The human ESCC cell lines EC9706 and KYSE70 (Cell Bank of Chinese Academy of Sciences, Shanghai Institute of Cell Biology, Shanghai, China) were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS; both Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) in 5% CO₂ at 37°C. Human umbilical vein endothelial cells (HUVECs; Zhong Qiao Xin Zhou Biotech, Shanghai, China) were cultured in high-glucose Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) containing 20% FBS in 5% CO₂ at 37°C. Aidi was purchased from Guizhou Yibai Pharmaceutical Co., Ltd. (Guizhou, China) with an initial concentration of 0.3 g/ml. 5-Fluorouracil (5-Fu; 0.025 g/ml) was purchased from Jin Yao Amino Acid Co., Ltd. (Tianjin, China). The drugs were diluted to the required concentration in a sterile environment with the cell culture media mentioned above, and were used in in vitro experiments; drugs diluted with sterile saline were used in in vivo experiments.

The cells (2×10³) were cultured in 96-well plates for 24 h at 37°C. Culture medium (100 µl) containing 1.5, 3, 6, 12, 24, 48 or 96 mg/ml Aidi was added following removal of the original medium. The cells were cultured for 24 or 48 h at 37°C. A volume of 15 µl 5 mg/ml MTT (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was added and incubated for 4 h, and subsequently 180 µl dimethyl sulfoxide (Amresco, LLC, Solon, OH, USA) was added. Once the bluish violet crystalline formazan had dissolved completely with gentle shaking for 10 min, the absorbance (A) was detected at 490 nm using a microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The cell growth rate (%) was determined as follows: (A_{control group}-A_{observed group})/A_{control group} ×100.

A total of 1×10⁵ cells per well were cultured in 6-well plates for 24 h at 37°C. Then cells were treated with 3, 6 or 12 mg/ml Aidi, 32 mg/l 5-Fu, or without any drugs respectively for 24 h at 37°C. Cells were observed and photographed using bright field microscopy (magnification, ×100, BX41; Olympus Corporation, Tokyo Japan).

To detect the rate of migration, a wound-healing assay was performed. Cells were cultured on 6-well plates for 24 h, prior to being treated with 6, 12 or 24 mg/ml Aidi, 32 mg/l 5-Fu, or without drugs for 24 h. A pipette tip was used to scratch the monolayer in each well. Detached cells were removed by washing twice with PBS. The remaining adherent cells were cultured with complete culture medium for 24 h. The extent of space filling, representing the cell migration, was evaluated using bright field microscopy (magnification, ×100, BX41; Olympus Corporation, Tokyo Japan). The migratory capacity of cells was calculated as (scratch width of the treatment group-scratch width of the control group)/scratch width of the control group ×100.

To detect the invasive ability of the cells, Corning Matrigel Invasion Chambers (Corning Incorporated, Corning, NY, USA) were used. Matrigel was diluted with RPMI-1640 medium at a ratio of 1:8 and used to coat the upper chamber. RPMI-1640 or DMEM with 20% FBS (500 µl) was added to the lower chamber. Cells were collected following 24 h of treatment, as specified in the wound-healing assay method. Cells (5×10⁴) in serum-free RPMI-1640 medium were added to the upper chamber and cultured for 24 h at 37°C. The non-invading cells were removed by wiping gently with cotton swabs. The invading cells on the lower membrane were fixed with 100% methanol at room temperature for 15 min and stained with 0.1% crystal violet (Sigma-Aldrich; Merck KGaA) at room temperature for 30 min. The invading cells were detected by light microscopy (magnification, ×100, BX53M; Olympus Corporation). The invasive capacity of cells was calculated as (number of invading cells in the treatment group/number of invading cells in the control group).

In the tube formation assay, HUVECs and ESCC KYSE70 cells were used to determine the rate of angiogenesis in a vasculogenic mimicry (VM) formation assay. Matrigel diluted with serum-free medium (1:1) was applied to a 96-well plate at 50 µl/well, and incubated at 37°C for 1 h. A total of 5×10³ cells treated with drugs for 24 h suspended in 50 µl culture medium were added to the matrix gel and incubated in 5% CO₂ at 37°C for 16 h. The three-dimensional organization of the cells was examined and imaged under an inverted microscope (magnification, ×100).

