Human esophageal cancer cell line Eca-109 was provided by the Cell Resource Center of Shanghai Life Sciences Institute, Chinese Academy of Sciences (Shanghai, China). The cells were cultured in Roswell Park Memorial Institute (RPMI)-1640 medium (Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% fetal bovine serum (Zhejiang Kangchen Biotech Co., Ltd., Wuhan, China), at 37°C in an atmosphere containing 5% CO₂. When the cells reached a confluence of ~90% (~every three days), they were digested and passaged. The cells in passages 3–5 were used for experimental analyses.

The Eca-109 cells in logarithmic phase were prepared to a single cell suspension (2×10⁴/ml) and 100 µl of the cells were added in 96-well culture plates. Then, 50 µl of GEM (≥98%, high performance liquid chromatography; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was added in the wells to reach various final concentrations (1, 2, 4, 8, 16 µg/ml). Next, the treated cells were cultured in an atmosphere containing 5% CO₂ at 37°C for 24 or 48 h. Following this, the cells were collected and centrifuged at (150 × g) for 10 min. The supernatant was discarded, and the precipitate was washed with PBS once. A total of 200 µl of serum-free RPMI-1640 medium was added, followed by addition of 10 µl of MTT (5 mg/ml). Following culturing for a further 4 h, 100 µl of dimethyl sulfoxide solution was added to dissolve the formazan. The cell viability was assessed by measuring the optical density (OD) at a wavelength of 450 nm in a Multiskan SPECTRUM full spectrum microplate spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The untreated wells also served as a control. Data from triplicate samples were averaged. The cell inhibition rate=(OD_control-OD_treated)/OD_control × 100%.

The cells were seeded at a concentration of 2×10⁴ cells/ml, then cells were treated with GEM (4 µg/ml). After co-cultured for 12 and 24 h, the cells were removed from the plates by using 0.25% trypsin. The cells were centrifuged at 150 × g for 10 min and then collected. Following washing with PBS twice, 1×10⁵−5×10⁵ cells were resuspended in 500 µl binding buffer and sequentially mixed with 5 µl of Annexin V-FITC (Nanjing KeyGEN Biotech Co., Ltd., Nanjing, China) in the dark for 15 min at room temperature, followed by 5 µl of propidium iodide (PI) (Nanjing KeyGen Biotech Co., Ltd., Nanjing, China) for 5 min. The cells were incubated at room temperature in the dark for 5–10 min. Apoptotic cells were detected using a flow cytometer (Beckman Coulter, Inc., Brea, CA, USA) with an excitation wavelength at 488 nm and an emission wavelength at 530 nm and observed using a fluorescence microscope (Olympus Corporation, Tokyo, Japan). The data was analyzed using kaluza version 1.20 software (Beckman Coulter, Inc.).

The treated cells were seeded in 24-well plates (2×10⁵/ml). Following a culture period of 72 h, the cells were fixed using 3.7% neutral formalin for 10 min at room temperature. Then, they were rehydrated with a descending gradient of ethanol (100, 95, and 70%, 5 min each). Then, they were washed with PBS for 10 min, and 50 µl of cell suspension was added on a poly-l-lysine-pretreated slide. A total of 0.01 M PBS was then used to wash the cells, followed by the TdT labeling at room temperature for 5 min. The reaction was stopped with a stop buffer at room temperature for 5 min, followed by an incubation with 50 µl of FITC-labeled antibody (cat.: 129–10684; R&D Systems Inc., Minneapolis, MN, USA) at room temperature for 30 min, and then washed with PBS twice. The cell nuclei were examined under a laser scanning confocal microscope (at 490 nm excitation wavelength and 520 nm emission wavelength). A total of 20 random fields of view were selected for analysis.

The cells were treated with GEM (8 µg/ml) as previously described. Then, the cells were collected and prepared into specimens ~3.0×1.0 mm, and then fixed in 3% glutaraldehyde and osmic acid for 24 h at 4°C. These specimens were dehydrated using graded ethanol at 37°C for 30 min, followed by embedding in epoxy resin. Then, the specimens were cut into 50 nm-thick sections consecutively. They were doubly stained with 2% uranyl acetate and lead citrate for 12 h at room temperature. The sections were washed and then the alterations in endocardium, nucleus, cytoplasm, and matrix components were examined using a HT7700 transmission electron microscope (Hitachi, Ltd., Tokyo, Japan).

The cells were treated as previously described. Following a culture period of 24 h, the cells were harvested and collected. Then, a cell lysis solution was added containing 8 mol/l urea, 40 g/l CHAPS, 2 mmol/l TBP, and 2 ml/l Bio-Lyte. The cell lysates were collected and centrifuged at 13,400 × g at 4°C for 15 min. The supernatant was then harvested and the proteins were quantified using the Bradford method.

