Cilostazol, acetylcholine (Ach), sodium nitroprusside (SNP), angII and U46619 were gifts from Dr. Zhiqiang Yan (Department of Neurosurgery, Urumqi General Hospital of Lanzhou Military Command, Urumqi, China). LY294002 and 4′,6-diamidino-2-phenylindole (DAPI) were gifts from Dr. Wei Zhang (Department of Cardiology, Tangdu Hospital, Fourth Military Medical University, Xi'an, China). The HUVEC cell line was a gift from Mr. Xiaofei Zhu (Department of Neurosurgery, Tangdu Hospital, Fourth Military Medical University).

Male SD rats (weight, 200–220 g; age, 10–12 weeks) were purchased from the Animal Center of The Fourth Military Medical University (FMMU). They were maintained in a temperature-controlled room (24°C), on a 12 h light/dark cycle and given free access to food water. The experimental protocol was approved by the Institutional Care and Use Committee of the FMMU, which conforms to the Guidelines for the Care and Use of Laboratory Animals of the US National Institutes of Health (NIH publication no. 85–23, revised 1996) (31). Rats were divided into four groups (n=10): (i) Saline-treated group, rats treated with saline by intragastric administration (IA); (ii) Saline + Cilo-treated group, rats treated with saline and cilostazol (30 mg/kg/day) by IA; (iii) angII-treated group, rats continuously infused with angII (1,000 ng/kg/min) by subcutaneously implanted Alzet osmotic pumps (2004 model; Cupertino, CA, USA) as previously described (32); and (iv) angII (1,000 ng/kg/min) + Cilo-treated group, rats treated with cilostazol (30 mg/kg/day) in addition to angII (1,000 ng/kg/min) infusion.

After 4 weeks of angII infusion and cilostazol administration, rats were anesthetized with an intraperitoneal injection of sodium pentobarbital (45 mg/kg; Shanghai Rongbai Biological Technology Co., Ltd., Shanghai, China). Hemodynamic measurements were obtained as previously described (33). In brief, the right carotid artery of anesthetized rats was cannulated with a catheter connected to a microtip pressure transducer (Chengdu Instrument Factory, Chengdu, China) and the transducer was connected to a recording system (Rm6280C, Biological Instruments). The systolic and diastolic blood pressure (sBP and dBP, respectively) were then measured. Finally, the rats were sacrificed by cervical vertebral dislocation. The abdominal aortae were removed and divided into two sections by transverse section. The first was incubated in 10% buffered formalin (Sangon Biotech Co., Ltd., Shanghai, China) for 24 h and then paraffin-embedded (Sangon Biotech Co., Ltd.) for further assessment of apoptosis. The second was used for the detection of endothelium-dependent and -independent vasorelaxation, and assessment of superoxide anion and NO production.

Abdominal aortae were carefully dissected and mounted as ring preparations of ~3 mm on hooks in individual organ baths (Radnoti Glass Technology, Monrovia, CA, USA). Arterial integrity was assessed first by stimulation of vessels with 120 mM KCl, and then washed three times in physiological salt solution (130 mM NaCl, 14.9 mM NaHCO₃, 4.7 mM KCl, 1.18 mM KH₂PO₄, 1.18 mM MgSO₄7H₂O, 1.56 mM CaCl₂ H₂O, 0.026 mM EDTA and 5.5 mM glucose). The temperature was maintained at 37°C and 95% O₂ and 5% CO₂ was pumped into the physiological salt solution. Following washing and stabilization, by contracting the segments with phenylephrine (10 µmol/l), followed by relaxation with Ach (10 µmol/l). The contraction response was detected using an organ chamber containing an isometric Mulvany-Halpern myograph (model 610; DMT-USA, Inc., Marietta, GA, USA) and recorded using a PowerLab 8/SP data acquisition system (ADInstruments Ltd., Colorado Springs, CO, USA), as previously described (34–36). Contractile responses of abdominal aortic rings were evoked with 30 nmol/l U46619. At the plateau of contraction, Ach (1×10⁻⁸−1×10⁻⁴ mol/l) or SNP (1×10⁻¹⁰−1×10⁻⁶ mol/l)were progressively added to the organ bath to induce endothelium-dependent or -independent relaxation.

Superoxide anions from the aortae were measured using flow injection chemiluminescence, as previously described (37). The superoxide anion concentration is reported as chemiluminescence intensity (CI) per mg of tissue weight.

The total NO production in aortae was determined by measuring nitrite concentration, as previously described (37). The concentration of nitrite was calculated from a nitrite standard curve.