Total protein from EC9706 and KYSE70 cells treated for 24 h was extracted by lysis in solubilizing buffer with 1 mmol/l phenylmethanesulfonyl fluoride and a protease inhibitor cocktail (Beyotime Institute of Biotechnology, Haimen, China), and subsequently centrifuged at 13,000 × g at 4°C for 15 min. Protein concentration was detected by a Bicinchoninic Acid protein measuring kit (Beyotime Institute of Biotechnology). A total of 30 µg protein per lane was separated by 10% SDS-PAGE and transferred onto polyvinylidene fluoride membranes. The membranes were blocked with 5% skimmed milk at room temperature for 2 h and incubated with primary antibodies against cadherin-1 [Cell Signaling Technology (CST), Inc., Danvers, MA, USA; cat. no. 3195; dilution, 1:1,000], cadherin-2 (CST, Inc.; cat. no. 13116; dilution, 1:1,000), vimentin (CST, Inc.; cat. no. 5741; dilution, 1:1,000) and vascular endothelial growth factor (VEGF-A; Santa Cruz Biotechnology, Inc., Dallas, TX, USA; cat. no. sc-152; dilution, 1:500) overnight at 4°C. β-actin (CST, Inc.; cat. no. 4970; dilution, 1:1,000) was utilized as a loading control. A horseradish peroxidase-conjugated goat anti-rabbit antibody (CST, Inc.; cat. no. 7074; dilution, 1:2,000) was subsequently applied at room temperature for 2 h. An enhanced chemiluminescent detection reagent (Thermo Fisher Scientific, Inc.) was used to visualize the immunoreactive signals with a gel imaging system (Bio-Rad Laboratories, Inc.). Densitometry analysis of bands was calculated using Quantity One analysis software, version 4.62 (Bio-Rad Laboratories, Inc.).

To evaluate the anti-metastatic effects of Aidi injection, nude mouse peritoneal metastasis models were established. A total of 24 female BALB/c NU mice (5 weeks old) were purchased from the Comparative Medicine Laboratory Animal Center (license no. scxk (SU) 2012–0004) of Yangzhou University (Jiangsu, China). These nude mice weighed ~19 g. The nude mice were housed at a constant temperature (25–27°C), constant humidity (45–50%), specific pathogen free environment. Sterile purified water and food were provided ad libitum. EC9706 cells (2×10⁶) were inoculated into the peritoneal cavity of 20 mice, and the rest of 4 mice were not injected with the cells as the normal group. After 14 days following tumor implantation, the mice were treated with Aidi injection (1, 2 or 4 g/kg/day), 60 mg/kg 5-Fu, or the same volume of saline (control group) via intraperitoneal injection for 15 days. Each group contained four mice. At the end of treatment, the mice were sacrificed by cervical dislocation and a laparotomy was performed. The numbers of peritoneal cancer nodules were examined. All experimental procedures were performed following internationally accepted guidelines regarding the use of laboratory animals (18) and the protocol of the present study was reviewed and approved by the Institutional Animal Care and Use Committee of Yangzhou University (Jiangsu, China).

All tissues were fixed with 10% neutral buffered formalin at room temperature overnight and embedded in paraffin. Sections of 5-µm thickness of the tumor tissues temperature were cut for immunohistochemical staining. The slides were dewaxed in xylene and rehydrated a descending alcohol series. Hydrogen peroxide (3%) was added to quench the endogenous peroxidase activity for 10 min. Antigen retrieval was performed in citrate buffer at 95°C for 5 min. Goat serum (10%; Beyotime Institute of Biotechnology) was used to block for 20 min at room temperature. Subsequently, the aforementioned primary antibodies (vimentin, dilution 1:200; cadherin-1, dilution 1:200; VEGF-A, dilution 1:100) were incubated with the sections at 4°C overnight. Following this incubation, the slides were incubated with an appropriate secondary antibody conjugated with HRP (CST, Inc.; undiluted; cat. no. 8114) at 37°C for 20 min. The slides were stained with 3,3-diaminobenzidine at room temperature for 10 min following washing with PBS. The slides were counterstained with hematoxylin for 15 min at room temperature. The slides were subsequently dehydrated and mounted with neutral gum for imaging with inverted microscope (magnification, ×200).

Each experiment was repeated three times. All values are presented as the mean ± standard deviation. Data from the in vitro and in vivo experiments were compared using a one-way analysis of variance with Bonferroni's post hoc test, or a Student's t-test. P<0.05 was considered to indicate a statistically significant difference. All analysis was performed using SPSS software version 16.0 (SPSS, Inc., Chicago, IL, USA).

