Animal care and all experimental protocols involving animals were approved by the Animal Care and Use Committee of Wenzhou Medical University (Wenzhou, China). A total of 50 adult male Sprague-Dawley (SD) rats (specific pathogen-free; aged 6-8 weeks (200-320 g); Wenzhou Medical University Laboratory Animal Center, Wenzhou, China) were used in the study. All rats were housed at room temperature (21-25°C) with 45-50% humidity and a 12 h light/dark cycle, with free access to soft food and tap water.

The EPCs were isolated from the bone marrow of SD rats and cultured according to established methods with minor modifications (26,27). Briefly, five rats per batch were sacrificed by cervical vertebra separation following anesthesia by intraperitoneal injection of 320 mg/kg chloralhydrate. Bones (femur and tibia of hind legs) from the 6-8 week-old male SD rats were repeatedly flushed with homogenate rinse solution (Tianjin Hao Yang Biological Manufacture Co., Ltd., Tianjin, China). The washing fluid was centrifuged for 10 min at room temperature (700 × g) and the pellet was resuspended in 3 ml sample diluent containing 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), and then was combined carefully under 3 ml histopaque-1083 (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). Following centrifugation at 400 × g for 30 min at room temperature, the mixture had separated into three layers. The cloudy layer was gently removed and washed with PBS. Following three washes, the mononuclear cells (MNCs) were resuspended in endothelial growth factor-supplemented media (EGM-2 bullet kit; Lonza Group, Ltd., Basel, Switzerland) with 10% fetal bovine serum and plated on fibronectin-coated culture plates (Sigma-Aldrich; Merck KGaA). The cells were cultured at 37°C with 5% CO₂ in a humidified atmosphere. The medium was replaced following the first 48 h and every 3 days thereafter. Late EPCs appeared after 14 days in culture, and the third or fourth passage was used for later analysis.

To identify the bone marrow-derived-EPCs, immunofluorescence staining and flow cytometry were performed to analyze the expression of the specific cell markers FLK-1 and CD34. For the immunofluorescence staining, the EPCs that grew to 90% were washed with PBS three times prior to fixation with 4% paraformaldehyde. The cells were then incubated with FLK-1 (cat. no. sc-393163; Santa Cruz Biotechnology, Inc., Dallas, TX, USA; 1:400) and CD34 (cat. no. 14486-1-AP; ProteinTech Group, Inc., Chicago, IL, USA; 1:200) antibodies at 4°C overnight. Following incubation with the corresponding secondary antibodies at room temperature for 1 h (cat. no. BYE010; Boyun Biotech, Shanghai, China; 1:200) and DAPI nuclei dye, the EPCs were observed and images were captured under a fluorescence microscope (932706; Nikon Corporation, Tokyo, Japan). For flow cytometry (28), the adherent cells were digested with EDTA-free trypsin and resuspended in PBS at the density of 5×10⁶ cells/ml. The MNCs were then stained with FITC-conjugated mouse anti-vascular endothelial growth factor receptor 2 (cat. no. ab184903; Abcam, Cambridge, UK; 1:200) and PE-conjugated mouse anti-CD34 (1:200; cat. no. MA1-10205; Thermo Fisher Scientific, Inc.) for 30 min at room temperature. The same fluorescein-labeled isotype IgG was used as a negative control for each antibody. The cells were analyzed using a flow cytometer (BD Biosciences, San Jose, CA, USA) and data were analyzed with FlowJo 7.6.1 (TreeStar, Inc., Ashland, OR, USA).

A Cell Counting Kit-8 (CCK-8) assay was used to examine cell viability. The EPCs (5×10³ cells/well) were plated on 96-well plates and cultured in EGM-2 medium supplemented with 10% FBS to induce attachment overnight. The medium was replaced with EGM-2 plus the specified additives: i) 0.5 mM PA; ii) 50 µM metformin, and iii) metformin + PA. Following incubation at 37°C with 5% CO₂ for 48 h, the medium was aspirated and fresh DMEM without serum but containing CCK-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was added (10 µl/well). Following culturing for 2 h at 37°C, the absorbance was measured with a microplate reader (SpectraMax^® Plus³⁸⁴ Absorbance Microplate Reader; Molecular Devices, LLC, Sunnyvale, CA, USA) at a wavelength of 450 nm. Each group had five replicates and was assessed in triplicate.