Tissue samples were prepared from abdominal aortae. After being paraffin-embedded, the abdominal aorta was exposed by a transverse section and cut into 4 µm thick sections. The cell nuclei of the sections were stained with DAPI as previously described (26). The immunofluorescence data were analyzed using an Eclipse Ni-E microscope (Nikon, Tokyo, Japan) and NIS-elements imaging software (Nikon).

HUVECs were provided by Dr. Xiaofei Zhu from the Department of Neurosurgery, Tangdu Hospital, Fourth Military Medical University (Xi'an, China). HUVEC monolayers were grown as previously described (2,20). In brief, cells were plated into dishes with Gibco Dulbecco's Modified Eagle's medium (Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 1% penicillin-streptomycin (Beyotime Institute of Biotechnology, Shanghai, China) and 10% fetal bovine serum (Beyotime Institute of Biotechnology), under 5% CO₂ at 37°C, at a density of 1×10⁵ cells/ml. Cells were supplemented with 5 U/ml heparin (Sangon Biotech Co., Ltd.) and 100 ng/ml endothelial cell growth substance (Collaborative Research Inc., Bedford, MA, USA). When the cells reached confluence (90%), subcultures were prepared. Cells in actively growing conditions between the third and fifth passages were used for further experiments.

HUVECs were divided into four parallel groups: Untreated cells (control group); cells treated with 10 µmol/l angII (angII-treated group); cells pretreated with 10 µmol/l cilostazol prior to incubation with angII (angII + Cilo-treated group); and cells pretreated with a combination of cilostazol and LY294002 (a PI3K inhibitor) prior to incubation with angII (angII + Cilo + LY-treated group). The concentrations of cilostazol and angII were selected on the basis of previous studies (20,24).

Cultured HUVECs were fixed with 10% buffered formalin and permeabilized with 0.5% Triton X-100 (Sangon Biotech Co., Ltd.). The cell nuclei were stained with DAPI as previously described (33).

Cultured HUVECs in lysis buffer containing 50 mmol/l Tris-hydrochloride, 150 mmol/l sodium chloride, 1% Nonidet P-40, 0.25% superoxide dismutase, 1 mmol/l EDTA, 1 mmol/l NaF, 1 mmol/l Na₃VO₃, 1 mM phenylmethylsulfonyl fluoride and a proteinase inhibitor cocktail tablet (Roche Diagnostics, Basel, Switzerland). Protein samples were assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, as described previously (33). In brief, total protein concentration of each sample was determined prior to polyacrylamide gel electrophoresis, followed by the transfer of proteins to polyvinylidene fluoride (PVDF) membranes (Sangon Biotech Co., Ltd.). PVDF membranes were incubated with blocking buffer (LI-COR Biosciences, Lincoln, NE, USA) for 1 h at room temperature and subsequently immunoblotted with either of the following primary antibodies: Polyclonal rabbit total Akt (1:1,000; cat. no. SAB4500797; Sigma-Aldrich, St. Louis, MO, USA); polyclonal rabbit phosphorylated-Akt (p-Akt) (1:1,000; cat. no. SAB4504017; Sigma-Aldrich); cleaved and total monoclonal rabbit caspase-3 (1:1,000; cat. no. 9665; Cell Signaling Technology, Inc., Danvers, MA, USA); and monoclonal mouse β-actin (1:10,000; cat. no. A5441; Sigma-Aldrich). β-actin protein expression served as a loading control. Following being immunoblotted with primary antibodies overnight at 4°C, the PVDF membranes were washed three times (10 min each) in Tris-buffered saline containing Tween-20 and then incubated with IRDye 680RD goat anti-rabbit IgG (1:5,000; cat. no. 925-68071; LI-COR Biosciences) or IRDye 800CW goat anti-mouse IgM (1:5,000; cat. no. 926-32280; LI-COR Biosciences) for 60 min at room temperature, separately. The PVDF membranes were then washed three times (10 min each) in phosphate-buffered saline with Tween 20. Bands were evaluated by densitometry using an Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA).

The TUNEL assay kit (Beyotime Institute of Biotechnology) was used to detect apoptotic cells according to the manufacturer's instructions. TUNEL-positive cells were detected by microscopy (Eclipse Ni-E; Nikon). The apoptotic index is expressed as the number of positively stained cells per total number of endothelial cells.

Data are presented as the mean ± standard error of the mean. SPSS 13.0 software (SPSS, Inc, Chicago, IL, USA) was used to analyze the data. One-way analysis of variance was used to determine statistically significant differences among the four groups. P<0.05 was considered to indicate a statistically significant difference.

Previous studies have demonstrated that angII infusion at doses between 175 and 1,000 ng/kg/min may result in hypertension in rats (32,38,39). In order to demonstrate the hypertensive effect of angII, a dose of 1,000 ng/kg/min was infused into rats for the period of 4 weeks. Following the infusion, the sBP and dBP levels were measured and demonstrated to be significantly increased compared with the saline-treated group (^*P<0.05; Table I). Cilostazol was not identified to exhibit an effect on angII-infusion-dependent hypertension (Table I).

