4, Sox2, and Nanog, have been identified as major biomarkers for gastric cancer stem cells (GCSCs) [4,5]. GC cells positive for cell surface proteins CD133, CD44, and integrin α6 have higher tumorigenic and metastatic potential than the negative subpopulation [3]. ALDH1A1 overexpression induces cancer stem-like features and chemoresistance in GC cells [6]. The transcription factors Oct4, Sox2, and Nanog play critical roles in maintaining the pluripotency and self-renewal characteristics of GCSCs [7]. Therefore, the inhibition of these cancer stemness regulators is necessary to effectively eliminate GCSCs.
Drug repurposing is the application of drugs to diseases other than those approved for original indications [8]. Preclinical and clinical investigations on pharmacokinetics and pharmacodynamics of repurposed drugs have previously been completed; therefore, there is growing interest to discover new repurposed drugs with anticancer activity. Statins are potent competitive inhibitors of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase (HMGCR), which is a rate-limiting enzyme in the mevalonate pathway, approved by the Food and Drug Administration (FDA) as a lipid-lowering drug [9]. Recent studies have reported the potential of statins as repurposed drugs with anticancer activities [10,11]. Statin drugs, including simvastatin, lovastatin, and atorvastatin, inhibit proliferation and induce apoptosis of several different cancer cells by inhibiting the prenylation of various proteins, such as members of the Ras and Rho GTPase families [10-12]. Recently, statins have been reported to inhibit the proliferation and self-renewal ability of CSCs [13,14]. Lovastatin inhibited the stemness properties and induces apoptosis of nasopharyngeal carcinoma CSCs [13]. Simvastatin suppressed invasion and migration, and induced spheroid disassembly and cell death in ovarian CSCs by decreasing stemness and epithelial-mesenchymal transition (EMT) marker expression and inactivating the Hippo/YAP/RhoA pathway in a mevalonate synthesis-dependent manner [14]. Atorvastatin, in combination with celecoxib and tipifarnib, inhibited the proliferation of pancreatic CSCs by reducing the expression of stemness markers, including CD44, CD133, and ALDH1A1, through the inactivation of protein kinase B (AKT) and nuclear factor-κB (NF-κB) [15]. However, the anticancer effects of statins on GCSCs remain unclear. In this study, we investigated the anticancer effect and underlying molecular mechanism of atorvastatin, a lipophilic statin, on MKN45-derived GCSCs.
METHODS
Materials
Atorvastatin, DL-Mevalonolactone, and sorafenib were purchased from Tocris Bioscience (Bristol, UK), Tokyo Chemical Industry (Tokyo, Japan), and Sigma-Aldrich (Saint Louis, MO, USA), respectively. RPMI-1640 was obtained from Corning Cellgro (Manassas, VA, USA). DMEM/F12 and trypsin were obtained from HyClone (Marlborough, MA, USA). Epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) were obtained from Prospecbio (East Brunswick, NJ, USA). Fetal bovine serum (FBS), Matrigel, penicillin-streptomycin-amphotericin B, heparin, and Accutase were purchased from R&D systems (Minneapolis, MN, USA), BD Biosciences (San Jose, CA, USA), Lonza (Basel, Switzerland), Sigma-Aldrich, and EMD Millipore (Temecula, CA, USA), respectively. B-27 serum-free supplement, L-glutamine, and penicillin/streptomycin were obtained from Gibco (Grand Island, NY, USA). The CellTiter-Glo 2.0 Cell Viability Assay kit was purchased from Promega (Madison, WI, USA). The Muse Annexin V & Dead Cell and Muse Cell Cycle kits were purchased from Luminex Corporation (Austin, TX, USA). Anti-CDK4 (#sc-70831) and anti-HMGCR (#sc-271595) antibodies were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Antibodies against cyclin D1 (#2922), cleaved caspase-9 (#9501), cleaved caspase-3 (#9661), PARP (#9542), ALDH1A1 (#12035), integrin α6 (#3750), CD44 (#37259), CD133 (#64326), Nanog (#3580), Oct4 (#2750), Sox2 (#3579), phospho-GSK3β (#9322), GSK3β (#9315), β-catenin (#9562), phospho-STAT3 (#9145), STAT3 (#9139), phospho-ERK1/2 (#9101), ERK1/2 (#9102), phospho-AKT (#4060), AKT (#9272), phospho-NF-κB p65(#3033), NF-κB p65 (#8242), rabbit IgG (#7074), and mouse IgG (#7076) were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-β-actin (#ab6276) antibody was purchased from Abcam (Cambridge, UK).
