Mice were male and at age of 8 weeks old. All male C57BL/6 were from Jiangsu Animal Experimental for Medical and Pharmaceutical Research Center. The in vivo experiments were approval of Institutional Animal Care and Use Committees at the University of Nanjing Medical University (Number: 1912034) and carried out in accordance with the guidelines. Mice were kept with 12 h light/12 h dark cycle at 23 ± 3°C and 30–70% humidity.

Infarction mouse model was generated through ligature of the left anterior descending (LAD) as described previously with male mice (11). Briefly, mice were anesthetized with 1.5–2% isoflurane, and then a left thoracotomy and pericardiotomy was performed. The left anterior descending was completely occluded by a knot with 7–0 non-absorbable surgical suture. After the ribs and skin were closed with 5–0 silk, the mouse remained under anesthesia until succinylcholine was completely metabolized. For the procedure of sham surgery, mice underwent the same procedure without the ligation of the artery.

Dioscin (80 mg/kg/day, Di'ao group, Chengdu, China) was administrated with mice for 2 weeks beginning on the day of operation. Briefly, the dioscin was first dissolved in dimethyl sulphoxide (DMSO) and then added 5% sodium carboxymethyl cellulose (CMC-Na) to make the volume ratio for 1:19. Consequently, the mice were given the dissolution by intragastric administration for continuous 14 days after myocardial infarction 1 day. The treatment groups were given dioscin solution and the control groups were given same volume of 5% CMC-Na vehicle solution. The preparation and application of the drug's dissolution should be completed in 1 day.

Transthoracic echocardiographic analysis was performed 2 weeks after the sham or MI surgery as previously described (13). Echocardiography on unconscious mice was used with a Vevo 2100 system (VisualSonics). The echocardiography was blinded to the mice in this study.

Hematoxylin and eosin and Masson trichrome staining were performed as described previously (14). The whole hearts were collected and fixed in 4% paraformaldehyde, embedded in paraffin, and then cut into 5-μm serial sections. Image analysis and quantification of fibrosis was performed by a blinded investigator using Image J software.

The purification of cardiac mitochondrial fraction from the heart tissues or cultured primary cardiomyocytes was followed as previously described (15), using Mitochondrial Extraction Kit (Solarbio Life Sciences, SM0020) according to the manufacturer's instructions.

Dihydroethidium (DHE) staining (Beyotime, China) was performed as described previously (16). Briefly, heart samples were embedded in OCT compound and transverse sectioned at 5 μm thickness to perform staining and then conserved at −80°C. Next, heart sections were set in room temperature for 10 min and then washing by PBS three times for 5 min. Further, the sections were incubated with the DHE (1:1,000) for 30 min in wet box at 37°C. Finally, after washing by PBS three times for 5 min, co-stain with DAPI. The imagines are photographed by confocal, and we use the same intensity of laser.

The activity of glutathione peroxidase (GPx), superoxide dismutase (SOD), glutathione-S-transferase (GST), catalase (CAT), glutathione (GSH) was determined in cardiac tissues, using Micro Superoxide Dismutase (SOD) Assay Kit (Solarbio Life Sciences, BC0175, BC0205, BC1195, BC0355, and BC1175) according to the manufacturer's instructions.

Effect of dioscin on MI-induced mitochondrial damage was measured by detection of the activities of mitochondrial enzymes such as malate dehydrogenase (MDH), isocitrate dehydrogenase (ICDH), and succinate dehydrogenase (SDH) using Micro Isocitrate Dehydrogenase Mitochondrial (ICDHm) Assay Kit (Solarbio Life Sciences, BC2165), Micro Succinate Dehydrogenase (SDH) Assay Kit (Solarbio Life Sciences, BC0955) and Micro NADP-Malate Dehydrogenase (NAD-MDH) Assay Kit (Solarbio Life Sciences, BC1055) according to the manufacturer's instructions.

The nicotinamide adenine dinucleotide dehydrogenase (NADH) activity was determined in cardiac tissues as reported (17), using Micro Nicotinamide Adenine Dinucleotide Dehydrogenase (NADH) Assay Kit (Solarbio Life Sciences, BC0635) according to the manufacturer's instructions.

The cytochrome c-oxidase (NCR) activity was determined in cardiac tissues, using Micro NADPH-cytochrome C Reductase Assay Kit (Solarbio Life Sciences, BC2725) according to the manufacturer's instructions.

