The human colon cancer cell line HCT116 and the human lung fibroblast cell line HELF were obtained from the Cell Bank of the Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). HCT116 and HELF cells were maintained in RPMI-1640 medium (Gibco/Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Gibco) and penicillin-streptomycin mixed solution. The CSCs isolated from HCT116 cells were cultured in DMEM/F12 medium (Gibco) with 10 µg/ml EGF (Sigma-Aldrich, St. Louis, MO, USA), 10 µg/ml bFGF (Sigma-Aldrich) and 1% B27 supplement (Gibco). The media were changed every other day. The cells were incubated in a humidified incubator with 5% CO₂ at 37°C, and were passaged by a dilution of 1:3 every 4 or 5 days.

The cell selection method followed the protocol published by Zhang et al (21). Briefly, HCT116 cells were rinsed with phosphate-buffered saline (PBS) and detached with trypsin at 37°C. After centrifugation, cells were first incubated with anti-human CD44 monoclonal antibody conjugated with biotin (eBioscience, Inc., San Diego, CA, USA), and then fractionated using a CELLection Biotin Binder kit (Invitrogen, Carlsbad, CA, USA), according to the manufacturers recommendations. The isolated cells were further incubated with anti-human EpCAM monoclonal antibody conjugated with biotin (eBioscience) and positive cells were then isolated with the same kit. Cells were cultured in CSC medium, and the purity of such CSCs was evaluated by flow cytometry and immunohistochemistry.

The CSCs were dissociated into single cells, further fixed with fixation buffer (eBioscience) and prepared at a concentration of 2.0×10⁶/ml of PBS. Anti-human CD44 antibody conjugated with FITC (BD Biosciences, San Jose, CA, USA) and anti-human EpCAM antibody conjugated with PE (BD Biosciences) were added and incubated for 30 min at 4°C. After washing twice with PBS, the cells were acquired and analyzed by FACScalibur (BD Biosciences).

The single CSCs were seeded into plates covered in Matrigel and cultured for 24 h. They were fixed in 4% formaldehyde for 20 min at 4°C and permeabilized with 0.1% Triton X-100 for 10 min at 25°C. For immunohistochemistry, the primary antibodies used were anti-human CD44 monoclonal antibodies (1:50; eBioscience) and anti-human EpCAM monoclonal antibodies (1:50; eBioscience). After 12–14 h of incubation at 4°C, samples were washed three times with PBS and processed using an ABC kit and DAB solution (both purchased from Vector Laboratories, Inc., Burlingame, CA, USA). Finally, the sections were imaged with an Axio Scope A1 and AxioCAM MRc 5 (Carl-Zeiss, Oberkochen, Germany).

The CSCs were dissociated into single cells and seeded into a 96-well plate at a concentration of 0.5 cells/well. Briefly, the cell suspension was diluted at concentration of 50 cells/ml and 100 µl of cell suspension was added into each well. As a result, there is 1 cell or 0 cell seeded into each well. We observed and imaged cell cloning. The plate was incubated in a humidified incubator with 5% CO₂ at 37°C and half of the medium was changed every 3 days. Colonies gradually formed over the next 3 weeks.

In order to analyze the anticancer effect of THIO on CSCs derived from HCT116 cells, a CCK-8 assay was performed. In order to explore the effect of THIO on fibroblast cells, representing one of the largest amount of cells in humans, HELF cells were also selected in CCK8-assay. The cells were seeded in 96-well plates at a concentration of 1.0×10⁵ cells/well, and were treated with THIO (Sigma-Aldrich) at different concentrations (0, 10, 20 or 50 µM) for 24 h. The medium without any cells was used as the blank group, while 100 µM cisplatin (DDP) was used as the positive control group. The proliferation index of each group was determined using the Cell Counting kit-8 (CCK-8; Dojindo Laboratories, Tokyo, Japan) according to the manufacturers instructions (21). In brief, 10 µl of CCK-8 solution was added into each well (containing 100 µl of medium) and cultured for 1–2 h at 37°C. The absorbance at 450 nm, which was directly proportional to the number of living cells, was observed for each group (n=4). The inhibition ratio was used to measure cell proliferation in the present study and it was described as (absorbance of 0 µM group - absorbance of each experimental group)/(absorbance of 0 µM group - the absorbance of the blank group).

To assess cell invasion, a Transwell system was used (pore size: 8 µm; Corning, Inc., Corning, NY, USA) following the manufacturers protocol. The CSCs were seeded onto the upper insert covered with Matrigel at a concentration of 1×10⁵ cells per insert in serum-free medium. Outer wells were filled with RPMI-1640 medium containing 10% FBS as a chemoattractant. Then, cells were incubated for 48 h at 37°C. Non-invading cells were removed by swabbing the top layer, and cells able to migrate through the gel and attach to the lower surface of the membrane were stained with crystal violet. The number of cells in four randomly selected microscopy fields was counted for each filter.

