Abstract As a well-known classical Chinese medicine prescription, Shengxian Decoction (SXD) has been applied for a century to treat cardiovascular diseases, especially coronary heart disease (CHD), but the potentially effective compounds and underlying mechanisms remain unclear. With ultra-performance liquid chromatography-quadrupole-time of flight-mass spectrometry (UPLC-Q-TOF/MS) and network pharmacology analysis, the potential effective compounds of SXD and their pharmacological mechanisms against CHD were identified and revealed. 57 effective compounds with favorable pharmacokinetic characteristics and biological activities were screened through UPLC-Q-TOF/MS analysis, database and literature mining, interacting with 96 CHD-related targets to support potential synergistic therapeutic actions. Systematic analysis of the PPI network and microarray data further revealed six core targets, including TNF, IL-1β, IL-6, TP53, VEGFA and PTGS2, which were mainly involved in fluid shear stress and atherosclerosis, lipid and atherosclerosis, PI3K-Akt signaling pathway et al. Moreover, the proposed contribution indexes of effective compounds indicated these compounds, including isoferulic acid, quercetin, calycosin, ferulic acid, kaempferol, calycosin 7-O-glycoside, formononetin, astragaloside IV and saikosaponin D, as the core compounds of SXD. The molecular docking results confirmed that those core compound-target pairs exhibited strong binding energy. Furthermore, we validated that SXD significantly alleviated myocardial tissue injury in CHD rats and reversed H/R-induced decreases in H9c2 cell viability by attenuating the production of TNF, IL-6 and IL-1β, and reducing cardiomyocyte apoptosis via down-regulating the TP53, caspase3 and cytochrome C mRNA expression levels as well as caspase3, caspase9 and cytochrome C protein expression levels according to RT-qPCR and Western blot results. Our findings explained the pharmacological mechanisms underlying the effectiveness of SXD in the treatment of CHD, and laid a foundation for future basic and clinical research of SXD. Keywords: Coronary heart disease, Shengxian Decoction, UPLC-Q-TOF/MS, Network pharmacology, Inflammation, Apoptosis 1. Background Coronary heart disease (CHD) is one of the most common cardiovascular diseases, which is caused by stenosis, spasm or blockage of the coronary artery lumen, and leads to myocardial ischemia, hypoxia or necrosis [[37]1,[38]2]. Chronic ischemia from coronary artery stenosis or rupture and myocardial infarction can lead to heart failure and even death [[39]3]. Drugs for treating CHD usually act on individual targets, such as beta-blockers and angiotensin converting enzyme inhibitors. The current clinical drugs for CHD have good efficacy, but may produce inevitable side effects and drug resistance [[40]4]. In addition, the morbidity and mortality of CHD are on the rise, and gradually show a younger trend. It is particularly important to seek a more effective and safe way to combat CHD. CHD belongs to the category of “chest arthralgia” in traditional Chinese medicine (TCM), and is characterized by blood stasis, “qi” stagnation (vital energy retardation), and phlegm obstruction [[41]5]. In TCM theory, the “qi” stagnation is deemed as the main pathogenesis of CHD [[42]6]. Shengxian Decoction (SXD), a classic TCM prescription for the treatment of atmospheric subsidence syndrome, was first recorded in the text “Medical Treatise on the Integration of Chinese and Western Medicine” by the famous Chinese medicine master Zhang Xi-chun in the late Qing Dynasty and the early Republic of China (between the 19th and 20th centuries). SXD is composed of HuangQi (HQ, roots of Astragalus membranaceus var. mongholicus (Bunge) Hsiao), ZhiMu (ZM, rhizomes of Anemarrhena asphodeloides Bge.), ChaiHu (CH, roots of Bupleurum chinense DC.), JieGeng (JG, roots of Platycodon grandiflorum (Jacq.) A. DC.) and ShengMa (SM, rhizomes of Cimicifuga foetida L.) at a ratio of 6:3:1.5:1.5:1 [[43]7,[44]8]. In this prescription, HQ is the emperor, which can play the role of