YQHX (Astragali 50 g, Ginseng 10 g, Angelicae Sinensis 15 g, Chuanxiong Rhizoma 5 g and Notoginseng 6 g) used in animal experiments were purchased from Dongzhimen Hospital of Beijing University of Chinese Medicine. According to the preparation process of oral Chinese medicine decoction, a continuous reflux extraction was performed for 1 h using 10 times the amount of boiling distilled water (g/m), repeated three times. The aqueous extract was filtered and concentrated to a final volume of 100 ml, such that each milliliter of the medicine corresponds to 0.82 g of crude drug, and then administered to rat through gastric lavage at a dose of 8.2 g/kg/d, which corresponds to a daily dose for adults (20, 25–27). YQHX decoction for cell experiments was prepared by lyophilizing the supernatant from animal experiments into a powder. The powder was vortex-mixed and completely dissolved in a 37°C water bath, followed by filtration through a 0.22-μm membrane filter (Millipore, USA). The filtrate was sealed and stored at 4°C until use.

HPLC-LTQ Orbitrap MS and DDAMS2 data acquisition methods were employed to analyze the components of the aqueous extract of YQHX. 87 compounds have been identified, primarily consisting of triterpenoid saponins and flavonoids. Ginsenosides, anthracene glycosides, astragalosides, wugui glycosides, and panaxosides are included (26).

Adult male Sprague Dawley (SD) rats of SPF grade (200 ± 20 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (license number: SCXK2021-0011). Rats were acclimatized for 5 days in the barrier facility of the Animal Experiment Center at Beijing University of Chinese Medicine prior to MI surgery. Environmental conditions were maintained as follows: daily temperature fluctuation ≤4℃ (range: 20–26℃), relative humidity 40%–70%. Left anterior descending coronary artery (LAD) ligation surgery was performed for MI model rats as described previously (27–29). First, the rats were anesthetized with an intraperitoneal injection of 1% pentobarbital sodium at a dose of 40 mg/kg and shaved chest for surgical preparation. Then, rats were connected to ventilator after tracheal intubation (tidal volume of 0.6 ml, a respiratory ratio of 1:2, and a respiratory rate of 85 breaths per minute). Thoracotomy was then performed between the 3rd and 4th intercostal spaces, and LAD was ligated. Except the LAD ligation, the operation was the same procedure in Sham group. Finally, electrocardiographic (ECG) monitoring was conducted 24 h postoperatively. The rats were randomly divided into sham group (S, N = 9), and infarct group. Rats with ST segment elevation in the limb and more than 4 Q-wave numbers were confirmed the successful replication of MI model. The rats with the same Q-wave number were randomly assigned to 3 group: myocardial infarction model (M, N = 9), myocardial infarction model + YQHX (Y, N = 9), and myocardial infarction model + perindopril (P, N = 9). Administration dosage of YQHX was 8.2 g/kg/day while perindopril (Servier Co., Ltd., Tianjin, China) was administered 0.4 mg/kg/day via gavage at 9:00 am on the second day after the MI modeling, and the administration lasts for 7 and 28 days respectively as previously described (27).

H9c2 cells (Servicebio, Wuhan, China) are cultured in the H9c2 special medium (Servicebio, Wuhan, China) in a CO₂ incubator (Thermo, Scientific, Bremen, Germany). Cells were replaced with low-sugar serum-free DMEM (Invitrogen, California, USA) at the confluence of 80% and placed in a three-gas box (Thermo, Scientific, Bremen, Germany) for 24 h of culture (95% N₂, 5% CO₂, and O₂ concentration ≤1%) to prepare a model of glucose and oxygen deprivation cells. After the various concentrations of YQHX (ranging from 50 μg/mL to 1,000 μg/mL) CCK-8 assay and cell damage test, the experiment was divided into 5 groups: the control group (C), the model group (M), and Y10, Y5, Y2 groups. YQHX was diluted to concentrations of 1,000 μg/mL, 500 μg/mL, and 250 μg/mL for Y10, Y5, Y2 groups respectively. A mixture of MKT-077 (MedChemExpress, New Jersey, America) and Y5, which is medium concentration of YQHX was used for the complex research.

