Human umbilical vein endothelial cells (HUVECs) were purchased from the American Type Culture Collection (ATCC; Rockville, MD, USA) and cultured in RPMI-1640 (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) and 100 µg/ml penicillin/streptomycin (both from Invitrogen).

The isolation and culture of EPCs was performed as previously described (28). Mononuclear cells were isolated from the bone marrow of apolipoprotein E (ApoE) knockout (ApoE-KO or ApoE^−/−) mice by flushing the femurs and tibias of the mice. Briefly, the mice (n=5) were anesthetized by an intraperitoneal injection of 10% chloral hydrate (0.3 ml/100 g) and then sacrificed by cervical dislocation. After disinfecting the area with alcohol and performing muscle incision, the whirlbone of the thigh was clipped and a needle was inserted into the bone tube to flush out the bone marrow using sterilized PBS. The bone marrow was then cultured in selective medium (EGM-2; CC-3162; Lonza, Walkersville, MD, USA) for 14 days. The Institutional Animal Care and Use Committee of Central South University approved all the animal protocols. Following isolation, the cells were plated immediately onto 6-well plates pre-coated with fibronectin (Sigma-Aldrich, St. Louis, MO USA) at a density of 5×10⁶ cells/well and cultured in RPMI-1640 (Gibco Life Technologies, Grand Island, NY, USA) with 10% FBS (Gibco Life Technologies). The cells were incubated at 37°C in 5% CO₂. The culture medium was changed every 3 days. After 4 days in culture, the non-adherent cells were washed away with 0.01 mol/l phosphate-buffered saline (PBS; pH 7.4) and fresh medium was then added. The EPCs after 7 days of culture were used in the experiments. To confim the identity of the EPCs prior to use, western blot analysis was performed to detect EPC-specific surface markers [CD34, CD133 and vascular endothelial growth factor receptor (VEGFR)-2] (data not shown).

Human ox-LDL was obtained from Shanghai Luwen Biotech Inc. (LW-6002; Shanghai, China). LW Human LDL had been purified to homogeneity via ultra-centrifugation (1.019–1.063 g/cc) and had been oxidized using 5 µM Cu₂SO₄ (oxidant) ihn PBS at 37°C for 20 h. The reaction was terminated by the addition of EDTA-Na₂. The concentration of ox-LDL used in the present study was 100 mg/l.

Six-week-old male ApoE-KO mice were obtained from Xiangya Hospital of Central South University, Changsha, China. The mice were randomly divided into 3 groups (n=5) and administered for 7 weeks (via their drinking water) with the treatments: 2 groups were administered distilled water, and the other group was administered 0.8 g/l EGCG (Sigma-Aldrich). As previously described (15), the mice in the EGCG group and those in the distilled water groups were fed a high-fat diet (HFD). All the mice were monitored until sacrifce (by cervical dislocation) at the age of 15 weeks and the tissue samples of the ApoE^−/− mice were then collected. All the animal protocols were approved by the Institutional Animal Care and Use Committee of Central South University.

pCDNA3.1 vectors containing Jagged-1 (pCDNA3.1-Jagged-1) and pRNAT-U6.1/Neo vectors containing Jagged-1 shRNA (pRNAT-U6.1/Neo-Jagged-1-sh) were constructed and transfected into the HUVECs. To confirm the effects of the vectors on the expression of Jagged-1, western blot analysis was performed to measure the protein expression levels of Jagged-1 in the HUVECs. The transfected cells were expanded and harvested for further analysis. Untransfected cells were used as controls (Con) and cells transfected with empty carrier vectors (pCDNA3.1 or pRNAT-U6.1/Neo) served as the negative control (NC).

Total RNA was isolated from the HUVECs or the EPCs using TRIzol reagent (Invitrogen) and then reverse transcribed into cDNA using the RevertAid™ First Stand cDNA Synthesisi kit (Thermo Fisher Scientific, Waltham, MA, USA). The relative expression levels were detected using Real-Time PCR SYBR-Green Reagents (Dongsheng, Xian, China) in accordance with the manufacturer's instructions. Target RNA levels were normalized to β-actin. The primer sequences used in this study are listed in Table I.

The 3-(4,5-dimethylthiazal-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay was used to estimate cell viability, as prevoiusly described (29). Briefly, the cells were plated at a density of 1×10⁴ cells/well in 96-well plates. After being subjected to the specific treatments, the cells were incubated with MTT solution at a final concentration of 0.5 mg/ml for 4 h at 37°C. After the removal of the medium, 150 mM DMSO were added to dissolve the formazan crystals. The absorbance was read at 570 nm using a multi-well scanning microplate reader (Thermo Fisher Scientific). The cells in the control group were considered 100% viable.

