Human ATC cell lines 8505c and SW1736 were acquired from the Chinese Academy of Sciences Cell Bank (Shanghai, China) and used in the present study. Following thawing and recovery, 8505c cells were cultured in Eagle's minimum essential medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) while SW1736 cells were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.). The two media were supplemented with fetal bovine serum (10%; Gibco; Thermo Fisher Scientific, Inc.) and 1% antibiotic mix (penicillin/streptomycin; Invitrogen; Thermo Fisher Scientific, Inc.). Cells were cultured in an incubator providing a humidified atmosphere composed of 5% CO₂ and 95% fresh air at 37°C.

The cell viability of ATC cells was assessed by MTT assay. Cells were seeded into a 96-well plate at density of 7×10³ cells/well and cultured for 12 h. The medium was replaced with either control saline or emodin solutions (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) at incremental concentrations (0, 10, 15, 20, 25, 30, 35 and 40 µmol/l) for 12 h. Following washing, MTT solutions were added into each well to incubate the cells for 4 h at 37°C. Dimethyl sulfoxide (Sigma-Aldrich; Merck KGaA) was added to dissolve the formazan crystals. A microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used to measure the absorbance values at 540 nm. A total of three replicate readings were recorded and used for further analysis.

The concentration of vascular endothelial growth factor (VEGF) in cell culture supernatants was determined by ELISA using a Quantkine Human VEGF ELISA kit (cat. no. VDE00; R&D Systems, Inc., Minneapolis, MN, USA). Procedures were performed according to the manufacturer's protocol. The recorded absorbance values and plotted standard curves were used to calculate the concentration of VEGF in the cell culture supernatants.

The migratory ability of ATC cells was evaluated via a wound healing assay. 5×10⁵ ATC cells were cultured on a 60-mmdiameter culture dish (Corning Incorporated, Corning, NY, USA) to reach a confluence of 90%. The wound was made using a 2-mm razor blade and the injury line was marked. The peeled-off cells were removed and the remaining cells were cultured and allowed to migrate to heal the wound for 24 h. Following careful washing, cells were fixed and subjected to DAPI fluorescence labeling using a DAPI kit (Beyotime Institute of Biotechnology, Haimen, China), according to the manufacturer's protocol. The wound closure was observed with a fluorescence microscope at ×100 magnification.

The male Balb/c nude mice (6 weeks old; weight, 19–22 g; n=62) used in the present study were acquired from Charles River Laboratories, Inc. (Wilmington, MA, USA). Mice were kept in an artificial environment providing a 12-h light/dark cycle, a 50% humidity and a temperature of 20–24°C in laminar flow cleanrooms. Animals were had free access to sterile water and standard chow. All animal experimental procedures were performed according to the recommended guidelines for the care and use of laboratory animals issued by the Chinese Council on Animal Research. The protocol was reviewed and approved by the Animal Ethics Committee of the College of Medicine, Zhejiang University (Hangzhou, China).

Harvested ATC cells suspended in culture mediums (1×10⁷ cells in 200 µl per mouse for 10 mice) were inoculated into the left flank of each mouse. Mice continued to survive for 28 days. Mice were orally administered with emodin at 100 mg/kg, once per day for 28 days. Immunohistochemical staining of platelet endothelial cell adhesion molecule (CD31) was used to mark the neovessels in tumor tissues in the present study. The tumor tissues were fixed by 10% neutral buffered formaldehyde for 10 h and processed into sections at 5 µm using paraffin-embedding method. Following blocking by QuickBlock blocking buffer (cat. no. P0620; Beyotime Institute of Biotechnology, Shanghai, China) at room temperature for 15 min, the sections were incubated with a CD31 antibody (cat. no. ab32457; 1:1,000; Abcam, Cambridge, UK) at 4°C for 12 h. Following washing, a secondary antibody (anti-rabbit IgG HRP-conjugated; cat. no. ab191688; 1:2,000; Abcam, Cambridge, UK) was used to incubate the slides at room temperature for 20 min. The images were observed using alight microscope at ×200 magnification. Microvessel density (MVD) was evaluated by counting the number of CD31-positive vessels in 10 different fields.

In the lung metastasis experiment, 52 mice were used. A total of 5 control animals received injection of cell culture medium. An ATC cell suspension at a density of 2×10⁷ cells/ml was injected into the tail vein of the mice. Mice continued to survive for 28 days. Mice were orally administered with emodin at 100 mg/kg, once per day for 28 days. Lung tissue samples were obtained and then fixed with 10% formalin at room temperature for 10 h and then embedded in paraffin wax. Tissues were processed into 5 µm tissue sections. H&E staining was applied; the slides were stained with hematoxylin for 5 min and eosin for 30 sec at room temperature. Stained slides were subsequently observed with a light microscope at ×200 magnification. Metastatic rate was calculated by comparing number of mice with metastasis to the number of mice without metastasis.