Total RNA was extracted using TRIzol reagent (Invitrogen). For each sample, RNA (2 mg) was reverse-transcribed using an RT-PCR kit (Takara, Shiga, Japan), and qPCR was performed with a Thermal Cycler Dice™ Real-Time System and SYBR-Green Premix EX Taq™ (Takara). In the present study, GAPDH was used for qPCR normalization, and all measurements were performed in triplicate. The primer sequences used (5→3) are shown in Table I.

Western blotting was carried out to test for cleaved caspase-3 using the same protocol as the one we reported in a previous publication (21). Briefly, cells were lysed with lysis buffer (50 mM Tris pH 7.0, 1 mM EDTA, 150 mM NaCl, 1% NP40, 10 mM NaF, 1 mM Na₃VO₄) containing protease inhibitor cocktail (Roche, Basel, Switzerland), and protein concentrations were determined using a BCA assay kit (Beyotime Institute of Biotechnology, Nanjing, China). Protein bands were separated by electrophoresis in a 12% sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE) and electroblotted onto polyvinylidene fluoride (PVDF) membranes (Millipore, Bedford, MA, USA). After blocking them with 4% non-fat dry milk in Tris-buffered saline (TBS), the membranes were incubated overnight at 4°C with primary antibodies (rabbit anti-caspase-3; Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1:1,000 in TBS. They were washed three times with TBS containing 0.5% Tween-20, and then incubated for 1 h at 25°C with secondary antibodies conjugated with a 1:5,000 dilution of horseradish peroxidase (HRP) in TBS. Membranes were then washed three times in TBS containing 0.5% Tween-20 at 25°C. Finally, protein bands were visualized on X-ray film using enhanced chemiluminescence (ECL; GE Healthcare, Bethesda, MD, USA).

To analyze cell apoptosis, acridine orange/ethidium bromide (AO/EB) staining and Annexin V-FITC/PI staining were used according to the manufacturers instructions (BD Biosciences). Moreover, to determine the effect of THIO on mitochondrial membrane potential in CSCs, JC-1 apoptosis detection kit (BD Biosciences) was used according to the manufacturers instructions and assessed by fluorescence-activated cell sorting (FACS).

Statistical analysis was performed with the SPSS 17.0. The results are expressed as mean ± SEM. The differences between the groups were assessed by one-way ANOVA followed by t-tests. P<0.05 was considered statistically significant.

The CSCs isolated from HCT116 formed tumor spheres in suspension when cultured in vitro. Their appearance was distinct from the original HCT116 cells, which were spindle-shaped and grew in adherent state (Fig. 1A). After three to four passages, the cells were further characterized by immunohistochemistry and flow cytometry. The results indicated that the CSCs used in the present study were positive for EpCAM (97.9%) and CD44 (92.7%), while HCT116 cells used as the negative control group displayed a low expression of EpCAM (1.21%) and CD44 (0.87%) (Fig. 1A). To further confirm the colony-forming ability of the CSCs, we separated the CSC spheres into single cells and analyzed their colony-formation ability in 96-well plates. As expected, a single CSC could proliferate and grow; in fact, over 40% of the CSCs formed a tumor sphere after 21 days of single-cell culture (Fig. 1B). In addition, we confirmed that the serum-free condition was necessary for the colony-formation ability and the stem cell-specific gene expression of CSCs. We compared the in vitro cell culture in basal conditions plus 10% FBS and in serum-free conditions. Cultures under FBS conditions could lead to a layer of adherent confluent cells. Compared with HCT116 cells, the CSCs showed a high expression of stem cell-specific genes, NANOG, and CLDN-6 (P<0.05), while downregulation of these genes occurred in the CSCs cultured under FBS conditions (P<0.05) (Fig. 1C). These results suggest that the tumor sphere-like colonies could be obtained from the HCT116 cell line, and that these cells had some stem cell characteristics. In serum-free medium supplemented with EGF and bFGF, the CSCs differentiated even under conditions of an extra-low cell concentration, such as single cell conditions, ruling out the possibility that CSCs may aggregate owing to the high concentration of cells in cultures.

After treatment with THIO for 24 h at different concentrations, severe morphological alterations were observed in the majority of CSCs cells. We found that the CSCs could still form tumor sphere-like colonies when treated with THIO at 10 µM. However, the number of colonies formed at 10 µM was considerably lower than at 0 µM. No colony formation could be observed at THIO concentrations of 20 µM and above and after treatment with DDP (Fig. 2A).

