A total of 56 male Sprague-Dawley rats (age, 6 weeks; weight, 100–120 g), were purchased from the Hong Kong University of Science and Technology Medical Center, Shenzhen Peking University [SYXK (Yue) 2020–0106]. The animal experimental procedures were performed under the approval of the Experimental Ethics Committee of Hong Kong University of Science and Technology Medical Center, Shenzhen Peking University. The rats were maintained at a constant temperature of 22±2°C and 40–60% humidity on a 12-h light/dark cycle with free access to standard rodent food and water. They were kept adaptive for breeding for at least 1 week prior to surgery.

The rats belonging to the DCM model and control groups received a high-fat diet and a normal diet, respectively, for a period of 4 weeks. Subsequently, the rats were intraperitoneally administered 30 mg/kg streptozotocin (STZ) per day (MilliporeSigma) for 5 consecutive days to induce DCM and were fed a high-fat diet for 4 weeks. The high-fat diet consisted of 66.5% standard laboratory chow, 10% lard, 20% carbohydrate, 2.5% cholesterol, and 1% bile acid sodium (Syagen Biosciences). The control rats were administered a regular diet. The regular diet consisted of 75% corn, 20% wheat bran, 3% fish meal, 1.5% farina and 5% salt (Syagen Biosciences). The successful establishment of the model was confirmed when blood glucose levels ≥11.1 mmol/l were detected in blood acquired from the tail vein, evaluated using a contour glucose meter (Roche Diagnostics).

A total of 24 rats were randomly divided into three groups (n=8/group) as follows: The control, DCM and DCM treated with BBR (BBR-DCM) groups, as previously described (21). BBR (cat. no. B107342; Shanghai Aladdin Biochemical Technology Co., Ltd.) was dissolved in physiological saline and intragastrically administered daily (200 mg/kg) for a further 4 weeks following the induction of DCM (22). The rats in the control and DCM groups received an equal volume of physiological saline.

In order to evaluate the effects of miR-18a-3p overexpression on the heart functions of diabetic rats, 32 rats (n=8/group) were used for this experiment. The groups used were the following: The control, DCM, scrambled miR (Scm)-DCM and the overexpressing miR-18a-3p (OE-18a-3p)-DCM groups. The rats in the OE-18a-3p-DCM group were generated by an intravenous (i.v.) injection of adeno-associated virus (AAV) overexpressing miR-18a-3p (Shanghai GenePharma Co., Ltd.) for 4 weeks following the induction of DCM. The rats in the Scm-DCM group received AAV containing scrambled miR via i.v. injection. The rats in the control and DCM groups received physiological saline via i.v. injection.

After BBR and AAV intervention, all rats were sacrificed under anesthesia using 30 mg/kg pentobarbital sodium with cervical dislocation, pentobarbital sodium was only used for anesthesia prior to euthanasia. Subsequently, blood samples and mouse hearts were collected. The heart tissues were rapidly frozen in liquid nitrogen and stored at −80°C or fixed using 4% paraformaldehyde (cat. no. BL539A; Biosharp Life Sciences) at room temperature for 2 h. Humane endpoints were defined for this experiment, including the inability to access food or water, dyspnea, astasia, self-mutilation, a weight loss >15% within 5 days and death was confirmed by breath cessation.

293T cells (cat. no. 632180, Takara Bio, Inc.) were cultured in DMEM (cat. no. C11995500BT; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS. H9C2 cell (cat. no. CL-0089; Procell Life Science & Technology Co., Ltd.) suspensions containing glucose concentrations of 5.5 mmol/l (control) and 33 mmol/l high glucose (HG) were cultured at 37°C in the presence of 5% CO₂ for 48 h, as previously described (23). The cells in the culture plates were subsequently transfected with Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 24 h in accordance with the manufacturer's instructions. For this purpose, miR-18a-3p mimics (50 nmol), miR-18a-3p inhibitor (100 nmol), or the negative control (NC); all designed and synthesized by Guangzhou RiboBio Co., Ltd.) were diluted in 200 µl Opti-MEM (Thermo Fisher Scientific, Inc.). Following incubation for 20 min at room temperature, the mixture was added to the cell cultures. The sequences were as follows: miR-18a-3p mimic sense strand, 5′-ACUGCCCUAAGUGCUCCUUCU-3′ and antisense strand, 5′-UGACGGGAUUCACGAGGAAGA-3′; miR-18a-3p inhibitor, 5′-AGAAGGAGCACUUAGGGCAUG-3′; NC sense strand, 5′-UUUGUACUACACAAAAGUACUG-3′ and antisense strand, 5′-AAACAUGAUGUGUUUUCAUGAC-3′.

