Gastric cancer cell lines NCI-N87 and AGS were obtained from the Cell Bank of the Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). NCI-N87 cells were maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS), and AGS cells were cultured in F-12K with 10% FBS. Both types of cells were incubated at 37°C in an atmosphere with 5% CO₂.

The NCI-N87 and AGS cells were plated in 96-well plates at a density of 10⁴ cells/well, and were treated with thioridazine (Sigma) at indicated concentrations for 48 h. After incubation with 0.5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) (Beyotime Institute of Biotechnology, Haimen, China) for another 4 h, the supernatant was discarded, and 100 μl DMSO (dimethyl sulfoxide) was added to dissolve the produced formazan crystals. The absorbance was measured on a microplate reader (Thermo Fisher Scientific) at a dual wavelength of 595 nm and 630 nm. The difference in the outcome of OD₅₉₅ and OD₆₃₀ was considered as the measure of cell viability.

To directly observe the cytotoxic effects of thioridazine, NCI-N87 and AGS cells were seeded in 24-well plates and incubated with thioridazine at the indicated concentrations. Forty-eight hours post-treatment, the cells were directly observed using microscopy or stained with 2% crystal violet which was dissolved in 20% methanol solution.

The NCI-N87 cells were seeded into 6-well plates at 3,000 cells/well and treated with thioridazine at indicated concentrations after the cells recovered. The formed colonies were stained with 2% crystal violet 10 days later, washed and photographed. The AGS cells were seeded at 1,000 cells/well in 6-well plates, treated similarly to NCI-N87 cells, and were stained with crystal violet 6 days later.

Western blot analysis was performed according to standard procedures. In brief, cells were trypsinized following treatment for the indicated times, lysed using RIPA lysis buffer and boiled in SDS-PAGE sample loading buffer (both from Beyotime Institute of Biotechnology). Total protein (40 μg) was electrophoretically separated on 10 or 12% SDS-PAGE gel and transferred onto a polyvinylidene difluoride (PVDF) membrane. The PVDF membrane was blocked with milk, incubated with the primary antibodies and then with the secondary antibodies. Primary antibodies for caspase-9, caspase-8, caspase-3 and PARP were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and GAPDH was from CoWin Bioscience Co. (Beijing, China). All the secondary antibodies were obtained from Santa Cruz Biotechnology. Subsequently, the protein on the membrane was revealed using ECL reagents (Pierce Biotechnology, Rockford, IL, USA).

The AGS cells were treated with thioridazine at indicated concentrations, and fixed with 4% paraformaldehyde (PFA) for 15 min 48 h post-treatment. The cells were then stained with Hoechst 33258 (Molecular Probes, Eugene, OR, USA) at 1 μg/ml for 1 min, washed with PBS and photographed under a fluorescence microscope.

The NCI-N87 and AGS cells were trypsinized following treatment with thioridazine for 48 h, and then resuspended in PBS and fixed with 70% ethanol overnight at −20°C. The fixed cells were centrifugated, washed with PBS and incubated with 20 μg/ml RNAase in a water bath at 37°C for 30 min. After been stained with 50 μg/ml PI for another 30 min, they were subjected to a fluorescence-activated cell sorting (FACS) instrument for analysis.

The NCI-87 and AGS cells were treated with thioridazine for 48 h. Cells were trypsinized into single cells, washed with PBS and binding buffer, and then resuspended in 500 μl binding buffer. Annexin V-FITC (5 μl) and 10 μl propidium iodide (PI, 20 μg/ml) were subsequently added to the cells for incubation. Fifteen minutes later, the cells were washed with binding buffer and subjected to FACS for analysis.

Alteration in mitochondrial membrane potential is detected by JC-1 staining assay. Unimpaired cells retain high mitochondrial membrane potential, and JC-1 accumulates in the matrix of the mitochondria and propagates as aggregates, which emit red fluorescence in the presence of excitation. However, the mitochondrial membrane potential decreases as cells are impaired, and then JC-1 diffuses into the cytoplasm and forms monomers, which emit green fluorescence after excitation. The experiment was performed as follows.

The cells were seeded in 6-well plates at a density of 3×10⁶ cells/well and treated with thioridazine for 48 h. After been trypsinized, cells were resuspended in 500 μl DMEM, and then mixed with 500 μl JC-1 working solution and incubated for 20 min at 37°C in the dark. Afterwards, the cells were washed twice with JC-1 staining buffer, resuspended in 300 μl JC-1 staining buffer and assessed by FACS.

