The esophageal cancer cell lines KYSE-150, KYSE-70, and KYSE-450 were gifted by Dr. Johnny C.O. Tang of Hong Kong Polytechnic University (Hong Kong). KYSE-150 cell were cultured in Roswell Park Memorial Institute (RPMI) 1640 supplemented with 2% fetal bovine serum (FBS) while KYSE-70 and KYSE-450 cell lines were grown in RPMI 1640 supplemented with 10% FBS (Shimada et al., 1992). HCT116, SW480, Huh-7, HepG2, and MIHA cell lines were purchased from American Type Culture Collection (ATCC, Manassas, United States). Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% FBS according with the instruction. All the culture medium was supplemented with 100 units/ml penicillin and 100 μg/ml streptomycin (Thermo Fisher Scientific Inc., MA, United States). Cells were maintained in a humidified atmosphere containing 5% CO₂ at 37°C.

AZOX (illustrated in Figure 1A) was purchased from Sigma–Aldrich (MO, United States) and dissolved in dimethyl sulfoxide (DMSO) (Sigma–Aldrich, MO, United States) with stock concentration at 25 mg/ml and stored in -20°C. Bax channel blocker was purchased from Tocris Bioscience (Bristol, United Kingdom). Antibodies for β-actin, cleaved caspase-3, cleaved caspase-9, cleaved caspase-8, β-actin, Bcl-2, Bax, Bad, and cleaved-poly ADP ribose polymerase (PARP) were purchased from Cell Signal Technology (MA, United States). The reagents for mitochondrial protein extraction were purchased from Merck Millipore (MA, United States).

The effects of AZOX on cell proliferation and viability of different cell lines were assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. MTT powder was purchased from Sigma–Aldrich (MO, United States). Briefly, cells with a density of 3000 cells/well was seeded in 96-well plates 24 h prior to AZOX treatment. Then the medium was removed and cells were exposed to different concentration of AZOX for another 48 h followed by MTT assay. Independent experiments were performed in triplicate.

The cell cycle phase distribution was determined by fluorescence-activated cell sorting (FACS) analysis of cellular DNA content. KYSE-150 cells were treated with AZOX in different dosages and time points. Then cells were dissociated with 0.05% trypsin-ethylenediaminetetraacetic acid (EDTA) (Thermo Fisher Scientific Inc., MA, United States), washed with cold phosphate-buffered saline (PBS), and fixed in 70% ethanol at 4°C overnight. The fixed cells were washed twice with PBS and stained with propidium iodide (PI) working buffer (PI 50 μg/ml, RNase 2.5 μg/ml) in the dark at room temperature for 1 h. The stained cells were assessed by flow cytometry (Beckman Coulter, CA, United States) and analyzed with ModFit 4.0 software.

The pro-apoptosis effect of AZOX on KYSE-150 was accessed with FITC Annexin V Apoptosis Detection Kit (BD Biosciences, San Jose, CA, United States) following the standard protocol. Briefly, cells were treated with AZOX with different concentration or time points, and dissociated into single cells with 0.05% trypsin. After washing twice with PBS, cells were stained with annexin V/PI for 20 min at room temperature in the dark and then detected by flow cytometry (Beckman Coulter, United States). The percentage of apoptosis cells were expressed as: (a) early apoptosis cancer cells (annexin +ve; PI -ve), (b) late apoptosis cancer cells (annexin +ve; PI +ve) (Paul et al., 2013).

