The primary antibodies used in the present study included: VEGF (Bioworld Technology, Inc., St. Louis Park, MN, USA), phosphorylated AKT (p-AKT), p-mechanistic target of rapamycin (p-mTOR), p-phosphatase and tensin homolog (p-PTEN), p-extracellular signal-regulated kinase (p-ERK1/2), and AKT, mTOR, PTEN, ERK and GAPDH (Cell Signaling Technology, Inc., Danvers, MA, USA). The reagents used in the present study included: PI103 (Selleck Chemicals, Houston, TX, USA), the Human Angiogenesis Array Q1 (RayBiotech, Norcross, GA, USA), sorafenib (Selleck Chemicals) and bufalin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany)

The in vitro experiments were performed with HUVECs from the American Type Culture Collection (Manassas, VA, USA), and SMMC-7721 and PLC/PRF/5 HCC cells, which were purchased from the Cell Bank of Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). The cells were propagated with DMEM containing 10% fetal bovine serum (FBS; HyClone, GE Healthcare Life Sciences, Logan, UT, USA) and 1% penicillin/streptomycin at 37°C in an atmosphere containing 5% CO₂.

Six-week old male Balb/c nude mice were used in the present study. The study was approved by Fudan University Shanghai Cancer Center (Shanghai, China). The mice were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), and were raised under the following pathogen-free conditions: Room temperature, 20°C; relative humidity, ~50%. The mice were given ad libitum access to food and water, and received humane care according to the principles of animal care of Fudan University (Shanghai, China). The mice were maintained under a 12-h light/dark cycle. The principles were as described in the Guide for the Care and Use of Laboratory Animals issued by the National Institutes of Health (Bethesda, MD, USA). The mice were randomly divided into four groups (n=6 mice/group). The mice in the experimental group were intraperitoneally injected with 1 mg/kg bufalin (5 days/week), oral uptake of 30 mg/kg/day sorafenib (5 days/week), or combination treatment with the two drugs. The control mice were treated with the vehicle only. The treatment lasted for 16 days, following which the mice were sacrificed, and the tumors and blood were obtained. All experiments conformed to the ethical principles of animal experimentation stipulated by Fudan University.

A total of 5×10⁶ SMMC-7721 cells in 0.2 ml phosphate-buffered saline (PBS) were injected into the right flank of each mouse to form subcutaneous tumors. When the volume of these subcutaneous tumors reached a size of 100–300 mm³, the mice were treated with intraperitoneal injections of mg/kg bufalin (5 days/week), 30 mg/kg oral sorafenib (5 days/week), or a combination of the two drugs. The control mice were treated with saline. Tumor size was measured every 4 days. The tumor-bearing mice were sacrificed following 16 days of treatments, and tumors and blood were extracted. The tumors were weighed and subjected to immunohistochemistry. All procedures conformed to the ethical principles of animal experimentation as stipulated by Fudan University.

To evaluate the effects of different treatments on cell viability, the HCC cells and HUVECs were plated in 96-well plates at a density of 5,000 cells per well. A total of 20 nM bufalin, 10 μM sorafenib, or a combination of the two were added to the wells and incubated at 37°C for 24, 48 or 72 h. The cells were subjected to a Cell Counting Kit-8 assay (Dojindo Molecular Technologies, Inc., Kumamoto, Japan), used according to the manufacturer’s protocol, and the absorbance was measured at a wavelength of 450 nm with a microplate reader to determine the cell viability rate.

To determine the effects of sorafenib, bufalin, or a combination of the two on cell migration, an assay was performed using Transwell inserts (8-μm pore; Corning Incorporated, Corning, NY, USA). The lower side of the chamber was filled with condition medium (CM) from the HCC cell lines (SMMC-7721 and PLC/PRF/5) following different treatments (control, bufalin, sorafenib, and combination). The HUVEC cells were plated on the upper side of the chamber. Following a 48-h incubation period, the migrated cells were fixed with methanol and stained with 0.1% crystal violet. The numbers of migrated cells were calculated in three different fields of view, at ×20 magnification using an Olympus light microscope (Olympus Corporation, Tokyo, Japan).

The SMMC7721 and PLC/PRF/5 cells were treated with bufalin, sorafenib, or the two in combination. At 48 h post-treatment, the cell supernatants were collected as CM. The collected CM from each group was stored at −80°C if not used immediately and was thawed prior to use.