Echocardiographic exams were conducted at day 7 and day 28 after MI modeling (Visual Sonics Inc., Toronto, Canada). The rats were anesthetized with intraperitoneal injection of 1% sodium pentobarbital, and the chest skin was prepared in the supine position. An average of three consecutive cardiac cycles were captured for each rat. The following parameters were measured: left ventricular short axis rate (FS), left ventricular ejection fraction (EF), left ventricular end-systolic volume (LVESV), left ventricular end-diastolic volume (LVEDV) and left ventricular end-systolic diameter (LVDs), left ventricular end-diastolic diameter (LVDd).

After perfusion of the rat heart with 0.9%NaCl (Servicebio, Wuhan, China, G4702–500ML) and 4%paraformaldehyde (Coolaber, Beijing, China, SL331828400), tissues in the infarct border zone were excised into less than 1 mm³ cube and fixed in a pre-cooled 2.5% glutaraldehyde (Labcom, Fuzhou, China, 20230921) solution for 2 h. Similarly, H9c2 cells were fixed with 2.5% glutaraldehyde for 1 h, then scraped off, centrifuged at 500–800 rpm for 5 min, and fixed for 1 h. After three washes, heart tissues and H9c2 cells underwent further processing. Copper grids were stained with 2% uranyl acetate, followed by embedding in epoxy resin. The embedded samples were then heated to 60℃ for polymerization. Finally, images were captured using a transmission electron microscope (Hitachi-H-7650B, Tokyo, Japan).

Rat hearts were fixed in 4% paraformaldehyde for 24 h, followed by paraffin embedding and sectioning into 5-μm-thick slices for subsequent analyses. For general histology, tissue sections were deparaffinized, subjected to hematoxylin and eosin (HE) staining, dehydrated through graded ethanol series, cleared in xylene, and coverslipped. Masson trichrome staining was performed following the manufacturer's instructions: after completing the staining protocol, sections were rinsed with distilled water, differentiated with 1% glacial acetic acid, and permanently mounted with neutral balsam for microscopic examination. HE staining serves as a fundamental technique for evaluating myocardial pathological changes, whereas Masson trichrome staining is employed to assess myocardial fibrosis by visualizing collagen deposition.

H9c2 cells were cultured in a 96-well plate, 10% CCK-8 (Biorigin, Beijing, China) was added to assess the effects of different concentrations of YQHX and MKT-077 on cell viability. After incubating the cells with CCK8 for 2 h, the absorbance values of the samples were accurately measured at 450 nm using a microplate reader.

H9c2 cells were cultured in a 96-well plate, the Fluo-4 AM working solution, diluted at a ratio of 1:500 was added for Ca²⁺ detection (Beyotime, Shanghai, China). Briefly, the Fluo-4 AM working solution was added to the cell culture wells and incubated at 37°C for 30 min in the dark. After washing 3 times with PBS, the cells were incubated for an additional 15–30 min at 37°C under 5% CO₂. Fluorescence images were then captured using a confocal microscope.

H9c2 cells were cultured in a 96-well plate. Pre-prepared Calcein AM staining solution, fluorescence quenching working solution, and Ionomycin control solution were added separately (Beyotime, Shanghai, China). Cells were incubated at 37°C in the dark for 30 min, followed by replacement with fresh culture medium pre-warmed at 37°C to ensure sufficient hydrolysis of Calcein AM by intracellular esterases, generating green fluorescent Calcein. Then the cells were washed 3 times with PBS, and buffer was added. Images were then recorded using a fluorescence microscope. All cell experiments, including CCK-8 assay, Ca²⁺ detection, and mPTP opening assay, were performed in triplicate.

The rat heart was fixed, embedded in paraffin, and cut into thin slice. After cell membrane permeabilization and serum blocking, the following primary antibody, Mfn2 (Proteintech, Chicago, USA), IP3R2 (Santa Cruz, California, USA), GRP75 (Proteintech, Chicago, USA), VDAC1 (Affinity Biosciences, Inc., USA) were incubated overnight in the dark at 4°C. The secondary antibody (goat anti-rabbit IgG 1:200, goat anti-mouse IgG, 1:400, goat anti-rabbit IgG 1:400) were then incubated for 50 min at room temperature in the dark. After counterstaining the cell nuclei with DAPI using the DAPI staining solution, the slides were mounted with an anti-fade mounting medium. Images were recorded using a fluorescence microscope.

