Utilize the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP, https://old.tcmsp-e.com/index.php) to search for the chemical components of Astragalus membranaceus (Huangqi), Salvia miltiorrhiza Bunge (Danshen), Ligusticum striatum (Chuanxiong), Agastache rugosa (Huoxiang), and Coptis chinensis (Huanglian) in QXJYG. Based on pharmacokinetic properties such as absorption, distribution, metabolism, and excretion, screen for active ingredients using the criteria of oral bioavailability (OB) >30% and drug likeness (DL) >0.18. Obtain the isomeric SMILES of QXJYG-related active ingredients from the PubChem website (https://pubchem.ncbi.nlm.nih.gov/). Use the obtained SMILES to search on the SwissTarget Prediction website (http://www.swisstargetprediction.ch) to predict the potential targets of the active ingredients. All predicted targets with a non-zero probability were retained and considered as potential targets of the compounds.

Retrieve and filter potential targets related to MI through the DisGeNET database (https://www.disgenet.com) and the GeneCards database (https://www.genecards.org). Remove duplicate targets after merging the databases to obtain the related targets of MI. Use Venny 2.1.0 (http://www.liuxiaoyuyuan.cn) to identify the intersection targets between MI and the potential targets of QXJYG, and construct a Venn diagram. Import the screened active ingredients of QXJYG and the intersection targets from the Venn diagram into Cytoscape 3.10.0 to construct a “compound-target-disease” network map.

Import the intersecting genes of QXJYG and MI into the STRING network platform (https://string-db.org) and select the species “Homo sapiens”. Set the confidence score to >0.4 as a condition for screening and generate the interaction relationships of the intersection targets. Import this network diagram into Cytoscape 3.10.0 for analysis. Use the CentiScape 2.2 plugin to calculate the degree and betweenness of each node. Identify the key nodes as those with degree and betweenness values higher than the average of these two metrics. The intersection targets were subjected to gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses using the DAVID platform (https://david.ncifcrf.gov/summary.jsp). GO functional enrichment was categorized into three domains: biological processes (BP), cellular components (CC), and molecular functions (MF). The KEGG pathway enrichment analysis aimed to elucidate the underlying mechanisms by which QXJYG exerts its effects on MI. The resulting data were visualized using bar charts and bubble diagrams, created with the assistance of the micro-information visual cloud platform (https://www.bioinformatics.com.cn).

QXJYG was prepared and provided by Jiangyin Tianjiang Pharmaceutical Co., Ltd (Jiangsu, China; batch number: 2112308). The product is a water-soluble granular formulation manufactured through standardized extraction, purification, and spray-drying procedures. The manufacturer conducted a multi-indicator quality control assessment and established UPLC fingerprints for QXJYG. Details regarding its preparation, chemical profiling, and quantification of active constituents have been previously published (8–10). We used the same batch of QXJYG as in the previous studies (9). Isosorbide mononitrate (ISMN; MedChemExpress, USA) was prepared in 0.9% saline for animal use and in sterile PBS for in vitro experiments. Stock solutions were stored at −80 ℃ and diluted with serum-free DMEM prior to use. SB203580 (HY-10256, p38 MAPK inhibitor; MedChemExpress, USA) was dissolved in DMSO to prepare a 10 mM stock solution and diluted to 10 μM for cell treatment. This concentration was selected based on previous studies that demonstrated effective inhibition of p38 MAPK activity (11).

Male C57BL/6 mice (10–12 weeks old, 20–22 g) were purchased from Shanghai Slack Experimental Animals Co., Ltd. All animals were housed under standard laboratory conditions at 24 ± 2  °C with 50%–60% relative humidity and a 12-h light/dark cycle. Mice were acclimatized for 5 days prior to experimentation. All animal procedures were reviewed and approved by the Animal Ethics Committee of the Research Institute at Fujian University of Traditional Chinese Medicine (FJTCM IACUC 2022197) and conducted in accordance with institutional guidelines and internationally accepted principles for the care and use of laboratory animals.

