Formalin-fixed and paraffin-embedded specimens of vascular anomalies for IHC were obtained randomly from the Department of Pathology of Qilu Hospital (Shandong University, Jinan, China) between January 2000 and December 2013. Specimens were from 20 patients; 8 males and 12 females, aged 4 months-39 years (mean, 5 years). Patients were included if they had undergone no other treatments, such as drug therapy, laser therapy, and cryotherapy, in the past but were excluded if they presented with infection, systemic diseases, had undergone sclerotherpy or had a history of drug use. Specimens included lymphatic malformations (n=10), proliferating hemangiomas (n=8), and involuting hemangiomas (n=2). The diagnosis of these anomalies was confirmed by patient medical histories, physical examinations, magnetic resonance imaging (MRI) and final pathological examinations. The present study was approved by Qilu Hospital Ethics Committee of Shandong University and all the patients provided their written informed consent.

Paraffin- embedded 4-µm serial tissue sections were obtained and IHC was performed to assess the expression of PDE-5 in different types of vascular anomalies using an ABC kit (Gene Tech Co., Ltd., Shanghai, China). Lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1), cluster of differentiation 34 (CD34) and glucose transporter-1 (GLUT-1) were used as markers for lymphatic endothelial cells, vascular endothelial cells and hemangioma tissues, respectively, to confirm the type of vascular anomalies. Subsequently, the expression of PDE-5 was assessed in the tissues of all the vascular anomalies.

Tissue sections were deparaffinized in xylene (Guangcheng Chemical Reagent Co., Ltd., Tianjin, China) and rehydrated in graded alcohol, followed by three rinses (each for 5 min) in phosphate-buffered saline (PBS). Endogenous peroxides were quenched by incubating the tissue sections with 10% (v/v) methanol and 3% (v/v) hydrogen peroxide for 15 min at room temperature. A citrate buffer solution (10 mM; pH 6.0; Beijing Solarbio Life Sciences Co., Ltd., Beijing, China) was used to unmask antigens at 100°C for 15 min and was slowly cooled down to room temperature. Following blocking with 10% normal goat serum (Beijing Zhongshan Jinqiao Biological Technology Co., Ltd., Beijing, China) for 1 h at 37°C, the slices were incubated with goat anti-rabbit LYVE-1 (ab36993; 1:100; Beijing Zhongshan Jinqiao Biological Technology Co., Ltd.), anti-rabbit CD34 polyclonal IgG (sc-9095; 1:100; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), rabbit anti-human GLUT-1 polyclonal antibody (GR-004G1; 1:200; Sun Bio Technology Co., Ltd., Shanghai, China); and anti-rabbit PDE-5 polyclonal IgG (sc-32884; 1:100; Santa Cruz Biotechnology, Inc.), which were diluted in PBS overnight at 4°C. Following several rinses with PBS, tissue sections were incubated for 1 h at room temperature with anti-mouse and anti-rabbit horseradish peroxidase-conjugated secondary antibodies (GK500705; undiluted; Gene Tech Co., Ltd., Shanghai, China). Following three rinses in PBS, tissue sections were developed by incubation with 3,3′-diaminobenzidine tetrahydrochloride (Gene Tech Co., Ltd.) for 10 min at room temperature. Following counterstaining with hematoxylin (Beijing Dingguo Changsheng Biotechnology Co., Ltd., Beijing, China) to stain the nuclei, the tissue sections were dry mounted and cover slips were treated with neutral balsam (Yi Yang Instrument Co., Ltd., Shanghai, China).

A specimen of proliferating hemangioma was obtained from a 3-month-old female by surgical resection and was confirmed by pathological examinations in the Department of Oral & Maxillofacial Surgery of Qilu Hospital, Shandong University. Tissue samples were rinsed with PBS and cut into pieces 1cmx1cmx1 cm large. The majority of the pieces were used for culturing HemECs and the residue was maintained to estimate the expression of GLUT-1, which served as the molecular marker for hemangioma.