EC9706 and KYSE70 cells treated with varying concentrations (3, 6, 12, 24, 48 or 96 mg/ml) of Aidi for 24 or 48 h revealed significantly inhibited growth (P<0.05, P<0.01; Fig. 1A) in a concentration and time-dependent manner. The cell growth rates of EC9706 for 24 h were 1.5 mg/ml (95.95±2.67%), 3 mg/ml (86.79±2.64%), 6 mg/ml (82.76±3.10%), 12 mg/ml (66.27±2.74%), 24 mg/ml (58.84±2.12%), 48 mg/ml (44.99±4.29%), 96 mg/ml (37.44±3.07%), respectively. The cell growth rates of KYSE70 for 24 h were 1.5 mg/ml (95.68±2.91%), 3 mg/ml (89.95±4.34%), 6 mg/ml (86.61±2.94%), 12 mg/ml (74.36±4.07%), 24 mg/ml (68.65±3.41%), 48 mg/ml (50.11±3.24%), 96 mg/ml (39.57±3.87%), respectively. The half maximal inhibitory concentration at 24 h for EC9706 and KYSE70 cells was 25.89 mg/ml and 32.58 mg/ml, respectively. Aidi was administered via intravenous injection once a day, so 24 h treatment was selected and Aidi concentrations of 6, 12 and 24 mg/ml which growth inhibition rate was <50% were used to avoid cytotoxic effects of excessive concentration in subsequent experiments; 32 mg/ml 5-Fu was used as a control. The morphological characteristics of the cells were observed following the selected treatments. The intercellular space increased, cell numbers were reduced and the cytoplasm shrank in response to Aidi and 5-Fu treatments compared with in the control group of the two cell lines (Fig. 1B).

The migratory capacity of the cells was tested with a wound-healing assay. Treatment with Aidi for 24 h significantly inhibited the migration of EC9706 and KYSE70 cells in a dose-dependent manner (P<0.01; Fig. 2A). The wound-healing rates of the EC9706 cells following treatment with 6, 12 and 24 mg/ml Aidi and 32 mg/ml 5-Fu were 84.77±3.46%, 76.88±2.27%, 46.70±1.65% and 30.44±3.85%, respectively, while the wound-healing rates of the KYSE70 cells were 77.55±3.63%, 47.27±5.54%, 32.84±1.95% and 23.06±2.83%, respectively. The trends of the rates in the two cell types, with different degrees of differentiation, are consistent. The number of invading cells was significantly reduced by treatment with Aidi, in a dose-dependent manner (P<0.01; Fig. 2B). The invasion inhibition rates of treatment with 6, 12 and 24 mg/ml Aidi and 32 mg/ml 5-Fu were 76.46±1.45%, 51.97±4.15%, 26.90±3.52% and 20.85±4.86% in EC9706 cells, and 67.03±5.36%, 43.26±4.29%, 23.67±7.77% and 16.86±6.86% in KYSE70 cells, respectively. These results demonstrated that treatment with Aidi induced an inhibition of ESCC migratory and invasive capacity in a concentration-dependent manner, suggesting an inhibitory effect on metastasis.

An angiogenesis assay was performed to investigate the effect of treatment with Aidi on angiogenesis in vitro. HUVECs resemble the principal constituent of tumor blood vessels; it was identified that treatment with Aidi may reduce the formation of a network of capillary-like tubes in HUVECs (Fig. 3A). To better understand the mechanism of angiogenesis, VM formation in ESCC cells was examined. As EC9706 cells are highly differentiated and less malignant, they may not form capillary-like tubes well, while KYSE70 cells have higher malignancy, and were therefore selected. The results showed treatment with Aidi downregulated VM formation in KYSE70 cells (Fig. 3A). Western blot analysis demonstrated that treatment with Aidi decreased the expression of VEGF-A in EC9706 and KYSE70 cells (Fig. 3B). These data indicated that treatment with Aidi may inhibit angiogenesis.

EMT signaling contributes to tumor metastasis. To investigate whether the inhibition of EMT was associated with the effects of treatment with Aidi, EMT signaling in EC9706 and KYSE70 cells was examined. The expression levels of cadherin-1, associated with epithelial morphology, were increased, while the expression of cadherin-2 and vimentin, associated with mesenchymal morphology, were decreased in a dose-dependent manner following treatment with Aidi (Fig. 4).

To evaluate the effects of Aidi injection on the metastasis of ESCC, a mouse abdominal tumor model was established. The weight of nude mice in the treatment groups was compared with the negative control group, which were injected with saline; there was no significant alteration (Fig. 5A), indicating that no toxicity had occurred in any of the treated groups. Aidi injection inhibited peritoneal dissemination, as the number of macroscopic tumor nodules in the Aidi injection groups was fewer compared with that in the control group (Fig. 5B). These data suggested that Aidi injection may inhibit ESCC metastasis.

To better understand the mechanism of Aidi injection on the suppression of metastasis in vivo, the expression levels of cadherin-1, vimentin and VEGF-A were examined in the tumor tissue of the nude mice by immunohistochemical analysis; cadherin-2 expression was not analyzed in the present study as the respective antibody could not be applied for immunohistochemical analysis. The results revealed that Aidi injection increased the expression of cadherin-1 and decreased the expression of vimentin and VEGF-A in the tumor tissue (Fig. 5C). The data collectively indicated that Aidi may suppress metastasis by inhibiting EMT and angiogenesis.