A total of 11 infantile hemangioma patients were recruited in the present study from June 2012 to May 2013 at the Second Affiliated Hospital of Xi’an Jiaotong University. The patients were orally treated with propranolol (Tianjin Lisheng Pharmaceutical Co., Ltd., Tianjing, China). The patients had received no surgery or drug treatments prior to the propranolol treatment. Before and after the propranolol treatment, venous blood, urine and tumor tissues (d=2 mm) were collected with the informed consent of the family members and the approval of the Ethics Committee of the Second Affiliated Hospital of Xi’an Jiaotong University.

Six-week-old female BAlB/c nude mice (20–30 g) were provided by the Experimental Animal Centre of the College of Medicine of Xi’an Jiaotong University. The mice were housed in a standard pathogen-free room according to the standard feeding protocols and the animal experimental procedures were handled in accordance with the Institutional Animal Care and Use Committee of Xi’an Jiaotong University. The hemangioma-derived endothelial cell line from proliferating infantile hemangioma tissues was previously prepared and established in our laboratory (28). The cell line was cultured in a RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10 ng/ml epidermal growth factor and 15% fetal bovine serum (FBS) plus 100 U/ml penicillin and 100 μg/ml streptomycin in a humidified atmosphere chamber containing 5% CO₂ at 37°C.

The collected blood and urine samples were centrifuged at 1,000 x g for 15 min. The supernatants were collected and the concentration of HIF-1α was measured using an ELISA reagent kit (R&D Systems, Minneapolis, MN, USA) as per the supplier’s instructions and analyzed by an ELISA reader (BioTek Instruments Inc., Winooski, VT, USA).

For propranolol treatment, hemangioma endothelial cells were treated with various concentrations of propranolol (0, 10, 50 and 100 μM) and incubated for 24, 48 and 72 h. For gene overexpression or gene knockdown, recombinant lentiviral vectors (Shanghai GenePharma Co., ltd., Shanghai, China) or specific siRNA (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were transfected with the cells and incubated for 48 h prior to being collected for analysis.

Total proteins were extracted from tumor tissues or cells and quantified using a BCA protein assay kit (Thermo Fisher Scientific, Rockford, Il, USA). Approximately 25 μg protein was run on a precast 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by electro-blotting onto a nitrocellulose membrane (Amersham, Little Chalfont, UK). The membranes were blocked by blocking buffer (3% skimmed milk solution) at 37°C for 1 h. The membranes were then blotted with primary antibodies diluted in blocking buffer overnight at 4°C. After being washed three times with Tris-buffered saline (TBS) and Tween (TBST) (each for 5 min), the membranes were incubated with horseradish peroxidase conjugated secondary antibody (Wuhan Boster Biological Technology, Ltd., Wuhan, China) in the blocking buffer for 1 h. The membranes were then washed three times with TBST and once with TBS, and the blots were developed with an enhanced chemiluminescence (ECL) detection system (Amersham). The following primary antibodies were used: anti-HIF-1α, anti-VEGF, anti-STAT3, anti-p-STAT3, anti-β1-AR, anti-β2-AR, anti-Bcl-2 and anti-GAPDH (Santa Cruz Biotechnology).

Cell growth was detected by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Briefly, the cells were grown in 96-well plates (1×10⁴ cells/well) and cultured until they reached 80% confluence. Thereafter, the cells were treated with HIF-1α, siRNA or lV-HIF-1α with 50 μM propranolol and incubated for 48 h. MTT diluted in PBS (5 mg/ml) was added at 20 μl/well and continually incubated for 4 h. Dimethylsulfoxide (150 μl/well) was added to dissolve the formazan crystals. The optical density value was measured using an ELISA reader at 490 nm.

A xenograft of hemangioma cells in the mice was prepared as previously described (28). Briefly, the cells (4×10¹⁰) that were diluted in 200 μl PBS were injected subcutaneously into the right groin. Two days after the tumor cell implantation, recombinant lentivirus (5×10¹⁰ plaque-forming units) in 200 μl PBS was injected subcutaneously into the left groin and this was performed every two weeks. The nude mice were intragastrically administered with propranolol solution (0.25 mg/ml) at 0.2 ml/10 g body weight every two days. After 40 days, the mice were euthanized by subcutaneous injection with sodium pentobarbital (40 mg/kg) and tumor tissues were harvested for analysis.

Cell apoptosis was detected by a TUNEL staining assay using a one-step TUNEL apoptosis assay kit (Beyotime, Haimen, China). Briefly, tumor tissues were dissected, fixed with 4% paraformaldehyde for 24 h at 4°C, gradient-dehydrated, embedded in paraffin and then serially cut into 5 μm-thick sections. The sections were dewaxed, gradient-rehydrated and digested with 20 μM proteinase for 10 min. After being washed with PBS, the apoptotic cells were examined using a TUNEl apoptosis assay kit according to the supplier’s instructions. The sections were observed under a fluorescence microscope (Olympus, Tokyo, Japan) and the TUNEl staining cells were calculated in five random fields and averaged per field.

The data were expressed as mean ± standard deviation (SD). Statistical analyses were performed using one-way ANOVA followed by the Bonferroni post hoc test among multiple groups. A value of P<0.05 was regarded as statistically significant.

