Curcumin was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). A Cell Counting Kit-8 (CCK-8) was purchased from Dojindo Molecular Technologies, Inc. (Kumamoto, Japan). An Annexin V/fluorescein isothiocyanate (FITC) kit was purchased from Vazyme Biotech Co., Ltd. (Nanjing, China). VEGF antibodies (cat no. sc-7269), used for western blotting and immunohistochemistry, were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). PI3K (cat no. ab151549) and AKT (cat no. ab38449) antibodies were purchased from Abcam (Cambridge, USA). The polymer horseradish peroxidase (HRP) detection systems, used for immunohistochemistry and western blotting, were purchased from Zhongshan Golden Bridge Biotechnology Co., Ltd (Beijing, China) and Cellular Signaling Technology (Danvers, MA). Recombinant human VEGF (rhVEGF) was from R&D Systems, Inc. (Minneapolis, MN, USA).

The H22 HCC cell line was purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences, Shanghai Institute of Cell Biology, Chinese Academy of Sciences (Shanghai, China) and cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (Hyclone; GE Healthcare Life Sciences, Logan, UT, USA), 100 U/ml penicillin, and 100 µg/ml streptomycin in a humidified incubator at 37°C with 5% CO₂.

Nude male mice (4–5 weeks old) were purchased from the Model Animal Research Center of Nanjing University (Nanjing, China) and kept under controlled conditions with a 12-h light/dark cycle with access to food and water ad libitum. All animal work was conducted in compliance with guidelines for ethical animal research and approved by the Animal Ethics Committee of Xiamen University (Xiamen, China).

H22 cells were treated with different concentrations of curcumin (0, 5, 10, 20, 40 and 80 µm) for 12, 24, or 48 h. In other experiments, H22 cells were stimulated by VEGF (20 ng/ml) for 24 h in serum-free medium, and curcumin (20, 40, and 80 µm) was added at the same time. Thereafter, a CCK-8 assay was used to determine cell viability, as per the manufacturer's protocol.

H22 cells were treated with different concentrations of curcumin (0, 20, 40, and 80 µm) for 24 h. Apoptosis was quantitatively measured by Annexin V/propidium iodide (PI) double staining which was carried out at room temperature for 2 h. Cell apoptosis was evaluated using FACSVerse™ (Beckman Coulter, Inc., Brea, CA, USA), and data were analyzed using FlowJo V10 (Flowjo LLC, Ashland, OR, USA).

H22 cells (2×10⁵) were subcutaneously injected into nude mice (4–5 weeks old, n=15). Seven days after tumor cell inoculation, the mice were randomly divided into three groups of five and treated with vehicle (0.1% DMSO) or curcumin at a dose of 50, 100 mg/kg daily for two weeks. At day 24, the tumors were removed and weighed.

Tumor tissues from the xenograft model mice were collected and fixed in formalin at 4°C for 48 h, decalcified in 10% ethylenediamine tetraacetic acid, and embedded in paraffin for histopathological analysis. Serial paraffin sections (5 µm) were stained with hematoxylin and eosin at room temperature for 10 min.

Tumor samples were fixed with formalin at 4°C for 48 h, embedded in paraffin, and sectioned into 5 µm sections. Sections were dewaxed and dehydrated in a descending series of ethanol dilutions: 100% ethanol (2 times, 2 min each), 95% ethanol (2 times, 2 min each) and 70% ethanol (1 time, 2 min). Once rehydrated, antigen retrieval was performed using a citrate buffer, and endogenous peroxidase activity was blocked at room temperature for 10 min using 3.0% hydrogen peroxide, and the sections were blocked at room temperature for 30 min using 10% goat serum, then incubated with the primary antibodies directed against VEGF (1:200) at 4°C overnight. Primary antibodies were detected using a horseradish peroxidase-conjugated anti-mouse secondary antibody (1:1,000; cat no. SP-9002; Zhongshan Golden Bridge Biotechnology Co., Ltd) for 30 min at 37°C. Following rinsing with PBS three times, the sections were exposed for 7 min to diaminobenzidine (DAB) + chromogen, then dehydrated in ethanol and xylene and mounted onto glass slides. The staining intensity and average percentage of positive cells were assayed for 10 independent fields using a light microscope at a magnification of ×400.

Tumor tissues were washed with PBS three times, then ground and lysed in RIPA cell lysis buffer (1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS in PBS). Following centrifugation at a speed of 14,000 × g for 5 min at 4°C, the supernatants were kept frozen at −80°C until use. Protein concentrations were measured using the BCA protein quantification Kit (Beyotime, Haimen, China) according to the manufacturer's protocol. A total of 50 µg of denatured protein was separated using 10% SDS-PAGE and transferred onto polyvinylidene fluoride membranes. The membranes were blocked with 10% non-fat milk at room temperature for 1 h and then incubated with primary antibodies for VEGF, PI3K, AKT (1:1,000) and mouse monoclonal anti-β-actin (1:5,000) at 4°C overnight, followed by incubation with appropriate HRP-conjugated anti-mouse secondary antibodies (1:1,000; cat nos. 7074 and 7076; Cell Signaling Technology, Inc., Danvers, MA, USA) for 1 h at room temperature. Protein signals were detected using the BeyoECL plus kit (Beyotime Institute of Biotechnology, Haimen, China) according to the manufacturer's protocol. Finally, the densities of the bands were quantified using Image J software (version 1.38; National Institutes of Health, Bethesda, MD, USA) (15). Equivalent protein loading and transfer efficiency were verified by staining for β-actin.