The cardiac tissues were fixed using 4% paraformaldehyde (cat. no. BL539A; Biosharp Life Sciences) at room temperature for 2 h and embedded in paraffin. Subsequently, the tissue samples were sectioned into 5-µm-thick slices and stained with H&E (cat. no. E607318-0200; Sangon Biotech Co., Ltd.) and Masson's (cat. no. G1346; Beijing Solarbio Science & Technology Co., Ltd.) trichrome staining at room temperature according to the manufacturer's instructions separately. The myocardial histomorphology and collagen deposition were examined using a fluorescence microscope (Cytation 5, BioTek).

Protein extraction from H9C2 cells and cardiac tissues was performed using RIPA lysis buffer (Thermo Fisher Scientific, Inc.) and 100X protease inhibitor cocktail (Thermo Fisher Scientific, Inc.). A BCA protein assay kit was used to determine the protein concentrations (cat. no. P0010S, Beyotime Institute of Biotechnology). Subsequently, 20 µg protein samples were electrophoresed on 10% SDS gels by PAGE. Following protein transfer to polyvinylidene difluoride membranes (0.22 µm; EMD Millipore) and blocking with 5% milk at room temperature for 1 h, the membranes were incubated with primary rabbit anti-β-actin (cat. no. ab49900; 1:20,000; Abcam), rabbit anti-Gapdh (1:10,000; cat. no. ab9482; Abcam), rabbit anti-Nlrp3 (1:1,000; cat. no. NBP2-67639; Novus Biologicals, LLC), mouse anti-caspase-1 (1:2,000; cat. no. NB100-56565; Novus Biologicals, LLC) and rabbit anti-Gsdmd (1:1,000; cat. no. ab209845; Abcam) at 4°C overnight. On the following day, goat with anti-rabbit antibody (1:20,000; cat. no. ab6276; Abcam) or goat anti-mouse antibody (1:20,000; cat. no. ab6789; Abcam) were applied at room temperature for 1 h. β-actin and Gapdh were used as internal controls. The band visualization was performed using enhanced chemiluminescence (cat. no. WBKLS0500, MilliporeSigma). The densitometry of immunoreactive bands was calculated by ImageJ 1.54b (National Institutes of Health).

Bioinformatics tools were used to predict the Gsdmd-targeting miRNAs (http://www.targetscan.org/, http://www.mirdb.org/, http://www.pictar.org/). Total RNA extraction from cardiac tissues and H9C2 cells was performed using TRIzol^™ reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Subsequently, total RNA was converted to cDNA using the Transcriptor First Strand cDNA Synthesis Kit (cat. no. 4897030001; Roche Diagnostics), with a mix of dT primers for mRNA and stem loop primers (Sangon Biotech Co. Ltd.) for miRNA, at 50°C for 1 h. An Faststart Universal SYBR Green (cat. no. A6001; Promega) was used to perform qPCR with the Bio-Rad CFX96 (Bio-Rad) and Real-Time PCR system. The PCR conditions were 10 min at 95°C followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. Relative quantification was calculated using the 2^−ΔΔCt method (24,25). U6 and β-actin were used as an internal control for the respective miRs and all other mRNAs. The primer sequences used are shown in Table I.