replenishing and raising “qi”; as the ministers, CH and SM play a supporting and harmonizing role; the HQ's medicinal properties are slightly hotter, so ZM plays the role of heat-clearing as the adjuvant herb; JG, as a courier role, can help to guide the other herbs in the formula to the heart and also modulate or harmonize the properties of other drugs. Together, these herbs are particularly effective in treating conditions such as “qi” stagnation, which can manifest as chest tightness, shortness of breath, fatigue, and other symptoms similar to those of CHD [[45]9]. And the main compounds of SXD are flavonoids, phenolic acids, polysaccharides, saponins and amino acids, which are closely related to regulating inflammatory factors, oxidoreductases, and cardiovascular indicators, etc. Based on a metabolomic study, SXD has a cardioprotective effect against chronic heart failure, which is involved in the metabolism of energy and sphingolipid [[46]7]. SXD, in particular, has been clinically used in China to cure a variety of heart diseases, including chronic heart failure and myocardial ischemia, especially CHD [[47]10,[48]11]. However, the potential pharmacological mechanism of SXD in the treatment of CHD, as well as the potential targets and pathways are still unclear. TCM has the multi-compound, multi-target and multi-pathway characteristics, thus it also belongs to polypharmacy therapy [[49]12]. In recent years, due to the complex composition of TCM prescription, accurate and complete characterization of the prescription is the core problem of TCM research. The ultra-performance liquid chromatography-quadrupole-time of flight-mass spectrometry (UPLC-Q-TOF/MS) is a modern analytical technique to analyze and identify the chemical composition of TCM prescription, which is characterized by rapid and effective qualitative analysis of complex compounds. Besides, network pharmacology can comprehensively and systematically reveal the complex relationship between the effective compounds of TCM and their mechanism of action [[50]13]. It has been successfully applied to interpret the mechanism of the prescription therapy for cardiovascular disease at the molecular network level, such as Radix Salviae and Buyang Huanwu Decoction [[51]14,[52]15]. In the present study, we first analyzed the compounds of SXD by UPLC-Q-TOF/MS, and based on which the potential compounds and targets for the treatment of CHD were collected by database and literature analysis. Then, we tried to establish network pharmacology models including protein-protein interaction (PPI), compound-target-disease (C-T-D), and target-pathway (T-P) networks, calculate contribution index (CI) of each effective compound, which was based on chemical, pharmacokinetic and pharmacological data, explore the interaction between compound-target pairs by molecular docking, and construct the rat CHD model and hypoxia/reoxygenation (H/R)-induced H9c2 cell model for further validation, so as to uncover the underlying mechanism of SXD in the treatment of CHD. 2. Methods 2.1. Preparation and analysis of SXD 2.1.1. Preparation of the SXD extract SXD was prepared as follows: roots of Astragalus membranaceus var. mongholicus (Bunge) Hsiao (HQ, Batch No. 20220502), rhizomes of Anemarrhena asphodeloides Bge. (ZM, Batch No. 20221201), roots of Bupleurum chinense DC. (CH, Batch No. 20221101), rhizomes of Cimicifuga foetida L. (SM, Batch No. 20221001), and roots of Platycodon grandiflorum (Jacq.) A. DC. (JG, Batch No. 20230201) were purchased from Shaanxi Sciendan Pharmaceutical Co., Ltd. (Tongchuan, China) and identified by Professor Yonggang Yan. The voucher specimens were deposited at the Key Laboratory of Shaanxi Administration of Traditional Chinese Medicine for TCM Compatibility, Shaanxi University of Chinese Medicine (Specimen number: 22-05-02, 22-12-01, 22-11-01, 22-10-01, 23-02-01). Seventeen reference standards including mangiferin (Batch No. Yz111920), caffeic acid (Batch No. Yz012423), rutin (Batch No. Yz051924), Calycosin 7-O-glucoside (Batch No. Yz102721), ferulic acid (Batch