The rat embryonic cardiomyocyte line (H9c2) was obtained from the Cell Bank of China Science Academy and cultured in high glucose Dulbecco's modified Eagle's medium (DMEM, Gibco, MA, USA) containing 10% fetal bovine serum (FBS, HyClone, UT, USA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Solarbio, China) at condition of 37°C with 5% CO₂. The H9c2 cells were used for the experimental procedures at 60–70% confluence.

Maslinic acid was purchased from Med Chem Express (purity ≥ 98%, MCE, NJ, USA) and dissolved in dimethyl sulfoxide (DMSO). At confluence, the H9c2 cells were treated with varying concentrations of commercially available MA (5–100 μM), prepared in DMEM medium for 24 h before H/R stimulation. In the untreated control groups, the H9c2 cells were pretreated with the same amount of DMSO as that in the treated groups.

The H/R model was established as previously described (23, 24). Briefly, the cells were cultured in glucose-free DMEM and then put into a near-anaerobic atmosphere (5% CO₂ and <0.1% O₂) produced using an Anaero Pack (Mitsubishi Gas Company, Tokyo, Japan). Followed by incubation in hypoxic conditions at 37°C for 4 h, the cells were cultured in normoxic condition (reoxygenation) for another 12 h. The cells were randomly assigned into the control group, H/R group, H/R+MA5 (H9c2 cells pretreated with 5 μM MA and subjected to H/R procedure) group, H/R+MA10 (the H9c2 cells pretreated with 10 μM MA and subjected to H/R procedure) group, and H/R+MA20 (the H9c2 cells pretreated with 20 μM MA and subjected to H/R procedure).

The experiments were approved by the Ethics Committee for Animal Experimental Center of Nankai University, Tianjin, China. All the experimental protocols were performed in compliance with the US National Institutes of Health guidelines. Fifty-five healthy adult Sprague Dawleys (SD) rats [weighing 220 ± 10 g, Animal License: SCXK (Hubei) 2017-0012, Center for Animal Experiment of China Three Gorges University, Hubei, China] were housed in a temperature-controlled room with 50% humidity under 12 h light/dark cycles, and free access to food and water. All the efforts were made to minimize the animal suffering and reduce the number of animals used.

Prior to myocardial ischemia/reperfusion (I/R) operation, the rats were treated with varying concentrations of MA (5, 10, and 20 mg/kg) or vehicle (the same amount of DMSO dissolved in normal saline) by intraperitoneal injection for two consecutive days. The rats were randomly assigned to the following group (=11 per group): (1) Sham group: sham-operation, (2) I/R group: myocardial ischemia/reperfusion-operation, (3) I/R + MA5: administered 5 mg/kg MA before I/R, (4) I/R + MA10: administered 10 mg/kg MA before I/R, and (5) I/R + MA20: administered 20 mg/kg MA before I/R.

An MI/RI model was established as described previously (6). The rats were anesthetized by intraperitoneal injection of 3% sodium pentobarbital (40 mg/kg), and then ventilated with room air using a small animal ventilator. Meanwhile, the limbs of rats were connected to ECG machine (VE-300, China). Then, the hearts were exposed gradually through a left thoracotomy at the fourth intercostal space and a 6-0 silk suture over a 1-mm medical latex tube was placed around the origin of left anterior descending coronary artery (LAD). After 30 min of ischemia, the suture was cut, allowing the myocardium to be reperfused for 3 h. The Sham group was subjected to the same surgical procedure without blood occlusion of LAD. At the end of reperfusion, the rats were anesthetized and euthanized in a CO₂ chamber, and the heart and blood specimens were collected.

The viability of H9c2 cells was measured as per the instructions of Cell Counting Kit-8 (CCK-8) assay (Dojindo, Japan). The H9c2 cells were seeded in 96-well plates at a density of 5,000 cells per well. After proper treatment, the CCK-8 reagents (10 μl/well) were added into each well and incubated for additional 2 h, and then the optical density (OD) values were measured at 450 nm by a microplate spectrophotometer.

The cell apoptosis was determined by Annexin V-PE/7-AAD (Becton Dickinson Co, NJ, USA) according to the instructions of the manufacturer. Approximately, 1 × 10⁵ H9c2 cells were placed on six-well plates. Followed by the two washes with PBS, the H9c2 cells were collected and analyzed using BD Accuri® C6 Plus Flow Cytometry (BD Bioscience, NJ, USA).

The levels of cardiac enzymes, such as lactate dehydrogenase (LDH) and creatine kinase isoenzyme-MB (CK-MB) in serum and culture medium, were detected by the ADVIA2400 automatic biochemical analyzer (SIEMENS, Germany).