A 96-well plate was coated with Matrigel and HUVEC cells were seeded at density of 4×10⁴ cells per well in 100 μl CM, with or without 1 or 6 ng/ml VEGF. The HUVECs were then incubated at 37°C for 12, 24, 36 or 48 h. The central region of each well and the tube networks were analyzed using an Olympus light microscope (Olympus Corporation).

To determine the cytokines in SMMC-7721 cells treated with bufalin, sorafenib, or the two in combination, the Human Angiogenesis Array Q1 kit was used to detect angiogenesis-related cytokines in the CM extracted from HCC cells in the different treatment groups, according to the kit protocol. Following this, the VEGF concentration was determined using an enzyme-linked immunosorbent assay (ELISA) kit, according to the manufacturer’s protocol and analyzed using a Labsystems Multiscan reader (Thermo Fisher Scientific, Inc., Waltham, MA, USA).

Western blot analysis was performed as previously described (23). Briefly, proteins were extracted from cells following the indicated treatments using radioimmunoprecipitation assay buffer (R0020; Beijing Solarbio, Science & Technology Co., Ltd., Beijing, China). Protein concentrations were assessed using a bicinchoninic acid kit. Subsequently, protein samples (12 μg) were loaded and separated by 10% SDS-PAGE for the detection of various proteins. Proteins were then transferred to polyvinylidene fluoride membranes, which were blocked in 5% skim milk for 1 h at 37°C. The expression levels of VEGF, AKT, p-AKT, mTOR, pmTOR, PTEN and p-PTEN were detected by incubation with their respective primary antibodies overnight at 4°C, after which the blots were washed three times with PBS. The primary antibodies used in the present study included: VEGF (20301555-1, 1:1,000; Bioworld Technology, Inc.), phosphorylated AKT (p-AKT; 4060, 1:2,000), p-mechanistic target of rapamycin (p-TOR; 5536, 1:1,000), p-phosphatase and tensin homolog (p-PTEN; 9551, 1:1,000), p-extracellular signal-regulated kinase (p-ERK; 4370, 1:2,000), and AKT (4691, 1:1,000), m-TOR (2983, 1:1,000), PTEN (9188, 1:1,000), ERK (9102, 1:1,000) and GAPDH (5174, 1:1,000) (all from Cell Signaling Technology, Inc.). Finally, the blots were incubated with a horseradish peroxidase-conjugated immunoglobulin G secondary antibody (HAF008, 1:5,000; Novus Biologicals, LLC, Littleton, CO, USA) for 1 h at 37°C, after which proteins were visualized by enhanced chemiluminescence (WBKLS0500; Merck KGaA). Semi-quantification of the blots was conducted using ImageJ software (version no. k 1.45; National Institutes of Health).

The mice were treated with the different treatments, as described above. Blood samples were collected from the eyeballs of the mice. The expression of VEGF in the blood was measured using an ELISA kit, according to the manufacturer’s protocol.

Immunohistochemistry protocols were performed as described previously (23). Samples were dehydrated in 70% ethanol, embedded in paraffin and sectioned (4 μm). Briefly, the tumor sections were stained with rabbit anti-VEGF (1:1,000; cat. no. AP0742; Bioworld Technology, Inc.) at 4°C overnight, followed by a goat-anti-rabbit secondary antibody (1:1,000; cat. no. 7054; Cell Signaling Technology, Inc.) for 1 h at 37°C. The expression of VEGF was calculated by multiplying the intensity score by the percentage score using an Olympus light microscope (Olympus Corporation). The staining intensity (0, negative; 1, weak; 2, intense) and the ratio of stained cells (0, ≤10%; 1, 11–25%; 2, 26–50%; 3, 51–75%; 4, >75%) were evaluated by investigators blinded to the treatment status.

The results were analyzed using SPSS 22.0 software for Windows (IBM SPSS, Armonk, NY, USA). The comparisons between two groups were made using Student’s t-test. Multi-group comparisons of the means were made using one-way analysis of variance with post hoc contrasts using the Student-Newman-Keuls test. All experiments were repeated three times. P<0.05 was considered to indicate a statistically significant difference.

The SMMC-7721 and PLC/PRF/5 cells were incubated with 10 μM sorafenib, 20 nM bufalin, or a combination of the two. At 12, 24, 36 and 48 h post-treatment, the CM was collected and incubated with HUVECs. A tube formation assay was performed to observe the effect of the treatments. It was observed that the combination-CM inhibited the blood vessel formation of the HUVECs most markedly, compared with the other groups (Fig. 2A and B). The migration rates of the HUVECs in the different treatments groups were also detected. The migration rate of the HUVECs treated with the combination-CM was reduced compared with that of the HUVECs in the other treatment groups for the SMMC-7721 and PLC/PRF/5 cells (Fig. 2C and D).