No. Yz011422), isoferulic acid (Batch No. Yz1009201), luteolin (Batch No. 116523), quercetin (Batch No. Yz011823), calycosin (Batch No. Yz120322), kaempferol (Batch No. Yz110987), isorhamnetin (Batch No. Yz090422), saikosaponin A (Batch No. Yz050723), astragaloside Ⅳ (Batch No. Yz081320), saikosaponin D (Batch No. Yz080522), neomangiferin (Batch No. Yz1027222), timosaponin AIII (Batch No. Yz111022) and timosaponin BII (Batch No. Yz072323) were obtained from Nanjing Plant Origin Biological Co., Ltd. (Nanjing, China). SXD water extraction solution was prepared as follows: the slices of HQ (22.38 g), CH (5.60 g), JG (5.60 g) and SM (3.73 g) were evenly mixed, added 10 times water (MiLLi-Q IQ 7000, MA, USA) was added, soaked for 30 min, ZM (11.19 g) was added after boiling for 8 min, and boiled for another 7 min, and then strained. Next, the residue was added with 8 times the amount of water, boiled for 10 min, and filtrate was mixed twice [[53]16]. The water extraction was filtered with gauze, and freeze-dried to obtain SXD freeze-dried extract (Labconco Freezone 4.5L, MO, USA) The SXD freeze-dried extract was diluted with water to a certain concentration, and then used for subsequent experiments. SXD ethanol extraction solution was prepared as follows: HQ (22.38 g), ZM (11.19 g), CH (5.60 g), JG (5.60 g) and SM (3.73 g) were immersed in 485 mL of 70 % ethanol solution for 1 h and refluxed three times for 1 h each [[54]17]. The combined ethanol extraction was filtered by a 0.22 μm filter, and then diluted to 1 mg/mL for chemical fingerprint analysis. 2.1.2. Chemical fingerprint of SXD by ultra-performance liquid chromatography-quadrupole-time of flight-mass spectrometry (UPLC-Q-TOF/MS) The chemical fingerprint of SXD was determined by UPLC-Q-TOF/MS. The concentration of SXD water and ethanol extraction solution was 1 mg/mL, filtered by 0.22 μm filter, and injected into UPLC-Q-TOF/MS for analysis. The samples were analyzed by Waters Acquity SDS system. The mobile phases of UPLC were 0.1 % formic acid and acetonitrile on a ACQUITY UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm) at a flow rate of 0.3 mL/min. Optimized gradient elution was as follows: 0–2 min, 10%–35 % acetonitrile; 2–15 min, 35%–95 % acetonitrile; 15–17 min, 95 % acetonitrile; 17–18 min, 95%–10 % acetonitrile; 18–20 min, 10 % acetonitrile. Column temperature was maintained at 35 °C. The relevant parameters of MS were set as follows: positive ion mode analysis, capillary voltage of 3.0 kV, source temperature of 100 °C, desolvent temperature of 350 °C, cone-hole gas flow of 50 L/h, nitrogen as atomizing gas (600 L/h), scanning mode of full scan, scanning range m/z 50∼1200 Da; negative ion mode analysis, capillary voltage of 2.5 kV, source temperature of 100 °C, desolvent temperature of 280 °C, cone-hole gas flow of 50 L/h, nitrogen as atomizing gas (600 L/h), scanning mode of full scan, scanning range m/z 50∼1200 Da. Data was collected and analyzed using MassLynx V4.2 software (Waters, USA). 2.2. Network pharmacology 2.2.1. Screening the effective compounds and predicting the potential targets of SXD The compounds of SXD were collected from multiple sources, including the compounds by UPLC-Q-TOF/MS, the Traditional Chinese Medicine System Pharmacology Database and Analysis Platform (TCMSP, [55]https://tcmspe.com/), and literature mining. The compounds were collected using “Huangqi”, “Zhimu”, “Chaihu”, “Shengma” and “Jiegeng” as search terms in TCMSP [[56]18]. These compounds were screened by oral bioavailability (OB) ≥ 30 % and drug-likeness (DL) ≥ 0.18, and their PubChem ID ([57]https://pubchem.ncbi.nlm.nih.gov/) were recorded. OB represents the oral availability of pharmaceutical compounds, and DL refers to the similarity between compound and a known drug [[58]19]. At the same time, the effective compounds of SXD were imported into TCMSP database to obtain targets. The compounds with fewer targets were imported into the PubChem database to obtain their Canonical SMILES numbers and then imported into the Swiss Target Prediction database ([59]http://www.swisstargetprediction.ch/) to predict the possible targets. All targets were integrated and entered the UniProt database ([60]https://www.uniprot.org/) for verification, and duplicated targets were deleted to obtain the SXD-related targets. 