Total RNA from the tumor tissues was isolated using TRIzol (Takara Bio, Inc., Otsu, Japan), according to the manufacturer's protocol. RETROscript™ Reverse Transcription kit (Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 1 µg RNA were used for cDNA synthesis. SYBR Green One-Step qRT-PCR kit (Thermo Fisher Scientific, Inc.) and a 7900 HT Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.) were used for qPCR. Data were normalized to the internal control GAPDH and presented as an expression fold change using the 2^−ΔΔCq method (16). Thermocycling conditions were as follows: 95°C for 3 min, followed by 40 cycles at 94°C for 30 sec, 60°C for 30 sec and 72°C for 60 sec. The gene-specific primers were selected according to the literature. The primers were synthesized by Sangon Biological Engineering Technology Services Co., Ltd. (Shanghai, China). The primer sequences were as follows: PI3K, forward 5′-CGTTTCTGCTTTGGGACAAC-3′ and reverse 5′-CCTGATGATGGTGGAG-3′; AKT, forward 5′-TGAGAGAAGCCACGCTGTC-3′ and reverse 5′-CGGAGAACAAACTGGATGAA-3′; VEGF, forward 5′-TAGACGTTCCCTGCCAGCAA-3′ and reverse 5′-AGCATCCGAGGAAAACATAAAATCTT-3′; and GAPDH, forward 5′-AACGGATTTGGTCGTATTGGG-3′ and reverse 5′-TCGCTCCTGGAAGATGGTGAT-3′.

Data were expressed as mean ± standard deviation. Statistics were presented in Prisms 6.1 software (GraphPad Software, Inc., La Jolla, CA, USA). All data were analyzed using one-way analysis of variance followed by the Tukey test. P<0.05 was considered to indicate a statistically significant difference.

To investigate the inhibitory effect of curcumin on HCC cells, a CCK-8 assay was performed to detect cell viability in vitro. The results demonstrated that curcumin inhibited H22 cell viability in a dose-dependent manner, and particularly at a very high dose of 80 µM (P<0.01; Fig. 1A). In addition, Annexin V/PI staining was performed to determine the apoptosis rate of H22 cells. Curcumin treatment significantly induced H22 cell apoptosis, especially at concentrations of 40 and 80 µm (P<0.05 and P<0.01 compared with untreated control respectively; Fig. 1B).

To further explore the anticancer activities of curcumin in vivo, a xenograft model was established in mice using H22 cells and treated with 50 or 100 mg/kg of curcumin by intragastric administration for two weeks. The results demonstrated that curcumin treatment resulted in a dramatic inhibition of H22 xenograft tumor growth compared with the vehicle-treated control group, particularly at a very high dose (50 mg/kg, P<0.05; 100 mg/kg, P<0.01; n=5; Fig. 2A). The tumor weight in the curcumin-treated groups were significantly decreased compared with the vehicle-treated control group (50 mg/kg, P<0.05; 100 mg/kg, P<0.01; n=5; Fig. 2B). Furthermore, histopathological analysis indicated that tumor cells in the curcumin-treated groups were visibly sparser compared with those in the vehicle-treated control group (Fig. 2C).

Next, the mechanisms by which curcumin exerts its antitumor effect in vivo were explored. Immunohistochemistry analysis indicated that curcumin treatment impaired the expression of VEGF in the tumor tissues obtained from the tumor-bearing mice in a dose dependent manner (50 mg/kg, P<0.05; 100 mg/kg, P<0.01; Fig. 3). Previous studies have implicated that VEGF expression is mediated via PI3K/AKT signaling pathway. Results from the western blot analysis of the tumor tissues indicated that protein expression levels of PI3K and AKT were inhibited following curcumin treatment (Fig. 3). In addition, the mRNA expression levels of VEGF, PI3K and AKT were also significantly downregulated by curcumin treatment (Fig. 4). These findings suggested that the anticancer activities of curcumin in vivo were associated with the downregulation of VEGF expression in human HCC cells, and most likely mediated through the PI3K/AKT signaling pathway.

Finally, in order to examine whether VEGF mediates the curcumin effect in inhibiting HCC cell viability, recombinant human VEGF (rhVEGF) was used. Compared with different concentrations of curcumin-treated cells, rhVEGF-treated group caused an increase in HCC cell growth (Fig. 5). Curcumin treatment resulted in a significant reduction of VEGF-induced proliferation (P<0.01; Fig. 5).