HCT-116 human CRC cells were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China), and maintained in McCoy's 5A (Sigma-Aldrich, St. Louis, MO, USA). Media contained 100 IU penicillin, 100 µg/ml streptomycin (Invitrogen, Carlsbad, CA, USA) and 10% fetal bovine serum (FBS; Gibco-BRL, Grand Island, NY, USA). The cells were incubated at 37°C in 5% CO₂. CD44⁺EpCAM⁺ HCT-116 cells (CRSCs) were maintained in Dulbecco's modified eagle's medium (DMEM)/F12 (Sigma-Aldrich) containing 2% B27 (Invitrogen), 20 ng/ml epidermal growth factor, and 20 ng/ml basic fibroblast growth factor (both from PeproTech, Rocky Hill, NJ, USA) in an incubator at 37°C in 5% CO₂.

Salinomycin (Sigma-Aldrich) was stored as a 500 mM dimethylsulfoxide (DMSO; Gibco, CA, USA) solution in the dark at −20°C. For experiments, salinomycin solutions were prepared by diluting the stock solution with DMEM/F12. The antibodies used for western blotting were as follows: rabbit anti-caspase-3 and anti-caspase-6, mouse anti-Bcl-2, anti-Bax, anti-caspase-8 anti-caspase-9 and anti-β-actin. All antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). CD44 and epithelial cell adhesion molecule (EpCAM) antibodies (eBioscience, San Diego, CA, USA) were used for magnetic-activated cell sorting (MACS).

MACS was performed using a CELLection™ Biotin Binder kit according to the manufacturer's instructions (Invitrogen). In brief, HCT-116 cells were collected and incubated with the CD44 antibody for 10 min. Dynabeads (750 µl) were added to the cells, which were then incubated for 20 min and separated using a magnet. Subsequently, releasing buffer (DNase I; 120 µl) was added, and the cells were incubated for 15 min at room temperature with gentle tilting and rotation to collect target cells (CD44⁺ cells). CD44⁺ cells were incubated with the EpCAM antibody for 10 min, and Dynabeads (25 µl) were added to the cells. The cells were incubated for 20 min and then separated using a magnet. Subsequently, releasing buffer (DNase I; 4 µl) was added and the cells were incubated for 15 min at room temperature with gentle tilting and rotation to collect target cells (EPCAM⁺CD44⁺ cells, CRSCs).

To induce differentiation, CRSCs were collected and maintained in DMEM/F12 medium supplemented with 10% FBS and incubated for 3 days. Results were analyzed using an inverted microscope.

One hundred living cells mixed with 0.3% agar liquor were immediately plated onto a 0.5% solidified agar-based 6-well plate. Cells were incubated for 3 weeks on soft agar culture medium. The resulting colonies were photographed with an optical microscope.

HCT-116 cells and CRSCs were plated at a density of 5,000 cells/well in flat-bottom 96-well plates (100 µl medium/well). After 24 h, the cells were treated with salinomycin at various concentrations (1, 10 and 25 µM), 100 µM cisplatin (DDP) as a positive parallel control or 0.1% DMSO as a solvent control. After 48 h, the Cell Counting Kit-8 (CCK-8; Dojindo, Tokyo, Japan) was used according to the manufacturer's instructions, optical density was measured using a microplate reader at 450 nm, and the cell viability was calculated. All experimental concentrations were assessed in triplicate. Inhibition ratio was calculated using the formula: Inhibition ratio (%) = 1 − OD_{treatment group}/OD_{solvent control} × 100.

An invasion assay was performed using 6.5-mm Transwell^® plates with sterile 8.0-µm pore polycarbonate membrane inserts (Corning, Steuben County, NY, USA). In brief, 5,000 CRSCs in DMEM/F12 medium were seeded in the insert and treated with salinomycin (0–25 µM). The lower chamber was filled with DMEM/F12 medium supplemented with 10% FBS as a chemotactic factor, and the plates were subsequently incubated at 37°C in 5% CO₂ for 48 h. The insert chamber contained an 8-µm pore polycarbonate membrane covered with a thin layer of BD Matrigel™ (BD, San Diego, CA, USA). The Matrigel layer blocks pores in the membrane, which prevents the migration of non-invasive cells; however, invading cells can migrate through the Matrigel layer and eventually attach themselves to the bottom of the polycarbonate layer. After 48 h, the number of cells in the lower chamber was quantified using the CCK-8 assay as described above.