Total proteins were extracted from tumor tissues and ATC cells using radioimmunoprecipitation assay lysis buffer (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and a protein extraction kit (Beyotime Institute of Biotechnology), according to the manufacturer's protocol. The concentration of proteins was determined using a bicinchoninic acid assay kit (Beyotime Institute of Biotechnology). Vertical electrophoresis was performed on SDS-PAGE gels loaded with protein samples. The gel was 10% and 40 µg protein sample was loaded per lane. Proteins were electro transferred to polyvinylidene fluoride/nitrocellulose membranes. Membranes were blocked by 10% defatted milk (in TBST) at 37°C for 30 min. The PVDF membranes were incubated with primary antibodies against TRAF6 (cat. no. ab33915; 1:2,000; Abcam), HIF-1α (cat. no. 79233; 1:2,000; Cell Signaling Technology, Danvers, MA, USA); VEGF (cat. no. ab53465; 1:2,000; Abcam); MMP9 (cat. no. sc21733; 1:2,000; Santa Cruz Biotechnology, Inc.); basigin CD147 (cat. no. 13287; 1:2,000; Cell Signaling Technology); GAPDH (cat. no. ab9485; 1:4,000; Abcam) at 4°C for 8 h. Membranes were subsequently washed with TBST 3 times. Secondary antibodies conjugated to HRP (cat. nos. 193651 and 99817; 1:2,500; Abcam and cat. no. sc2351; 1:2,500; Santa Cruz Biotechnology, Inc.) were used to incubate the membranes at room temperature for 20 min. The immunoblots were visualized with an Enhanced Chemiluminescence Prime Detection kit (GE Healthcare, Chicago, IL, USA) and the densities of the immunoblots were analyzed by ImageJ software version 1.48 (National Institutes of Health, Bethesda, MD, USA).

Data collected in the present study were presented as the mean ± standard error of the mean. Performed using the software SPSS (version 18.0; SPSS, Inc., Chicago, IL, USA), Student's t-test and one-way ANOVA were used to analyze the significant differences between groups. Post hoc tests were carried out by SNK tests. P<0.05 was considered to indicate a statistically significant difference.

A total of two different human ATC cells, namely 8505c and SW1736 cells, were incubated with emodin solutions at incremental concentrations. An MTT assay was performed to identify the non-cytotoxic concentrations. The results are presented in Fig. 1. Emodin began to inhibit the proliferation of 8505c and SW1736 cells at 30 µmol/l. Therefore, concentrations <30 µmol/l (0, 10, 15, 20 and 25 µmol/l) were chosen for the subsequent experiments to avoid the anti-proliferative effect of emodin on the metastatic and angiogenic abilities of ATC.

In the present study, it was observed that in cultured 8505c and SW1736 cells, the TRAF6/ HIF-1α/VEGF signaling pathway was activated. As a result, the concentration of VEGF in the culture supernatant was significantly elevated. VEGF is thought to be an important effecter in the initiation and promotion of tumor angiogenesis. Following incubation with emodin at non-cytotoxic concentrations, however, the activation of TRAF6/HIF-1α/VEGF was markedly suppressed in a concentration-dependent manner. Subsequently, the VEGF content in the culture supernatant was decreased in a concentration-dependent manner following incubation with emodin. These results are presented in Fig. 2.

As demonstrated in Fig. 3, the TRAF6/CD147/MMP9 signaling pathway was significantly activated in cultured 8505c and SW1736 cells. However, following incubation with emodin at non-cytotoxic concentrations, the activation of TRAF6/CD147/MMP9 signaling was markedly inhibited in a concentration-dependent manner. Additionally, a wound healing assay was performed to evaluate the metastatic ability of ATC cells. Presented in Fig. 3, the results indicated that incubation with emodin markedly impaired the motility of 8505c and SW1736 cells in a concentration-dependent manner.

In the present study, in vivo experiments were performed to reinforce the conclusion derived from the in vitro studies. ATC cells were used to inoculate the nude mice. In harvested ATC tissues, the TRAF6/HIF-1α/VEGF signaling pathway was activated. However, following treatment with emodin, the activation of the TRAF6/HIF-1α/VEGF and TRAF6/CD147/MMP9 signaling pathways in ATC tissues was significantly inhibited. The MVD assay demonstrated that emodin administration suppressed angiogenesis in ATC. In addition, a lung metastasis model was established by tail vein injection of ATC cells. Emodin administration decreased the lung metastatic rate. These results are presented in Fig. 4.