Traditional Chinese Medicine Systems Pharmacology Database v2.3 (TCMSP; https://old.tcmsp-e.com/index.php); UniProt database v2020_01 (https://www.uniprot.org/); Swiss Target Prediction v2019 (http://www.swisstargetprediction.ch/); PharmMapper v2017 (http://www.lilab-ecust.cn/pharmmapper/index.html); GeneCards database v5.5 (https://www.genecards.org/); E Venn Diagram website (http://www.ehbio.com/test/venn/); STRING database v11.0 (https://string-db.org/); DAVID v6.8 (https://david.ncifcrf.gov/tools.jsp); Microbiology website (http://www.bioinformatics.com.cn/) and Cytoscape software (version 3.7.2) (17) were employed.

The compound characteristics and physicochemical parameters of LIQ were obtained using CAS No. 551-15-5 from the TCMSP. Additionally, the corresponding targets were obtained. The SDF files of LIQ were acquired from the PubChem database and subsequently imported into the Swiss Target Prediction and PharmMapper websites to predict the targets associated with LIQ. The results from these three databases were then consolidated. The UniProt database was used to gather the gene names of the respective LIQ targets, with a species restriction set to ‘Homo sapiens’. Finally, a search was conducted in the GeneCards database using ‘Myocardial ischemia-reperfusion injury’ as a keyword to identify potential targets.

Venn diagrams were created by intersecting the targets associated with MIRI and those associated with LIQ, using an online Venn diagram tool. This facilitated the identification of common targets shared between the drug and the disease. Following this, a protein-protein interaction (PPI) network was constructed for these common targets using the STRING 11.0 database. The resulting TSV files were downloaded and then imported into the Cytoscape software for visualization.

The ‘NetworkAnalyzer’ plug-in within Cytoscape software was used to analyze the Degree values of each target in the PPI network. This analysis enabled the identification of core targets by prioritizing them based on their Degree values. It is important to highlight that targets with higher Degree values hold greater significance within the network.

The common targets were imported into the DAVID database to gain deeper insights into the functions of the common targets resulting from the intersection of LIQ and MIRI. Following this, GO and KEGG enrichment analyses were performed using the DAVID database by entering a list of target gene names and specifying the species as human. The GO analysis encompassed biological process (BP), cellular component (CC) and molecular function (MF) categories. The obtained results were subsequently analyzed and visualized using the Microbiology website.

LIQ was obtained from Chengdu Alfa Biotechnology Co., Ltd. CoCl₂ was purchased from Toronto Research Chemicals. Cell Counting Kit-8 (CCK-8; cat. no. G02111) was acquired from Dongren Chemical Technology Co., Ltd. High-glucose Dulbecco's modified Eagle medium (DMEM; cat. no. 12430047) and fetal bovine serum (FBS; cat. no. 10093188) were purchased from Gibco (Thermo Fisher Scientific, Inc.). TNF-α (cat. no. 88-7340-88) and IL-6 (cat. no. 88-50625-88) ELISA kits were obtained from Thermo Fisher Scientific, Inc. The remaining ELISA kits, including lactated dehydrogenase (LDH; cat. no. A02022), creatine kinase isoenzyme-MB (CK-MB; cat. no. H197-1-1), superoxide dismutase (SOD; cat. no. A00132), catalase (CAT; cat. no. A007-11), glutathione peroxidase (GSH-Px; cat. no. A005-1-2) and malondialdehyde (MDA; cat. no. A00311), were from Nanjing Jiancheng Bioengineering Institute. Unless stated otherwise, all other reagents used in this experiment were purchased from Sigma-Aldrich (Merck KGaA).

H9c2 cells (Bluefbio Biotechnology Development, Inc.; cat. no. BFN60804388) were maintained in high-glucose DMEM supplemented with 10% FBS and 100 U/ml penicillin/streptomycin. The cells were incubated at 37˚C, 5% CO₂ and 95% air in an incubator. The culture medium was refreshed every 2-3 days. When the cells reached ~70-80% confluence, they were detached using 0.25 g/l trypsin, passaged and diluted to the appropriate cell suspension concentration. Subsequently, the cells were cultured in 96-, 24- and 6-well plates until they adhered.

H9c2 cells cultured in 96-well plates were exposed to various concentrations of LIQ (1, 3, 10, 30, 100 and 300 µmol/l). After a 24-h incubation period, 10 µl of CCK-8 solution was added and absorbance was measured at 450 nm using an enzyme labeling instrument (Thermo Scientific Varioskan LUX; Thermo Fisher Scientific, Inc.). The protective concentrations of LIQ were determined based on CCK-8 calculations, confirming 3 and 10 µmol/l as the effective concentrations.