Apoptotic response was evaluated by detecting DNA fragmentation using an apoptosis DNA ladder detection kit (Keygen, Nanjing, China). CRSCs were treated with salinomycin (1, 10 or 25 µM), 100 µM DDP as a positive parallel control, or 0.1% DMSO as solvent control. After 48 h, CRSCs were collected and washed with PBS, resuspended in 20 µl lysis buffer, and lysed for 10 min on ice. Subsequently, enzyme A (10 µl) was added and incubated at 37°C for 1 h. Then samples were incubated with enzyme B (10 µl) at 50°C for 90 min. After 2% agarose gel electrophoresis, DNA was stained with ethidium bromide (EB) and photographed.

AO is used to stain normal cells and EB indicates apoptotic cells. In brief, CRSCs were plated into 6-well plates and treated with salinomycin (1, 10 and 25 µM), 100 µM DDP as a positive parallel control and 0.1% DMSO as a solvent control. After 48 h, each well was treated with AO (5 µl) and EB (5 µl), and subsequently incubated for 5 min at room temperature. The stained cells were analyzed using a fluorescence microscope (Olympus, Tokyo, Japan). The experiments were repeated 3 times.

Annexin V analysis was performed using an Annexin V-fluorescein isothiocyanate (FITC) kit (BD Biosciences, Franklin Lakes, NJ, USA) according to the manufacturer's instructions. Briefly, after CRSCs were incubated with 1, 10 or 25 µM salinomycin, 100 µM DDP as a positive parallel control or 0.1% DMSO as a solvent control, they were harvested by quick trypsinization to minimize potentially high Annexin V background levels in adherent cells. Cells were then washed twice with cold phosphate-buffered saline (PBS) and re-suspended in binding buffer at a concentration of 1×10⁶ cells/ml. Cells (100 µl) were stained with Annexin V-FITC (5 µl) and propidium iodide (PI; 5 µl) and incubated in the dark at room temperature for 15 min. Then, binding buffer (400 µl) was added, and the cells were analyzed using a flow cytometer (Beckman Coulter, Salt Lake, UT, USA) and a fluorescence microscope (Olympus). Cells negative for both Annexin V and PI were viable, Annexin V⁺/PI⁻ cells were in early apoptosis and Annexin V⁺/PI⁺ cells were necrotic or in late apoptosis. The experiments were repeated 3 times.

JC-1 and FITC staining were carried out using a commercial mitochondrial membrane potential detection (JC-1) kit (BD Biosciences). Briefly, CRSCs were incubated with salinomycin (1, 10 or 25 µM), 100 µM DDP as a positive parallel control or 0.1% DMSO as a solvent control for 48 h. CRSCs were harvested, then washed twice with cold PBS and re-suspended in binding buffer at a concentration of 1×10⁶ cells/ml. Cells (100 µl) were stained with JC-1 (5 µl) and pI (5 µl) and incubated in the dark at room temperature for 15 min. Then, binding buffer (400 µl) was added and cells were analyzed using a flow cytometer (Beckman Coulter). Cells negative for both JC-1 and PI were viable, JC-1⁺/PI⁺ cells were in early apoptosis and JC-1⁺/PI⁻ cells were necrotic or in late apoptosis. The experiments were repeated 3 times.

Total RNA was extracted using TRIzol reagent (Invitrogen). First-strand cDNA was reverse transcribed using PrimeScript RT kit (Takara, Otsu, Shiga, Japan) according to the manufacturer's protocol. Relative mRNA levels were quantitatively determined using a real-time PCR system. The primer sequences used for quantitative real-time PCR are shown in Table I. glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the endogenous reference. cDNA was subjected to PCR for 40 cycles of 94°C for 30 sec, 60°C for 30 sec and 72°C for 45 sec. qPCR was performed using a Thermo^® PikoReal 96 system (Thermo, Waltham, MA, USA). Real-time RT-qPCR was performed using FastStart universal SYBR-green Master (ROX) (Roche, Basle, Switzerland) and analyzed using PikoReal software 2.1 (Thermo). Experiments were performed in triplicate and the average CT values of target genes were normalized to control as ΔCT. Changes in expression levels are shown either as a fold increase or as a ratio (target gene/control gene).

Western blotting was carried out to test for caspase-3, -6, -8 and -9, Bcl-2 and Bax. To collect whole protein, cells were lysed with RIPA buffer containing protease inhibitor cocktail (Roche), and protein concentrations were determined using a BCA assay kit (Beyotime, Nanjing, Jiangsu, China). Protein bands were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and were then transferred to membranes (Millipore, Bedford, MA, USA) at 100 V for 45 min at room temperature. After blocking in 4% non-fat dry milk in tris-buffered saline (TBS), the membranes were incubated with primary antibodies at a 1:1,000 dilution in TBS overnight at 4°C, washed 3 times with TBS containing 0.5% Tween-20, and then incubated with secondary antibodies conjugated with horseradish peroxidase (HRP) at a 1:5,000 dilution in TBS for 1 h at room temperature. Membranes were washed again in TBS containing 0.5% Tween-20 for 3 times at room temperature. Protein bands were visualized on X-ray film using enhanced chemiluminescence (ECL; GE Healthcare, Bethesda, MD, USA).