A cell migration assay was conducted in a 24-well Millicell (Corning Incorporated, Corning, NY, USA) containing a microporous (8.0-mm) membrane. Briefly, the EPCs (3×10⁴ cells/well) were resuspended in EBM-2 medium (EGM-2 bullet kit; Lonza Group, Ltd.) and added to the upper chambers. The lower chambers (24-well plates) were mixed with EGM-2 medium in the presence of metformin (50 µM, Sigma-Aldrich; Merck KGaA), PA (0.5 mM, Sigma-Aldrich; Merck KGaA) or both. Following incubation for 24 h, the non-migrated cells were removed with cotton-tipped swabs. Those cells attached to the lower chambers were fixed with 10% paraformaldehyde for 30 min. Following DAPI staining for 30 min, the average cell number was counted manually in random fields (magnification, ×100) with an inverted microscope (Olympus CKX41; Olympus Corporation, Tokyo, Japan) in each well.

The in vitro angiogenic activity of the EPCs was assessed using a Matrigel tube experiment, as described previously (24,29). In brief, Matrigel (cat. no. 356234; BD Biosciences) was thawed at 4°C overnight, and 150 µl Matrigel per well was added to a 48-well plate and incubated at 37°C for 30 min to polymerize. The transfected EPCs (3×10⁴ cells/well) were cultured in the presence or absence of metformin, PA or their combination for 12 h. Tube-like structures were observed under an inverted microscope (Olympus CKX41; Olympus Corporation). The numbers of tube-like structures in the images were measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA). At least six random areas per well were analyzed, and the experiments were repeated three times using EPCs from three independent isolations.

For the overexpression or inhibition of miR-130a, the EPCs were transfected with miR-130a mimics (50 nmol/l, 5′-CAG UGC AAU GUU AAA AGG GCA U-3′) or inhibitor (200 nmol/l, 5′-AUG CCC UUU UAA CAU UGC A CU G-3′) and the corresponding negative controls (Guangzhou RiboBio Co, Ltd., Guangzhou, China) for 6 h using LipoFiter™ transfection reagent according to the manufacturer's protocol (Hanbio Biotechnology Shanghai Co., Ltd., Shanghai, China). Following 6 h of transfection, the medium was replaced with 10% FBS EGM-2 consisting of either PA or metformin for further analysis.

The target of miR-130a was predicted by using TargetScan 7.1 (http://www.targetscan.org). Combining previous literature on miR-130a and angio-genesis (22,30), PTEN was selected for further analysis as a potential target of this miRNA.

The EPCs were cultured in EGM-2 with 10% fetal bovine serum containing either PA (0.5 mM, Sigma-Aldrich; Merck KGaA) or metformin (50 μM, Sigma-Aldrich; Merck KGaA) for 24 h. For the miR-130a overexpression experiment, the medium was replaced with PA or metformin 6 h following transfection.

Total RNA was extracted from the harvested cells using TRIzol reagent (Thermo Fisher Scientific, Inc.). To quantify the mRNA levels of PTEN, RNA was quantified and reverse transcribed into complementary DNA using specific stem-loop reverse transcription primers. The mRNA first strand synthesis was performed using a HiScript II Q RT SuperMix for qPCR (Vazyme Biotech, Co., Ltd., Piscataway, NJ, USA), and qPCR was operated using a ChamQ™ Universal SYBR qPCR Master mix (Vazyme Biotech, Co., Ltd.) on a Bio-Rad CFX96™ R-T PCR machine (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The Vazyme cycling conditions were as follows: 95°C for 30 sec, followed by 39 cycles at 95°C for 10 sec and 60°C for 30 sec. A melting curve every 0.5°C between 65 and 95°C for 15 min was obtained. GAPDH was used as a loading control. To analyze the expression of miRNAs, cDNA was synthesized using the miRcute Plus miRNA First-Strand cDNA Synthesis kit (Tiangen Biotech Co., Ltd., Beijing, China) and RT-qPCR was performed using an miRcute Plus miRNA qPCR Detection kit (cat. no. FP411; Tiangen Biotech, Co., Ltd.). The Tiangen cycling conditions were as follows: 95°C for 15 min, followed by 44 cycles at 94°C for 20 sec and 60°C for 34 sec. A melting curve every 0.5°C between 65 and 95°C for 15 min was obtained. U6 was used as an internal control. The PCR primers used in the present study were synthesized by Sangon Biotech, Co., Ltd., (Shanghai, China) and the sequences are listed in Table I. The PTEN forward, 5′-ATG TTC AGT GGC GGA ACT TG-3′ primer and the reverse primers for miR-130a, miR-132, miR-34a, miR-221 and U6 were synthesized by Tiangen Biotech Co., Ltd., and were include in the kit applied for the synthesis of cDNA. The results were subjected to melting curve analysis and the relative gene expression levels of miRNAs were determined. The levels of PTEN genes were analyzed using the 2^−ΔΔCq method (31).

