All animal experiments were approved by the Center of Experimental Animals Committee of Xinqiao Hospital (Chongqing, China). The male C57BL/6J mice (6–8 weeks, 18–20 g, from Xinqiao Hospital Experimental Animal Center, Chongqing, China) were sacrificed by cervical dislocation, and spleens were explanted and thoroughly minced. Spleen-derived mononuclear cells were isolated by density gradient centrifugation (Histopaque 1083; Sigma-Aldrich, St. Louis, MO, USA) at 400 × g for 20 min in 4°C. The cells were collected, washed in phosphate-buffered saline (PBS) three times, and then resuspended in Dulbecco's modified Eagle's medium/nutrient mixture F12 (DMEM/F12; Gibco BRL, Gaithersburg, MD, USA) supplemented with 20% fetal calf serum (FCS; HyClone, Los Angeles, CA, USA), 20 ng/ml vascular endothelial growth factor (VEGF; R&D Systems, Inc., Minneapolis, MN, USA), 100 U/ml penicillin and 100 g/ml streptomycin. Cells were seeded into gelatin-coated cell culture flasks and incubated at 37°C under an atmosphere with 5% CO₂. Forty-eight hours later, non-attached cells were removed and adherent cells were cultured continuously. Only adherent cells were used in further experiments. The medium was refreshed with complete medium every 2 or 3 days.

To verify the phenotype of EPCs, cells were incubated with Dil-ac-LDL (Biomedical Technologies, Inc., Stoughton, MA, USA) for 3 h, fixed with 4% paraformaldehyde and then incubated with fluorescein isothiocyanate (FITC)-labeled lectin (UEA-I; Sigma-Aldrich) for 1 h, then washed with PBS three times. The cells were then observed under an immunofluorescence laser scanning confocal microscope (Leica TCS; Leica, Mannheim, Germany). Dual-stained cells positive for Dil-ac-LDL and UEA-I were identified as EPCs. Additionally, fluorescence-activated cell sorting (FACS) analysis was performed using FITC-conjugated antibodies against mouse Sca-1 (ab25031) and vascular endothelial growth factor receptor 2 (VEGFR2; ab11939), and the corresponding isotype control antibodies (ab18446, ab171870; all from Abcam, Cambridge, UK).

Transient silencing of E2-2 and ATG7 was induced by transfection with siRNAs (siRNA-E2-2 and siRNA-ATG7, respectively) (both from GenePharma, Shanghai, China). The selected siRNA duplex sequences specifically targeted mouse E2-2 and ATG7, and showed no homology to any other sequences during a BLAST search. The siRNA-E2-2 efficient sequence was: sense, 5′-CAAGAAGGAUAUCAAAUCA-3′ and antisense, 5′-UGAUUUGAUAUCCUUCUUG-3′. The siRNA-ATG7 sequence was: sense, 5′-GGAGUCACAGCUCUUCCUUTT-3′ and antisense, 5′-AAGGAAGAGCUGUGACUCCTT-3′. The corresponding non-silencing control (siRNA-con) sequences (GenePharma) were designed according to the sequence of a negative control. Transfection with siRNA-E2-2 or siRNA-ATG7 was carried out using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) with a molar ratio between DNA and lipid of approximately 1:3. Forty-eight hours after transfection, cells were collected and used for subsequent assays.

Adenovirus vector expressing E2-2 (Ad-E2-2) and adenovirus-carrying blank vector (Ad-vector) were both constructed by HanBio Technology Co., Ltd. (Shanghai, China). Before transfection, EPCs were incubated in DMEM/F12 medium without FCS and antibiotics for 6 h. Adenovirus vector was then added to the cells at a multiplicity of infection (MOI) of 100. The transfection medium was removed 2 h later, and the cells were maintained in the complete medium for 48 h. Ad-vector was used as a control. The mRFP-GFP-LC3 adenovirus vector for detecting autophagy was also purchased from HanBio Technology Co., Ltd. The process of mRFP-GFP-LC3 adenovirus transfection was similar to that of Ad-E2-2. Autophagy was observed under an immunofluorescence laser scanning confocal microscope (Leica TCS). Autophagy was detected by counting the number of green fluorescent protein (GFP) and red fluorescent protein (RFP) dots (dots/cell).

EPCs were collected from the cultures and replated into a 96-well plate (1×10⁵ cells/ml) and underwent different treatments. Cell proliferation was measured using a Cell Counting Kit-8 (CCK-8; Beyotime Biotech, Jiangsu, China) according to the manufacturer's instructions. Before reading the absorbance at 450 nm, 10 µl CCK-8 solution and 100 µl complete medium were sequentially added to each well.