2.2.2. Seeking out the potential targets of SXD in the treatment of CHD With “coronary heart disease” as the keyword, GeneCards Database ([61]https://www.genecards.org/), Online Mendelian Inheritance in Man Database ([62]https://www.omim.prg/), Therapeutic Target Database ([63]http://db.idrblab.net/ttd/) and DisGeNET Database ([64]https://www.disgenet.org/search) were used to search and screen the known CHD-related targets, and the repeated targets in the search results were discarded. However, there were too many results in those databases, so targets with a target score greater than 50 in the GeneCards database and a Score greater than 0.1 in the DisGeNET database were reserved. The UniProt knowledge base was used to get the standard target's name with the organism selected as “Homo sapiens”. Finally, the potential targets of SXD acting on CHD were obtained by using the Venny 2.1.0 platform ([65]https://bioinfogp.cnb.csic.es/tools/venny/index.html). 2.2.3. Construction and analysis of the protein-protein interaction (PPI) network These potential targets were imported into the STRING database ([66]https://string-db.org/) to construct the PPI network. The organism was set to “Homo sapiens” and the minimum required interaction score was “medium confidence (0.4)”, and then the PPI network was constructed. Subsequently, the PPI network was saved in “tsv” file format, which was imported into Cytoscape 3.8.2 software ([67]https://cytoscape.org/), an open-source software package project, to analyze the results by the network topology analysis function. In the PPI network, the degree value is an important index, and these targets above the average value usually play an important role. So, we ranked these targets according to degree value. At the same time, the important potential targets were screened out according to the average of degree and betweenness centrality of the node. Then, the target sub-networks were identified from the PPI network by using the MCODE plugin. Finally, the overlapping of the core sub-networks cluster and important potential targets was taken as the key targets. Furthermore, we analyzed the PPI network obtained with the CytoHubba plugin. Four target-ranking methods, including Edge Percolated Component, Eccentricity, Closeness, and Radiality, were used to analyze targets' weight (S-value) [[68]20]. Each method allowed the targets to receive a score, and then we ranked the targets according to the score. Targets with higher scores were ranked higher. For example, the highest scoring target was ranked first. However, many targets contained the same score, and it didn't matter how many targets contained the same score, the rank of the next target increased by one. Subsequently, all potential targets were ranked according to these four methods. If the rank of each target exceeded its median rank, the S-value of this target was increased by 1. Therefore, the maximum S-value of each target was 4 and the minimum was 0. Finally, we analyzed the targets' weight according to the S-value of targets for subsequent analysis. 2.2.4. Microarray data analysis of core targets Since myocardial infarction (MI) is one of the most serious and harmful diseases in CHD, the published microarray data (accession number [69]GSE66360 ) of circulating endothelial cells of patients with MI and control were searched through the GEO database to characterize the significance of key targets ([70]https://www.ncbi.nlm.nih.gov/geo). The gene expression profile dataset [71]GSE66360 comprised the gene expression data of circulating endothelial cells samples from 50 healthy donors and 49 patients with MI. This data was based on the [72]GPL570 Affymetrix Human Genome U133 Plus 2.0 array. For the expression profile dataset, we first obtained the annotation information of the probes according to the corresponding