H9C2 cells were cultured in DMEM (Servicebio) medium with 10% FBS (Gibco, Life Technologies) and 1% penicillin/streptomycin (Beyotime). Final concentration of 100 μM CoCl₂ (Sigma-Aldrich) was added to simulate hypoxia condition. After 24 h of treatment, the extraction of mRNA was performed. And after 48 h, the corresponding staining was performed.

Total RNA was extracted using RNeasy Mini Kit (Qiagen). cDNA was generated using HiScript III 1st Strand cDNA Synthesis (Vazyme) and then qPCR was performed using on QuantStudio 5 applied biosystems (Thermo Fisher Scientific) with Hifair III One Step RT-qPCR SYBR Green Kit (Yeasen), all according to the manufacturer's instructions. Actin was regard as control and the data analysis was using 2CT. Obtaining form three independent experiments, and each sample was repeated duplicated in each experiment. Primer sequence are listed below.

For mouse qPCR:

Anp: TTCTTCCTCGTCTTGGCCTTT, GACCTCATCTTCTACCGGCATCT;

Bnp: CACCGCTGGGAGGTCACT, GTGAGGCCTTGGTCCTTCAAGGTCACT;

β-mhc: ATGTGCCGGACCTTGGAA, CCTCGGGTTAGCTGAGAGATCA;

Bax: CCCGAGAGGTCTTTTTCC, GCCTTGAGCACCAGTTTG;

Bcl₂: CCTGGCTGTCTCTGAAGACC, CTCACTTGTGGCCCAGGTAT;

Actin: CATCGTCCACCGCAAATG, CACCTTCACCGTTCCAGTT.

Mitochondrial membrane potential was measured by the method of fluorescent probe, using the MitoTracker® Green FM (Yeasen, 40742ES50) according to the manufacturer's instructions.

TUNEL staining was performed as previously report (18). The suitable density H9C2 cells were induced to CoCl₂ (100 uM, Sigma, 449776) under the condition of treatment with dioscin (5 mg/L) for 48 h, compared to control group. Further, the H9C2 cells were fixed in 4% paraformaldehyde and stained with One Step TUNEL Apoptosis Assay Kit (Beyotime, C1086). TUNEL positive were counted from different images per sample, and the averages were presented.

The cysteine-requiring aspartate protease 1 (Caspase-1) activity was determined in cardiac tissues, using Caspase-1 Reductase Assay Kit (Hypertime, China) according to the manufacturer's instructions.

Data were presented as mean ± standard error of the mean (SEM) using GraphPad Prism 8. Two-tailed unpaired t-test were used to determine the differences between two groups, and two-ANOVA was employed for comparisons of the differences between multiple groups. Statistical significance is indicated by ^* p < 0.05.

Mitochondrial dysfunction is one of the primary causes for cardiac death in response to infarction. Firstly, we performed in vivo experiment to assess the mitochondrial injury after dioscin treatment in MI heart tissues. The protein level of Pink1 revealed that dioscin recovered the expression of Pink1 after MI, indicating dioscin has a protective role in mitochondria (Figure 1A). Then, the protective effect of dioscin was evaluated on mitochondrial membrane potential. The disturbance of the mitochondrial membrane potential (MMP) caused by CoCl₂, and this injury was mitigated by dioscin (Figure 1B). Considering the impaired mitochondria could lead to cell death, the dead cells were detected by using TUNEL assay. This showed that dioscin recovered the numbers of H9C2 cells in response to hypoxic condition (Figure 1C). Additionally, the mRNA levels of Bax and Bcl₂ as apoptotic markers were detected in hypoxic H9C2 cells after dioscin treatment. The results showed that dioscin exerts anti-apoptotic effects and prevents cells from death by downregulating Bax expression and increasing Bcl₂ expression (Figure 1D). Furthermore, the activities of cysteine-requiring aspartate protease 1 (caspase-1) was repressed in hypoxic cells after dioscin incubation, suggesting that dioscin repressed apoptosis caused by hypoxia (Figure 1E). These data suggest that dioscin plays protective roles against hypoxia induced mitochondrial injury to mitigate cellular apoptosis.