KYSE-150 cells treated with AZOX for different concentration or different time course. Whole cell lysates were obtained by suspending the cells with pre-cold RIPA lysis buffer (50 mM Tris–HCl, 0.1% SDS, 150 mM NaCl, 2 mM EDTA, 50 mM NaF, 0.5% sodium deoxycholate, 1% NP-40, pH = 7.4) for 30 min on ice, followed by centrifugation at 14,000 rpm for 10 min at 4°C. To detect the expression level of Bax, Bcl-2, and Bad, mitochondria fractions were extracted. Briefly, cells were collected and washed with ice-cold PBS, centrifuged at 600 ×g for 5 min. The cell pellets were resuspended and homogenized in 1× cytosol extraction buffer mix. After centrifugation steps, the supernatant was isolated as cytosolic fraction. The cell pellets were then dissolved in mitochondria extraction buffer mix and vortex for 10 s to obtain the mitochondrial fraction. Western blot analysis was conducted as described previously (Lin et al., 2015). Protein concentration was detected using Pierce BCA protein assay kit (Thermo Fisher Scientific Inc, MA, United States), subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes. After blocking with 5% bovine serum albumin in PBS-Tween 20 buffer (PBST) for 1 h at room temperature, the membranes were incubated with the primary antibodies overnight at 4°C. After three-time washing with PBST, the membranes were re-incubated with corresponding secondary antibodies for 1 h at room temperature and then subjected to electrochemiluminescence immunoassay.

Male BALB/c nude mice (6-week-old) were purchased from the Laboratory Animal Services Centre, The Chinese University of Hong Kong. The mice were bred in barrier facilities with 12 h light/dark cycle environment and free access to food and water ad libitum. Each mouse was inoculated subcutaneously on the flank of the mice with 2 × 10⁶ KYSE-150 cells in 100 μl PBS. Once tumor size reached about 80 mm³, the mice were divided randomly into two groups with six mice per group: AZOX group received AZOX (40 mg/kg/day) dissolved in 0.5% carboxymethylcellulose sodium solution by intragastric administration, while vehicle group received daily 0.5% carboxymethylcellulose sodium solution by intragastric administration. Tumor volume was measured with calipers and calculated by width²× length × 0.4 as previously described (Chen et al., 2015). General health and body weight were also monitored every 2 days. The mice were anesthetized with 7% chloral hydrate and sacrificed 14 days later, and the tumor xenografts were dissected, weighed and fixed in 4% Paraformaldehyde (PFA) for further examination. All experimental protocols were approved by The Animal Ethics Committees of Hong Kong Baptist University, and in accordance with “Institutional Guidelines and Animal Ordinance” from Department of Health, Hong Kong Special Administrative Region.

The data was evaluated as means ± standard error of the mean (SEM). Statistical differences between individual groups were evaluated using Student’s t-test or one-way analysis of variance (ANOVA). All experiments were performed at least three times independently. GraphPad Prism 6.0 software (GraphPad Software Inc., San Diego, CA, United States) was used for the calculations, and p < 0.05 was considered statistically significant.

The effect of increasing AZOX concentrations in cancer cells was examined by MTT assay. The results showed that the cytotoxicity of AZOX on cancer cells are cell-type dependent after 48 h treatment. As shown in Figure 1B, in esophageal cancer cell lines, the inhibitory concentration 50% (IC₅₀) values were calculated to be 2.42, 40.76, and 44.88 μg/ml for KYSE-150, KYSE-450, and KYSE-70, respectively. In contrast, KYSE-520 cell line did not show any effect after AZOX treatment (data not shown). In hepatocellular carcinoma cancer cell lines, AZOX did not alter the Hep3B cell proliferation but could decrease HuH-7 and HepG2 cell viability with IC₅₀ values of 10.86 μg/ml and 22.52 μg/ml, respectively (Figure 1C). It is worth noted that AZOX inhibit MIHA cell growth with IC₅₀ of 68.87 μg/ml. In parallel experiments on colon cancer cell lines, HCT116 cell were more sensitive to AZOX with IC₅₀ at 8 μg/ml than SW480 cell (IC₅₀, 45.44 μg/ml) (Figure 1D).