Protein lysates were separated by 10% SDS-polyacrylamide gel electrophoresis and then electroblotted onto PVDF membranes. Primary antibodies against inducible nitric oxide synthase (iNOS), endothelial nitric oxide synthase (eNOS), NADPH (p47^phox), NADPH (p67^phox), NADPH (p22^phox), Jagged-1, hairy and enhancer of split (HES)1, HES5 and Math1 were used, with β-actin antibody as an internal control. Densitometric analysis was performed using LabWorks Image Acquisition and Analysis software (UVP, Inc., Upland, CA, USA).

The HUVECs were cultured in 6-well plates and exposed to ox-LDL (50 µg/ml) for 0, 12, 24 and 48 h. The cells were harvested and stained using the Annexin V-FITC Apoptosis Detection kit (Beyotime Biotech, Jiangsu, China) according to the manufacturer's instructions. Briefly, 5×10⁵ cells were suspended in 500 µl 1X binding buffer (10 mM HEPES pH 7.4, 140 mM NaCl, 2.5 mM CaCl₂). The cells were then incubated with Annexin V (1:20) for 5 min followed by incubation with PI (1 mg/ml) for 15 min. The apoptotic rate was evaluated by flow cytometry.

The HUVECs (1×10⁵ cells/ml) were cultured in 96-well flat-bottom plates (0.1 ml/well) for 1–2 days. The cells were then pre-treated with the indicated concentrations of EGCG and incubated with ox-LDL. The wells were incubated at 37°C for 50 min in a 5% CO₂ incubator and washed 3 times with PBS to remove the non-adherent cells.

The root of the aorta was obtained from the ApoE^−/− mice and fixed in 4% paraformaldehyde overnight. Tissue specimens were then cut at 5 µm thickness for subsequent hematoxylin and eosin (H&E) staining or immunohistochemical analysis. The method for H&E staining of the aortic tissues was conducted as previously described (30). Immunohistochemical analysis was performed according to the manufacturer's instructions. The staining results were observed and captured using an AE31 light microscope (Motic, Xiamen, China).

Each experiment was repeated at least 3 times. Data are presented as the means ± SE and analyzed using SPSS 18.0 software (SPSS, Inc., Chicago, IL, USA). Statistical comparisons between the groups were analyzed using a Student's t-test and a two-tailed value of P<0.05 was considered to indicate a statistically significant difference.

It has been previously demonstrated that ox-LDL induces endothelial cell apoptosis (31). As shown in Fig. 1A, MTT assay revealed that the incubation of HUVECs with ox-LDL (100 mg/l) enhanced endothelial cell apoptosis in a time-dependent manner. This result was also confirmed by flow cytometry (Fig. 1B and C). The HUVECs that were incubated with ox-LDL for 72 h were used in the subsequent experiments.

To determine the potential effect of Jagged-1 on endothelial cell apoptosis, we transfected pCDNA3.1 vectors containing Jagged-1 (pCDNA3.1-Jagged-1) and transfected pRNAT-U6.1/Neo vectors containing Jagged-1 shRNA (pRNAT-U6.1/Neo-Jagged-1-sh) into the HUVECs in order to induce the overexpression or to silence Jagged-1, respectively. Western blot analysis revealed that the Jagged-1 levels were effectively downregulated following transfection of the cells with Jagged-1 shRNA, and significantly enhanced following transfection with pCDNA3.1-Jagged-1 (Fig. 2A). As shown in Fig. 2B and C, the silencing of Jagged-1 significantly enhanced the apoptosis of the HUVECs compared with the control group, and this effect was reversed by the overexpression of Jagged-1.

It has been previously demonstrated that the activation of NADPH oxidase is associated with ox-LDL-induced endothelial dysfunction (32,33). As active NADPH oxidase is assembled on the membrane, the effects of EGCG on the membrane translocation of p22^phox, p47^phox and p67^phox were examined by RT-qPCR. As shown in Fig. 3, the levels of membrane-bound p22^phox and p47^phox were markedly decreased in the cells treated with ox-LDL for 72 h compared with the untreated cells (Fig 3). Of note, this decreasing effect on p47^phox was enhanced by treatment with EGCG in a dose-dependent manner. In addition, treatment with EGCG enhanced p22^phox expression, with a significant increase in expression being observed following treatment with 50 µM EGCG (P<0.01). As regards p67^phox, the expression levels were slightly increased by ox-LDL and then decreased following treatment with 12.5 µM EGCG. However, treatment with 50 and 200 µM EGCG increased the levels of p67^phox even further than ox-LDL did. The most significant effect on p67^phox expression was observed following treatment with 200 µM EGCG (Fig. 3A–C). Moreover, ox-LDL decreased the expression of Jagged-1, and this effect was attenuated by treatment with 50 µM EGCG (Fig. 3D). Western blot analysis revealed that treatment with 12.5 µM EGCG further decreased Jagged-1 protein expression. However, treatment with 50 µM EGCG attenuated this effect (Fig. 4).

