Endostar was provided by Simcere Pharmaceutical Group (Nanjing, China). RPMI-1640 medium and trypsin were provided by Nanjing KeyGen Biotech Co., Ltd. (Nanjing, China). Fetal bovine serum (FBS) was purchased from Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Antibodies against CD31 (cat. no. ab28364), hypoxia-inducible factor-1α (HIF-1α; cat. no. ab2185), MMP-2 (cat. no. ab37150), MMP-9 (cat. no. ab38898), VEGF (cat. no. ab46154), IFN-γ (cat. no. ab198801), IL-17 (cat. no. ab79056) and CD86 (cat. no. ab119857) were provided by Abcam (Cambridge, MA, USA). Mouse ELISA kits for tumor necrosis factor-α (TNF-α; MK1169), IFN-γ (MK1205), IL-17 (MK1445), α-fetoprotein (AFP; MK2689) and cyclic adenosine 5′-phosphate (cAMP; MK2893) were ordered from Wuhan Boster Biological Technology, Ltd. (Wuhan, China). A monoclonal β-actin antibody (sc-47778) was bought from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). The secondary antibody (horseradish peroxidase conjugated-anti-rabbit IgG) was obtained from Invitrogen (A-11034; Thermo Fisher Scientific, Inc.).

The mouse H22 cell line was purchased from the Tumor Institute of the Chinese Academy of Medical Sciences (Shanghai, China). H22 cells were incubated in RPMI-1640 medium containing 10% FBS, 80 U/ml penicillin and 0.08 mg/ml streptomycin. Cells were cultured in an incubator at 5% CO₂ and 37°C. The medium was renewed every other day. These cells were detached by 0.25% trypsin containing 0.01% EDTA and used for seeding. Cells passaged between three and four times were used for subsequent experiments.

A total of 86 male C57B L/6 (C57 black 6) mice at 6–8 weeks (18–22 g) were purchased from the Shanghai Laboratory Animal Center (Shanghai, China). All mice were maintained at the clean standard of laboratory animals. Mice were housed for 7 days to adapt to the environment and maintained at a temperature of 25°C with a relative humidity of 45%. Animals were housed in the Experimental Animal Center of Jiangsu Province (Nanjing, China) on a 12/12 h light/dark cycle with food and water ad libitum. The animal experiment protocol was reviewed and approved by the Institutional Animal Care and Use Committee of the Jiangsu Provincial Academy of Chinese Medicine (Jiangsu, China). Cell suspension containing H22 cells (1×10⁶ cells/ml; 0.2 ml) was injected into liver tissues of C57BL/6 mice. The mice with H22 cells were then randomly divided into two groups, with 40 mice in each group, and administered for 12 consecutive days. The other 6 mice were set as a blank control. The Endostar group was administered Endostar every day at the dose of 0.5 mg/kg. Mice in the model group and blank control group were treated with normal saline at the same amount as Endostar. The model group mice and Endostar-treated mice were sacrificed at days 1, 3, 6, 9 and 12, and HCC tissues were removed immediately.

HCC tissues were fixed in 4% (w/v) paraformaldehyde at room temperature for 24 h. A microscope slide was used to bear cryosections or rehydrated hepar tissue sections (4 µm thick). The sections were dewaxed in xylene and rehydrated through decreasing concentrations of ethanol. The slide was immersed in 0.3% H₂O₂ for 30 sec with agitation by hand. The slide was then dipped into a Coplin jar containing Mayer's hematoxylin and agitated for 30 sec at 50°C. The slide was rinsed in water for 1 min and stained with 1% eosin Y solution for between 10 and 30 sec at 30°C with agitation. Next, two changes of 95% alcohol and two changes of 100% alcohol for 30 sec each were used to dehydrate the sections. The alcohol was extracted by washing twice with xylene. Drops of neutral gum (Bioworld Technology, Inc., St. Louis Park, MN, USA) were then added and covered with a coverslip. Liver tissues were stained with H&E and images were captured using a digital camera.

ELISA was used to determine the levels of AFP, cAMP, TNF-α, IFN-γ and IL-17 in serum or HCC tissues of H22-bearing C57BL/6 mice. Blood was obtained from mice retinal veins under sterile conditions and then centrifuged at 1,000 × g for 10 min at 4°C. HCC tissues were lysed by adding 400 µl lysis buffer containing 20 mM sodium phosphate buffer, pH 7.4, 1% Triton X-100, 150 mM NaCl and 5 mM EDTA (23). The lysate was left on ice for 30 min, vortex-mixed and centrifuged at 14,000 × g at 4°C for 15 min. The supernatant was collected for subsequent use. ELISA was performed according to the manufacturer's protocols. At the end of the reaction, the optical density value of samples was read at 450 nm using a microplate reader (SPECTRAmax 19.0; Molecular Devices, LCC, Sunnyvale, CA, USA).