Cell culture
The human gastric adenocarcinoma cell line, MKN45 was obtained from the Korean Cell Line Bank (Seoul, Korea). The cells were grown in RPMI-1640 supplemented with 10% FBS and 1% penicillin–streptomycin–amphotericin B. To propagate GCSCs, MKN45 cells were cultured in DMEM/F12 containing 1× B-27, 5 µg/ml heparin, 2 mM L-glutamine, 20 ng/ml EGF, 20 ng/ml bFGF, and 1% penicillin/streptomycin as described previously [16,17]. Tumorspheres grown in serum-free media were subcultured by dissociation using Accutase. The adherent and tumorsphere cells were maintained at 37°C in a humidified CO2 incubator with 5% CO2 (Thermo Scientific, Vantaa, Finland).
Cell proliferation assay
MKN45-derived GCSCs (5 × 103 cells/well) were seeded in a 96-white-well culture plate using serum-free media and treated with the different concentrations of each compound for 7 days. Cell proliferation was measured using the CellTiter-Glo 2.0 Cell Viability Assay as described previously [16,17]. Luminescence was detected using a multimode microplate reader (BioTek, Inc., Winooski, VT, USA). The IC50 value from the obtained data was analyzed using the GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA).
Tumorsphere forming assay
MKN45-derived GCSCs (3 × 103 cells/well) were seeded in a 96-well culture plate using serum-free media and treated with the different concentrations of each compound for 7 days. The number of tumorspheres > 70 µm in diameter was counted under an optical microscope (Olympus, Tokyo, Japan).
Cell cycle analysis
MKN45-derived GCSCs (1 × 105 cells/well) were seeded in a 60-mm culture plate using serum-free media and treated with the different concentrations of each compound for 4 days. The cells were harvested, fixed with 70% ethanol, and stained with 200 µl of Muse Cell Cycle reagent as described previously [16,17]. Cell cycle plots were obtained using the Guava Muse Cell Analyzer (MuseSoft_V1.8.0.3; Luminex Corporation).
Apoptosis analysis
MKN45-derived GCSCs (1 × 105 cells/well) were seeded in a 60-mm culture plate using serum-free media and treated with the different concentrations of atorvastatin for 4 days. The cells were collected and stained with 100 µl of Muse Annexin V & Dead Cell reagent as described previously [16,17]. Cellular apoptosis was quantitatively analyzed using the Guava Muse Cell Analyzer.
Western blot
The cell lysates were separated using 7.5%–15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The separated proteins were transferred to polyvinylidene difluoride membranes (EMD Millipore, Hayward, CA, USA), and the blots were blocked and immunolabeled with the primary antibodies (dilution 1:500–1:10,000) as described previously [16,17]. After washing with TBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (dilution 1:3,000). Immunolabeling was detected using an enhanced chemiluminescence kit (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer’s instructions. The band density was analyzed using ImageJ 1.5 software (NIH, Bethesda, MD, USA). Expression levels were determined as the normalized ratio of each target protein to β-actin or phospho-protein to total protein.
Chick embryo chorioallantoic membrane (CAM) assay
To evaluate the effects of the compounds on GC tumorigenesis in vivo, a modified CAM assay was performed as described previously [16]. Briefly, fertilized chicken eggs were incubated at 37°C in a humidified egg incubator for 6 days, and the eggshell membrane was carefully peeled away. MKN45-derived GCSCs (1 × 106 cells/egg) were mixed with Matrigel (30 µl/egg) in the absence or presence of the compounds and implanted onto the CAM inside the silicone ring. After incubation for 7 days, the tumor formed on the CAM was retrieved, and the tumor weight was measured.
Statistical analysis
The results are expressed as the mean ± standard deviation from at least three independent experiments. Statistical analysis was performed by analysis of variance with Tukey’s post-hoc test using SPSS 9.0 software (SPSS Inc., Chicago, IL, USA). Statistical significance was set at p < 0.05.
RESULTS
Atorvastatin inhibits proliferation of MKN45 GCSCs in a mevalonate pathway-independent manner
To propagate the CSC population from the GC cell line, MKN45 cells were grown in a serum-free spheroid suspension culture. The stem-like properties of GC cells are accompanied by upregulation of stemness-related factors. The expression levels of several key GCSC markers, including CD133, Oct4, and Sox2, were significantly increased in MKN45 tumorsphere cells cultured in serum-free media containing EGF and bFGF compared to MKN45 adherent cells cultured in 10% FBS-supplemented media (Fig. 1A). These data indicate that serum-free spheroid culture can expand the MKN45-derived GCSC population. In this study, we used MKN45 GCSCs at passage 3 with the highest expression levels of stem cell markers that promote the cancer stem-like features.