HemECs were isolated from adherent tissue and the expression levels of CD34, von Willebrand factor (vWF) and PDE-5 were determined by immunocytochemistry. Cells (5.0×10³ cells/well) were seeded in gelatin-coated 24-well plates in endothelial basal medium (EBM-2; Lonza Inc., Allendale, NJ, USA), supplemented with 20% heat-inactivated fetal bovine serum (Gibco; Thermo Fisher Scientific Inc., Waltham, MA, USA), SingleQuot (Lonza Inc.), penicillin (100 U/ml) and streptomycin (100 µg/ml; both Gibco; Thermo Fisher Scientific Inc.). Isolated cells were cultivated in a humidified and sterilized incubator, in an atmosphere containing 5% CO₂ at 37°C. Fresh medium was added following removal and washing of the initial medium was performed every 2–3 days. All experiments were performed using endothelial cells between generations three and five.

An MTT assay was used to assess the effects of sildenafil on the viability of HemECs. Sildenafil (Selleck Chemicals; Houston, TX, USA) stock solution at 100 µM was prepared by dissolving sildenafil in dimethyl sulfoxide (DMSO) and was stored at 4°C. HemECs were seeded in 96-well plates (5.0×10³ cells/well) and were cultivated in EBM-2, containing all the supplements mentioned, for 24 h with five replicates per concentration. The medium was subsequently changed to serum-free EBM-2 for another 24 h. The medium was removed and replaced with 200 µl fresh medium, containing increasing concentrations of sildenafil (0, 1, 3, 5, 10 and 15 µM) and cultured for different durations (24, 48 and 72 h). Cells in the control groups (0 µM sildenafil) were cultivated in the medium with 0.1% DMSO (sildenafil-free medium). Following treatment, HemECs were incubated with MTT (20 µl/well) for 4 h at 37°C, and the optical density at 490 nm was measured after the cell contents were dissolved in DMSO. Absorbance was measured using an automatic microplate reader (BioTek Instruments, Inc., Winooski, VT, USA), and a minimal effective concentration and duration was calculated for late apoptosis and the expression of Id-1.

Following treatment with 0 and 5 µM sildenafil for 24 h, HemECs (1×10⁵ clones) were trypsinized, rinsed with cold PBS and blended in binding buffer using the Annexin V-fluorescent isothiocyanate (FITC)/propidium ioidide (PI) Apoptosis Detection kit (Merck Millipore, Darmstadt, Germany). Annexin V-FITC and PI were added to the fixed cells for 20 min at room temperature in the dark. Subsequently, Annexin V-FITC binding buffer was added to the mixture prior to the fluorescence being measured on a FACSort flow cytometer (BD Biosciences, San Jose, CA, USA). Cellular apoptosis was analyzed using FlowJo 7.6.1 software (Tree Star, Ashland, OR, USA).

HemECs were exposed to 0 and 5 µM sildenafil for 24 h. Total RNA in HemECs was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific Inc.) and the Id-1 RNA was reverse transcribed into cDNA using a PrimeScript RT-PCR kit (Takara Biotechnology Co., Ltd., Dalian, China). cDNA was amplified by PCR using specific primers for Id-1 (GeneChem Co., Ltd. -Shanghai, China): Forward, 5′-CATTCTGTTTCAGCCAGTCG-3′ and reverse, 5′-AGCTCCTTGAGGCGTGAGTA-3′; (120-bp fragment). Amplification conditions were as follows: 35 cycles at 95°C for 5 min, 95°C for 30 sec, 57°C for 30 sec, 72°C for 30 sec, and 72°C for 5 min, using SYBR-Green I dye real-time fluorescent quantitative detection (Takara Biotechnology Co., Ltd) on a Mastercycler ep realplex assay system (Eppendorf; Hamburg, Germany). A primer pair for β-actin (Genechem) was used as an internal control: Forward, 5′-GTGGGGCGCCCCAGGCACCA-3′ and reverse, 5′-CTCCTTAATGTCACGCACGATTTC-3′. Differences in the threshold cycles between the target genes and the reference gene (β-actin) were calculated. The value of the relative mRNA quantity for the control group was arbitrarily set to one for normalization and the data were analyzed using the comparative Cq method (2^−ΔΔCq) (9).