Western blotting was used to determine the protein expression of CD31, HIF-1α, MMP-2, MMP-9, VEGF, TNF-α, IFN-γ, IL-17 and CD86. Following treatment with Endostar, HCC tissues of the model group mice and Endostar-treated mice were washed with saline. Subsequently, these tissues were lysed with ice-cold radioimmunoprecipitation assay buffer containing 50 mM Tris, 1% Nonidet P-40 and 2 mM EDTA, 11.5 µg/ml aprotinin, 120 mM NaCl, 11.5 µg/ml leupeptin, 100 mM sodium fluoride, 50 µg/ml phenylmethylsulfonyl fluoride and 0.2 mM sodium orthovanadate for 30 min at 4°C. The concentration of protein was quantified using the bicinchoninic acid method (Pierce; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Subsequently, the same amount of protein (50 µg) from each sample was resolved by 12% SDS-PAGE and transferred to polyvinylidene fluoride membranes. Membranes were blocked with 5% bovine serum albumin (BSA; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 2 h at room temperature, washed three times in TBS-Tween-20, and incubated with primary antibodies against CD31 (dilution, 1:1,000), HIF-1α (dilution, 1:500), MMP-2 (dilution, 1:500), MMP-9 (dilution, 1:500), VEGF (dilution, 1:500), TNF-α (dilution, 1:1,000), IFN-γ (dilution, 1:1,000), IL-17 (dilution, 1:1,000), CD86 (dilution, 1:1,000) and β-actin (dilution, 1:500) overnight at 4°C. Subsequently, membranes were incubated with the previously described secondary antibody (dilution, 1:5,000) for 1 h at room temperature. Finally, the protein bands were treated with enhanced chemiluminescent reagents (Pierce; Thermo Fisher Scientific, Inc.) for visualization. β-actin was used as a loading control.

RT-qPCR was used to determine the mRNA expression of HIF-1α and VEGF. Total RNA of hepar samples was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). It was then reverse-transcribed with a SuperScript III First-Strand Synthesis System for RT-qPCR following the manufacturer's protocol. The primers were as follows: HIF-1α forward, 5′-AAACCTAAATGTTCTGCCTAC-3′ and reverse, 5′-GGATGTTAATAGCGACAAAGT-3′; VEGF forward, 5′-AGGGAAGAGGAGGAGATGAGA-3′ and reverse, 5′-GGCTGGGTTTGTCGGTGTT-3′; GAPDH forward, 5′-ATGACATCAAGAAGGTGGTG-3′ and reverse, 5′-CATACCAGGAAATGAGCTTG-3′. Primers were selected and constructed using Primer Premier 5 (PREMIER Biosoft, Palo Alto, CA, USA). qPCR was performed with an ABI 7900 sequence detector (Thermo Fisher Scientific, Inc.) using the SYBR Green method and d (N)₆ random hexamers. PCR thermocycling parameters were 95°C for 10 min, and 40 cycles of 95°C for 15 sec and 60°C for 1 min. GAPDH was used as an internal competitive reference standard; the comparative Cq (quantitative cycle) method was used to calculate the relative changes in gene expression using the 2^−ΔΔCq method (24). Each sample was run in triplicate, each experiment was repeated three times.

The detailed operational procedure was performed as described previously (25). Briefly, mice were anesthetized with ether for 60 sec and then sacrificed by cervical dislocation. The tumor-bearing liver tissues were taken for fixing with 4% freshly prepared paraformaldehyde at room temperature in 0.1 M PBS (pH 7.2) for 24 h. The liver tissues were dehydrated by serial gradient concentrations of alcohol and then embedded in paraffin blocks. Sections of 4 µm thickness were cut and mounted on specific slides for immunohistochemistry. Following deparaffinization and dehydration with a graded series of alcohol (100, 95, 80 and 70% for 2 min each) 3% hydrogen peroxide was used to block endogenous peroxidase for 20 min at room temperature. Following washing with PBS (pH 7.2), the sections were treated with 5% BSA (Sigma-Aldrich; Merck KGaA) for 30 min at 37°C to prevent non-specific reactions. Sequentially, these sections were incubated with the primary antibodies, including anti-IL-17 (dilution, 1:500), anti-CD86 (dilution, 1:500) and anti-IFN-γ (dilution, 1:500) at 4°C overnight. Sections were washed with PBS three times, and then incubated with the previously described secondary antibody (dilution, 1:1,000) at 37°C for 2 h. Finally, sections were stained with 3,3′-diaminobenzidine for 1 min at room temperature. The sections were visualized and images were captured under light microscopy at ×200 magnification (IX71) using an Olympus image capturing system (Olympus Corporation, Tokyo, Japan), and the images were then analyzed and quantified using Image-Pro Plus software 6.0 (Media Cybernetics, Inc., Rockville, MD, USA). PBS was used as primary antibody for normalization to the loading control.

Data are expressed as the mean ± standard deviation. Statistical analysis was performed by one-way analysis of variance based on Student's two-tailed unpaired t-test or Dunnett's multiple comparisons test using SPSS software (version 16.0; SPSS, Inc., Chicago, IL, USA). Image-Pro Plus software (version 6.0; Media Cybernetics, Inc.) was used for processing images. P<0.05 was considered to indicate a statistically significant difference.

