HUVECs (ATCC, USA) were maintained in Dulbecco's modified Eagle's medium (DMEM) (HyClone; GE Healthcare Life Sciences, Logan, UT, USA) that contained 10% FBS (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and penicillin/streptomycin (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) at 37°C in an incubator with 5% CO₂. The normal blood glucose level ranged from 3.9 and 5.5 mM (32) in non-diabetics. Blood sugar levels increased more than normal range may be an indicator of diabetes. According to some researches on hyperglycemia injury model in vitro (33,34), HUVECs were respectively treated with glucose at 5 and 33 mM for the control and injury model for 48 h. The incubation concentration of OA was determined in reference to a previous study (35). The research groups in this study were classified as follow: Blank group (blank): no glucose treatment; control group (Con): 5 mM glucose treatment; High glucose model group (GC): 33 mM glucose (Sangon Biotech Co., Ltd., Shanghai, China) treatment; positive control group (OA-H/Con): 40 mM OA (purity >98%; Beijing Solarbio Science & Technology Co., Ltd.) (dissolved in ethanol) for incubation for 24 h prior to 5 mM glucose treatment; low OA treatment group (OA-L/GC): Cells were treated with 20 mM OA at 37°C for 24 h prior to 33 mM glucose treatment; high OA treatment group (OA-H/GC): Cells were treated with 40 mM OA at 37°C for 24 h prior to 33 mM glucose treatment.

The cell survival rate was examined using Cell Counting Kit-8 (CCK-8; Beyotime Institute of Biotechnology, Haimen, China) assay, according to the manufacturer's protocols. To explain further, 1×10⁴/each well were seeded into a 96-well plate and incubated at 37°C for 24 h. CCK-8 (10 µl/well) was added into the 96-well plate, and then the cells were incubated at 37°C for 4 h. Absorbance at 450 nm was detected using a microplate reader (PerkinElmer, Inc., Waltham, MA, USA).

Intracellular ROS level was detected using fluorescent probe DCFH-DA probe (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). Cells were collected and washed by PBS. Next, the cells were maintained with 10 µM DCHF-DA in darkness at 37°C for 30 min. Flow cytometry analysis was carried out to examine the fluorescence signals corresponding to DCHF-DA on flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) with Cell Quest software version 5.1 (BD Biosciences). At least 10,000 events were analyzed in each examination.

The Rho 123 accumulation was determined by flow cytometry analysis as previously described (36). Following the treatment above, the cells in 24-well plates (2×10⁵ cells/well) were incubated with 50 mM Rho 123 (Sigma-Aldrich; Merck KGaA) at 37°C for 30 min. The reaction was terminated at 5 min of incubation on ice. Subsequently, the fluorescent of Rho 123 was detected by FACS Calibur flow cytometer (BD Biosciences). Cell Quest software version 5.1 (BD Biosciences) was used to perform the analysis.

Annexin V-FITC (Sigma-Aldrich; Merck KGaA)/PI (Sigma-Aldrich; Merck KGaA) double-staining for apoptosis analysis was used to determine the aopotosis by flow cytometry analysis. To be more specific, the cells in 6-well plate at a density of 2×10⁶/well were treated as above. Subsequently, the cells were incubated with 5 ul Annexin V-FITC for10 min at room temperature and with 5 µl PI at room temperature for 15 min in the darkness. FACS Calibur flow cytometer (BD Biosciences) and Cell Quest software version 5.1 were used to count the apoptotic cells.

The generation of NO was examined with commercial Nitrate/Nitrite Fluorometric Assay kit (Cayman Chemical Company, Ann Arbor, MI, USA) through nitrate reductase method via its breakdown products, Nitrate/Nitrite, After conducting treatment as designed, the determination was performed following manufacturer's protocols.

Cells (1×10⁴/well) were seeded into a 96-well plate and incubated at 37°C for 24 h. Then, the cells were centrifuged at 200 g for 5 min at 4°C. After being washed with PBS, the cells were resuspended and collected for the detection of the enzymes activities. The cellular SOD and CAT enzyme were determined using corresponding kits (Beijing Solarbio Science & Technology Co., Ltd.), according to the protocols. The MDA content was detected with thiobarbituric acid substance provided by Malondialdehyde Assay Kit (Beyotime, China).