HemECs were exposed to 0 and 5 µM sildenafil for 24 h, and were collected for protein extraction using radio immunoprecipitation assay lysis buffer/phenylmethanesulfonyl fluoride (100:1; Beijing Solarbio, Beijing, China). Protein concentration was determined using the Micro BCA Protein Assay kit (Thermo Fisher Scientific, Inc). Protein samples were separated using an SDS-PAGE gel kit (15% separating gel and 5% stacking gel; CoWin Biotech, Beijing, China) and were electrophoretically transferred onto a nitrocellulose membrane (Wuhan Boster Biological Technology Co., Ltd., Wuhan, China). Following blocking with 5% non-fat milk for 1 h at room temperature, the membrane was incubated with the following antibodies: Anti-rabbit Id1 polyclonal IgG (sc-488; 1:400; Santa Cruz Biotechnology, Inc.) and anti-mouse β-actin polyclonal antibody (TA-09; 1:1,000; Beijing Zhongshan Jinqiao Biological Technology Co., Ltd., Beijing, China) in Tris-buffered saline containing Tween-20 [TBST; 10 mM Tris (pH 7.4), 138 mM NaCl and 0.05% Tween-20] at 4°C overnight. Subsequently, the membrane was washed with TBST three times and incubated with the appropriate secondary antibody: Goat anti-rabbit IgG conjugated to horseradish peroxidase (ZB-2301; 1:3,000; Beijing Zhongshan Jinqiao Biological Technology Co., Ltd) or goat anti-mouse IgG (GAM0041; 1:4,000; Lianke Biotech Co., Ltd., Hangzhou, China), for 1 h at room temperature. Immunopositive bands were examined using an enhanced chemiluminescence (ECL) detection system FluorChem Q (Alpha Innotech Corp., San Leandro, CA, USA) and an ECL kit (EMD Millipore, Billerica, MD, USA). Images were captured on X-ray film, according to the instructions of FluorChem Q. A human TCA 8113 tongue squamous cell carcinoma cell line (American Type Culture collection, Manassas, VA, USA) was used as a positive control for the protein expression of Id-1 (10).

The SPSS 20.0 software package (IBM SPSS, Armonk, NY, USA) was used to analyze the data and all the figures were generated by GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, California, USA). Paired data from the two groups were analyzed using a paired Student's t test. Repeated measurement data with two variances, two-way analysis of variance was used and an intergroup t-test was applied. P<0.05 was considered to indicate a statistically significant difference and data are expressed as the mean ± standard error of the mean.

CD34, LYVE-I and GLUT-1 were considered as positive markers for vascular endothelial cells, lymphatic malformation endothelial cells and hemangiomas endothelial cells, respectively (11–13). Expression levels of PDE5 demonstrated that the expression of PDE5 was present in the cytoplasm of endothelial cells of hemangiomas, but not in the lymphatic malformation endothelia (Fig. 1).

Specimen-derived HemECs were identified using vWF and CD34 markers, and positive expression of PDE-5 was also determined (Fig. 2). Viability of HemECs was determined via MTT assay following treatment with sildenafil (0, 1, 3, 5, 10 or 15 µM) for different durations (24, 48 and 72 h). As the concentration increased, the suppressive effects gradually increased in a dose-dependent manner (Fig. 3A). It was demonstrated that 5 µM sildenafil exerted significant suppressive effects on the viability of HemECs following incubation for 24 h, compared with the control group (0 µM sildenafil; t=3.220, **P<0.01; Fig. 3B). However, insignificant suppression of viability was observed in HemECs treated with 10–15 µM sildenafil (P>0.05). Therefore, the present study considered 5 µM and 24 h to be the minimal effective concentration and duration of sildenafil treatment, respectively, for the optimal suppression of HemECs.

Following treatment with 5 µM sildenafil for 24 h, the morphology of HemECs changed, indicated by the destruction of intercellular junctions and increasing particles in the cytoplasm observed under light microscopy (Fig. 3C). Results of the FITC-PI labeled flow cytometric analysis demonstrated a higher apoptosis rate of 10.5% compared with 2.53% in the control group (Fig. 3D).

Following treatment with 5 µM sildenafil for 24 h, the relative mRNA expression levels of Id-1 in the HemECs markedly decreased by 44.2%, according to the 2^−ΔΔCq method (t=9.749; df=4; P=0.0006; Fig. 4A). Additionally, the protein expression levels of Id-1 markedly decreased following treatment with sildenafil, according to the grey scale of obtained immunoblotting bands (Fig. 4B). These results suggest that, the Id-1 gene was downregulated at the mRNA and protein expression levels following treatment with 5 µM sildenafil. Therefore, it was suggested that sildenafil may inhibit the proliferation of HemECs by downregulating the gene expression of Id-1 in HemECs.