After various treatments, the levels of IL-1β, (IL-6) and TNF-α in the serum and cell culture supernate were detected by specific enzyme-linked immunosorbent assay ELISA kits (R&D Systems, MN, USA) based on the instructions of the manufacturer. The value of absorbance was detected using a microplate spectrophotometer at 450 nm. Then, the concentrations of cytokines were calculated by reference to the standard curves.

The H9c2 cells were fixed with 4% paraformaldehyde for 15 min, blocked with 0.3% Triton X-100, incubated with the anti-NF-κB p65 antibody (CST, 1:500) at 4°C overnight, and subsequently incubated with a Cy3 secondary antibodies for 2 h at room temperature in the dark. After incubation, the cells were washed three times in PBS, and stained with 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, China) for nucleus identification. The images were captured using a fluorescence microscope (Leica DMi8, Germany). The data illustrated were representative of at least three independent repeats. The relative fluorescence intensity in the nucleus (NF-κB) was analyzed by ImageJ software.

The myocardial tissue was collected and fixed with 4% paraformaldehyde and embedded in paraffin. Subsequently, all the tissue samples were cut into 6-μm sections and stained with H&E as previously described (25). Then, the pathological changes of myocardial tissue were inspected under a light microscope.

The apoptosis of cardiomyocytes in tissue sections was assessed by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining (Beyotime, China) according to the instructions of the manufacturer. Briefly, a double staining technique was used: TUNEL-positive cells displayed green fluorescence and the nucleus of all cells were stained with DAPI which produced blue fluorescence. The apoptotic index was determined by the equation: TUNEL-positive cells/ DAPI positive cells × 100%.

The myocardial infarct size was evaluated by 2,3,5-triphenyltetrazolium chloride (TTC, Sigma-Aldrich, MO, USA) staining as previously described (6). The hearts of rats were harvested immediately after 3 h reperfusion, rinsed with saline, and frozen at −20°C for 20 min. Then, the hearts were crosscut into 1.5 mm thick slices and incubated in 1.5% TTC solution at 37°C in the dark for 15 min. Afterward, the slices were taken out, fixed with 4% paraformaldehyde overnight, and photographed. The proportion of the infarcted area (white) was obtained via Image-Pro Plus 7.0 software (Media Cybernetics, Inc., MD, USA).

Total proteins from the H9c2 cells or rat cardiac tissues were extracted using radioimmunoprecipitation (RIPA) buffer (Beyotime, China). The protein concentration was assessed by Bicinchoninic Protein Assay Kit (Beyotime, China). All extracted proteins were boiled at 95°C in loading buffer for 5 min. Identical quantities of protein were divided by 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with the voltage from 80–120 V, and then, transferred to PVDF membranes (200 mA). Afterward, the membranes were hindered with 5% skimmed milk for 2 h and hatched overnight at 4°C with the diluted primary antibodies for Bax (1:1000, Proteintech Group, China), Bcl-2 (1:1000, Proteintech Group, China), caspase-3 (1:1000, CST, MA, USA), HMGB1 (1:1000, CST, MA, USA), IκBα (1:1000, CST, MA, USA), P-IκBα (1:1000, CST, MA, USA), p65 (1:1000, CST, MA, USA), P-p65 (1:1000, CST, MA, USA), MyD88 (1:1000, CST, MA, USA), TLR4 (1:1000, Proteintech Group, China), GAPDH (1:5000, Proteintech Group, China), and β-actin (1:2000 Proteintech Group, China). After three tris-buffered saline tween (TBST) washes, the horseradish peroxidase-conjugated secondary antibodies (CST, MA, USA) were used for hatching the membranes for 1.5 h. Subsequently, an enhanced chemiluminescence detection system was used for the analysis of protein bands and gray intensity analysis was quantified using Image J software.

The binding sites of MA into HMGB1 protein molecular was achieved using AutoDockTools-1.5.6 and Python-2.4 software. Information regarding MA and HMGB1 protein was downloaded from the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database (https://old.tcmspe.com/tcmsp.php) and the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (https://www.pdbus.org/) respectively, which were saved in mol2 or pdb format. Prior to docking, the ligand and acceptor structures were preprocessed through the deleting water molecules, adding hydrogens, and Gasteiger charge to create. Then, AutoGrid program was used to construct the grid maps and AutoDock needs pre-calculated grid maps for performing the docking calculations fast. The final pose was determined based on the lowest binding energy. The results were visualized by PyMOL 3.8 software, and the hydrogen bonds as well as their binding sites were observed and analyzed.

To assess for the effect of HMGB1 on NF-κB activation by MA, a verification experiment using rHMGB1 (R&D Systems, MN, USA) was conducted. After pretreated with 20 μM MA, the H9c2 cells were treated with recombinant HMGB1 (rHMGB 1) protein (200 ng/ml) or vehicle for 2 h and then, subjected to H/R procedure. The application concentration of rHMGB1 was determined according to the previously published articles (26, 27). After various treatments, the supernatants and cells were collected to detect the concentration of inflammatory factors by ELISA and the level of apoptotic-related proteins by western blot.