Evidence from previous studies has suggested that angII induces aberrant oxidative stress in the vascular wall and, therefore, intracellular ROS production, resulting in excessive apoptosis and dysfunction of the epithelium and endothelium (40–42). Whether cilostazol suppresses angII induced endothelial dysfunction remains unknown. Thus, the percentage of Ach-induced vascular relaxation was investigated (Fig. 1). Compared with the saline-treated rats, abdominal aorta rings from the angII-infused rats demonstrated significantly impaired Ach-induced endothelium-dependent relaxation (^*P<0.05; Fig. 1). Cilostazol significantly reduced the impairment in vasorelaxation in the angII +Cilo-treated group compared with the angII-treated group (^#P<0.05; Fig. 1). No significant difference among the four groups of rats was identified upon investigation of SNP-induced endothelium-independent relaxation (Fig. 2).

Furthermore, the apoptosis of endothelial cells was investigated. Compared with the saline-treated group, endothelial apoptosis was significantly increased in the angII-infused rats (^*P<0.05; Fig. 3). Endothelial apoptosis was significantly decreased in angII-infused rats treated with cilostazol compared with the angII-treated group (^#P<0.05; Fig. 3). The cilostazol + saline-treatment had no effect on endothelial apoptosis (Fig. 3) or the Ach-induced vasorelaxation (Fig. 1). These results suggest that cilostazol alleviates the endothelial dysfunction in angII-induced hypertension rats, possibly through an anti-apoptotic effect.

AngII acts via the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase-derived ROS to trigger endothelial cell apoptosis and impair endothelium-dependent relaxation (34). Cilostazol may therefore alleviate endothelial cell apoptosis and attenuate impairment in vasorelaxation by inhibiting superoxide production. Thus, the superoxide levels produced in the aortic tissue from the four groups of treated rats were determined and compared. AngII significantly increased the superoxide production compared with the saline-treated group (^*P<0.05; Fig. 4). However, cilostazol significantly suppressed the superoxide anion production compared with the angII-only treated group (^#P<0.05; Fig. 4).

To detect whether cilostazol has an effect on NO production and thus improves vasorelaxation in response to Ach, total NO production in aortae was determined from the concentration of nitrite, a stable metabolite of NO in vitro (38). AngII significantly increased the total NO production compared with the saline-treated, control group (^*P<0.05; Fig. 5). Cilostazol had no effect on the angII-induced NO production.

AngII treatment (10 µmol/l) resulted in a significant increase in the number of TUNEL-positive (apoptotic) HUVECs compared with the control group (^*P<0.05; Fig. 6). Pretreatment with cilostazol (10 µmol/l) reduced the number of TUNEL-positive HUVECs produced on exposure to angII, compared with the angII-treated group (^#P<0.05; Fig. 6). LY294002, a specific inhibitor of PI3K, was used to detect whether the PI3K/Akt pathway was involved in the effect of the cilostazol treatment. Compared with the angII + Cilo-treated group, HUVECs pretreated with a combination of cilostazol and LY294002 demonstrated increased numbers of TUNEL-positive cells (^ηP<0.05; Fig. 6).

Cilostazol (1, 10 and 100 µmol/l) was administered to HUVECs to investigate the effect concentration. Cilostazol at 1 µmol/l mildly inhibited angII-induced HUVEC apoptosis, compared with concentrations of 10 or 100 µmol/l which significantly attenuated apoptosis, to similar levels, in the angII treated group (^*P<0.05; Fig. 7).

To elucidate the mechanism underlying the cilostazol-dependent reduction of angII-induced apoptosis, the effect of cilostazol on Akt phosphorylation was examined using western blot analysis (Fig. 8A). As demonstrated in Figure 8B, angII significantly reduced the phosphorylation of Akt compared with the control group (^*P<0.05) and cilostazol attenuated the reduction of Akt phosphorylation compared with the angII-treated group (^#P<0.05). Furthermore, the effect of cilostazol on Akt phosphorylation was reduced by combinational treatment with LY294002 compared with the angII + Cilo-treated group (^ηP<0.05; Fig. 8B).

AngII treatment upregulated the cleaved caspase-3 protein expression levels compared with the control group (^*P<0.05; Fig. 8C) and cilostazol treatment suppressed this effect compared with the angII-treated group (^#P<0.05; Fig. 8C). LY294002 attenuated the effect of cilostazol on cleaved caspase-3 protein expression compared with the angII + Cilo-treated group (^ηP<0.05; Fig. 8C).