A total of 48 mice were randomly assigned into six groups (n = 8 per group) based on body weight. The dosages of QXJYG and ISMN were calculated from their respective adult clinical doses and converted to mouse-equivalent doses using a body surface area -based method, with the converted QXJYG dose defined as the medium dose and the low and high doses set at half and twofold of the medium dose, respectively. The mice were divided into the following groups: ① Sham (received saline, ig), ② Model (MI, received saline, ig), ③ ISMN (MI+5.2 mg/kg/day ISMN, ig), ④ QXJYG-L (MI+1.43 mg/g/day QXJYG, ig), ⑤ QXJYG-M (MI+2.86 mg/g/day QXJYG, ig), and ⑥ QXJYG-H (MI+5.72 mg/g/day QXJYG-H, ig). Drug administration was initiated on the first day after MI surgery and continued once daily for four weeks. The Sham and Model groups received normal saline, while the other groups received either ISMN or QXJYG at the indicated doses.

Mice subjected to left anterior descending branch (LAD) coronary artery ligation was performed to establish the model of MI following the methodology previously described by Shen et al. (12). Briefly, mice with pre-plucked chest hair were anaesthetised with 1% isoflurane gas at a flow rate of 0.6–1 L/min, and ligated 2–3 mm below the left auricle with a sterile non-absorbable banded suture needle, size 6–0. Surgical procedures for the Sham group mirrored those of the Model group, with the exception that no ligation was executed.

Cardiac function in mice was assessed after a 4-week treatment period using the high-resolution Vevo2100 ultrasound system. Following chest hair removal, the mice were anesthetized with isoflurane and positioned supine on an animal operating platform. The MS400 ultrasound probe was oriented at a 45 ° angle relative to the midline on the left side of the sternum to capture a long-axis view of the left ventricle. Parameters of cardiac function, such as ejection fraction (EF) and fractional shortening (FS), were evaluated using Vevo LAB offline software.

After four weeks of drug intervention, mice were euthanized following blood collection. The hearts were then removed, photographed, and cut longitudinally into two halves, with one half containing the entire apex region and the other half containing the entire base region. The apical portion was washed with saline, fixed into a 4% paraformaldehyde solution for 48 h, and then embedded with paraffin for histological analysis. The remaining half of the heart specimens designated for western blot analysis were frozen at −80 ℃. The blood was centrifuged (4 ℃, 3,000 rpm, 15 min) to separate the serum, which was then stored at −80 ℃ for subsequent experiments.

Paraffin-embedded sections (4 μm thick) were stained with hematoxylin and eosin (H&E) and Masson's trichrome (Solarbio, Beijing, China) to visualize myocardial pathology and assess collagen volume fraction (CVF) for fibrosis quantification, respectively. Examination and imaging were performed using an optical microscope (Leica, Germany) at 400× magnification, with CVF analysis conducted via ImageJ software (NIH, USA).

For immunohistochemical staining, antigen retrieval was performed on mouse heart sections using 0.01M sodium citrate buffer (MVS-0066, Maixin, Fuzhou, China). Sections were incubated with CD31 antibody (1:200 dilution, 28083-1-AP, Proteintech Group, Inc, China) overnight at 4 ℃, followed by washing with PBS and secondary antibody incubation with HRP-polymer-conjugated anti-Mouse/Rabbit IgG (Maixin, KIT-9921, Fuzhou, China). Visualization was achieved using a DAB kit (Maixin, DAB-2031) and hematoxylin counterstaining. Slides were observed under a light microscope at 400× magnification, and positive expression was analyzed using ImageJ software in six random fields of view.

Apoptosis in myocardial tissue was detected using a TUNEL assay kit (MK1025, Boster, Wuhan, China) following the manufacturer's protocol. Briefly, sections were treated with 3% hydrogen peroxide to inactivate endogenous peroxidases, followed by Proteinase K digestion. DNA fragmentation was labeled using terminal deoxynucleotidyl transferase (TdT) and DIG-dUTP, with signal amplification achieved through biotinylated anti-DIG antibodies and a streptavidin-HRP complex. Visualization was performed with DAB and hematoxylin counterstaining. Positive cells were quantified in randomly selected fields using ImageJ software.