In addition, the cell cycle was analyzed by flow cytometric analysis. EPCs were harvested and resuspended in PBS solution at 1×10⁶ cells/ml. The cells were then stored in 70% ethanol at 4°C overnight. Before being analyzed, the cells were mixed with 1 ml propidium iodide (PI; 1 mg/ml) (BD Biosciences, Bedford, MA, USA) and then incubated at 4°C for 30 min. After centrifugation, the cells were resuspended and the cell cycle was analyzed immediately using a flow cytometer (MoFloTMXDP; Beckman Coulter, Co. Fullerton, CA, USA). The proliferation index was defined as the total percentage of cells in S and G2/M phases. All groups of experiments were undertaken in triplicate.

EPCs were lysed in ice-cold lysis buffer (RIPA) containing a protease inhibitor (0.5 mM PMSF) (both from Beyotime Biotech). The cell lysates were further centrifuged at 15,000 × g for 15 min at 4°C. Proteins from EPC lysates were measured using the Bradford method. The same amount of protein was loaded in each lane, separated by 10–15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to 0.22 µm polyvinylidene difluoride (Millipore Corp., Boston, MA, USA) membranes. Membranes were blocked with 5% non-fat milk 2 h followed by incubated with primary antibodies overnight at 4°C. The primary antibodies included: rabbit anti-mouse LC3 polyclonal antibody (1:1,000; Cat. no. 2775), rabbit anti-mouse p62 monoclonal antibody (1:1,000; Cat. no. 5114), rabbit anti-mouse ATG7 polyclonal antibody (1:1,000; Cat. no. 8558) (all from Cell Signaling Technology, Inc., Boston, MA, USA) and rat anti-mouse β-actin monoclonal antibody (1:1,000; Cat. no. AA128; Beyotime Biotech). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5,000; A0208, A0216; Beyotime Biotech) for 1 h at 37°C. Protein bands were visualized by enhanced chemiluminescence (ECL) (Thermo Fisher Scientific, Waltham, MA, USA), and densitometric signals were quantified using ImageQuant TL software (General Electric, Co., Fairfield, CT, USA).

Total RNA was isolated from EPCs using RNAiso Plus (Takara Bio, Dalian, China) according to the manufacturer's instructions. Single cDNAs was synthesized from RNA using a PrimeScript™ RT reagent kit (Takara Bio). The following primer sequences were used for ATG7 (sense, 5′-ACCCAGAAGAAGCTGAACGA-3′, and antisense, 5′-CTCATTTGCTGCTTGTTCCA-3′); β-actin (sense, 5′-TTCTACAATGAGCTGCGTGTGG-3′ and anti-sense, 5′-GTGTTGAAGGTCTCAAACATGAT-3′). SYBR^® Premix Ex Taq™ II (Takara Bio) was used according to the manufacturer's instructions, and RT-qPCR was performed using a StepOnePlus™ Real-Time PCR system (Applied Biosystems Life Technologies, Waltham, MA, USA). The reaction conditions were as follows: 95°C for 30 sec, 40 cycles of 95°C for 5 sec, and 60°C for 30 sec. Relative expression of target mRNA was normalized to β-actin using the ΔΔCt method, as previously described (19).

Data from at least three independent experiments are expressed as the means ± SD. SPSS 16.0 software was used for statistical analysis. Student's t-tests were used for comparisons between groups. A P-value <0.01, or P<0.05 was considered to indicate a statistically significant difference.

After 7 days of being cultured, round, oval and spindle shapes of attached spleen-derived mono-nuclear cells were noted under the microscope (Fig. 1A), similar to morphologies which have been previously described (20). Immunocytochemical (ICC) staining demonstrated that 92.33±2.63% (n=3) of attached cells were stained positively for FITC-UEA-I and Dil-ac-LDL (Fig. 1B), validating the theory that the FITC-UEA-I/Dil-ac-LDL double-positive cells possessed the characteristics of ECs. Furthermore, the attached cells were identified by analyzing the expression of the mouse stem-cell marker Sca-1 and EC marker VEGFR2 using flow cytometry. The percentage of positive cells was 88.93±1.31 and 85.13±2.83% (n=3), respectively (Fig. 1C). Taking these results into account, the attached cells were considered EPCs.