The cellular content of ROS was evaluated using a dichloro-dihydro-fluorescein diacetate (DCFH-DA) assay kit (Beyotime Institute of Biotechnology). Briefly, the H9C2 cells were seeded at a density of 2×10⁵ cells/plate in six-well plates at 37°C and incubated for 24 h. The cells were transfected in three groups as follows: The first group was transfected with miR-18a-3p mimics, the second with miR-18a-3p inhibitor, and the third with NC. Subsequently, the cells were cultured with 33 mmol/l HG, and following 48 h, 10 µmol/l DCFH-DA solution were added to the wells. The samples were incubated at 37°C for 20 min in the dark. Following this incubation, the ROS content of the cells in different groups was evaluated by measuring the fluorescence activity at 495 nm excitation wavelength and 529 nm emission wavelength in real-time (Synergy H1, BioTek).

A WT or mut-Gsdmd fragment was constructed and cloned after the luciferase reporter gene of the psiCHECK2 vector (Promega Corporation). Briefly, 293T cells were seeded in 24-well plates, incubated overnight at 37°C and transfected with Gsdmd WT-vector, Gsdmd Mut-vector by Lipofectamine 3000 (Invitrogen; Thermo Fisher Scientific, Inc.). The cells were then co-transfected with the miR-18a-3p mimic or inhibitor. The relative Firefly and Renilla luciferase activity were detected using a dual-luciferase reporter system (cat. no. #E3971; Promega Corporation) at 24 h after transfection. Firefly luciferase served as an internal reference.

The miR-18a-3p promoter fragments (−2,000, −1,500, −1,000 and −500 bp) were synthesized and cloned into the XhoI site of the pGL3 vector (Promega Corporation). 293T cells were co-transfected the constructs with the TK vector by Lipofectamine 3000 with or without BBR. The samples were incubated for 24 h at 37°C. The Firefly and Renilla luciferase activity were detected, Firefly luciferase served as an internal reference.

The supernatants were collected from all cell cultures and tissue samples, and the released inflammatory cytokine amount in each sample was evaluated using a rat IL-1β ELISA kit (cat. no. ELR-IL1b-1; Raybiotech, Inc.) according to the manufacturer's instructions. The levels of cardiac troponin-I (cTn-I) and creatine kinase-MB (CK-MB) in serum were determined using a Rat Troponin I type 3/cTn-I (cat. no. E-EL-R1253c; Elabscience Biotechnology, Inc.) and a rat CK-MB ELISA kit (cat. no. MM-61468r1; MEIMIAN; http://www.mmbio.cn/goods.php?id=18052), respectively following the manufacturer's instructions. The absorbance was measured at 450 nm (Epoch2, BioTek Instruments, Inc.).

The cell death rate (pyroptosis) was examined using calcein-AM staining. For this purpose, the cells were initially plated in a six-well plate overnight and subsequently transfected with miR-18a-3p mimics or miR-18a-3p inhibitors for 24 h. Subsequently, the cells were stimulated with HG for an additional 48 h and stained with 5 µΜ propidium iodide (PI) plus 5 µΜ calcein-AM for 20 min at 37°C. The images of non-viable and viable cells were captured by fluorescence microscopy (Cytation 5, BioTek Instruments, Inc.). PI-positive cells revealed the non-viable cells and calcein AM-stained cells represented the viable cells.

The obtained data are reported as the mean ± standard error. An unpaired Student's t-test was used for two-group comparisons and one-way analysis of variance (ANOVA) followed by the Tukey's post hoc test was used for multiple group comparisons. Analyses were performed with Prism 7 (Dotmatics). The number of replicates is outlined in the figure legends. P<0.05 was considered to indicate a statistically significant difference. All non-significant data are represented as ns throughout the manuscript.

To determine whether BBR can ameliorate myocardial injury in DCM, rats with DCM were treated with this compound (Fig. 1A). The expression levels of cardiac biomarkers, including cTn-I and CK-MB were increased in the DCM model group rats and were reduced in the rats in the BBR-treated group, indicating that BBR may exert a cardioprotective effect (Fig. 1B and C). Moreover, the pathological changes of the cardiomyocytes were assessed using H&E staining and collagen deposition was assessed using Masson's staining. The results indicated that the cardiomyocytes in the rats with DCM contained a highly increased number of vacuoles with a disordered arrangement and distortions, which indicated an extensive destruction in the myocardial structures of these models. By contrast, reduced pathological manifestations were observed in the myocardial tissue of the BBR-treated animals compared with those in the DCM group (Fig. 1D). Masson's staining indicated that collagen deposition was increased in the rats with DCM, while it was reduced in the BBR-treated rats (Fig. 1E).

