Bioinformatic analysis

The Gene Expression Omnibus (GEO) Database with accession number GSE160717 was used to download the circRNA expression data for three AMI patients and three healthy controls (Zhao et al. 2021). The differentially expressed circRNAs between AMI patients and controls and the clinical information of patients and normal controls are summarized in the previous publication (Zhao et al. 2021).

Myocardial I/R mouse model

The Model Animal Research Institute of PLA Southern Theater Command General Hospital provided 20 male C57BL/6J mice used in this study. The mice were randomly divided into two groups: sham group (n = 10) and AMI group (n = 10). The AMI mouse model was developed as previously mentioned (Wang et al. 2021). Briefly, Pentobarbital sodium (50 mg/kg, P3761, Sigma-Aldrich) was administered intraperitoneally to anaesthetize mice and then cut the left sternum to fully expose the heart. To induce ischemia, a 5 to 0 (2 mm) suture was used to ligate the left anterior descending (LAD) coronary artery. Electrocardiographic ST elevation was used to confirm AMI. The sham group consisted of mice with the same surgical procedure except without ligation of the anterior descending branch of the left coronary artery. ShRNA-carrying adenovirus against circDiaph3 was myocardially injected into mice to construct a model of AMI + si-circDiaph3, while the comparable control was AMI + si-NC. After two days of surgery, mice were euthanized by increasing the CO2 flow into their cages to a rate of 3 L/min until they die. Subsequently, hearts were collected for assays.

Cell culture

Rat cardiomyocyte H9C2 cells were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The cells were maintained in DMEM medium (Gibco Healthcare Life Sciences, Logan, UT, USA) with 10% fetal bovine serum (FBS, Thermo Fisher Scientific, Inc., Waltham, MA, USA). In the control group, H9C2 cells were cultured at 37 °C and 5% CO2. To construct a cell model of myocardial H/R injury, H9C2 cells were kept under a hypoxic condition (95% N2 and 5% CO2 at 37 °C) for 2 h and then reoxygenated (75% N2, 20% O2 and 5% CO2) for 12 h.

Cell transfection

H9C2 cells were initially seeded at a density of 1 × 105 cells/well in 6-well plates and incubated at 37°C overnight until about 60% density. After replenishing with the serum-free medium, the cells were transfected with si-NC (negative control) sense, 5’-UUCUCCGAACGUGUCACGUTT-3’ and antisense, 5’-ACGUGACACGUUCG GAGAATT-3’ or si-circDiaph3 sense, 5’-CGGCAGGCAUUAGAGAUGAACAGCA − 3’ and antisense, 5’-UGCUGUUCAUCUCUAAUGCCUGCCG-3’ using the Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), or with miR-338-3p inhibitor (Cat; 4,464,079, Life Technologies) using Lipofectamine RNAiMAX Transfection Reagent (Invitrogen), according to the manufacturer’s protocol. Cells that did not need H/R treatment were transfected for 48 h, while cells that did require H/R treatment were treated for 48 h after transfection. The cells were then harvested for subsequent assays.

Cell proliferation assay

The viability of H9C2 cells was measured by Cell Counting Kit (CCK)-8 Assay Kit (Solarbio, M1020-500T). Briefly, the cells were seeded in the 96-well plates at a density of 5 × 103 cells/well and incubated at 37 °C for predetermined time intervals of 24, 48, and 72 h. After incubation, 10 µl of CCK-8 working reagent was added to each well and incubated for 30 min. Finally, a final concentration of 10% Sodium dodecyl sulfate (SDS, Solarbio, S8010) was added to each well, and the optical density (OD) values at 570 nm were recorded.

Apoptosis assay

We used the Annexin V-FITC apoptosis detection kit (Sigma-Aldrich, St. Louis, MO) to study apoptosis. Shortly, cardiac cells and H9C2 cells were collected and resuspended in 1 suspended in 1× binding buffer and stained with Annexin V-FITC (Elabscience, China) and propidium iodide (Elabscience, China) for 20 min at room temperature under dark conditions. Finally, the cells were acquired by FACSCanto II flow cytometry (BD Biosciences) and analyzed with FlowJo software.

Immunofluorescence detection of myocardial ROS levels

For evaluating the oxidative stress level in H9C2 cells, the DCFDA/H2DCFDA - Cellular ROS Assay Kit (ab113581, Abcam) was used. Briefly, H9C2 cells were plated into a 6-well plate at a density of 3 × 105/well. After indicated treatment, cells were washed with PBS and then stained with 2′,7′-dichlorofluorescin diacetate (DCFDA) solution (20 µM) for 45 min at 37 °C. The nuclei were stained with DAPI solution. The fluorescence signal was detected and observed using fluorescence microscope (BX53, Japan) and images were acquired. The intracellular ROS levels were observed in red color, while the nuclei of the cells appeared blue under the microscope. The Indica Labs - Area Quantification FL v2.1.2 module in Halo v3.0.311.314 analysis software was used to quantify the target area of each section separately, and the mean intensity was calculated.