To determine the therapeutic role of dioscin in vivo, mice were subjected to MI surgery. Two weeks after dioscin administration, the echocardiography was performed to monitor cardiac function. Representative echocardiograph showed that the infarcted mice exhibited left chamber dilation, and dioscin administration alleviated the cardiac deformation (Figure 2A). Then, cardiac function analysis showed that the left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS) were recovered after administration of dioscin in infarcted mice (Figure 2B). Additionally, haematoxylin and eosin (HE) staining of heart sections from mice in each group showed that the normal arrangement of cells was destroyed in infarcted heart increased, which could be prevented by dioscin (Figure 2C). Representative images exhibited that less collagen deposition after dioscin treatment by the size of blue region compared to MI group, which was decreased about 50% (Figure 2D), Suggesting dioscin alleviated cardiac fibrosis in infarcted hearts. Additionally, the expression level of heart failure markers was detected in hearts, including natriuretic peptide A (Anp), natriuretic peptide B (Bnp) and beta-myosin heavy chain (β-Mhc). It showed that the increased levels of Anp, Bnp and β-Mhc mRNA levels in heart tissues from infarcted mice were suppressed by dioscin administration (Figure 2E). These results indicate that dioscin administration preserved the cardiac function by alleviating cardiac fibrosis after MI.

Considered the recovery of cardiac function occurs in infarcted hearts after disocin administration, cardiac remodeling was measured. There is no significant difference of heart rate among these four groups (Figure 3A). It was consistent with previous report that infarction increases left ventricular mass. Such increase was significantly repressed after administration of dioscin (Figure 3B). Other cardiac structure parameters, such as left ventricular posterior wall end-diastolic thickness (LVPW; d), left ventricular inner diameter end-diastolic thickness (LVID; d), left ventricular septum (IVS), and left ventricular end-systolic volume (LV Vol; d) were also sufficiently persevered after dioscin administration in infarcted hearts (Figures 3C–F). Additionally, mice subjected to MI showed a significant increase in heart weight/body weight (HW/BW) ratio and heart weight/tibia length (HW/TL), which could be ameliorated by dioscin e (Figures 3G,H). These results demonstrate that dioscin attenuated MI-induced heart remodeling.

The impaired function of creatine kinase and decreased mitochondrial respiration are often observed in myocardium after pathological injury, including myocardial infarction and hypertrophy (19, 20). In attempt to explore the effect of dioscin in the process of energy metabolism, the activities of mitochondrial enzymes were measured in mice hearts. The activities of isocitrate dehydrogenase (ICDH), succinate dehydrogenase (SDH) and malate dehydrogenase (MDH) in mitochondria were significantly downregulated in the infarcted hearts compared to the sham group (Figures 4A–C). The activities of these enzymes were remarkable enhanced by treatment with dioscin in MI group (Figures 4A–C). These proved the beneficial role of dioscin on Kreb's cycle enzymes, which are the major source of ATP production in the cells. We further examined the activities of respiratory chain enzyme responsible for the synthesis of ATP. The activities of nicotinamide adenine dinucleotide dehydrogenase (NADH) (the Complex I) and cytochrome c-oxidase (the Complex III) were significantly reduced in the mitochondrial isolated from cardiac tissues obtained from the mice subjected to MI. Conversely, the decline was inhibited by treatment with dioscin (Figures 4D,E). Overall, these findings suggest that dioscin promotes the activities of mitochondrial Kreb's cycle in myocardial infarction.

In the infarcted hearts, subcellular organelles undergo stress caused by the released of reactive oxygen species (ROS) from mitochondria. Hence, prevention of ROS accumulation in the subcellular compartments are essential for the restoration of the normal cardiomyocyte function. Representative images of dihydroethidium (DHE) staining showed that dioscin attenuated the increase of ROS in response to MI surgery (Figure 5A). We further detected the expression of antioxidant enzymes in mitochondria. Consistently, the activities of superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPx) and glutathione-S-transferase (GST) were declined in the animals subjected to MI, but recovred after dioscin treatment (Figures 5B–E). The decrease in the endogenous antioxidant defense enzyme system was corroborated to a diminished glutathione (GSH) content (Figure 5F). Similarly, the ROS accumulation was measured in H9C2 cells. Consistent with in vivo experiments, dioscin suppressed the enhancement of ROS caused by CoCl₂ (Figure 5G). These results confirmed that dioscin abrogated the oxidative stress, enhancing the endogenous antioxidant defense system. Taken together, our study demonstrates that dioscin protects heart dysfunction against myocardial infarction by suppressing the oxidative stress.