To explore the effects of BBR on pyroptosis, the expression levels of pyroptosis-related proteins and inflammatory cytokines were evaluated. According to the results of western blotting for Nlrp3, caspase-1 and Gsdmd, a significant increase was observed in the expression levels of these proteins in the rats with DCM. BBR suppressed the Gsdmd levels, whereas it had no marked effect on the expression levels of Nlrp3 and caspase-1 (Fig. 1F and G). The expression levels of inflammation-related proteins were investigated using ELISA. The data indicated that the expression levels of IL-1β were increased and this increase was blocked in the BBR-treated rats (Fig. 1H). These data revealed that BBR protected cells from pyroptosis through the obstruction of Gsdmd expression and the release of IL-1β in rats with DCM. To further investigate the mechanisms through which BBR regulates Gsdmd expression, the expression levels of the precursor mRNA (pre-mRNA) and the mature mRNA (mat-mRNA) of Gsdmd were assessed. The results indicated that the expression levels of mat-mRNA of Gsdmd were significantly enhanced and inhibited by BBR (Fig. 1I). However, no significant difference was noted in the expression levels of pre-mRNA in the DCM and BBR-DCM groups (Fig. 1J), which suggested that BBR did not affect the transcription of Gsdmd. Taken together, these data indicated that the downregulation of Gsdmd expression in the BBR-DCM group may be attributed to a post-transcriptional regulation.

miRNAs have been demonstrated to regulate gene expression through a post-transcriptional mechanism of action (26). Herein, the effects of BBR were examined on Gsdmd expression by measuring the expression levels of major miRNAs targeting Gsdmd. These Gsdmd-targeting miRs were predicted using TargetScan, miRDB and PicTar prediction servers (http://www.targetscan.org/, http://www.mirdb.org/, http://www.pictar.org/). Subsequently, the results of RT-qPCR demonstrated that the expression of five miRNAs, including miR-18a-3p, miR-134-5p, miR-182, miR-485-5p and miR-3084b-3p was significantly attenuated in the H9C2 cells treated with HG (Fig. 2A). Among these, the expression of miR-18a-3p was dose-dependently upregulated by BBR (Fig. 2B), indicating that it may be crucially involved in the mechanisms of action of BBR. Specifically, the expression levels of miR-18a-3p were significantly decreased in rats with DCM in comparison with the control group, whereas they were increased in the BBR-treated rats compared to the rats with DCM (Fig. 2C).

To further evaluate the possible role of decreased miR-18a-3p levels in the increased expression of Gsdmd, miR-18a mimic and inhibitor were transfected into H9C2 cells. The results revealed that the expression of miR-18a-3p was significantly increased in the miR-18a-3p mimic group and decreased in the miR-18a-3p inhibitor group compared with that in the negative control cells (Fig. 2D). A bioinformatics analysis was applied to analyze the potential targets of miR-18a-3p, which indicated that miR-18a-3p was a potential regulator targeting the Gsdmd 3′ untranslated region (UTR) binding sites (Fig. 2E). The Gsdmd 3′UTR with wild-type (WT) or the base-pair mutant miR-18a-3p binding regions were cloned into the psiCHECK2 luciferase reporter to determine whether the miR-18a-3p binds directly to the 3′UTR of Gsdmd. The luciferase activity of the clones expressing the WT miR-18a-3p binding regions was markedly decreased when it was co-transfected with miR-18a-3p mimics and increased in the inhibitor co-infected group, but not in the mutant reporter (Fig. 2F).

In addition, RT-qPCR and western blotting confirmed that the expression levels of Gsdmd in the miR-18a-3p mimic-infected H9C2 cells were decreased as compared with Gsdmd expression in the control group. In addition, Gsdmd expression was increased following transfection of the cells with a miR-18a-3p inhibitor (Fig. 2G-I). These findings suggested that the inhibitory effects of miR-18a-3p on Gsdmd are mediated by direct interaction with its 3′UTR terminal.

