Hypoxia-inducible factor (HIF-1α; cat. no. ab82832), caspase-3 (cat. no. ab13847), VEGFR1 (cat. no. ab32152) and β-actin (cat. no. ab8227) were purchased from Abcam (Cambridge, UK). VEGF-A (cat.no. 31274-1) and VEGFR-2 (cat. no. 21079-1) were purchased from Signalway Antibody LLC., (College Park, Maryland, USA). The EOMA cell line was purchased from the American Type Culture Collection, (ATCC, Manassas, VA, USA). Fetal calf serum (FCS) and Dulbecco's Modified Eagle's Medium (DMEM) were purchased from Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA); artesunate was purchased from Guilin Pharmaceutical Co. Ltd., (Guilin, China); MTT was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Matrigel-coated Transwell chambers were purchased from Qiagen GmbH (Hilden, Germany). Apoptosis kits, including propidium iodide (PI) and fluorescein isothiocyanate (FITC)-Annexin V work solution, were purchased from Thermo Fisher Scientific, Inc. PrimeScript™ RT reagent kit with gDNA Eraser and SYBR^® Premix Ex Taq™ II were purchased from Takara Bio, Inc., (Otsu, Japan). TRIzol Reagent was purchased from Invitrogen; Thermo Fisher Scientific, Inc.

Cells were cultured in DMEM supplemented with 10% FCS and incubated at 37°C with 5% CO₂, prior to being used at the logarithmic growth phase.

An MTT assay was used to assess cell proliferation. The EOMA cells (1×10⁵/ml) were seeded onto 96-well plates. Artesunate was diluted in 1 ml 5% sodium bicarbonate solution, shaken for 2–3 min producing 0, 100, 200, 300, 400 and 500 µg/ml concentrations and then 100 µl was added to the EOMA cells in each well. At 0, 24, 48 and 72 h, an equal volume of MTT solution was added to each well and cultured for another 4 h. The incubation conditions in this assay were all 37°C in an atmosphere containing 5% CO₂. MTT-treated cells were fixed with 150 µl dimethyl sulfoxide for 30 min at room temperature and then assayed with an Evolution™ 201/220 UV-Vis spectrophotometer at 490 nm.

After 24 and 48 h, artesunate-treated (300 µg/ml) cells were collected and washed with cold PBS. Stained cells were assessed with a FACSCalibur™ flow cytometer and analyzed using BD CellQuest™ software (version 5.1; BD Biosciences, Franklin Lakes, NJ, USA). The percentage of early apoptotic cells (stained with Annexin V only) and late apoptotic cells (stained with Annexin V and PI) was recorded. Furthermore, Hoechst 33342 staining solution was used to identify the apoptotic cells, and later detected with fluorescence microscopy. Unstained cells were included as the control.

A Transwell assay was performed by using Matrigel-coated Transwell chambers (pore size of 8.0 µm). A total of 1×10⁵ cells were re-suspended in 200 µl serum-free medium and seeded into the upper compartment of the chamber. The lower compartment was loaded with 800 µl DMEM containing 10% FCS. After incubation at 37°C for 36 h, the membranes were fixed with formaldehyde at room temperature for 5 min, and counter-stained with DAPI at room temperature for 30 min in the dark. The staining was examined under a vertical fluorescence microscope. Trans-membrane migrated cells from each sample were counted in three random fields at ×200 magnification. The assay was performed in triplicate.