TRIzol reagent was used to extract total DNA used for reverse transcription from myocardial tissue. The reverse transcription system are as follows: 5× Reaction Buffer (4 μl), Oligo(dT)18 Primer (50 μM) (1 μl), RT Enzyme Mix (1 μl), RNase-free water (12 μl), and mRNA (2 μl). The reverse transcription reaction is performed in PCR instrument (25℃,300s; 42℃,1800s; and 85℃, 5s.) The sequences for the primers used are as follows: MCU: forward, 5'-CCCCTGGAGAAGGTACGGAT-3', reverse,5'-GGTGACCGGTTCCATGATGT-3', CypD: forward, 5’-CGCTTTCCTGACGAGAACTT-3', reverse,5'-ACATCCATGCCCTCTTTGAC-3', Sig-1R: forward, 5’-GCAGTGGGTGTTTGTGAACG-3', reverse,5'-GCCCAGTATCGTCCCGAATG-3', NOGO-B: forward, 5’-GAACTGAGGCGGCTTTTCTT-3', reverse,5'-TGATCTATCTGCACCTGATGCC-3', IP3R2: forward, 5’-ACCTCTGCGTGTCCAATAGC-3', reverse,5'-CATGGACACCAGCTTCGTCT-3', GRP75: forward, 5’-GACGAGGATGCCCAAGGTTC-3', reverse,5'-CAGCCAACACACCTCCTTGA-3', VDAC1: forward, 5’-AGCCTCCTCGCCGCAA-3', reverse,5'-GCACAGCCATGTTCTCGGA-3', GAPDH: forward, 5’-AGACAGCCGCATCTTCTTGT-3', reverse,5'-CTTGCCGTGGGTAGAGTCAT-3'. qPCR was performed on the Quantagene q225 PCR instrument. The qPCR denaturation involved 42 cycles (95℃, 30s; 95℃, 15s; 60℃, 30s; 75℃, 30s) GAPDH served as the internal reference in this experiment, and the relative mRNA levels were calculated using the 2-ΔΔCt method.

Marginal heart tissues were washed with pre-cooled PBS and added lysis buffer to extract the proteins. Similarly, lysis buffer was added to H9c2 cells. Both the tissue and cell samples were then assayed for protein concentration using the BCA method. The protein samples were first loaded onto gels and then transferred to membranes, blocked for 2 h. The membranes were following incubated overnight at 4°C with the primary antibodies, CypD (Affinity Biosciences, Inc., USA), MCU, Sig-1R and NOGO-B (Proteintech, Chicago, USA). GAPDH and β-actin were employed as loading controls. Following the incubation, the membranes were washed with TBST and incubated with secondary antibodies (selected based on the type of primary antibody, either goat anti-rabbit or goat anti-mouse, diluted in 3% skim milk at a ratio of 1:50,000) for 30 min. Finally, the membranes were added with the ECL mixed solution and placed in a developing tray for exposure and image capture. Grayscale values of the blots were analyzed by Image J software processing system. For each protein, Western blot and qPCR experiments were performed in triplicate.

Experimental data were analyzed using SPSS 22.0 software (IBM, Armonk, USA). Measurement data were presented as mean ± SD. Student's t-test, one-way ANOVA, or repeated-measures ANOVA were employed to compare values between groups. A statistically significant difference was determined when the P-value was less than 0.05.

Cardiac structure and function were evaluated through echocardiogram. 28 days after MI, the ejection fraction (EF) and fractional shortening (FS) were significantly reduced, while the left ventricular end-diastolic volume (LVEDV), end-systolic volume (LVESV), and left ventricular end-systolic diameter (LVDs) were significantly increased. However, both YQHX and perindopril did not show significant effects on end-diastolic diameter (LVDd) and LVEDV MI (Figures 1A,F). In S group, the cardiac appearance showed no obvious infarct area, with clear coronary vessels and appropriate heart size. While M group, a distinct pale infarct center was observed near the ligation site, accompanied by thinning and depression of the ventricular wall. Both YQHX and Perindopril group showed a smaller infarct area and clear and visible coronary vessels in the infarct border zone (Figure 1B). Similar, 7 days after MI, significant improvements in corresponding cardiac function and structure were also observed, suggesting that protective effects of both medicine on cardiac function and structure may be present during the subacute and chronic phases of MI.