The collected serum samples were utilized for enzyme-linked immunosorbent assay (ELISA) to measure the levels of various cardiac biomarkers. Specifically, the ELISA kits were conducted to detect the levels of Superoxide Dismutase (SOD, catalog number U96-3518E), Lactate Dehydrogenase (LDH, catalog number MM-44145M1), respectively. These ELISA kits were procured from U-BioA (Shanghai) Trading Co., Ltd. China. The ELISA detection procedure was performed in strict accordance with the guidelines provided in the respective ELISA kit instructions.

To assess oxidative stress and cell injury, MDA levels were measured in both serum and cell culture supernatants, while LDH release was analyzed in cell supernatants. MDA levels were determined using a thiobarbituric acid (TBA) reaction-based assay (ZC-S0343, Shanghai Zcibio Technology Co., Ltd., China). Briefly, samples were incubated with TBA reagent in a boiling water bath, allowing MDA to react with TBA and form a red-colored complex. After cooling, the absorbance was measured at both 532 nm and 600 nm using a spectrophotometer. The difference in absorbance at these two wavelengths was used to calculate MDA levels, following the manufacturer's instructions.

LDH activity was quantified using a commercial LDH assay kit (A020-2-2, Nanjing Jiancheng Bioengineering Institute, China). Following the manufacturer's instructions, samples were incubated with reaction reagents at room temperature. Absorbance was measured at 450 nm using a microplate reader, and LDH activity was calculated according to the standard curve provided with the kit.

H9C2 cells (Rattus norvegicus, CVCL 0286, SCSP-5211) were cultured in DMEM with 10% FBS and 1% Penicillin-Streptomycin (100×) at 37 ℃, 5% CO₂, and saturated humidity until 80%–90% confluence for passaging and seeding. For experiments, cells were seeded at 10³ cells/well in 96-well plates and treated with a range of concentrations of QXJYG or ISMN for the CCK-8 assay to determine non-cytotoxic and effective concentrations. Based on the results of cell viability, the selected concentrations of QXJYG (low, medium, and high) and ISMN were used for subsequent in vitro experiments. Then, they were seeded at 8 × 10⁴ cells/well in 6-well plates for western blot or qPCR experiments, grouped as follows: ① Control (normoxia); ② Model (hypoxia); ③ ISMN (hypoxia+900 μM ISMN); ④ QXJYG-L (hypoxia+7.5 μg/mL QXJYG); ⑤ QXJYG-M (hypoxia+15 μg/mL QXJYG); ⑥ QXJYG-H (hypoxia+30 μg/mL QXJYG).

Meanwhile, cells were divided into five groups: ① Control (normoxia); ② Model (hypoxia); ③ QXJYG (hypoxia + 30 μg/mL QXJYG); ④ SB203580 (hypoxia + 10 μM SB203580); ⑤ QXJYG + SB203580 (hypoxia + 30 μg/mL QXJYG + 10 μM SB203580). Oxidative stress and apoptosis levels in each group were assessed using MDA assay kits and Hoechst 33,258 staining. After 24 h of plating, cells were treated with medication and subjected to 24 h of hypoxia according to their respective groups, the anoxic conditions were based on previous laboratory data: 1%O₂, 94% N₂ and 5% CO₂. For the SB203580 group, cells were pretreated with 10 μM SB203580 for 1 h, after which the medium was replaced with fresh DMEM before hypoxia. For the QXJYG + SB203580 group, cells were pretreated with 10 μM SB203580 for 1 h, followed by medium replacement with DMEM containing 30 μg/mL QXJYG prior to hypoxia. Following the completion of hypoxic culture, cells and the corresponding culture medium are harvested for further experimental procedures.

Cell viability was evaluated using the Cell Counting Kit-8 (CCK-8; Abbkine, USA; Cat. KTC011001). The H9C2 cells were seeded into 96-well plates and allowed to adhere for 24 h, followed by treatment with various concentrations of QXJYG or ISMN for an additional 24 h. After incubation, 10 μL of CCK-8 reagent was added to each well, and cells were further incubated for 1 h at 37 °C. The optical density (OD) was measured at 450 nm using a microplate reader. Cell viability was calculated based on OD values, and the results were used to determine the optimal concentrations of QXJYG and ISMN for subsequent H9C2 cell intervention experiments.