All data were expressed as the mean ± SD, and analyzed using SPSS 20.0 software (IBM Corp., NY, USA). One–way ANOVA followed by the least significant difference (LSD) post-hoc test was used to analyze the differences between the groups. The value of p < 0.05 was considered statistically significant. The histograms were graphed using GraphPad Prism 7.00 software (San Diego, CA, USA).

The chemical structure of MA is shown in Figure 1A. The cytotoxicity effects of MA on the H9c2 cells were detected at a wide range of concentrations (0, 5, 10, 20, 50, and 100 μM) for 24 h using the CCK-8 assay. As shown in Figure 1B, MA treatment led to a significant decrease in cell viability when the concentration of MA reached 50 μM (P < 0.05) and 100 μM (P < 0.01). Thus, 5, 10, and 20 μM MA were utilized for the following experiments. It is well known that H/R, a reliable model for in vitro to mimic I/R, is widely used in the MI/RI research and the development of cardiovascular drugs. The H/R model of the cardiomyocytes can be directly observed the morphological structure, physiological, and biochemical indexes of the injured cardiomyocytes (24, 28). Then, the H9c2 cells were pre-incubated with MA for 24 h and subjected to H/R.

The cell morphology was detected by a light microscope. After H/R injury, the H9c2 cells lost their elongated spindle-shape morphology, exhibiting round, shrunken, and typical apoptotic morphologies, such as blebbing, apoptotic bodies, and detachment, while 5, 10, and 20 μM of MA treatment could attenuate the above changes in the cell morphologies (Figure 1C). Based on the above results, we subsequently proposed to explore the effects of MA on the cell viability and LDH release after H/R injury. The CCK-8 assay results showed that the cell viability was significantly decreased in the H/R group, compared with the control group (Figure 1D). Conversely, MA treatment increased the cell viability after H/R injury (5, 10, and 20 μM of MA vs. H/R: 70.5 ± 5.8%, 74.7 ± 8.9%, 79.4 ± 5.9% vs. 58.2 ± 5.8%), showing a dose-dependent relationship (Figure 1D). Additionally, these results were confirmed by LDH release assay. As shown in Figure 1E, the level of LDH in the H/R group was markedly increased relative to the control group, whereas MA treatment (5, 10, and 20 μM) in a dose-dependent manner strongly reduced the H/R-induced LDH release. Thus, these results indicated that MA attenuated H/R-induced cardiomyocyte injury.

In our previous study, we confirmed that apoptosis contributes to cardiomyocyte death during the MI/RI (6). So there and then, we measured the effects of MA on H/R-induced apoptosis by flow cytometric analysis. As shown in Figures 2A,B, the percentage of apoptotic H9c2 cells was obviously increased in H/R group, compared with the control group (2.8 ± 0.4% vs. 41.2 ± 4.9%). Instead, 5, 10, and 20 μM MA treatment significantly reduced the apoptosis rate to 26.3 ± 5.3, 17.7 ± 5.4, and 15.9 ± 5.8%, respectively. Next, we further examined the expressions of several apoptosis-related proteins by western blotting analysis (Figures 2C–F). Compared with the control group, the cleaved caspase-3 and Bax protein expression were significantly increased, while the Bcl-2 protein expression was obviously decreased in the H/R group. Conversely, MA treatment (5, 10, and 20 μM) obviously reduced the levels of cleaved caspase-3 and Bax expression, while increased Bcl-2 expression in a dose-dependent manner after H/R.

To determine whether MA could inhibit the inflammatory cytokines, we analyzed the expression of IL-1β, IL-6, and TNF-α in cell supernatants during H/R. The results from the ELISA analyses (Figures 3A–C) showed that the levels of IL-1β, IL-6, and TNF-α were significantly increased in the H/R group, relative to the control group. While compared with the H/R group. MA treatment (5, 10, and 20 μM) markedly reduced the levels of IL-1β, IL-6, and TNF-α in a dose-dependent manner.

An NF-κB, a nuclear transcription factor, acts as an important mediator of apoptosis and inflammatory response (29). We performed the immunofluorescence staining to observe the nuclear translocation of NF-κB p65. As shown in Figures 3D,E, H/R treatment obviously promoted the translocation of p65 into nucleus compared with the control group, and administration of MA (5, 10, and 20 μM) significantly blocked the nuclear translocation of p65 after H/R. All the above results suggested that MA could inhibit the H/R-induced cardiomyocyte apoptosis and inflammation, which was related to NF-κB transcription.