HE staining is the most fundamental method for observing the pathological condition of myocardial tissue. After 28 days of MI, the myocardial cells in the S group were arranged neatly, and no significant inflammatory infiltration was observed in the extracellular matrix. Compared with the S group, the myocardial fibers in the M group were arranged chaotically, with structural ruptures and a decrease in the number of myocardial cells, accompanied by varying degrees of neutrophil infiltration. In contrast, the myocardial fibers in the Y and P groups were relatively neatly arranged, with regular cell morphology and significantly reduced inflammatory infiltration in the extracellular matrix (Figure 1C).

The degree of cardiac fibrosis was observed through masson staining. 28 days after MI, the myocardial fibers in S group mostly appeared red and were neatly arranged, with minimal collagen fibers. In contrast, the myocardial fibers in the M group were arranged chaotically, with widened fiber spaces and extensive blue stained collagen fibers in the infarcted area. Both YQHX and perindopril significantly improved the fibrosis of the myocardial tissue (Figures 1C,E). Mitochondria and endoplasmic reticulum are highly dynamic structures whose quality, morphology, localization, and composition undergo constant changes in response to the cellular needs. The regions where these two organelles exhibit physical and functional interactions are known as MAMs. Ultrastructural observations of myocardial tissue and cellular MAMs were conducted. S group exhibited neatly arranged myocardial fibers, with a small amount of scattered endoplasmic reticulum surrounding the mitochondria, and their connections were relatively loose. After MI, the myocardial fibers became disorganized, mitochondria were swollen with dilated endoplasmic reticulum surrounded, and the connections between mitochondria and endoplasmic reticulum became tighter. Following treatment with YQHX and perindopril, the myocardial fibers became more organized, mitochondrial morphology recovered, endoplasmic reticulum dilation decreased, and the connection distance between the two became relatively wider (Figure 1D). We also identified the same alteration trends of myocardial tissue, myocardial fibrosis, and mitochondrial-endoplasmic reticulum ultrastructure in the 7 days group.

Drawing from our prior research (29), 100–400 μg/mL concentrations of YQHX significantly enhanced cell viability under hypoxic conditions for 24 h. This study expanded the concentration range and observed the effects of more concentrations on H9c2 cells. Under normoxic and glucose/oxygen-deprived conditions for 24 h, different concentrations of YQHX were used to intervene with H9c2 cardiomyocytes. Under normoxic conditions for 24 h, various concentrations of YQHX (ranging from 50 μg/mL to 1,000 μg/mL) had no significant impact on the vitality of H9c2 cells. Under glucose/oxygen-deprived conditions for 24 h, all concentrations of YQHX significantly improved the vitality of H9c2 cells. Considering the progressive concentration gradients set up in the experiment, we selected the concentrations of 1,000 μg/mL, 500 μg/mL, and 250 μg/mL for further research (Figures 2C,E). Like the heart tissues, in H9c2 cells, we found the same ultrastructural changes of MAMs. C group exhibited regularly shaped and full mitochondria, smooth and flat endoplasmic reticulum, with a certain gap between them. After the cells underwent glucose and oxygen deprivation, the mitochondria became swollen, with cristae fractured or even disappearing, and the endoplasmic reticulum became dilated and expanded, with a closer arrangement with the mitochondria. In the YQHX groups (Y10, Y5, Y2), the microstructure of MAMs showed significant improvement (Figure 2D).

Calcium, serving as a second messenger, occupies a pivotal role in the pathogenesis of cardiovascular diseases. In cardiomyocytes, calcium overload can induce cell apoptosis, leading to cell death and arrhythmias after MI (30). Through the study of Ca²⁺ levels in H9c2 cells, we found that glucose and oxygen deprivation conditions significantly increased intracellular Ca²⁺ concentration, leading to Ca²⁺ overload. YQHX significantly reduced the Ca²⁺ level in cardiomyocytes, indicating improvement function of Ca²⁺ overload in cardiomyocytes and reduction of cell damage (Figures 2A,F).