As it was previously identified that combination-CM-treated HUVECs exhibited reduced blood vessel formation and inhibit migration, it was hypothesized that the expression of cytokines associated with angiogenesis may be altered in the CM from the combination-treated SMMC-7721 cells. Therefore, the CM from the untreated and combination-CM-treated SMMC-7721 cells was analyzed using an angiogenesis-specific array for cytokine detection. Angiogenin, platelet-derived growth factor (PDGF-BB) and VEGF were altered most of the cytokines assessed, indicative of a potential role in facilitating angiogenesis (Fig. 3A). To further validate the alterations in cytokine expression detected by the human angiogenesis array, the concentration of VEGF was measured in the CM from SMMC-7721 and PLC/PRF/5 cells in the different treatment groups using ELISA (Fig. 3B and C). The results of the western blot analysis confirmed the ELISA results for the SMMC-7721 and PLC/PRF/5 cells (Fig. 3D).

CM from the SMMC7721 and PLC/PRF/5 cells collected at 48 h was added to HUVECs with or without the addition of 1 or 6 ng/ml VEGF, and tube formation was observed. The combination-CM suppressed HUVEC vessel formation. The incubation of HUVECs with VEGF disrupted the inhibitory effect on tubule formation induced by the addition of combination-CM. The higher concentration of VEGF resulted in enhanced vessel formation (Fig. 3E).

To measure the effect of the combination treatment on tumor growth in vivo, a subcutaneous tumor model was established. Mice received 1 mg/kg/day bufalin, 30 mg/kg/day sorafenib, or a combination of the two for 16 days. The tumors were then excised and weighed. The results revealed that the subcutaneous tumors from mice that received the combination treatment were smaller, compared with those in the other groups (Fig. 4A). The tumor weights were significantly attenuated in mice that received the combination treatment, compared with those treated with bufalin or sorafenib alone (Fig. 4B). Further analyses of the expression of VEGF in tumors revealed that the expression of VEGF was lowest in the tumors and blood from mice that received the combination treatment, as determined with immunohistochemistry and ELISA (Fig. 4C and D).

As it was identified that a decrease in the expression of VEGF occurred due to the combined treatment and that this may reduce the extent of angiogenesis in HCC, the pathways associated with VEGF were examined. The proteins upstream of VEGF were measured in HCC cells in the different treatments groups, including the phosphorylated and total levels of AKT, mTOR, PTEN and ERK, with western blot analysis. The HCC cells incubated with either bufalin or the combination treatment demonstrated a marginal decrease in the expression of p-AKT. The expression of p-mTOR was also reduced in the HCC cells incubated with bufalin or the combination treatment. However, no significant alterations were identified in the phosphorylation of PTEN or ERK (Fig. 5A).

As a decrease was observed in the expression of p-AKT and p-mTOR in the combination-treated SMMC-7721 cells, their involvement in the combination-induced reduction of VEGF was examined. PI103 (2 μM) was selected as a specific pharmacological inhibitor of the phosphoinositide 3-kinase (PI3K)/AKT/mTOR pathway. The results demonstrated that pretreatment with PI103 partially reversed the phosphorylation of mTOR (Fig. 5B). It was also observed that the expression of VEGF was significantly downregulated in the HCC cells pretreated with PI103, indicating the potential role of the mTOR pathway in modulating VEGF (Fig. 5C). The present study also evaluated the effect of the CM from cells pretreated with PI103 on HUVEC migration and tube formation. Compared with the CM from the non-pretreated cells, HUVEC migration and tube formation were suppressed by pretreatment with PI103 (Fig. 5D and E). Collectively, these results demonstrated that the mTOR pathway was involved in the reduced expression of VEGF induced by combination treatment.

The present study identified bufalin as asynergistic agent for HCC cells treated with sorafenib by regulating the mTOR/VEGF pathway, leading to the reduced secretion of VEGF. The reduced binding of VEGF to VEGFR on the HUVECs led to reduced tumor angiogenesis. To the best of our knowledge, this is the first evidence of the synergistic anti-hepatoma effect of bufalin combined with sorafenib via the tumor vascular microenvironment by targeting mTOR/VEGF signaling (Fig. 6).