Cell treatment
293T cells (CL-0005), human retinal microvascular endothelial cells (HRMECs, CP-H130) were purchased from Procell (Wuhan, China). 293T cells were cultured in DMEM containing 10% fetal bovine serum with 1% penicillin/streptomycin. HRMECs were cultured in HRMECs complete medium (CM-H130, Procell, Wuhan, China). HRMECs were treated with D-glucose (HG, 25 mM) for 48 h to construct a DR cell model [21]. Control group was treated with normal glucose 5 mM and the isotonic control group (Osm) was treated with 25 mM L-glucose for 48 h. HRMECs were treated with apigenin (0, 5, 10, 25, 50, 100 µM) for 24 h and then treated with HG [22].
Cell transfection
si-HDAC3, miR-140-5p mimic, miR-140-5p inhibitor and its negative controls were provided by GenePharma (Shanghai, China). Above si-RNA sequences were transfected into cells by using lipofectamine 3000 reagent according to the instruction. Cells were used for subsequent detection after transfection for 48 h.
Quantitative real-time PCR (qRT-PCR)
Total cell RNA was extracted by Trizol (15596026, Thermo, USA). The cDNA reverse transcription kit (4368814, Invitrogen, USA) reverses transcribed RNA into cDNAs, and genes expression was examined on ABI 7900 system using SYBR Green qPCR Mix (HY-K0501A, MedChemExpress, USA). Using β-actin or U6 as an internal reference, gene level was calculated by 2−ΔΔCt method. Primers were as follows:
miR-140-5p (F): GGGCCAGTGGTTTTACCCTA
miR-140-5p (R): GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACCTACCA
HDAC3 (F): CATCCAGATGTCAGCACCCG
HDAC3 (R): AAAGTAGGCTGAAGTCCCTGC
PTEN (F): CGACGGGAAGACAAGTTCAT
PTEN (R): AGGTTTCCTCTGGTCCTGGT
β-actin (F): CCCTGGAGAAGAGCTACGAG
β-actin (R): CGTACAGGTCTTTGCGGATG
U6 (F): CTCGCTTCGGCAGCACA
U6 (R): AACGCTTCACGAATTTGCGT
Western blot
Total protein was extracted from cells using RIPA lysis buffer (P0013B, Beyotime, China). Protein quantification of samples was performed based on BCA protein assay kit (BL521A, Biosharp, China). Proteins were mixed with SDS-PAGE loading buffer (MB2479, meilunbio, China), and adsorbed on PVDF membranes by gel electrophoresis. Membranes were blocked with 5% nonfat milk. HDAC3 (1:1000, ab137704, Abcam, UK), PTEN (1:1000, ab267787, Abcam, UK), PIK3 (1:1000, 4257, Cell Signaling Technology, USA), p-PIK3 (1:1000, 4228, Cell Signaling Technology, USA), AKT (1:1000, 9272, Cell Signaling Technology, USA), p-AKT (1:2000, 4060, Cell Signaling Technology, USA) primary antibodies and β-actin (1:1000, ab8227, Abcam, UK) were incubated with membranes at 4°C overnight. Secondary antibody was incubated with membranes. The target bands were visualized using a chemiluminescence imaging system (Chemiscope6100, CLiNX, China). β-actin was acted as internal reference to measure protein expression level.
Bioinformatics prediction and dual-luciferase reporter assay
Starbase database (https://starbase.sysu.edu.cn/) predicted miR-140-5p to HDAC3 binding sites. ThTFtarget database (http://bioinfo.life.hust.edu.cn/hTFtarget#!/) predicted HDAC3 to PTEN upstream promoter region binding site.
To verify miR-140-5p to HDAC3 3’UTR binding, wild-type (WT) or mutant (MUT) HDAC3 fragments were inserted in pmirGLO vector (E1330, Promega, USA). Rcombinant vector was transfected into 293T cells by pofectamine 3000 reagent (L3000015, Thermo, USA), and mimics NC and miR-140-5p mimics were transfected nto cells. Finally, luciferase activity was measured through Nano-Glo dual luciferase reporter assay.
Chromatin immunoprecipitation (ChIP)-qPCR
Cells were immobilized with 1% formalin and then DNA was fragmented to 200–800 bp randomly by ultrasound and immunoprecipitated with a target protein-specific antibody to HDAC3 or IgG. Then we de-cross-linked the target protein-DNA complex at 65°C overnight. Next, 100 µL H2O was applied to purify and elute ChIP DNA, and 2.5 µL ChIP-DNA was used for qPCR determination [23].
MTT assay
Cells were digested and counted, and seeded in the 96-well plate (1×104 cells/well). After culturing for the time corresponding to the above grouping, 5 mg/mL MTT (10 µL) was added, incubated at 37°C, 5% CO2 for 4 h. Then we took out the 96-well plate, aspirated the MTT-containing medium, and added 150 µL DMSO to each well. After shaking slowly at room temperature (10 min), absorbance (OD = 490 nm) was assessed on the Bio-Tek microplate reader (MB-530, Heales, China).
Wound-healing assay
Cells were digested and counted with trypsin. Each well was inoculated with about 5×105 cells. When cells growth reached 90% fusion rate, cells were washed once with sterile PBS and scratched with a sterile pipette tip. And medium was added. Photographs were taken to record scratches at 0 h. After culturing at 37°C and 5% CO2 for 24 h, pictures were recorded again.
Transwell migration assay
Cell movement was measured by cell migration method. 1×106/mL cells were re-suspended in serum-free medium, 100 µL cell suspension was added into the upper chamber of Transwell chamber (33318035, Corning), and 600 µL complete medium was added into lower. Culture medium in chamber was discarded. Cells on the upper chamber surface were wiped away with a damp cotton swab. 4% paraformaldehyde was fixed, stained with 0.5% crystal violet and eluted with water. Cells on the outer surface of upper chamber were monitored under the microscope (Olympus, Japan) and photographed.
Tube formation assay
Matrigel (356234, Biocoat) was putted on an ice box. Matrigel (70 µL) was added to each well of a pre-cooled 96-well plate until wells were evenly covered, let stand at 4°C for 10 min, and placed it in the 37°C incubator for 30 min. Cells were digested with 0.25% trypsin, counted with a cell counting plate, and 10,000 cells were added to each well. The number of meshes, tube length, and number of nodes were calculated.
Statistical Analysis
Statistical analysis was performed by Graphpad Prism8.0 software. Measurement data were expressed as mean ± standard deviation (SD). The comparison between groups was analyzed by Student’s t-test. One-way analysis of variance (ANOVA) and Tukey’s tests were carried out for multiple group comparisons. p < 0.05 indicated difference was statistically significant.