Notch signaling has been proven to be critical for arterial specification, sprouting angiogenesis and vessel maturation (24–27). Thus, in this study, we examined the effects of Notch signaling on the EGCG-mediated protection against ox-LDL-induced endothelial cell dysfunction. As shown in Fig. 4, the expression level of HES1 was significantly decreased following the exposure of the cells to ox-LDL. However, EGCG had no effect on HES1 expression in the ox-LDL + EGCG (12.5, 50 and 200 µM) groups compared with the ox-LDL group. The expression level of HES5 was increased following exposure of the cells to ox-LDL, and this effect was attenuated by treatment with 50 µM EGCG. EGCG (12.5 and 200 µM) did not affect HES5 expression in the ox-LDL + EGCG (12.5 and 200 µM) groups compared with the ox-LDL group. Exposure of the cells to ox-LDL did not affect Math1 expression; however, treatment with EGCG decreased Math1 expression in a dose-dependent manner (Fig. 4).

It has been previously demonstrated that ox-LDL reduces the expression of eNOS, thereby altering endothelial biology (34). In the present study, we examined the effects of EGCG on the protein expression levels of eNOS and iNOS in endothelial cells exposed to ox-LDL. Our results revealed that ox-LDL significantly reduced eNOS protein expression and increased iNOS protein expression compared with the control group. This effect was attenuated significantly in the cells treated with 50 µM EGCG. Treatment with 50 µM EGCG signficantly increased eNOS protein expression compared with the cells in the ox-LDL group. Treatment with EGCG at 12.5 µM promoted ox-LDL-induced endothelial dysfunction, whereas EGCG at 50 µM protected the cells against ox-LDL-induced endothelial cell dysfunction through the Notch signaling pathway. Thus, the concentration of 50 µM EGCG was used in the subsequent experiments.

The overexpression and silencing of Jagged-1 was induced in order to determine the role of Jagged-1 in the EGCG-mediated protection against ox-LDL-induced endothelial dysfunction. EPCs were obtained from ApoE^−/− mice and treated with EGCG (50 µM) for 24 h followed by exposure to 100 mg/l ox-LDL for 72 h. As shown in Fig. 5, the overexpression of Jagged-1 markedly decreased the expression levels of HES5, iNOS and Math1, and increased the expression levels of eNOS in the Jagged-1 + ox-LDL + EGCG group compared with the ox-LDL + EGCG group. By contrast, the silencing of Jagged-1 markedly increased the expression levels of HES5, iNOS and Math1, and decreased the expression levels of eNOS in the Jagged-1-sh + ox-LDL + EGCG group compared with the ox-LDL + EGCG group.

Furthermore, we examined the effects of Jagged-1 on apoptosis, as well as on the adhesiveness of EPCs. As shown in Fig. 6A and B, MTT assay and flow cytometric analysis revealed that EGCG suppressed ox-LDL-induced cell apoptosis. This effect was enhanced by the overexpression of Jagged-1, but inhibited by the silencing of Jagged-1. In addition, EGCG attenuated the decrease in cell adhesion induced by ox-LDL, and this effect was inhibited by the silencing of Jagged-1 (Fig. 6C). Thus, our results indicate that Jagged-1 is the key effector protein through which EGCG exerts its protective effects against ox-LDL-induced endothelial dysfunction.

To directly determine the protective effects of EGCG against the development of atherosclerosis, the characteristics of arterial lesions were examined by pathological section H&E staining using light microscopy (Fig. 7). In the control group, the vessel walls were round with even thicknesses. The endothelial cell core was stained and evenly arranged. In the HFD group, the vessel walls were uneven, and significant intimal hyperplasia was present. The inner elastic plates were broken. Treatment with EGCG resulted in more even blood vessels and smoother intima. Histomorphological analysis revealed that EGCG attenuated the HFD-induced accumulation of atherosclerotic plaque. Furthermore, we found that the expression of Jagged-1 and HES5 was upregulated in the HFD group (shown by increased dark brown staining), and this effect was attenuated in the HFD + EGCG group. These results indicated that impairment of the vascular endothelium induced the activation of the Jagged-1/Notch pathway, which was associated with significant intimal hyperplasia.