Western blot analysis was performed to detect protein expression in heart tissue and H9C2 cells. Samples were homogenized and lysed in Western & IP Lysis Buffer (P0013, Beyotime, China) supplemented with protease inhibitor cocktail (HY-K0010, MedChemExpress, USA), phosphatase inhibitor (PhosSTOP, 04906845001, Roche Diagnostics GmbH, Germany), and phenylmethylsulfonyl fluoride (PMSF, P0100-10, Beijing Solaibao Technology Co., Ltd., China). Lysates were incubated on ice for 15 min and centrifuged at 14,000× g for 15 min at 4 °C. Protein concentrations were determined using a BCA protein assay kit (23225, Thermo Fisher Scientific, USA).

Equal amounts of total protein (40 μg) were separated by 10% SDS-PAGE and transferred to 0.45 µm PVDF membrane (AmershamHybond, GE Healthcare, Munich, Germany). Membranes were blocked with a rapid protein-free blocking solution (PS108P, Shanghai Yemei Bio-Medical Technology Co., Ltd.) for 30 min at room temperature, and then incubated overnight at 4 ℃ with the following primary antibodies [rabbit anti-p38 MAPK (1:1,000 dilution, #8690, Cell Signaling Technology, USA), rabbit anti-Phospho-p38 MAPK antibody (1:1,000 dilution, #9,211, Cell Signaling Technology, USA), rabbit anti-Bcl-2 (1:1,000 dilution, #3498, Cell Signaling Technology, USA), rabbit anti-Bax (1:1,000 dilution, #2772, Cell Signaling Technology, USA), mouse anti-β-Actin (1:10,000 dilution, 81115-1-RR, Proteintech Group, Inc, China) and rabbit anti-Vinculin (1:5,000 dilution, GTX113294, GeneTex, USA)]. After incubation, the membrane was washed with TBST and then incubated with secondary antibodies [rabbit secondary antibody (L3012, SAB, USA) and mouse secondary antibody (L35665, SAB, USA)] at room temperature for 1 h. Following incubation, the membrane was washed with TBST and then detected with an ECL detection kit (Thermo Fisher Scientific, USA). The band intensities were analyzed using ImageJ software.

To assess the rate of cell apoptosis, H9C2 cells in each group were harvested and stained using the Annexin V/PI staining kit (KGA107, Jiangsu Kaiji Biotechnology Co., Ltd., Nanjing, China) following the manufacturer's instructions, and analyzed by flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA).

Apoptosis in H9C2 cells was assessed by examining nuclear morphological changes using the Hoechst Staining Kit (C003, Shanghai Beyotime Biotechnology Co., Ltd., China). Normal cell nuclei appeared blue, while apoptotic cell nuclei displayed a dense bright color, either densely stained or fragmented. Using ImageJ software to count the numbers of apoptotic (bright) and non-apoptotic (blue) nuclei. The apoptosis index was calculated as follows: Apoptosis rate = (apoptotic nuclei count/total nuclei count) × 100%.

All data were analyzed using SPSS 23.0. For metric data that conform to a normal distribution, the mean ± SD was used for representation. For non-normally distributed data, we reported the median and interquartile range. Normality of the metric data was assessed using the Shapiro–Wilk test. If data were normally distributed, we used one-way ANOVA to compare multiple groups, followed by post-hoc analysis with the LSD test for homogeneous variances and the Games-Howell test for heterogeneous variances. Nonparametric analysis (Kruskal–Wallis One-Way ANOVA) was conducted for data that did not conform to a normal distribution. The Bonferroni method was used to adjust for significance in cases of multiple comparisons. And the results were considered statistically significant when P < 0.05.

A total of 114 active components and their corresponding 988 targets were collected (Figure 1). A list of the top 20 active components based on OB values was compiled (Table 1). We identified a total of 3,029 MI-related genes from the GeneCards database and the DisGeNET database (Figure 2A). By calculating the intersection of drug targets and disease targets, 434 overlapping targets were identified (Figure 2B). A PPI network was constructed using the STRING database (Figure 2C), and core targets were identified and ranked by degree and betweenness using Cytoscape 3.10.0. The core target network (Figure 2D) suggests that QXJYG may primarily exert its therapeutic effects on MI by regulating these targets, which play crucial roles in various biological processes including inflammation, apoptosis, metabolism, and tissue remodeling.