To explore the molecular basis of propranolol in the treatment of infantile hemangioma, we evaluated the effect of propranolol on the levels of HIF-1α in infantile hemangioma patients. By using ELISA detection methods, we found that the concentrations of HIF-1α in the serum and urine were significantly increased in infantile hemangioma patients compared with those from healthy subjects. Surprisingly, the serum and urine levels of HIF-1α were significantly decreased after treatment with propranolol (Fig. 1A and B). To further validate that propranolol regulated the expression of HIF-1α, we detected the expression of HIF-1α in the hemangioma tumor tissues by western blot analysis. Similar results were obtained showing that propranolol treatment significantly decreased the protein levels of HIF-1α, which had been upregulated in infantile hemangioma tissues without treatment (Fig. 1C and D). Collectively, the results suggested that propranolol inhibited HIF-1α in infantile hemangioma patients.

To further verify the regulated effect of propranolol on HIF-1α expression, we used the hemangioma-derived endothelial cell line to investigate the effect of propranolol on HIF-1α expression in vitro. Hemangioma endothelial cells were treated with varying concentrations of propranolol (0, 10, 50, 100 and 200 μM) for 48 h and the HIF-1α protein expression was determined by western blot analysis. The results showed that propranolol administration decreased the HIF-1α expression in a dose-dependent manner (Fig. 2A and B). Next, we treated the cells with 100 μM propranolol and incubated them for 24, 48 and 72 h (Fig. 2C and D). The results showed that HIF-1α expression was increased at 24 h and continuously increased up to 72 h. In conclusion, these results further indicated that propranolol regulated the expression of HIF-1α in the infantile hemangioma.

Propranolol is a non-selective antagonist for β1-AR and β2-AR (21). To determine which β-AR played an important role in mediating the inhibitory effect of propranolol on the expression of HIF-1α, we added the β1-AR agonist (dobutamine) or the β2-AR agonist (salbutamol) to propranolol-treated cells and examined their effects on the expression of HIF-1α. The results showed that the β1-AR agonist had no apparent effect on the propranolol-induced HIF-1α decrease, whereas the β2-AR agonist significantly reversed the inhibitory effects of propranolol on HIF-1α expression (Fig. 3A and B). Additionally, inhibition of β2-AR by specific β2-AR siRNA had the same effect as propranolol on HIF-1α. However, knockdown of β1-AR had no apparent effect on HIF-1α expression (Fig. 3C and D). Collectively, these results suggested that propranolol repressed the expression of HIF-1α mainly through acting on β2-AR, not β1-AR.

To gain insight into HIF-1α in propranolol-induced cell growth arrest, we examined the effect of HIF-1α overexpression on propranolol-treated cells. The results showed that propranolol treatment inhibited the protein expression of HIF-1α and VEGF, whereas overexpression of HIF-1α blocked the inhibitory effects of propranolol on VEGF expression (Fig. 4A and B). Furthermore, propranolol-induced cell growth arrest was also abrogated by HIF-1α overexpression (Fig. 4C). Similarly, knockdown of HIF-1α resulted in a decreased VEGF expression (Fig. 4D and E) and cell growth inhibition (Fig. 4F), which had the same effect as propranolol treatment. These results implied that propranolol suppressed hemangioma cell growth by regulating HIF-1α.

STAT3, a critical molecule for regulating oncogenic signaling, has been previously reported to be overexpressed in proliferating infantile hemangiomas (29). To investigate whether propranolol had an inhibitory effect on the expression of STAT3, we detected the protein expression of STAT3 in propranolol-treated hemangioma cells by western blot analysis. The results showed that different concentrations of propranolol significantly inhibited the expression of total STAT3 and phosphorylated STAT3 (p-STAT3). Furthermore, the downstream gene Bcl-2, an anti-apoptotic gene, was also markedly decreased by propranolol (Fig. 5). These data indicated that propranolol was capable of inhibiting STAT3 signaling.

To investigate whether HIF-1α was involved in the regulation of STAT3 signaling, we detected the effect of HIF-1α overexpression on STAT3 signaling activation in propranolol-treated cells by western blot analysis (Fig. 6A). The results showed that overexpression of HIF-1α significantly abrogated the inhibitory effects of propranolol on the protein expression of p-STAT3, STAT3 and Bcl-2 (Fig. 6B). Knockdown of HIF-1α by siRNA markedly decreased the protein expression of p-STAT3, STAT3 and Bcl-2, which mimics the effect of propranolol on STAT3 signaling. Collectively, these results suggested that HIF-1α played an important role in propranolol-mediated inhibition of STAT3 signaling in hemangioma cells.

To further validate the important role of HIF-1α involved in propranolol treatment for hemangiomas, we infected a mouse xenograft hemangioma model with lV-HIF-1α overexpressing HIF-1α. Using the TUNEL method, we found that propranolol-induced hemangioma apoptosis was apparently inhibited by HIF-1α overexpression (Fig. 7A). Furthermore, the protein expression of VEGF, p-STAT3, STAT3 and Bcl-2, downregulated by propranolol in tumor tissues, was significantly upregulated by HIF-1α overexpression (Fig. 7B and C). Collectively, these results further confirmed that propranolol repressed hemangiomas in a HIF-1α-dependent manner.