To evaluate whether atorvastatin affects the proliferation of GCSCs, the MKN45-derived GCSCs were treated with atorvastatin at various concentrations for 7 days, and cell proliferation was measured using an ATP-monitoring luminescence assay. Atorvastatin dose-dependently inhibited the proliferation of MKN45 GCSCs with 2.1 µM of IC50 value (Fig. 1B). Moreover, atorvastatin reduced the number and size of tumorspheres formed by MKN45 GCSCs (Fig. 1C, D). These results demonstrate that atorvastatin inhibits the proliferation of GCSCs. Based on the IC50 value, we further assessed the effects of atorvastatin on MKN45 GCSCs at concentrations ranging from 0–10 µM.
Next, we investigated whether the antiproliferative effect of atorvastatin on MKN45 GCSCs is related to HMGCR activity inhibition. No significant difference in HMGCR protein expression level was observed between MKN45 adherent and tumorsphere cells (Fig. 1E). Interestingly, treatment with mevalonate did not affect the suppressive effect of atorvastatin on the proliferation and tumorsphere formation of MKN45 GCSCs (Fig. 1F–H). These data suggest that atorvastatin exerts an anticancer effect against MKN45 GCSCs in a manner independent of the mevalonate-mediated pathway.
Atorvastatin induces cell cycle arrest and apoptosis in MKN45 GCSCs
To determine whether atorvastatin exerted an antiproliferative effect on MKN45 GCSCs by regulating the cell cycle and apoptosis, we first investigated the effect of atorvastatin on the cell cycle distribution of MKN45 GCSCs by flow cytometry. Atorvastatin caused a significant increase in the cell population in the G0/G1 phase, along with a decrease in the cell population in the S and G2/M phases, compared to the untreated control cells (Fig. 2A). Next, apoptosis of MKN45 GCSCs was analyzed by flow cytometry using Annexin V-FITC and PI dual staining. Atorvastatin increased the proportion of apoptotic cells in a dose-dependent manner compared with that in the untreated control group (Fig. 2B).
Furthermore, we elucidated the molecular mechanism of cell cycle arrest and the apoptosis-inducing effect of atorvastatin in MKN45 GCSCs. Atorvastatin significantly decreased the protein expression of cyclin-dependent kinase 4 (CDK4) and cyclin D1, which are important regulators of the transition from the G0/G1 phase to the S phase of the cell cycle, in MKN45 GCSCs (Fig. 2C). Moreover, atorvastatin markedly increased the protein levels of cleaved caspase-9, cleaved caspase-3, and cleaved poly (ADP-ribose) polymerase (PARP) in MKN45 GCSCs. Collectively, these results demonstrated that the antiproliferative effect of atorvastatin on MKN45 GCSCs resulted from the induction of G0/G1 cell cycle arrest and caspase-dependent apoptosis.
Atorvastatin downregulates the expression of cancer stemness markers and activation of signaling pathways in MKN45 GCSCs
We investigated whether atorvastatin affected the expression of key stemness-related markers in GCSCs. The results showed that atorvastatin effectively suppressed the expression of cancer stemness regulators, including CD44, CD133, integrin α6, ALDH1A1, Oct4, Sox2, and Nanog, in MKN45 GCSCs (Fig. 3A).
Several signaling pathways, such as Wnt/β-catenin, JAK/STAT, AKT, ERK1/2, and NF-κB, contribute to the self-renewal, tumor progression, and chemoresistance of GCSCs by upregulating the expression of cancer stemness markers [4,18]. Thus, we assessed the effect of atorvastatin on the upstream signaling pathways in MKN45 GCSCs. Atorvastatin increased the protein expression level of glycogen synthase kinase 3β (GSK3β) by inhibiting its phosphorylation, thereby resulting in a significant decrease in β-catenin protein expression (Fig. 3B). In addition, atorvastatin significantly inhibited the expression of phosphorylated STAT3 and AKT compared to their total protein expression levels. However, atorvastatin did not cause an obvious decrease in the expression levels of phosphorylated ERK1/2 and NF-κB. These results imply that atorvastatin inhibits the expression of cancer stemness-related markers by downregulating the Wnt/β-catenin, STAT3, and AKT signaling pathways in MKN45 GCSCs.
Atorvastatin increases chemosensitivity of MKN45 GCSCs to sorafenib
Sorafenib is a tyrosine kinase inhibitor that shows anticancer effects against various tumors, including GC; however, its clinical application is often limited by drug resistance [19]. To further explore the therapeutic potential of atorvastatin in GCSCs, we evaluated the effect of atorvastatin on the chemosensitivity of MKN45 GCSCs to sorafenib. The combination of atorvastatin and sorafenib synergistically inhibited both the proliferative and tumorsphere formation abilities of MKN45 GCSCs (Fig. 4A–C). Moreover, the combined treatment of the two drugs more strongly induced cell cycle arrest at the G0/G1 phase compared to single-drug treatment (Fig. 4D).