Western blotting

Proteins from cardiomyocytes and H9C2 cells were extracted with RIPA Lysis buffer (Beyotime Biotechnology) containing Protease Inhibitor Cocktail (Roche Applied Science, Pleasanton, CA, USA). The total protein concentration was measured by BCA Protein Assay Kit (Solarbio, PC0020) and then loaded onto sodium dodecyl sulfate-10% polyacrylamide gel for electrophoresis (SDS-PAGE). The proteins were then transferred to Immobilon-NC Membranes (Triton-free, mixed cellulose esters, 0.45 μm, Solarbio, YA1711). Thereafter, the membranes were incubated with primary antibodies against BAX, BCL-2, cleaved-caspase-3, SRSF1, and GAPDH (Abcam) overnight at 4℃. Next day, the membranes were incubated with HRP-labeled secondary antibody and finally, the expression of the relative protein was detected with ECL chemiluminescent solution. The expression level of specific protein was normalized to GAPDH level and quantified using Image LabTM Software (Bio-Rad, Shanghai, China).

RNA isolation, cDNA synthesis, and real-time- qPCR

Total RNA from cardiomyocytes and H9C2 cells was extracted with Trizol™ Plus RNA Kit (ThermoFisher Scientific, IS10007, Waltham USA) and then reverse transcribed into cDNA (Superscript II reverse transcriptase, Life Technologies). The relative levels of circDiaph3, miR-338-3p, IL-1β, IL-6, TNF-α, and β-actin were determined with FastFire qPCR PreMix (SYBR Green, Tiangen, FP207, China) in a ProFlex™ PCR system (ThermoFisher Scientific) using the standard curve method. Each PCR mixture was initially denatured at 95 °C for 5 min and then cycled 40 times at 95 °C for 10 s, 60 °C for 15 s, and 72 °C for 8 s. The expression of genes was normalized to β-actin or U6 and calculated by 2−ΔΔCt method. Primer sequences used in this study are given in Table 1.

Table 1 Primer sequences used in the studyELISA

The supernatant of the cell culture medium was collected and then centrifuged at 2458 g, for 15 min. Commercial kits were used to measure the levels of IL-1β, IL-6, and TNF-α according to the manufacturer’s instructions (R&D Systems).

Dual luciferase activity assay

The StarBase online tool was used to predict the binding site of the circDiaph3 and miR-338-3p or SRSF1 and miR-338-3p. Luciferase reporters were constructed via cloning wildtype (WT) or mutant (MUT) potential binding sites for miR-338-3p in circDiaph3 or the 3’UTR of SRSF1 into the pmirGLO vectors (Promega). H9C2 cells were then co-transfected with either negative control (NC) mimic (miR-NC) or with miR-338-3p mimics and WT or MUT luciferase reporters (circDiaph3-WT, circDiaph3-MUT, SRSF1-3’UTR-WT, and SRSF1-3’UTR-MUT) for 48 h. Thereafter, the cells were harvested, and the luciferase activity was detected using the Dual-Luciferase Reporter Assay Kit (Promega).

RNA pull-down

RNA pulldown assay was performed with biotinylated antisense oligos. In brief, H9C2 cells were rinsed with cold PBS once, and resuspended in RIPA buffer for 10 min. Thereafter, cells were harvested and sonicated for 10 min, and then centrifugated at 13,000 rpm for 20 min. Then the supernatant was added with 100 pmol probes and cultured for 2 h at 4 °C. The magnetic RNA-protein pulldown kit (Thermo, Waltham, MA, USA) was used to pull down the biotin-coupled RNA complexes in accordance with the manufacturer’s instructions. 3′ end biotin-labeled circDiaph3 (Bio-circ_ circDiaph3) together with Bio-NCs (negative controls mixed with streptavidin magnetic beads) were then co-incubated with cell lysates at 4 °C overnight. After washing with buffer, RT-qPCR was carried out to measure the abundance of coprecipitated RNAs.

Statistical analysis

Data were presented as mean ± standard deviation (S.D.), and the analyses were conducted with SPSS 22.0 software (SPSS Inc., Chicago, US). Notably, the differences between groups were compared with the student’s two-tailed t-test and analysis of variance (ANOVA) followed by the post-hoc test at a defined level of statistical significance of P < 0.05.