After crystal violet staining, we observed that the CSC invasion across the membrane from the upper chamber to the lower surface of the membrane was suppressed by THIO. The number of cells on the lower surface of the membrane clearly decreased in the THIO-treated groups at concentrations of 20 and 50 µM, while a slight inhibition of the cell invasion ability was observed in the 10 µM concentration group (Fig. 2B). After counting the number of cells on the lower surface of the membrane, we concluded that THIO could significantly suppress the invasion of CSCs (P<0.05). This effect occurred in a concentration-dependent manner (52.0±8.0 cells for the 0 µM group, 35.6±7.3 cells for the 10 µM group, 22.0±5.4 cells for the 20 µM group, 10.3±4.0 cells for the 50 µM group, and 14.3±3.3 cells for the DDP group) (Fig. 2B).

To further analyze the anticancer effect of THIO on CSCs derived from HCT116 cells, a CCK-8 assay was performed using treated cells (normal human lung fibroblast cell line, HELF cells, HCT116 cells and CSCs). The results indicated that the proliferation of HELF and HCT116 cells was significantly inhibited by THIO at 50 and 100 µM, while that of CSCs was significantly inhibited by THIO at a concentration as low as 1 µM. The inhibition ratio of THIO at 1 and 10 µM was 19.2 and 39.5%, respectively. However, THIO concentrations of 20, 50 and 100 µM exhibited inhibition at very high rates (94.5, 100 and 100%, respectively). The positive control group (10 µM DDP) showed an inhibition rate of 82.7%. Interestingly, THIO at 20 µM did affect the vitality of HELF cells, but the inhibition rate was only 10.6%, while that of CSCs was 94.5% (Fig. 2C). These results indicate that, at a suitable concentration, THIO may be an optimal novel agent for colon cancer treatment. However, further studies are necessary to elucidate the mechanism of action driving this anticancer activity.

THIO has been reported to induce cervical and endometrial cancer cell apoptosis (7). To detect whether THIO could induce apoptosis in CSCs derived from HCT116 cells, we carried out AO/EB staining. The CSC spheres were labeled by AO/EB after treatment with THIO at different concentrations, and dual staining was examined using confocal microscopy. No significant apoptosis was detected in the negative control group. However, in the group treated with THIO at a concentration of 10 µM, many early-stage apoptotic cells (marked by yellow-green AO nuclear staining) and some late-stage apoptotic cells (marked by orange-red nuclear EB staining) were detected. At higher concentrations (20 and 50 µM), the number of late-stage apoptotic cells was even higher (Fig. 3A). At 50 µM, CSC spheres were much smaller than at lower concentrations, and many single cells could be observed in late-stage apoptosis (Fig. 3A). Further detection by the Annexin V/PI double staining assay revealed that THIO treatment could lead to an increase in the proportion of early-stage apoptotic cells (Annexin V-positive and PI-negative cells), as well as late-stage apoptotic cells (Annexin V-positive and PI-positive cells) in CSCs (Fig. 3B). Further analysis indicated that this induction of apoptosis by THIO occurred in a concentration-dependent manner. The percentage of early-stage apoptotic cells was 5.46, 8.94, 18.8, 40.7 and 14.3% at THIO concentrations of 0, 10, 20 and 50 µM, and DDP treatment, respectively. The percentage of late-stage apoptotic cells was 1.06, 27.4, 32.5, 34.6 and 42.5% at THIO concentrations of 0, 10, 20 and 50 µM, and DDP treatment, respectively (Fig. 3B).

To elucidate the mechanisms of cell apoptosis induced by THIO, cell apoptosis genes and mitochondrial membrane potential were assayed as part of this study. qPCR results indicated that apoptosis genes such as caspase-3 and Bax were significantly upregulated in THIO-treated CSCs (P<0.05), while the anti-apoptosis gene Bcl-2 was significantly downregulated under the same conditions (P<0.05) (Fig. 3C). Moreover, the expression of the caspase-3 gene was confirmed by the western blot analysis. To evaluate the THIO-induced cell apoptosis at the subcellular organelle level, JC-1 staining was carried out to determine the alterations in mitochondria. The percentage of cells with loss of mitochondrial membrane potential increased in a concentration-dependent manner (45.6% for the 0 µM group, 47.6% for the 10 µM group, 51.6% for the 20 µM group, 86.7% for the 50 µM group, and 73.6% for the DDP group) (Fig. 3D). The results suggest that THIO may induce apoptosis in CSCs from HCT116 cells via the loss of mitochondrial membrane potential.