Severe combined immunodeficiency (SCID) mice (CB17/Icr-Prkdc^scid/IcrlcoCrlVr) were bred in specific-pathogen-free microisolator cages that were purchased from the Animal Institute of the Chinese Academy of Medical Science. All experiments were performed according to the regulations of the Animal Care Committee of Jilin University. To generate CRSC xenografts, 1×10⁵ CRSCs were resuspended in PBS (100 µl) and injected subcutaneously into the right flank of the mice. The weight and size of the tumors were measured every other day. When tumors reached a volume of 40–60 mm³, the mice were randomized into therapy and control groups. After that, the mice were treated with vehicle (DMSO), 5-fluorouracil (5-Fu), salinomycin or a combination of salinomycin and 5-Fu. 5-Fu was administered once a week (100 mg/kg) 3 times, while salinomycin was injected every other day (4 mg/kg) 5 times. Both agents were injected intraperitoneally. The animal weight and tumor volumes were monitored every other day. Using Vernier calipers, tumor volume was calculated according to the formula A × B × B/2, where A is the length of the tumor and B is the width.

All analyses were performed using Origin8 (Origin8 Technologies Ltd., London, UK) and data are presented as means ± standard deviation (SD). Values of p<0.05 were considered to be statistically significant and were evaluated using the Student's t-test.

The proportion of CRSCs in HCT-116 cells was found to be 3%, and the CRSCs formed spheres after culturing for 10 days in CRSC medium (Fig. 1A). The generation of spheres was observed over the next 3 weeks in soft agar using an inverted microscope (Fig. 1B). CRSC spheroid cells became adherent cells when serum was added to the culture medium (Fig. 1C).

To initially assess the anticancer effect of salinomycin on CRSCs, a CCK-8 assay was performed on the treated cells. Salinomycin reduced the cell viability of HCT-116 cells and CRSCs in a concentration-dependent manner (Fig. 2A). Additionally, the Transwell migration assay indicated that treatment with 25 µM salinomycin significantly reduced the number of invasive cells (Fig. 2B).

AO/EB staining indicated that the number of apoptotic CRSCs increased in response to salinomycin treatment in a concentration-dependent manner (Fig. 3A). The Annexin V/PI double staining assay revealed that salinomycin treatment increased the percentage of Annexin V-positive cells in the CRSCs (Fig. 3B and C). A DNA ladder assay indicated that severe nuclear fragmentation took place in the CRSCs following treatment with 10 or 25 µM salinomycin, but not in the cells treated with DMSO or 1 µM salinomycin (Fig. 3D).

To further clarify the mechanism of salinomycin-induced apoptosis, the expression of a number of apoptosis-related proteins and their mRnAs were analyzed. Upregulation of caspase-3, -8 and -9 mRNA was observed in CRSCs treated with salinomycin for 48 h (Fig. 4A). Additionally, the protein expression of cleaved caspase-3, -8 and -9 was enhanced following treatment with salinomycin for 48 h (Fig. 4B).

An increased level of cleaved caspase-9 implied the breakdown of mitochondria in the salinomycin-induced apoptosis. The JC-1 staining assay showed that the percentage of cells experiencing a loss of mitochondrial membrane potential increased from 17.29 to 78.81% in the CRSCs following treatment with 25 µM salinomycin. The loss of mitochondrial membrane potential, along with increased cleaved caspase-9 (Fig. 4C and D), suggests that salinomycin-induced CRSCs death via the mitochondrial apoptosis pathway.

We measured the expression of Bcl-2 and Bax using western blotting and real-time RT-qPCR. Salinomycin treatment for 48 h significantly suppressed the expression of Bcl-2 and upregulated the expression of Bax in the CRSCs (Fig. 5A and B).

To evaluate the in vivo anticancer activity of salinomycin, CRSCs were subcutaneously injected into SCID mice in the right flank. Delayed tumor growth was observed in the salinomycin-treated group, the 5-Fu-treated group and the combined treatment group as compared to the control group. Tumor volumes in the salinomycin-treated group and the combined treatment group decreased significantly compared to the 5-Fu-treated group (Fig. 6A). After 3 weeks of treatment, the animals were sacrificed and CRSC xenografts were dissected and weighed (Fig. 6B). The weight of tumors in the salinomycin-treated group and the combined treatment group were both significantly smaller than those in the 5-Fu-treated and control groups (P<0.05).