An RT-qPCR assay was used for the quantitative estimation of the mRNA expression of VEGF-A, VEGFR-1, VEGFR-2 and HIF-1α in EOMA cells, in the artesunate and in the control groups. Total RNA from ~1×10⁷ cells was extracted using TRIzol^® reagent, according to the manufacturer's protocol. Total RNA then underwent RT using the PrimeScript™ RT reagent kit with gDNA Eraser. The expression of VEGF-A, VEGFR-1, VEGFR-2 and HIF-1α mRNA was determined using the CFX96 Real-Time Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using SYBR^® Premix Ex Taq™ II. β-actin was used as the internal reference. Data were analyzed using the 2^−ΔΔCq method (26). Three separate experiments were performed for each clone. Primers used for PCR analysis were as follows: VEGF-A forward, 5′-CAGGCTGCTGTAACGATGAA-3′ and reverse, 5′-TTTCTTGCGCTTTCGTTTTT-3′; VEGFR-1 forward, 5′-GAGGAGGATGAGGGTGTCTATAGGT-3′ and reverse, 5′-GTGATCAGCTCCAGGTTTGACTT-3′; VEGFR-2 forward, 5′-TTCTGGACTCTCCCTGCCTA-3′ and reverse, 5′-AAGGACCATCCCACTGTCTG-3′; HIF-1α forward, 5′-TGAGCTTGCTCATCAGTTGC-3′ and reverse, 5′-CCATCTGTGCCTTCATCTCA-3′; β-actin forward, 5′-AAGATGACCCAGATCATGTTTGAGACC-3′ and reverse, 5′-GCCAGGTCCAGACGCAGGAT-3′.

EOMA cells were treated with 300 µg/ml artesunate for 24, 48 and 72 h, washed with cold PBS, harvested and extracted using lysis buffer. The protein concentration was detected using a NanoDrop Lite spectrophotometer (Thermo Fisher Scientific, Inc.). A total of 100 µg protein were separated using SDS-PAGE (10% gel) and then transferred to polyvinylidene fluoride membranes. To eliminate background disturbance, membranes were blocked by incubation in blocking solution (5% low fat milk and 0.1% Tween 20 in TBS) at room temperature for 1 h. Cells were then incubated with a rabbit monoclonal primary antibody specific for VEGF-A (dilution, 1:500), VEGFR-1 (dilution, 1:1,000), VEGFR-2 (dilution, 1:500), HIF-1α (dilution, 1:1,000), caspase-3 (dilution, 1:500) and β-actin (dilution, 1:2,000) at 4°C overnight, followed by incubation with a secondary antibody for 1 h at room temperature. Following this, enhanced chemiluminescence protein analysis was performed.

To observe efficacy of intra-tumor injection of artesunate for mouse K-MS and later a comparison with pingyangmycin (PYM), a total of 20 five-week-old female BALB/c nude mice (weight, between 18 and 20 g) were obtained from Chongqing Medical University (Chongqing, China). The mice were housed under specific pathogen-free conditions in an animal facility with free access to pelleted regular rodent diet (Experimental Animal Centre of the Children's Hospital of Chongqing Medical University, Chongqing, China) and water ad libitum. The room temperature was between 22–25°C, humidity was between 45–55%, and a 12-h light-dark diurnal cycle (lights on from 7:00 to 19:00) was used. The Experimental Animal Centre of Children's Hospital of Chongqing Medical University approved the experimental protocols. All procedures were carried out according to the Institutional Animal Care and Use Committee Guide in Child Development and Disorders' Laboratories (Chongqing, China; license no. SYXK (YU)2012-0001; sydwzx.cqmu.edu.cn).

An inoculation of 1×10⁷ EOMA cells in 100 µl PBS was injected into the subcutaneous tissues of the right flank of the female nude mice. The female nude mice were randomly divided into the following four groups of five each: Artesunate group, PYM group, PBS group and no treatment group. Tumor-bearing mice were treated with intra-tumoral injections every three days until the tumor was visible to the naked eye. The PBS group (intra-tumoral injection with the same dose of PBS every three days) was used as the negative control group and the no treatment group (tumor growth was not modified) was considered as the blank control group. As PYM is typically used as a treatment modality for hemangioendothelioma, the curative effect of artesunate was compared with that of PYM. The tumors were measured on alternate days using tissue calipers. Tumor volume was determined using the following formula: (Width)²x length ×0.52. Mice were sacrificed by cervical dislocation when difficulty with ambulation and lethargy had set in.