The animal experiments were performed according to the U.S. Public Health Service Policy on Humane Care and the Use of Laboratory Animals, and were also approved by the Institutional Animal Care and Use Committee. Four-week-old female BALB/c nude mice were purchased from Shanghai Laboratory Animal Center (SLAC; Shanghai, China). To test the ability of thioridazine to inhibit tumor growth, the NCI-N87 cells were incubated with DMSO or 5 μM thioridazine for 24 h. The treated cells were trypsinized and counted after been stained with trypan blue. Viable DMSO-pretreated (1×10⁶) and thioridazine-pretreated NCI-N87 cells were subcutaneously injected into the left and right rear of mice, respectively. After visible tumors emerged, they were measured every three days. The volume was calculated as length × width × width/2.

Comparison of groups was analyzed by the Student’s t-test. A p-value of less than 0.05 was considered to indicate a statistically significant difference.

To initially assess the anticancer effect of thioridazine on gastric cancer cells, MTT assay was performed on treated cells. Thioridazine reduced the cell viability of NCI-N87 and AGS cells in a concentration-dependent manner (Fig. 1A). Moreover, severe morphological alterations were observed in the majority of the cells treated with 15 μM thioridazine (Fig. 1B). Results of the crystal violet staining clearly disclosed the cytotoxic effect of thioridazine on NCI-N87 and AGS cells (Fig. 1C). Furthermore, 5 μM thioridazine, which showed a slight effect on the cell viability of NCI-N87 and AGS cells, markedly inhibited their colony formation ability (Fig. 1D). Thioridazine at a concentration of 2 μM also exhibited a slight inhibitory effect on the colony formation of NCI-N87 and AGS cells.

Thioridazine was previously found to induce cervical and endometrial cancer apoptosis. To detect whether thioridazine induces gastric cancer cell apoptosis, Hoechst 333258 staining was carried out. Severe nuclear fragmentation was observed in the AGS cells following treatment with thioridazine at 10 and 15 μM, but not in the cells treated with DMSO or thioridazine at a low dosage (Fig. 2A). As the appearance of nuclear fragmentation indicates the occurrence of apoptosis, the cell cycle distribution was analyzed in the treated and untreated cells. The proportion of sub-G1 phase cells in the NCI-N87 and AGS cells increased as the dosage of thioridazine was increased (Fig. 2B). Further detection by Annexin V/PI double staining assay revealed that thioridazine treatment resulted in an increase in the percentage of Annexin V-positive cells in both the NCI-N87 and AGS cells (Fig. 2C). The results indicate the occurrence of apoptosis in the thioridazine-treated gastric cancer cells.

To further clarify the mechanism of thioridazine-induced apoptosis, expression of a number of apoptosis-related proteins was analyzed. PARP cleavage was observed in the presence of thioridazine, which further supported the occurrence of apoptosis. Precursors of caspase-9, caspase-8 and caspase-3 were decreased in the NCI-N87 cells following the treatment of thioridazine for 48 h. Moreover, downregulation of precursors of caspase-9, caspase-8 and caspase-3 was observed in the AGS cells treated with 15 μM thioridazine for 24 or 48 h (Fig. 3A). These results indicate the involvement of caspases in thioridazine-induced apoptosis in a time- and dose-dependent manner. Furthermore, caspase inhibitor Z-VAD-FMK was able to reverse the cytotoxicity of thioridazine both in the NCI-N87 and AGS cells (Fig. 3B), suggesting that thioridazine induced gastric cancer cell apoptosis in a caspase-dependent manner.

As caspase-9 is involved in the mitochondrial apoptosis pathway, a decreased level of caspase-9 precursor implied the breakdown of mitochondria in the thioridazine-induced apoptosis. JC-1 staining assay was then carried out to examine the alteration of mitochondria. The percentage of cells with loss of mitochondrial membrane potential increased from 21.7 to 40.2% in the NCI-N87 cells following treatment with 15 μM thioridazine (Fig. 4A), and this proportion increased from 12.5 to 49.8% in AGS cells (Fig. 4B). The loss of mitochondrial membrane potential, along with decreased caspase-9 (Fig. 3A), suggest that thioridazine induced gastric cancer cell death via the mitochondrial apoptosis pathway.

To test the in vivo activity of thioridazine on gastric cancer cells, NCI-N87 cells were pretreated with DMSO or 5 μM thioridazine for 24 h, and the cells were then injected subcutaneously into the nude mice at the left or right rear to compare their tumor formation ability. Thioridazine (5 μM) elicited little change in the cell viability of NCI-N87 cells (Fig. 1A); however, pretreatment with 5 μM thioridazine inhibited the growth of NCI-N87 cell-derived tumors (Fig. 5A), and this was consistent with thioridazine’s ability to inhibit colony formation (Fig. 1D). Obviously smaller tumors were observed at the right rear region of the mice which were derived from the thioridazine-pretreated cells, and the resected tumors exhibited a similar result (Fig. 5B and C). Moreover, the weight of the tumors derived from the thioridazine-pretreated cells was lower than the tumors derived from the cells pretreated with DMSO (Fig. 5D).