Following treatment for 5 days, H&E staining of liver tissues was performed to observe liver histological morphology injury. As presented in Fig. 1A, H&E staining revealed that liver tissues of H22 cell-injected mice (HCC mice) possessed typical basic characteristics of tumor cells, with larger volumes and irregular shaped nuclei compared with mice in the blank control group (Fig. 1B). Following treatment with Endostar (0.5 mg/kg/day), the proliferation and infiltration of tumor cells was inhibited (Fig. 1C).

As shown in Table I, the level of serum AFP produced by the majority of HCCs was significantly increased in HCC mice when compared with blank controls (28.19±3.26 ng/ml) (P<0.05). A positive association was observed between AFP level and treatment time; the level of AFP gradually increased with the extension of time from day 3 to 9. Although no significant difference was observed in the AFP level between the Endostar-treated group and the model group on day 12, serum AFP levels were decreased significantly by the treatment of Endostar from days 3 to 9, compared with the model mice (P<0.05).

cAMP has previously been associated with HCC (26). The serum cAMP level was markedly decreased in H22-bearing mice from a peak on day 1 of 0.51±0.07 nmol/ml to a low of 0.07±0.01 nmol/ml on day 12 (P<0.05). Endostar treatment resulted in a significant increase in the cAMP level compared with the model group at the same day until day 12 (Table I). Therefore, it was hypothesized that Endostar may improve the levels of AFP and cAMP in H22-bearing mice following administration at days 3 to 9, meaning that the effective work time continued for 6 days.

In order to screen the optimal normalization window, the effects of Endostar on angiogenesis-associated protein expression in liver tissues were investigated during the different treatment times, including on days 1, 3, 6, 9 and 12. Compared with the blank mice, the expression levels of VEGF, CD31, HIF-1α, MMP-2 and MMP-9 protein in the model group were increased significantly, followed by a gradual increase in protein expression levels with increasing time periods. Notably, expression of these proteins decreased significantly in the Endostar treatment group (P<0.05; Fig. 2A).

In the Endostar-treated group, a significant decrease in VEGF was observed at a later day compared with the earlier day (day 6 compared with day 3, and day 9 compared with day 6; P<0.05). However, the level of VEGF was reversed to increase on day 12 (Fig. 2B). The level of CD31, an indicator of intra-tumoral microvessel density (MVD), was downregulated significantly by Endostar treatment from days 3 to 9, compared with previous days (P<0.05; Fig. 2C), whereas HIF-1α, MMP-2 and MMP-9 were downregulated significantly on day 9 (P<0.05; Fig. 2D-F). Of note, the expression levels of angiogenesis-associated proteins, including VEGF, CD31, MMP-2 and MMP-9, in Endostar-treated group were significantly decreased compared with that in the model group (P<0.05), but no significant difference was observed when compared with the blank control. These results indicated that days 3 to 9 may be the optimal normalization window for Endostar treatment.

In order to confirm the aforementioned results, RT-PCR was used to determine the mRNA levels of HIF-1α and VEGF in tumor-bearing liver tissues. As presented in Fig. 3A, the mRNA level of HIF-1α was increased significantly on the third day in the model group when compared with the blank control group (P<0.05). Following treatment with Endostar, the HIF-1α mRNA level was decreased significantly when compared with the model group (P<0.05). The levels of VEGF mRNA also exhibited a similar trend (Fig. 3B). For HIF-1α and VEGF mRNA levels in the Endostar-treated group, the level of HIF-1α mRNA was decreased significantly on days 3, 6 and 9 (P<0.01), whereas VEGF mRNA levels were decreased on day 9 (P<0.01). These results also suggest that days 3 to 9 may be the optimal vascular normalization window for Endostar.

Immune factors are involved in the angiogenesis of HCC and contribute to the vascular normalization (27). TNF-α, an inflammatory cytokine, has been shown to be associated with HCC (28). The hepatic TNF-α level increased progressively in the model group with increasing time, and increased from low levels of 0.95±0.12 pg/mg at day 1 to 3.51±0.91, 5.36±0.74, 7.18±0.71 and 7.31±0.13 pg/mg at days 3, 6, 9 and 12, respectively (Table I). The TNF-α level was decreased significantly following administration at days 3 to 9 (P<0.05).

In order to explore the effect of Endostar on angiogenesis in HCC mice, the levels of immune factors IFN-γ, IL-17 and CD86 were evaluated in the liver tissues on the sixth day. The levels of IFN-γ and CD86 in the model group were identified to be significantly lower compared with the blank control group (P<0.01). Results of immunohistochemistry (Fig. 4A), ELISA (Fig. 4B) or western blotting (Fig. 4C and D) demonstrated that Endostar was able to enhance the decreased IFN-γ level (P<0.01). Results of Fig. 5 showed that Endostar was able to enhance the decreased CD86 level (P<0.01). For IL-17, a significant decrease was observed in model group, whereas this decrease was enhanced by the treatment of Endostar (Fig. 6). These results indicated that the immunomodulatory factors were involved in the angiogenesis of HCC mice and Endostar was able to promote vascular normalization by regulating these factors.