To determine whether E2-2 regulated the proliferation of EPCs, we overexpressed or silenced E2-2 by transfecting Ad-E2-2 or siRNA-E2-2 into the cells, and proliferation was then detected by CCK-8 and cell cycle assays in vitro. The CCK-8 assay demonstrated that the proliferation of EPCs significantly decreased in the E2-2 overexpression (Ad-E2-2) group compared with the adenoviral control vector (Ad-vector) group (Fig. 2A). Cell cycle analysis showed that the proliferation index (S + G2/M) of the EPCs was lower in the Ad-E2-2 group than in the Ad-vector group (Fig. 2B and C). The trend of the cell cycle was similar to the results of the CCK-8 assay. However, in the knockdown E2-2 (siRNA-E2-2) group, the proliferative capacity of EPCs significantly increased compared with that in the negative control (siRNA-con) group (Fig. 2D). The EPC proliferation index (S+G2/M) of the siRNA-E2-2 group was higher than that of the siRNA-con group (Fig. 2E and F). These results suggest that E2-2 suppressed the proliferation of EPCs in vitro.

To investigate whether E2-2 regulates autophagy in EPCs, we firstly studied the expression of LC3II and p62 while overexpressing or knocking down E2-2. We noted a significant decrease in LC3II expression, but a marked increase in the bands of p62, in the Ad-E2-2 group compared with the Ad-vector group (Fig. 3A), suggesting that overexpression of E2-2 induced suppression of autophagy in EPCs. By contrast, LC3II expression increased and the protein level of p62 decreased in the knockdown E2-2 group (Fig. 3B). Thus, knockdown of E2-2 upregulated autophagy of EPCs. In addition, we used the mRFP-GFP-LC3 adenovirus to confirm the regulatory effects of E2-2 on autophagy. This assay is used to monitor progression from autophagosome to autolysosome based on the pH difference between the acidic autolysosome and the neutral autophagosome and the pH sensitivity differences displayed by GFP and RFP. When autolysosomes are formed by fusing autophagosome with a lysosome, the GFP moiety degrades from the tandem protein, but RFP-LC3 maintains the dots (21,22). The red dots indicate autophagosomes, and yellow dots (merge of RFP and GFP) represent autolysosomes. When we co-transfected mRFP-GFP-LC3 adenovirus with Ad-E2-2 or siRNA-E2-2 in EPCs for 48 h, we observed that overexpression of E2-2 reduced the numbers of autophagosomes and autolysosomes (Fig. 3C). Conversely, knockdown of E2-2 increased not only the number of autophagosomes but also that of autolysosomes (Fig. 3D). The results further confirmed that E2-2 negatively regulated autophagy of EPCs.

To validate the role autophagy plays in EPC proliferation, we treated EPCs with chloroquine (CQ) at different concentrations and different periods of time. It was observed that CQ inhibited the proliferation of EPCs in a concentration-dependent manner. After treatment with CQ for 24 h at 20 µM, the proliferation of EPCs as measured by CCK-8 decreased markedly compared with the group which was not treated with CQ, and the degree of decrease changed slightly with the increasing periods of time (Fig. 4B). Subsequently, to determine whether autophagy was involved in E2-2-mediated EPC proliferation, we added 20 µM CQ to EPCs after transfection with siRNA-E2-2 for 24 h. After transfection for 48 h, the CCK-8 assay showed that the increase in proliferation in the siRNA-E2-2 group was partially reversed by the administration of CQ (Fig. 4C). The alteration of the proliferation index, as measured by flow cytometry, was similar to the results of the CCK-8 assay (Fig. 4D). Furthermore, we transfected the different groups of EPCs with siRNA-con, siRNA-E2-2 or siRNA-E2-2 and CQ and found that CQ inhibited the increase in autophagosomes and autolysosomes (Fig. 4E), suggesting that EPC autophagy induced by E2-2 knockdown was reversed by CQ. These results demonstrated that inhibition of basal autophagy reduced the proliferation of EPCs, and E2-2 inhibited the proliferation of EPCs through its effects on autophagy.

To further explore the mechanism by which E2-2 regulates autophagy in EPCs, we detected the expression of the autophagy-related gene ATG7 after E2-2 overexpression or knockdown. We found that overexpression of E2-2 decreased the expression level of ATG7 mRNA (Fig. 5A), and knockdown of E2-2 upregulated the expression of ATG7 mRNA (Fig. 5B). Subsequently, we noted that overexpression of E2-2 downregulated ATG7 protein expression, and knockdown of E2-2 upregulated ATG7 protein expression (Fig. 5C and D). Furthermore, to demonstrate whether ATG7 was involved in E2-2-mediated autophagy, we transfected the cells with siRNA-E2-2 and siRNA-ATG7 simultaneously or respectively. The transfection revealed that autophagy of EPCs was downregulated by ATG7 knockdown alone, and the increased autophagy induced by siRNA-E2-2 was reversed by ATG7 knockdown. Interestingly, the level of autophagy was still higher in the co-transfected siRNA-E2-2 and siRNA-ATG7 group than in siRNA-ATG7 group (Fig. 5E). We demonstrated that E2-2 regulated autophagy of EPCs via partially mediating the expression of ATG7.