To detail the biological characteristics of putative QXJYG targets on MI, GO and pathway enrichment analyses were conducted using the DAVID database. This yielded 1,171 biological processes (BP), 141 cellular components (CC), 291 molecular functions (MF) terms and 196 pathways. Ranked by significance, the top 10 GO terms and the top 20 pathways are shown in Figure 3. QXJYG appears to modulate various biological processes associated with post-MI, including the regulation of protein phosphorylation, which is crucial for signal transduction and cellular response, positive regulation of RNA polymerase II transcription, and the G-protein coupled receptor signaling pathway. Additionally, it may interfere with cell death mechanisms following MI. At the cellular component level, QXJYG is involved in the regulation of extracellular vesicles, cytoplasm, nucleus, nucleoplasm, plasma membrane, and mitochondria post-MI. At the molecular level, QXJYG participates in metal ion binding, zinc ion binding, and enzyme binding, particularly protein serine/threonine kinase activity, indicating its potential to interact with key molecular players in cellular mechanisms. This interaction may influence cellular outcomes post-MI, promoting recovery or adaptation (Figure 3A). Through KEGG pathway enrichment analysis, QXJYG in treating MI potentially regulates multiple critical biological pathways, including lipid metabolism and atherosclerosis, the calcium signaling pathway, the MAPK signaling pathway, the cAMP signaling pathway, and several others. This suggests that QXJYG may have a multifaceted therapeutic effect on MI by modulating various interconnected signaling pathways and biological processes (Figure 3B).

Compared to the Sham group, mice in the Model group exhibited significantly enlarged hearts, with the myocardium in the infarcted area of the left ventricular wall appearing pale and depressed, and the ratio of heart weight to tibia length (heart/tibia ratio) was markedly increased (Figures 4A,B, *P < 0.05, vs. Sham Group). Interventions with ISMN and high-dose QXJYG significantly reduced both heart size and the heart/tibia ratio in mice (Figures 4A,B, ^#P < 0.05, vs. Model Group). Echocardiography revealed that both the EF and FS were significantly reduced in the Model group (Figures 4C,D, *P < 0.05, vs. Sham Group), indicating the successful establishment of the MI model. Treatment with ISMN attenuated the reductions in EF and FS observed in the Model group (Figures 4C,D, #P < 0.05, vs. Model group). High doses of QXJYG led to significant increases in EF (Figure 4C, ^#P < 0.05, vs. Model group). Moreover, both medium and high doses of QXJYG significantly increased FS (Figure 4D, ^#P < 0.05, vs. Model group).

The HE staining results indicated that in the Sham group, the myocardial cells were organized in a neat spindle shape, with centrally located oval nuclei. The cytoplasm appeared red, and the nuclei were stained purple-blue. Conversely, the Model group displayed thickened myocardial fibers, disorganized cellular arrangement, widened interstitial spaces, and pathological alterations such as vacuolation and infiltration of inflammatory cells within the interstitium (as indicated by arrows). These pathological changes showed improvement following intervention with low, medium, and high doses of QXJYG and ISMN (Figure 4E). Masson's trichrome staining revealed that in the Sham group, myocardial cells were arranged in an orderly fashion with minimal collagen fiber deposition, which did not significantly disrupt myocardial structure. In contrast, the Model group exhibited substantial collagen fiber deposition, irregular myocardial cell arrangement, and structural disarray (Figures 4F,G, *P < 0.05, vs. Sham Group). Following intervention with ISMN and low, medium, and high doses of QXJYG, there was a notable reduction in collagen fiber deposition in cardiac tissue and an improvement in myocardial structure (Figures 4F,G, ^#P < 0.05, vs. Model Group). Quantitative assessment of mouse cardiac microvascular density (MVD) was conducted using immunohistochemistry with specific antibodies against CD31 antigen to visualize the cardiac microvascular network. Compared to the Sham group, the Model group exhibited a significant increase in MVD (Figures 4H,I, *P < 0.05, vs. Sham Group). Furthermore, compared to the Model group, both the ISMN group and the high-dose QXJYG group showed a significant increase in MVD ((Figures 4H,I, ^#P < 0.05, vs. Model Group).