Experiments were conducted with H9c2 cells seeded in 96-well plates. The cells were exposed to different concentrations (200, 400, 600, 800 and 1,000 µmol/l) of CoCl₂ solutions prepared in complete medium for 24 h. The optimal hypoxic concentration was determined based on the results obtained from the CCK-8 assay. Following the hypoxic phase, the cells were incubated in complete medium for varying time intervals (1, 2, 3, 4, 5 and 6 h) to determine the appropriate reoxygenation duration. The optimal reoxygenation time was established using the results from the CCK-8 kit. In summary, the H/R injury model was established by subjecting the cells to 400 µmol/l CoCl₂-induced hypoxia for 24 h, followed by reoxygenation for 3 h. The experimental groups were categorized as follows: i) Control (CON); ii) H/R; iii) H/R + 3 µmol/l (L-)LIQ; iv) H/R + 10 µmol/l (H-)LIQ; v) LIQ (10 µmol/l).

Following the aforementioned treatment, the cell culture medium supernatant was collected. Subsequently, the levels of LDH and CK-MB were measured in each group following the instructions provided with the respective kits.

At the conclusion of the experiment, the cells were scraped and resuspended in 500 µl of PBS solution. A cell homogenate was then prepared and the levels of SOD, CAT, GSH-Px and MDA were measured using the respective kits, following the provided instructions.

2,7-Dichlorodihydro-fluorescein diacetate (DCFHDA, Cayman Chemical Company; cat. no. 85155) itself does not emit green fluorescence. However, intracellular ROS oxidizes the non-fluorescent DCFH to green-fluorescent DCF. The intensity of the resulting green fluorescence indirectly reflects the level of intracellular ROS. H9c2 cells were seeded in 24-well culture plates and allowed to grow until they reached ~80% confluence. Subsequently, the cells were treated as required. After the treatment, the cells were rinsed twice with PBS and incubated with 10 µmol/l DCFH-DA dye at 37˚C for 20 min. Images were captured using a fluorescence microscope (Nikon Eclipse C1; Nikon Corporation) and the green fluorescence intensity was semi-quantitatively analyzed using ImagePro Plus 6.0 software (Media Cybernetics, Inc.). The green fluorescence intensity was then analyzed semi-quantitatively and statistically for each sample within each group.

Rhodamine 123 (Shanghai Yuanye Biotechnology Co., Ltd; cat. no. S19123) is a fluorescent dye that selectively accumulates in mitochondria based on their transmembrane potential. When the cells reached ~80% confluence, various conditioned media were applied as per the experimental requirements. Following treatment, the cells were washed twice with PBS and then incubated at 37˚C for 20 min in a serum-free medium containing 100 µg/l Rh123. The green fluorescence observed around the nuclei indicated the uptake of Rh123 by the mitochondria. The green fluorescence intensity, representing Rh123 uptake by mitochondria, was semi-quantitatively analyzed using ImagePro Plus 6.0 software (Media Cybernetics, Inc.). Statistical analysis was conducted on the green fluorescence intensity for each sample within each group.

H9c2 cells were seeded in 6-well plates and treated according to the experimental group requirements once they reached the desired confluency. The treated cells were collected as the test specimens. ELISA procedures were carried out following the instructions provided with the respective kits. After the reaction was completed, the absorbance values of each well were measured using a multifunctional microplate reader. The levels of TNF-α and IL-6 were determined following the provided kit instructions.

H9c2 cells were cultured in 24-well plates and treated as required once they reached the desired density. Hoechst-33258 dye (Beijing Solarbio Science & Technology Co., Ltd; cat. no. IH0060) buffer at a concentration of 5 mg/l was applied to the cells, followed by a 15-min incubation in a cell incubator. The nuclei of apoptotic cells displayed either condensed solidified forms or granular fluorescence. Quantification of apoptotic cells was performed using ImagePro Plus 6.0 software (Media Cybernetics, Inc.).