To determine the mechanism by which MA suppressed the inflammation and apoptosis in MI/RI, we analyzed the effects of MA on the signaling pathways involving HMGB1, a vital mediator in the pathophysiology of the MI/RI through TLR4-MyD88-NF-κB mediated signaling, and may represent an important therapeutic target (12, 15). As shown in Figures 4A–F, the protein levels of HMGB1, TLR4, MyD88, and the phosphorylation of NF-κB p65 and IκBα were markedly upregulated in the H/R group compared with those in the control group. However, 5, 10, and 20 μM of MA treatment markedly suppressed the protein expression of HMGB1, TLR4, MyD88, and the phosphorylation of NF-κB p65 and IκBα in a dose-dependent manner after H/R. Therefore, our results suggested that MA could inhibit the H/R-induced activation of HMGB1-TLR4-MyD88-NF-κB signaling pathway.

Through the molecular docking experiments, it was found that MA and HMGB1 were docked in the predicted binding sites. The docking simulations yielded ideal binding conformation with the lowest binding free energy of −8.95 kcal/mol and formed two hydrogen bonds (Figure 5A). For further confirming that MA attenuated H/R-induced inflammation and apoptosis via regulating HMGB1, a verification experiment using rHMGB1 was conducted. As shown in Figures 5B–D, 20 μM of MA treatment significantly inhibited the release of proinflammatory factors during H/R, such as IL-1β, IL-6, and TNF-α, related to the H/R group. However, co-incubation with rHMGB1 abolished the inhibitory effects of MA on the release of IL-6, IL-1β, and TNF-α. As the same time, compared with the H/R group, 20 μM of MA treatment also markedly suppressed the level of cleaved caspase-3 and Bax expression, while these aforementioned effects of MA were reversed by the rHMGB1 treatment (Figures 5E–H). These results indicated that the cardioprotective effect of MA on inflammation and apoptosis after H/R might, at least partly, be due to the inhibition of HMGB1 axis.

An MI/RI rat model was established to further confirm the effect of MA in vivo. As shown in Figures 6A,B, the level of LDH and CK-MB in serum was dramatically increased after MI/RI, compared with the sham group. As expected, MA treatment (5, 10, and 20 mg/kg) obviously decreased I/R injury-aroused the release of LDH and CK-MB in a dose-dependent manner. Then, we examined the myocardial infarct size using TTC staining (Figures 6C,D). The results showed that the infarct size was significantly increased to 44.7 ± 5.0% in I/R group, whereas MA treatment (5, 10, and 20 mg/kg) significantly reduced I/R injury-caused infarct size to 27.5 ± 3.6, 24.5 ± 3.0, and 16.4 ± 3.3%, respectively, showing a dose-dependent relationship.

To further examine the role of MA on I/R-induce inflammation in vivo, H&E staining and ELISA analysis was performed. As shown in Figures 6E,F, compared with Sham group, I/R obviously increased the infiltration of inflammatory cells into the injured myocardium. However, 5, 10, and 20 mg/kg of MA treatment significantly decreased the infiltration of inflammatory cells in I/R myocardium. Meanwhile, we also measured the concentrations of pro-inflammatory cytokine IL-1β, IL-6 and TNF-α in serum (Figures 6G–I). The results showed that the levels of IL-1β, IL-6 and TNF-α were dramatically elevated in serum after MI/RI relative to the sham group. As expected, MA treatment (5, 10, and 20 mg/kg) obviously reduced the release of IL-1β, IL-6, and TNF-α in serum after MI/RI, which was consistent with the results in vitro. These data suggested that MA preconditioning attenuated I/R-induced myocardial injury and inflammation.

Next, to further examine the role of MA on I/R-induce apoptosis in vivo, we used TUNEL staining and detected the expression of apoptosis-related proteins. As shown in Figures 7A,B, the apoptotic cells were increased to 46.4 ± 8.5% in the I/R group compared with the sham group, while 5, 10, and 20 mg/kg of MA treatment significantly reduced the number of apoptotic cells to 33.5 ± 8.4%, 25.1 ± 5.5%, and 15.6 ± 3.6% in I/R myocardium, showing a dose-dependent relationship. Moreover, the results of the western blot (Figures 7C–F) indicated that I/R markedly increased the expression of Bax, and cleaved caspase-3 while decreased the Bcl-2 expression. As expected, MA treatment (5, 10, and 20 mg/kg) obviously reduced the levels of Bax and cleaved caspase-3 expression, while elevated the level of Bcl-2 expression (Figures 7C–F). Above results showed a similar trend to that obtained in the H/R H9c2 cells. Thus, these results indicated that MA preconditioning attenuated I/R-induced myocardial injury by inhibiting myocyte apoptosis in vivo.