To test whether AZOX can disturb cell cycle distribution, the effect of AZOX was tested in KYSE-150 cells by PI staining, followed by FACS analysis. Cell treated with DMSO were used as control. As shown in Figures 2A,B, the cells treated with AZOX (15 μg/ml) resulted in an apparent increase in the proportion of S phase at 36 h after the exposure. Further, KYSE-150 cells were treated with AZOX at concentration of 5, 10, and 15 μg/ml. FACS analysis were performed 36 h after the treatment. Significant difference of S phase cell proportion was observed at the dose of 10 μg/ml, and more effective at 15 μg/ml.

Apoptosis is a known cell process linked with mitochondrial respiration (Pelicano et al., 2003). To test the pro-apoptosis effect of AZOX, KYSE-150 cells were treated with AZOX at different time points or with different concentrations. After staining with annexin V/PI, the cells were subjected to quantitative analysis of the apoptotic cell percentage. The result indicated that AZOX could induce cell apoptosis time dependently and dose dependently (Figures 3A,B). The shorted effective time was 24 h and the minimal effective concentration was 10 μg/ml. Treatment of 15 μg/ml AZOX for 36 h can induce approximate 10% cell apoptosis.

To investigate the underlying mechanisms of AZOX-induced apoptosis of KYSE-150 cell, the apoptotic proteins were measured by western blot analysis after the treatment with AZOX at different time points or with different concentrations. Cleavage of PARP in KYSE-150 cells, the indicator of cell apoptosis, was increased by AZOX both time-dependently and dose-dependently (Figure 4A). Since PARP is the critical substrate of caspase family proteins, we next detected whether AZOX would further affect the activities of caspase-3 and caspase-9. As shown in Figure 4A, the cleavage of caspase-3 and caspase-9, showed time-dependent and dose-dependent increases, indicating that AZOX induced KYSE-150 cell apoptosis via activating the caspase-3 and caspase-9. Interestingly, the expression of cleaved caspase-8 was significantly increased by AZOX in KYSE-150 cells (Supplementary Figure 1A). Furthermore, we explored the effect of AZOX on the apoptosis-related proteins in intrinsic mitochondrial pathway. The pro-survival protein Bcl-2 showed slightly decrease by AZOX with a corresponding increased level of Bax (Figure 4A). As a result, the ratio of Bcl-2 to Bax (Bcl-2/Bax) was significantly decreased in KYSE-150 cells after 36 h exposure to AZOX with the concentration of 10 and 20 μg/ml (Figure 4B). Non-phosphorylated Bad, however, was not significantly affected by AZOX in vitro (Figure 4A). Cytochrome c released from the mitochondria has been proposed to be a potential event which would trigger caspase-3, 7/9-dependent cell apoptosis (Jiang and Wang, 2000; Li et al., 2000). In our results, the expression of cytochrome c, unsurprisingly, was increased after AZOX treatment in a time-dependent and dose-dependent manner (Figure 4A). In parallel studies, Bax channel blocker could effectively block AZOX-induced cytochrome c expression and caspase-3 activation (Figure 4C).

The in vivo anti-cancer effect of AZOX was assessed in tumor xenografted mouse model. KYSE-150 cells were injected on the flank of the mice subcutaneously to reach about the size of 80 mm³, then the mice received 40 mg/kg/day AZOX or carboxymethylcellulose sodium solution for 14 consecutive days. The mice body weight and tumor volume were recorded every day. AZOX treatment significantly retarded tumor growth in nude mice as shown in Figure 5A (p = 0.003) and Figure 5C. Accordingly, compared to the control group, mice body weight in AZOX group was significantly decreased (p = 0.0125; Figure 5B). Hematoxylin and eosin staining indicated that AZOX can decrease the cell proliferation compared to control group (Figure 5D). To further confirm our in vitro results, we detected the expression of related proteins in tumor tissues. The results showed that cleaved PARP was increased along with the cleavage of caspase-3, caspase-9 (Figure 6A), and caspase-8 (Supplementary Figure 1B), which was in accordance with the in vitro findings. Moreover, Bcl-2/Bax ratio was significantly suppressed (Figure 6B) along with the increased cytochrome c release from the mitochondria (Figure 6A).