Next, we assessed the combination effects of atorvastatin and sorafenib on GC tumorigenesis using a CAM tumor model injected with MKN45 GCSCs. Co-treatment with atorvastatin and sorafenib more potently reduced the tumor weight in comparison with single-drug treatment, indicating that the combined treatment of the two drugs synergistically suppressed the growth of MKN45 GCSCs both in vitro and in vivo (Fig. 4E). These findings suggest that atorvastatin can be used in combination with sorafenib to treat GC by targeting GCSCs.
DISCUSSION
GC is a common malignant tumor worldwide, with a poor prognosis due to recurrence, metastasis, and drug resistance [3]. GCSCs significantly contribute to the refractory features of GC and are thus regarded as an important target of anticancer drugs to improve the outcomes of patients with GC [2]. Accumulating evidence has shown that statins exhibit potential anticancer activity in GC, alone or in combination therapy [20,21]. In GC cell lines, statins inhibited proliferation and activated apoptosis [20]. They synergistically reduced tumor growth in conjunction with radiation or chemotherapy in GC animal models [21]. In addition, statins improved overall survival of patients with GC after surgery and adjuvant chemotherapy [22]. However, the anticancer effects of statins on GCSCs remain unclear. In the present study, we demonstrated, for the first time, the antiproliferative effect and underlying molecular mechanism of atorvastatin against MKN45 GCSCs. Atorvastatin inhibited the proliferation of MKN45 GCSCs by inducing cell cycle arrest at the G0/G1 phase and promoting caspase-dependent apoptosis.
Although statins are approved only for the treatment of hypercholesterolemia, increasing evidence suggests that they are useful against other diseases, including cancer [23-25]. Atorvastatin is administered as a clinical treatment for hypercholesterolemia at 10–80 mg/day [26]. Several nationwide cohort studies have revealed that the administration of statins reduced the risk of GC and its overall mortality [27,28]. In several human cancer types, the antitumor mechanisms of statins are associated with the inhibition of HMGCR, which converts HMG-CoA to mevalonate [23,24]. The blockade of mevalonate synthesis by statins consequently interrupts post-translational modifications, such as farnesylation and geranylgeranylation, of several proteins that are essential for cancer cell proliferation and survival, including Ras and Rho GTPases [10-12]. Moreover, activation of the mevalonate pathway promotes EMT and the stemness properties of cancer cells [14]. However, recent studies have revealed that statins exert HMGCR-independent therapeutic effects [25]. Simvastatin and atorvastatin exhibit mevalonate metabolism-independent antibacterial and anti-inflammatory activities, respectively [25]. Interestingly, our results showed that the inhibitory effect of atorvastatin on the proliferation of MKN45 GCSCs might not result from the suppression of mevalonate-mediated pathways. Therefore, further studies are needed to elucidate the mevalonate pathway-independent mechanism responsible for the anti-proliferative effect of atorvastatin in GCSCs.
Several signaling pathways, such as Wnt/β-catenin, STAT3, AKT, ERK1/2, and NF-κB, play important roles in maintaining the stem-like features of GC cells by increasing the expression of major cancer stemness markers, including CD44, CD133, integrin α6, ALDH1A1, Sox2, Oct4, and Nanog [4,5,18]. Our data demonstrated that atorvastatin inhibited the expression of cancer stemness regulators by downregulating the Wnt/β-catenin, STAT3, and AKT signaling pathways in MKN45 GCSCs. Recent studies have reported the crosstalk between these signaling molecules in various cancers, including GC. The PI3K/AKT pathway activates the Wnt/GSK3β/β-catenin signaling pathway, which in turn upregulates the transcriptional activity of STAT3 [29,30]. Therefore, atorvastatin may exhibit anticancer effects against MKN45 GCSCs by impeding the crosstalk between the Wnt/β-catenin, STAT3, and AKT signaling pathways.
In recent studies, fluvastatin and lovastatin have shown synergistic anticancer effects in combination with a FDA-approved multi-kinase inhibitor, sorafenib, in hepatocelluar and advanced renal cell carcinomas, respectively [19,31]. Atorvastatin also shows enhanced anticancer effects in combination with other clinical drugs, such as docetaxel, aspirin, celecoxib, and tipifarnib, in several cancer cells [15,32,33]. However, combination therapy with atorvastatin and sorafenib in GCSCs has not yet been evaluated. Our results demonstrated that atorvastatin significantly increased the chemosensitivity of MKN45 GCSCs to sorafenib both in vitro and in vivo. Therefore, these findings suggest that atorvastatin is an attractive anticancer drug to eradicate GCSCs, either alone or in combination with sorafenib.