These experimental findings indicate that QXJYG treatment effectively mitigates cardiac enlargement, enhances myocardial structure, and reduces collagen deposition in a mouse model of MI, thereby preserving cardiac function.

The levels of MDA and LDH in the serum of mice were found to be significantly increased in the Model group (Figures 5A,B, *P < 0.05, vs. Sham group), whereas the SOD level was decreased (Figure 5C, *P < 0.05, vs. Sham group). The MI-induced mice treated with ISMN and high-dose QXJYG showed significantly reduced MDA levels (Figure 5A, ^#P < 0.05, vs. Model group). LDH levels exhibited a significant reduction in the ISMN group and in all dose groups of QXJYG (Figure 5B, ^#P < 0.05, vs. Model group). Serum SOD levels were significantly increased in the ISMN and QXJYG-M, QXJYG-H groups (Figure 5C, ^#P < 0.05, vs. Model group).

We further determined the effect of QXJYG on cardiomyocyte apoptosis in MI-induced mice using TUNEL staining. The results revealed that compared to the Sham group, the Model group exhibited a significant increase in the rate of apoptotic cells (Figure 5D, *P < 0.05, vs. Sham group). Conversely, mice treated with ISMN and high-dose QXJYG showed a significant reduction in the rate of cell apoptosis in cardiac tissue (Figure 5D, ^#P < 0.05, vs. Model group). Western blot analysis was then used to detect the expression of apoptosis-related proteins. The results showed that the protein expression of pro-apoptotic Bax in the cardiac tissue of the Model group were significantly increased (Figures 5E,G, *P < 0.05, vs. Sham group), while the protein expression of anti-apoptotic Bcl-2 was significantly decreased (Figures 5E, H, *P < 0.05, vs. Sham group). The protein expression of Bax was significantly decreased in the ISMN, QXJYG-M, and QXJYG-H groups compared to the Model group (Figures 5E,G, ^#P < 0.05, vs. Model group). Additionally, the protein expression of Bcl-2 was significantly increased in the ISMN, QXJYG-L, QXJYG-M, and QXJYG-H groups compared to the Model group (Figures 5E,H, ^#P < 0.05, vs. Model group).

Emerging evidence has shown that the activated p38 MAPK signaling pathway is involved in oxidative stress and apoptosis during the progression of MI (13, 14). Our findings displayed that the ratio of p-p38 MAPK/p38 MAPK was significantly increased in the cardiac tissue of MI-induced mice (Figures 5E,F, *P < 0.05, vs. Sham group). However, there was a decrease in the ratio of p-p38 MAPK/p38 MAPK in the ISMN and QXJYG-L, QXJYG-M, and QXJYG-H groups in comparison to the Model group (Figure 5E,F, ^#P < 0.05, vs. Model group). Taken together, our results indicated that QXJYG inhibited oxidative stress and apoptosis in cardiac tissue of MI-induced mice via suppression of the p38 MAPK signaling pathway.

To further elucidate the efficacy and mechanism of QXJYG against MI, we established an in vitro model of hypoxia-induced H9C2 cells. To determine the optimal concentrations of QXJYG and ISMN for subsequent in vitro experiments, we first performed a dose-response analysis using the CCK-8 assay. H9C2 cells were treated with QXJYG at concentrations of 0, 7.5, 15, 30, and 40 μg/mL for 24 h. Cell viability remained above 85% and showed no significant difference compared to the control group at 7.5, 15, and 30 μg/mL (P > 0.05), indicating that these concentrations were non-cytotoxic. However, a significant reduction in viability was observed at 40 μg/mL (Figure 6A, *P < 0.05, vs. Control group), suggesting the onset of cytotoxicity. Therefore, 7.5, 15, and 30 μg/mL were selected as the low, medium, and high concentrations, respectively, for subsequent experiments. Similarly, ISMN was tested at concentrations of 0, 700, 800, 900, and 1,000 μM. ISMN at 700–900 μM did not significantly affect cell viability compared to the control group, while 1,000 μM led to a significant reduction in viability, indicating cytotoxicity at this concentration. Therefore, 900 μM was selected as the working concentration for subsequent in vitro experiments due to its favorable safety profile and stable biological activity (Figure 6B, *P < 0.05, vs. Control group).