The protein expression levels of PTEN, AKT and phosphorylated (P-)AKT were detected by western blot analysis. The EPCs were washed with pre-cooled PBS, lysed in RIPA solution, and centrifuged at 12,000 g for 30 min (4°C). Protein concentration were measured using the BCA method (Beyotime Institute of Biotechnology, Beijing, China). The proteins (30 μg/lane) were separated on a 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto a polyvinylidene difluoride membrane. Following blocking of the nonspecific binding sites with 5% non-fat milk in Tris-buffered saline-Tween 20 (TBS-T), the membranes were incubated overnight with primary antibody against PTEN (1:1,000; cat. no. 9188; Cell Signaling Technology, Inc., Danvers, MA, USA), AKT (1:1,000; cat. no. 9272; Cell Signaling Technology, Inc.), and P-AKT (1:1,000; cat. no. AF0016; Affinity Biosciences, Cell Signal Transduction, Cincinnati, OH, USA) at 4°C overnight. The membranes were washed with TBS-T and incubated with goat anti-rabbit IgG, peroxidase-conjugated secondary antibodies (1:5,000; Biosharp Life Sciences, Hefei, China) at room temperature for 1 h. The signals were visualized using chemiluminescence detection reagents (Pierce ECL Plus; Thermo Fisher Scientific, Inc.). The relative band intensities were determined by the Image Lab software 4.1 (Bio-Rad Laboratories, Inc.). GAPDH was used as an internal control. The results of the analysis were are expressed as a relative ratio of the target protein to the internal reference. All experiments were repeated three times.

All data were analyzed with the statistical software GraphPad Prism v.7.0 (GraphPad Software, Inc., San Diego, CA, USA). The statistical analysis was performed using one-way analysis of variance among groups. All continuous data are expressed as the mean ± standard deviation, and P<0.05 was considered to indicate a statistically significant difference.

The EPCs derived from rat bone marrow were isolated by density gradient centrifugation and cultured in fibronectin-coated six-well plates. Following 3 days of culture in endothelial cell growth medium EGM-2, most cells were adherent to the well surfaces (Fig. 1A); non-adherent cells were removed by replacing the medium. On day 7 of culture, the EPCs were in spindle-like cell colonies, which indicated the formation of early EPCs (Fig. 1B). Following another 7 days (on the 14th day of culture), the cell colonies gradually became larger, and the cells exhibited a cobblestone-like morphology indicating their progression into late EPCs (Fig. 1C). The first generation of late EPCs (Fig. 1D) were used for phenotypic identification and subsequent experiments. To demonstrate that these primary cells were EPCs, immunofluorescence assays and flow cytometry were performed using the first generation of EPCs. FLK-1-positive cells were stained green and CD34-positive cells were stained red. Most EPCs were positive for both FLK-1 and CD34, as shown by their yellow coloration in Fig. 1E (Merge). The flow cytometry confirmed that most of the cells were positive for CD34 and FLK-1 (Fig. 1F). This result is consistent with that of previous studies (8,32).

To examine the function of EPCs, the cells were cultured in EGM-2 with or without PA. Transwell and Matrigel assays were used to measure the migration and angiogenic potential of the EPCs exposed to PA. As shown in Fig. 2A, the migration potential of the EPCs exposed to PA was inhibited, as the number of cells migrating to the lower chambers was significantly lower in the PA-treated EPCs than in the control cells (Fig. 2A and B). The PA-treated EPCs also formed fewer numbers of tubes than the non-treated control EPCs (Fig. 2C and D), indicating that PA reduced the angiogenic potential of EPCs. Cell proliferation was also analyzed using CCK-8. The EPCs exposed to PA exhibited significantly lower proliferation than the control EPCs that were not exposed to PA (Fig. 2E). However, treatment of metformin eliminated the effects of PA on the EPCs (migrating ability, Fig. 2A and B; tube formation, Fig. 2C and D; proliferation, Fig. 2E). These data showed that metformin not only improved EPC proliferation (Fig. 2E), but also migration and angiogenesis (Fig. 2A-D), which indicated that metformin may prevent EPCs from PA-induced impairment.