The opening of mPTP can lead to dissipation of mitochondrial membrane potential, organelle swelling, and ultimately rupture. mPTP is defined as a large Ca²⁺-activated channel involved in mitochondrial damage and cell death (31). The research results showed that under normoxic conditions, mPTP was closed. Under glucose and oxygen deprivation conditions for 24 h, the opening of the mPTP in H9c2 cells was evident, leading to a significant decrease in fluorescence intensity. YQHX significantly improved the excessive opening of mPTP in H9c2 cells, thereby alleviating intracellular calcium overload (Figures 2B,G).

The mitochondrial fusion protein 2 (Mfn2) is not only targeted to mitochondria but also localized to the mitochondrial-associated endoplasmic reticulum membranes, regulating Ca²⁺ transport from the ER to mitochondria (32, 33). Both in vivo and in vitro, our experiments found that Mfn2 levels significantly increased under ischemic and glucose- and oxygen-deprivation conditions, while YQHX significantly improved the upregulation of Mfn2, thereby reducing the connections between MAMs (Figures 3A,B,E,F). Concurrently, in vivo, myocardial ischemia led to a significant increase in the calcium regulatory proteins cyclophilin D (CypD) and mitochondrial Ca²⁺ uniporter (MCU) on MAMs, while sigma-1 receptor (Sig-1R) and neurite outgrowth inhibitor B (NOGO-B) were significantly reduced. Both YQHX and perindopril reversed this state (Figures 3D,G). Similarly, in vitro, we observed the same results, indicating that YQHX may improve intracellular calcium overload by regulating relevant proteins on MAMs (Figures 3C,H).

GRP75, as a member of the HSP-70 family, can be inhibited by specific inhibitors of HSP70 (34). It has been previously shown that rhodacyanine dye MKT-077, a GRP75-specific inhibitor, binds to the nucleotide-binding domain of mortalin/GRP75, inducing tertiary structural changes that inactivate the ATPase activity of mortalin/GRP75 (35) preventing the calcium overload of nerve cell mitochondria in the oxygen and glucose deprivation model, leading to improved mitochondrial structure and function (36, 37). Notably, the GRP75 protein within the IP3R2-GRP75-VDAC1 complex belongs to the HSP-70 family and can also be inhibited by MKT-077. Given that the research application of MKT-077 in H9c2 cells is still limited, we first conducted concentration screenings using 5 μM and 10 μM of MKT-077 based on previous experimental studies (38, 39). Under both normoxia and glucose- and oxygen-deprivation conditions for 24 h, 5 μM/mL MKT-077 had no significant impact on the vitality of H9c2 cardiomyocytes. However, the vitality after intervention with 10 μM/mL MKT-077 was significantly lower compared to C and 5 μM/mL group, indicating that 10 μM/mL MKT-077 exhibits a certain degree of cytotoxicity towards H9c2 cardiomyocytes. Therefore, a concentration of 5 μM/mL was selected for the subsequent experiments (Figure 4E).

Compared with C group, the intracellular Ca²⁺ level in M group was significantly elevated. However, compared with M group, the intracellular Ca²⁺ levels in both Y5MKT and Y5 group were significantly reduced. This suggests that after GRP75 is blocked, the connection of the IP3R2-GRP75-VDAC1 complex is weakened, leading to a decrease in Ca²⁺ flow on the MAMs (Figure 4B).

The results of co-immunoprecipitations indicates that immunoprecipitation of both IP3R2 and VDAC leads to the co-precipitation of GRP75, and immunoprecipitation of GRP75 results in the co-precipitation of VDAC and IP3R2, suggesting that GRP75 plays a central role in the formation of the complex (40). Co-localization of IP3R2, GRP75, and VDAC1 was observed in C, M, and Y5 groups. However, compared to the first three groups, the co-localization of IP3R2 and GRP75, as well as IP3R2 and VDAC1, was significantly reduced in the Y5MKT group. (Figures 4A,D).

in vivo, the gene and protein expressions of IP3R2, GRP75, and VDAC1 were significantly upregulated 7 and 28 days after MI, while both YQHX and perindopril reversed this upregulation. Similarly, in vitro, YQHX significantly downregulated the high expressions of IP3R2, GRP75, and VDAC1 in H9c2 cells under glucose- and oxygen-deprivation conditions, and the decrease was even more significant in the Y5MKT group compared to the Y5 group, further confirming the blocking effect of MKT-077 on the complex (Figures 4C,F).