The core targets and key signaling pathways predicted through network pharmacology were validated using the western blotting. H9c2 cells were seeded in 6-well plates and treated according to the experimental requirements. The cells were washed twice with pre-cooled PBS and then lysed using cell lysate (Beijing Solarbio Science & Technology Co., Ltd.; cat. no. R0020). The lysates were centrifuged at 12,000 x g for 10 min at 4˚C. The supernatant was quantified using the BCA method. Total protein (10 µg) was separated using 10% SDS-PAGE and transferred onto PVDF membranes. The PVDF membranes were subsequently incubated in 5% skimmed milk powder for 1.5 h to block nonspecific binding at 37˚C. Next, the membranes were incubated overnight at 4˚C with the corresponding primary antibodies, as follows: TNF-α receptor type 1 (TNFR1, Affinity Biosciences; diluted at 1:1,000; cat. no. AF0282), NF-κB (Affinity Biosciences; diluted at 1:1,000; cat. no. BF8005), phosphorylated (p-)NF-κB (Affinity Biosciences; diluted at 1:1,000; cat. no. AF2006), MMP9 (Affinity Biosciences; diluted at 1:1,000; cat. no. AF5228) and β-actin (Affinity Biosciences; diluted at 1:10,000; cat. no. AF7018). The following day, the membranes were washed three times with TBST (0.1% Tween 20) for an average of 10 min per wash, followed by incubation with the corresponding secondary antibody at room temperature for 1 h. After an additional three washes, grayscale values were measured using Visioncapt software (Fusion FX5 Spectra; Wilbur Bioimaging).

GraphPad Prism 8.0.2 software (Dotmatics) was used for statistical analyses. Data were presented as mean ± standard error of mean and one-way analysis of variance followed by Tukey's post hoc test was employed to assess statistically significant differences between groups. P<0.05 was considered to indicate a statistically significant difference.

The data pertaining to the physicochemical and pharmacological molecular properties of LIQ were sourced from the TCMSP database. Drug-like properties (DL) serve as a valuable indicator for assessing the potential of an ingredient in drug development, while oral availability (OB) is employed to gauge the ease of absorption of the ingredient into the bloodstream. The obtained OB value was 65.69% and the DL properties were measured at 0.74. These results demonstrated that LIQ exhibited an OB value of ≥30% and a DL value of ≥0.18, indicating its strong potential for drug development, as it is well-absorbed and possesses favorable drug-like properties.

A comprehensive analysis identified 315 targets for LIQ through predictions derived from the TCMSP database, Swiss Target Prediction and PharmMapper databases. In parallel, 1,052 targets for MIRI were predicted based on data from the Genecards database. The intersection of these two datasets revealed a total of 99 shared targets, as illustrated in Fig. 2A. Detailed information pertaining to specific target sites can be found in Fig. 2B.

The PPI network was established by entering the shared targets into the STRING database with subsequently analysis to identify its core network, as depicted in Fig. 2C. Within this core network, the top 10 targets were identified (Fig. 2D), which included AKT1, cystatin protease 3 (CASP3), albumin (ALB), proto-oncogene tyrosine-protein kinase Src (SRC), MMP9, heat shock protein 90α family class A member 1 (HSP90AA1), membrane-linked protein A5 (ANXA5), prostaglandin-endoperoxide synthase 2 (PTGS2), nitric oxide synthase 3 (NOS3) and peroxisome proliferator-activated receptor γ (PPARG). MMP9 was selected as the primary focus of this research among these targets.

GO analysis, conducted using the DAVID database, yielded a total of 541 GO entries. Examination of the 394 BP categories unveiled the involvement of processes such as hypoxia, apoptosis, inflammation and oxidative stress. Within the CC entries, a total of 48 terms were enriched, primarily connected to extracellular exosomes, extracellular regions, plasma membranes, cytosols and mitochondria. The MF entries were also enriched, featuring 99 terms primarily associated with functions such as identical protein binding, peptidase activity, protein binding, zinc ion binding and cysteine-type endopeptidase activity involved in the execution phase of apoptosis, as depicted in Fig. 3A-C.

The KEGG pathway enrichment analysis identified a total of 125 pathways (P<0.01). These pathways were predominantly related to lipid and atherosclerosis, pathways in cancer, the TNF signaling pathway, the IL-17 signaling pathway and apoptosis. To enhance clarity, the top 20 pathways were selected for visualization, as depicted in Fig. 3D. Notably, among these pathways, 11 targets of LIQ were found to be involved in the regulation of the TNF signaling pathway, with the core target MMP9 among them. Details of the predicted targets associated with the TNF signaling pathway can be found in Fig. 4.