Secondly, the detection results of MDA and LDH levels in cell culture supernatant showed that the levels of MDA and LDH in the Model group were significantly increased (Figures 6C,D, *P < 0.05, vs. Control group). Compared to the Model group, LDH levels of hypoxia-induced H9C2 cells significantly were decreased following intervention with ISMN and low, medium, and high doses of QXJYG (Figure 6C, ^#P < 0.05, vs. Model group). Simultaneously, treatment with ISMN, as well as medium and high doses of QXJYG, substantially reduced MDA levels (Figure 6D, ^#P < 0.05, vs. Model group).

Thirdly, the cell apoptosis was determined via Annexin V/PI staining followed by FACS analysis. The results revealed a significant increase in the rate of apoptotic cells in the Model group (Figure 6E, *P < 0.05, vs. Control group). Conversely, cells treated with ISMN and low, medium, and high doses of QXJYG showed a significant reduction in apoptosis (Figure 6E, ^#P < 0.05, vs. Model group).

We further used western blot analysis to evaluate whether the inhibitory effect of QXJYG on oxidative stress and apoptosis via regulating the p38 MAPK signaling pathway in hypoxia-induced H9C2 cells. The results showed that the ratio of p-p38 MAPK/p38 MAPK and the protein expression of Bax were significantly increased in the hypoxia-induced H9C2 cells (Figures 6F–H, *P < 0.05, vs. Control group), while the protein expression of Bcl-2 was significantly decreased (Figures 6F,I, *P < 0.05, vs. Control group). The ratio of p-p38 MAPK/p38 MAPK was significantly decreased in the QXJYG-L, QXJYG-M, and QXJYG-H groups compared to the Model group (Figures 6F,G, ^#P < 0.05, vs. Model group). Additionally, Bax protein expression showed a significant reduction in the ISMN and QXJYG-H groups relative to the Model group (Figure 6F,H, ^#P < 0.05, vs. Model group). In contrast, the protein expression of Bcl-2 was markedly increased in the ISMN, and QXJYG-H groups when compared to the Model group (Figures 6F,I, ^#P < 0.05, vs. Model group). Collectively, we conclude that QXJYG alleviated the hypoxia-induced oxidative stress and apoptosis through downregulation of p38 MAPK signaling pathway.

To further investigate the mechanism, SB203580, a p38 MAPK inhibitor, and its combination with high-dose QXJYG were used. Both treatments significantly decreased MDA levels compared to the Model group (Figure 7A, #P < 0.05, vs. Model group), with a more pronounced reduction observed in the combined treatment group compared to QXJYG treatment alone (Figure 7A, ^ΔP < 0.05, vs. QXJYG group), suggesting a potential synergistic antioxidative effect. Hoechst 33,258 staining further revealed a significant increase in apoptotic nuclei in hypoxia-induced H9C2 cells compared to the Control group (Figures 7B,C, *P < 0.05 vs. Control group). Treatment with QXJYG markedly reduced the number of apoptotic cells compared to the hypoxia group (Figures 7B,C, ^#P < 0.05 vs. Model group), indicating its protective effect against hypoxia-induced apoptosis. Notably, pretreatment with the p38 MAPK inhibitor SB203580 also significantly decreased apoptosis levels (Figures 7B,C, ^#P < 0.05 vs. Model group). Furthermore, combined treatment with SB203580 and QXJYG resulted in an even greater reduction in apoptotic cells (Figures 7B,C, ^ΔP < 0.05 vs. QXJYG group). These findings offer further mechanistic support for the involvement of the p38 MAPK signaling pathway in mediating the anti-apoptotic effects of QXJYG under hypoxic conditions.