Total RNA was exacted using RNAiso Plus (Invitrogen; Thermo Fisher Scientific, Inc.). M-MLV (Promega Corporation, Madison, WI, USA) and oligo (dT) primers (Takara Bio, Inc., Otsu, Japan) were used to synthesis cDNA from RNA (1 µg). The temperature and incubation protocol was set at 25°C for 10 min, at 42°C for 50 min, at 70°C for 10 min. Amplification of the related genes was conducted using LightCycler^® 480 SYBR-Green I Master (Roche Diagnostics, Indianapolis, IN, USA). The thermal cycle conditions was as follows: At 95°C, 4 min; at 35 cycles of 95°C, 15 Sec; at 60°C, 45 Sec; at 72°C, 7 min. The expression of target genes was normalized using the expression levels of GAPDH according to 2^−ΔΔCq method (37). The primers used were as follows: Fas forward 5′-GTGCTTTGCTTAGGGTTCCC-3′, Fas reverse 5′-AACTTGCACTTCTGGCCATG-3′; Fasl forward 5′-GTCCAACTCAAGGTCCATGC-3′, Fasl reverse 5′-TTGTTGCAAGATTGACCCCG-3′; Bax forward 5′-GTGCCGGAACTGATCAGAAC-3′, Bax reverse 5′-CCAAAGTAGGAGAGGAGGCC-3′; Bcl-2 forward 5′-GCCTTCTTTGAGTTCGGTGG-3′, Bcl-2 reverse 5′-GAAATCAAACAGAGGCCGCA-3′; GAPDH forward: 5′-CACAGTCCATGCCATCACTG-3′; GAPDH reverse: 5′-ATCTCGCTCCTGGAAGATGG-3′;

Cells were lysed using 200 µl lysis buffer NP40 (Beyotime Institute of Biotechnology). The concentration of the samples was determined by BCA protein quantitative kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Proteins were separated using 10% SDS-PAGE gels and transferred onto a PVDF membrane (EMD Millipore, Billerica, MA, USA). After being blocked with 5% non-skimmed milk at room temperature for 2 h, the membrane was incubated with primary antibodies as listed: cleaved caspase-3 (ab32042, 1:500; Abcam, Cambridge, UK), Fas (ab82419, 1:1,000), Fasl (ab15285, 1 ug/ml), Bax (ab32503,1:2,000), Bcl-2 (ab32124, 1:1,000), p-AKT (ab812831:5,000), AKT 1/2 (ab182729, 1:5,000), p-eNOS (ab184154, 1:1,000), eNOS (ab76198, 1:1,000), and GAPDH (ab8245, 1:5,000), at 4°C overnight. Then, horseradish peroxidase-conjugated secondary antibody (ab6721, 1:4,000) was added and maintained at room temperature for 1 h. The blot bands were developed using BeyoECL Star (Beyotime Institute of Biotechnology). The gray density was calculated with Quantity One software version 4.6 (Bio-Rad Laboratories, Inc.).

Statistical analysis was calculated using GraphPad Prism software 6.0 (GraphPad Software, Inc., La Jolla, CA, USA). One-way ANOVA following Tukey's post hoc test was used to compare the difference between groups. The results were shown as means ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

We first determined the effect of OA on cell viability of HUVECs. The CCK-8 test showed that the cell survival rate of HUVECs decreased obviously in the GC group in comparison to that in the control group, and that the pretreatment of OA improved the cell survival rate in a dose-dependent manner (Fig. 1A). Moreover, the nitric oxide (NO) generation was rescued by the pretreatment of OA (Fig. 1B). We also found that as an indicator of lipid peroxidation, the production of MDA caused by high glucose treatment was inhibited by the pretreatment of OA (Fig. 1C). Similarly, the activities of SOD and CAT were recovered in the pretreatment groups (Fig. 1D and E).

The increased ROS level is one of the sources causing mitochondria damage (38). The loss of MMP is an early hallmark event of mitochondria damage leading to apoptosis (39). Thus, the ROS production and the MMP of HUVECs were measured by flow cytometry analysis. Results showed that the ROS production was triggered by high glucose treatment. Noticeably, the ROS level was lower in the OA pretreatment groups than that in the GC group (Fig. 2A and B). Meanwhile, the loss of MMP caused by high glucose was recovered by the pretreatment of OA (Fig. 2C and D).

ROS accumulation and the loss of MMP were both considered to be closely associated with apoptosis (40). Therefore, we further detected the apoptosis of HUVECs using flow cytometry analysis. The results indicated that the apoptosis of HUVECs induced by high glucose was apparently inhibited in the OA pretreatment groups (Fig. 3A and B). Moreover, the RT-qPCR and western blot assays showed that the expressions of pro-apoptotic genes including caspase-3, Fas, Fasl and Bax were lower in the OA pretreatment groups than those in the model group. By contrast, the expression of Bcl-2 was decreased in the model group but increased by the pretreatment of OA (Fig. 4C-E).

To study the underlying mechanisms, the activity of AKT/eNOS signals was determined by Western blot. Data showed that protein level of phosphorylated AKT (p-AKT) was elevated in the OA pretreatment groups (Fig. 4A and B). Meanwhile, we noticed that the phosphorylation of eNOS (p-eNOS) was higher in the OA pretreatment groups than that in the model group (Fig. 4C and D).