As shown in Fig. 5A, no significant changes in cell survival were observed when H9c2 cells were exposed to various concentrations of LIQ. To determine the optimal pre-protective concentrations for LIQ, 3 µmol/l and 10 µmol/l were selected. As depicted in Fig. 5B, cell viability decreased as the CoCl₂ concentration increased when H9c2 cells were subjected to varying CoCl₂ concentrations. To maintain cell survival at ~50% following hypoxia/reoxygenation, 400 µmol/l of CoCl₂ was selected to induce hypoxia (P<0.01), followed by a 3-h reoxygenation period (Fig. 5C; P<0.01). As illustrated in Fig. 5D, the viability of cells in the H/R group was significantly lower than that in the CON group (P<0.01). However, LIQ pre-treatment effectively mitigated the inhibitory effect of CoCl₂ on H9c2 cell viability (P<0.01).

As depicted in Fig. 6A and B, the levels of LDH and CK-MB exhibited a significant increase in the H/R group compared with the CON group (P<0.01). However, treatment with LIQ resulted in a notable reduction in the levels of LDH and CK-MB in cardiomyocytes when compared with the H/R group (P<0.05 or 0.01). Additionally, it was notable that the reductions in LDH and CK-MB levels were more pronounced at higher LIQ doses, suggesting that LIQ pre-treatment effectively mitigated the damage to cardiomyocytes caused by H/R.

As illustrated in Fig. 7, the levels of MDA, SOD, CAT and GSH-Px and can serve as indicators of oxidative stress damage. In comparison to the CON group, the H/R group displayed notably lower levels of SOD, CAT and GSH-Px (P<0.01), while the levels of MDA were significantly higher (P<0.01). However, the pre-protective effect of LIQ alleviated the damage induced by H/R, as evidenced by significantly higher levels of SOD, CAT and GSH-Px in the LIQ group in comparison to the H/R group (P<0.05 or 0.01), along with significantly lower levels of MDA (P<0.01).

The intracellular ROS level and the intensity of green fluorescence demonstrated a direct proportional relationship. As depicted in Fig. 8, the intracellular fluorescence intensity and ROS level in H9c2 cells within the H/R group were markedly higher than in the CON group (P<0.01). By contrast, the fluorescence intensity and ROS level in the LIQ group were significantly lower than those in the H/R group (P<0.01). Additionally, it was observed that the antagonistic effect on H/R injury became more pronounced with higher doses of LIQ.

The extent of MtMP damage showed an inverse correlation with the aggregation of Rh123. As depicted in Fig. 9, the aggregation of Rh123 in H9c2 cells of the H/R group was significantly lower compared with the CON group, indicating substantial MtMP damage (P<0.01). In contrast, the aggregation of Rh123 in H9c2 cells was significantly increased in the LIQ-treated group compared with the H/R group (P<0.05 or 0.01). This suggests that LIQ can effectively reduce MtMP damage in H/R injury with a concentration-dependent effect. The higher the concentration, the more pronounced the effect.

During the process of apoptosis, morphological abnormalities often manifest, including cell shrinkage, chromatin condensation, cell membrane rupture and nuclear fragmentation (18). To verify the presence of apoptosis, the morphology of abnormal cells was examined using Hoechst 33258 staining and fluorescence microscopy. In normal cells, the nuclei displayed uniform blue fluorescence, while apoptotic cells exhibited densely stained blue nuclei or fragmented densely stained blue nuclei.

As illustrated in Fig. 10, when compared with the CON group, cells in the H/R group exhibited pronounced apoptotic morphology when observed under fluorescence microscopy following ultraviolet light excitation (P<0.01). However, in the LIQ-treated group, the apoptotic morphology was significantly reduced compared with the H/R group (P<0.01), indicating that LIQ attenuated H/R-induced apoptosis and enhanced the survival rate of H9c2 cells.

The results depicted in Fig. 11 indicated a significant increase in the secretion levels of TNF-α and IL-6 in the H/R group when compared with the CON group (P<0.01). However, LIQ treatment notably reduced the secretion levels of TNF-α and IL-6 (P<0.05 or 0.01). These findings suggested that LIQ can effectively mitigate the inflammatory damage induced by H/R, with a more pronounced antagonistic effect observed at higher doses.

The results of western blotting, showing TNFR1, p-NF-κB/NF-κB and MMP9 protein expression in H9c2 cells, are presented in Fig. 12. The expression levels of TNFR1, p-NF-κB/NF-κB and MMP9 proteins were markedly higher in the H/R group compared with the CON group (P<0.01). However, in the LIQ-treated group, the expression levels of these proteins were significantly lower than in the H/R group, demonstrating significant differences (P<0.05 or 0.01). These findings indicated that LIQ effectively mitigated the abnormally elevated expression of TNFR1, p-NF-κB/NF-κB and MMP9 proteins.