Based on an extensive literature review, four miRNAs (miR-132, miR-34a, miR-221 and miR-130a) were identified as being associated with angiogenesis (24,33,34), the levels of which may be affected by EPC exposure to PA and/or metformin. In brief, the EPCs were maintained in PA (0.5 mM) for 24 h, following which the expression levels of the four miRNAs were measured using RT-qPCR analysis. As shown in Fig. 3A, the levels of miR-130a were significantly lower in the PA-treated EPCs than in the EPCs of the control group. It was also found that the PA-treated EPCs exhibited significantly higher mRNA and protein levels of PTEN than the control EPCs (Fig. 3B-D). However, the levels of p-AKT/AKT were significantly decreased following treatment with PA compared with the control (Fig. 3E). To investigate how metformin affected the expression levels of miR-130a, PTEN and p-AKT in EPCs exposed to PA, the EPCs were cultured in EGM-2-containing PA (0.5 mM) and metformin (50 μM) for 24 h. The EPCs exposed to both metformin and PA had significantly higher levels of miR-130a and p-AKT than the EPCs exposed to PA alone (Fig. 3A and E). By contrast, the mRNA and protein expression levels of PTEN were lower in the EPCs treated with metformin and PA compared with those in the EPCs exposed to PA only (Fig. 3B-D). These data implied that metformin may attenuate the PA-induced reduction of miR-130a/p-AKT and increase of PTEN in EPCs.

Analyses using TargetScan softwareindicated that PTEN is a potential target of miR-130a regulation. The association between miR-130a and PTEN was confirmed in a previous study, which showed that miR-130a specifically inhibited the expression of PTEN via a 3′UTR region in the PTEN gene (22). To confirm whether miR-130a regulates the expression levels of PTEN in EPCs, the present study investigated the mRNA and protein expression levels of PTEN in EPCs transfected with miR-130a mimics or miR-130a inhibitor. EPCs overexpressing miR-130a had significantly lower mRNA and protein levels of PTEN than those transfected with the corresponding negative control (NC) (Fig. 4A-C). By contrast, EPCs transfected with miR-130a inhibitors shown an opposite trend (Fig. 4E-G). These data indicated that miR-130a negatively regulated the expression of PTEN in EPCs.

As mentioned above, metformin improved the functions of EPCs when exposed to PA. However, whether the miR-130a/PTEN and AKT pathways were involved remained unclear. The EPCs exposed to PA or metformin were treated with miR-130a mimics or inhibitors. The increased expression of miR-130a was correlated with the activated PI3K/AKT pathway, as the p-AKT ratio was higher (Fig. 4D and I). Regarding the expression of PTEN, opposite results were observed at both the mRNA and protein levels (Fig. 4A-C). As shown in Fig. 5A-E, the impaired tube formation, migration and cell viability of the EPCs induced by PA was attenuated in the EPCs transfected with miR-130a mimics. To further investigate whether the endothelial-protective action of metformin was miR-130a-dependent, the effects of metformin in EPCs were examined following transfection with an miR-130a inhibitor. As expected, the inhibition of miR-130a in EPCs resulted in a reverse effect of metformin in the expression of PTEN (Fig. 4E-G) and miR-130a (Fig. 4I). However, no significant effects on PI3K/AKT were observed in EPCs transfected with miR-130a inhibitor (Fig. 4E and H). Similarly, the effects of metformin on angiogenesis and cell viability significantly decreased in EPCs transfected with the miR-130a inhibitor (Fig. 6A-E). In summary, the ameliorative effect of metformin on PA-induced changes in EPC proliferation, cell migration and tube formation may be associated with an increase in the expression levels of miR-130a and a decrease in the levels of PTEN, which may be associated with activation of the PI3K/AKT signaling pathway.