SPSS 13.0 software was used for all data analysis (SPSS, Inc., Chicago, IL, USA). Numerical data are represented as the mean ± standard deviation. Differences between the control and treated groups were determined using the Student's t-test and one-way analysis of variance. Each experiment was performed in triplicate and P<0.05 was considered to indicate a statistically significant difference.

Artesunate treatment significantly inhibited the growth of EOMA cells in a time- and dose-dependent manner (Fig. 1A). The inhibition of cell proliferation increased correspondingly with the increase in drug concentration during the same time. Also, on extension of exposure time, there was an increase in cell proliferation inhibition with the same concentration. However, this inhibitory effect became apparent as ≤50% (IC₅₀) at a concentration of 300 µg/ml artesunate; therefore, this concentration was used throughout the study.

Based on the results of the MTT assay, 300 µg/ml of artesunate was used for determining apoptosis. To elucidate the mechanism by which artesunate exerts its anti-proliferative effect on EOMA cell lines, an Annexin V/PI binding assay was performed. Results demonstrated an increased percentage of early apoptotic cells (Annexin V/PI) after treatment of EOMA cells with 300 µg/ml artesunate for 24 h (24 h, P<0.05 vs. 0 h), and this percentage of early apoptotic cells increased significantly following incubation for 48 h (48 h, P<0.01 vs. 0 h; Fig. 1B and C). The rate of apoptosis for EOMA cells treated with artesunate increased in a time-dependent manner.

To investigate the inhibitory effect of artesunate on the invasion of EOMA cells in vitro. The invasion efficiency of EOMA cells in response to artesunate was tested using a Transwell invasion system. The invasion efficiency was calculated as the number of invaded cells to the control. In the presence of artesunate (300 µg/ml), the number of invaded EOMA cells was significantly decreased compared with that of the control group, which has the same number of EOMA cells in the initial stage of this experiment (Fig. 2; P<0.05).

RT-qPCR was performed to examine the expression of VEGF-A, VEGFR-1, VEGFR-2 and HIF-1α mRNA in EOMA cells treated with artesunate at 300 µg/ml for 24, 48 and 72 h. The results demonstrate that VEGF-A, VEGFR-2 and HIF-1α mRNA expression levels in EOMA cells treated with artesunate were significantly decreased in a time-dependent manner (*P<0.05, ^#P<0.01; Fig. 3A). There was no statistically significant difference in the VEGFR-1 mRNA expression level among the treatment groups (Fig. 3A).

Effect of artesunate on VEGF-A, VEGFR-1, VEGFR-2, HIF-1 and, caspase-3 protein level. Western blotting was performed to examine the expression of VEGF-A, VEGFR-1, VEGFR-2, HIF-1α and caspase-3 protein in EOMA cells treated with artesunate at 300 µg/ml for 24, 48 and 72 h. The results demonstrated that HIF-1α, VEGF-A and VEGFR-2 protein expression levels in EOMA cells were significantly decreased, but caspase 3 protein expression levels were significantly increased (*P<0.05, ^#P<0.01; Fig. 3B). There were no statistically significant differences in VEGFR-1 protein expression levels among the treatment groups (Fig. 3C).

The outcome of artesunate expression on tumor growth in mice was examined by establishing a subcutaneous K-MS transplantation model. The tumor size was measured every three days (Table I and Fig. 4). The results demonstrated that the volume of the tumor in the artesunate-treated group had significantly reduced, and the effect was approximately equal to that recorded for PYM treatment. Artesunate and PYM can significantly reduce Kasabach-Merritt tumor growth compared with the no treatment and PBS groups (P<0.05). There were no statistically significant differences in tumor size between the artesunate and PYM groups, which was the same between the no treatment and PBS groups (P>0.05).