Considering the regulatory role of ROS in cell growth, pyroptosis, inflammation, fibrosis and even heart failure, the total ROS levels were measured in the cells (27). The results of DCFH-DA-based fluorescence detection indicated that the ROS levels were significantly increased in the HG-treated groups, whereas their production was downregulated by transfection with miR-18a-3p mimics (Fig. 3B). The results of the viable/non-viable cell staining revealed that the percentage of HG-induced cells (red) was increased following HG treatment and it was downregulated by miR-18a-3p mimics (Fig. 3A).

To identify whether miR-18a-3p regulates Gsdmd expression, the mRNA and protein levels of Gsdmd were measured. The expression levels of Gsdmd were significantly increased in the HG-treated cells and were reduced following transfection with miR-18a-3p mimics (Fig. 3C-E). Moreover, the results of ELISA indicated that the HG-induced cells expressed high levels of IL-1β. miR-18a-3p suppressed the release of IL-1β into the supernatant of H9C2 cells treated with HG (Fig. 3F). According to these results, miR-18a-3p reduced Gsdmd expression and protected the cells from pyroptosis.

In order to identify whether miR-18a-3p can ameliorate cardiac trauma in DCM in vivo, the rats with DCM were subjected to an injection of AAV to overexpress miR-18a-3p or scrambled miR. RT-qPCR analysis revealed that miR-18a-3p was increased in the OE-18a-3p group compared with the DCM group, which indicated that the miRNA-18a-3p was successfully overexpressed in the heart tissues of the rats (Fig. 4E). The results of H&E and Masson's staining indicated that the pathological process of the heart tissues was alleviated compared with the DCM model rats; in addition, collagen deposition was reduced in the rats in the OE-miR-18a-3p group (Fig. 4A and B). Moreover, the expression levels of cTn-I and CK-MB were elevated in the rats with DCM compared with those of the control rats. miR-18a-3p overexpression markedly decreased the levels of these cardiac biomarkers (Fig. 4C and D).

Subsequently, the efficiency of miR-18a-3p in inhibiting Gsdmd expression was investigated in DCM model rats. Following miR-18a-3p overexpression in the cardiac tissue, the mRNA and protein levels of Gsdmd were decreased compared with those of the model group (Fig. 4F-H). By contrast, miR-18a-3p overexpression indicated the absence of a significant effect on caspase-1 and Nlrp3 expression levels, as compared with those of the DCM model rats (Fig. 4G). Moreover, the expression level of IL-1β was investigated in heart tissues, and it was observed that miR-18a-3p inhibited IL-1β expression (Fig. 4I). In summary, miR-18a-3p partially suppressed Gsdmd expression and IL-1β production in vivo, demonstrating that it may be considered as a promising candidate for developing targeted therapeutics for DCM.

The regulatory effect of BBR on miR-18a-3p expression was assessed by activating its promoter using a dual-luciferase reporter assay. To explore the promoter active region, the authors concentrated on the region 2,000 bp upstream of the transcription start site (TSS) of miR-18a-3p. For confirmation and identification of the core promoter region, a 2,000 bp 5′ upstream promoter sequence and multiple deleted fragments of this sequence were constructed using the pGL3-Basic vector, which was subsequently transfected into 293T cells (Fig. 5A). The luciferase activity of these cells was assessed. The results demonstrated that the luciferase activity was significantly increased in the fragment of 1,000/500 bp group compared with that of the control, which suggested that the core promoter was in the −1,000 to −500 BP region relative to the ATG start codon (Fig. 5B). The regulatory function of BBR on the miR-18a-3p promoter was assessed by a luciferase construct containing the miR-18a-3p promoter fragment (−1,000/500 bp), which was co-transfected with BBR in 293T cells. The results indicated an enhanced miR-18a-3p promoter activity at a BBR concentration-independent manner (Fig. 5C), suggesting that BBR enhanced the luciferase activity of the miR-18a-3p promoter.