The protein samples (~400 µg) were subjected to immobilized pH gradients (IPG) isoelectric focusing (IEF) and then run on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Next, the proteins were isolated by vertical electrophoresis. The gel was stained with coomassie brilliant blue R-250 for 2 h and destained for 2 h until the protein spots were clear. The stained proteins were scanned using a FXMolecular Imager (Bio-Rad Laboratories Inc., Hercules, CA, USA).

The differential protein spots (>than 3-fold alteration in OD) were cut off from the gel and digested with 0.25% trypsin for 20 h. The digested peptide fragments were isolated and desalinated with ZipTipTM (EMD Millipore, Billerica, MA, USA). Amino acid sequence analysis was then performed. According to the mass of the digested peptide fragment, the Mascot score was automatically obtained to verify the potential characteristics and names of the differential proteins by checking the Swiss-Prot protein database.

The cells were treated and lysed using the aforementioned procedure. The proteins were separated by 12% SDS-PAGE and then transferred onto a sheet of polyvinylidene fluoride membrane. Following blocking with 5% skim milk for 4 h and washing with Tris-buffered saline (TBS ×3, 5 min), the membrane was respectively incubated overnight with anti-human ASC (cat no. sc-33958), Mcl-1(cat no. sc-53951) and Bax-a (cat no. sc-70408) polyclonal antibody (1:200; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) at 4°C, followed by incubation with horseradish peroxidase-conjugated rabbit anti-goat IgG (cat no. TA130028; 1:3,000; Origene technologies Inc., Beijing, China) for 2 h at room temperature. Immunoreactive bands were detected by an enhanced chemiluminescence reagent (Pierce; Thermo Fisher Scientific, Inc.), visualized by autoradiography, and quantified by the QuantityOne analysis system (Bio-Rad Laboratories, Inc.). β-actin (primary antibody dilution 1:200; cat no. sc-130656; Santa Cruz Biotechnology, Inc.) served as an internal control.

Data are expressed as the mean ± standard error of the mean, and statistical comparisons were carried out by Student's t-test or one-way analysis of variance followed by Tukey's post hoc test, with the SPSS statistical software, version 17.0 (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

The MTT results demonstrated that GEM significantly inhibited the proliferation of Eca-109 cells. Furthermore, the inhibitory effect was observed to have been exhibited in a time- and concentration-dependent manner. The greatest inhibitory effect of GEM was at 8 µg/ml for 24 h (Fig. 1).

Following treatment with GEM (4 µg/ml) for 12 h, the apoptosis proportion (the cells doubly stained with AnnexinV-FITC and PI represent apoptotic cells) of the cancer cells significantly increased compared with the control (17.5±5.1% vs. 8.1±2.4%, P<0.05; Fig. 2A and B). With the prolongation of the treatment time, the apoptosis proportion was significantly elevated in the treated cancer cells. The FACS result demonstrated that the apoptosis proportion in the GEM-treated cells was increased compared with controls (36.1±8.8% vs. 16.4±5.8%, P<0.01; Fig. 2C and D).

Furthermore, TUNEL combined with laser scanning confocal microscope observations, demonstrated that GEM (4 µg/ml) significantly induced the apoptosis of the cancer cells. The majority of control cells were only stained with PI and few with green fluorescence (Fig. 3A), whereas the apoptotic cells (AnnexinV-FITC and PI double staining) were stained with green fluorescence (Fig. 3B). In addition, nuclear staining density of the untreated cells was lower and the nuclear shapes were larger (Fig. 3A), whereas the treated cells had dense chromatins and comparatively smaller nuclear shapes (Fig. 3B).

The alterations in the mitochondrial ultrastructure of the treated Eca-109 cells were notable (Fig. 4). Following treatment with GEM for 24 h, swelling mitochondria, fading matrix and a lack of mitochondrial crista were observed in the Eca-109 cells (Fig. 4).

The 2-DE assay result demonstrated that there were 12 protein spots between the acidic region (pH 4.7–6.5). These protein spots were cut off and digested with trypsin (Fig. 5). The MALDI-TOF-MS assay was performed in the 12 protein spots to gain access to peptide fingerprints and charge to mass ratio. Then, three differentially expressed proteins including spots 3, 7 and 12 (Fig. 5) were identified by Mascot and a ProteinProspector peptide fingerprint matching software. The three differential proteins were validated to be Bax-α (spot 3), myeloid cell leukemia sequence (Mcl-1; spot 7), and apoptosis-associated speck-like protein containing a CARD (ASC) (spot 12). The gray scales were increased in Bax-α and ASC and decreased in Mcl-1 in the treated cells compared with the controls (P<0.01; Fig. 6).

The western blotting result demonstrated that Bax-α and ASC protein levels increased in the treated cells at 12 and 24 h following the GEM treatment (Fig. 7). However, the Mcl-1 protein levels in the treated cells significantly decreased compared with the control (Fig. 7)