platform, mapped the probes to genes, removed multiple matches, used the mean value as the gene expression when multiple probes matched to a gene, and finally obtained the gene expression profile. The RStudio software's Limma package was used for the identification of the differences in expression level of key targets between control and patients with MI. The cutoff criteria in this analysis were set as p < 0.05 and |log FC| > 1. The key targets of expression difference were regarded as the core target of SXD in the treatment of CHD. 2.2.5. Construction of the compound-target-disease (C-T-D) network The effective compounds and their potential targets in SXD were made into qualified target table and type table, and imported into Cytoscape 3.8.2 software for visualizing and analyzing the interaction networks. Subsequently, a C-T-D network was constructed, while the size and color depth of the potential targets were adjusted according to the degree value. Finally, we analyzed the parameter rank-sum ratio (RSR) of this network, which was calculated from network topology parameters, including Betweenness, Closeness, Degree, Eccentricity, Neighborhood Connectivity and Average Shortest Path Length. 2.2.6. Contribution index (CI) calculation In order to screen the core compounds of SXD acting on CHD, a CI based on the intrinsic properties (content) of the effective compound in the corresponding herb and the importance of the effective compound in the C-T-D network was calculated [[73]21,[74]22]. The analysis was organized as follows: [MATH: Wj=iNSiN :MATH] [MATH: CIj=miinmi×Ci,jMj×OBj×RSRj×Wj×107 :MATH] where S[i] is the S-value of the potential target i; N is the number of potential targets acted by compound j; W[j] is the average S-value of all potential targets acted by compound j; m[i] is the weight of herb i in SXD; n is the total count of herbs in SXD; [MATH: mi inmi< /mfrac> :MATH] is the proportion of each herb in SXD (a common compound is the sum of the proportions of the herbs, which belongs in these herbs); C[i,j] is the content of the compound j in herb i (the content of effective compounds was collected by literature mining, which was expressed as the average content reported in the literature); M[j] is the molecular weight of the compound j; [MATH: Ci,jMj :MATH] represents the molar concentration of 1 g of the compound in SXD; OB[j] is the oral bioavailability of compound j; RSR[j] is the rank-sum ratio of the integrated network topology parameters of the compound j in the C-T-D network. This method was used to reflect the ratio of the compounds in SXD and the weight efficiency of their effect on potential CHD-related targets. Thus, a compound with a higher CI means a higher probability of the effect on CHD. 2.2.7. Gene ontology (GO) and Kyoto Encyclopedia of genes and genomes (KEGG) enrichment analysis and target-pathway (T-P) network construction The potential targets were imported into the DAVID platform ([75]https://david.ncifcrf.gov/tools.jsp) for GO function enrichment analysis and KEGG pathway enrichment analysis, and the “select species” was set to “Homo sapiens”. GO enrichment analysis involved the biological process (BP), cellular component (CC) and molecular function (MF) of the potential targets, and the top 10 results were visualized as bar graphs. Based on the S-values of all potential targets, the weight of the pathway P was the sum of the corresponding targets. Pathway analysis was performed with KEGG, and the top 20 results were visualized as a bubble diagram by using bioinformatics platform ([76]http://www.bioinformatics.com.cn/). Subsequently, the T-P network was constructed by Cytoscape 3.8.2 software according to the top 20 of P-weight [[77]23]. 2.2.8. Molecular docking First, we downloaded the 3D structures of the core compounds in SXD from the PubChem database and saved them in “sdf” format, and then downloaded the 3D structures of the core targets from the PDB database ([78]https://www.rcsb.org/) and saved them in “pdb” format. We would use Chem 3D software to construct their 3D structures for some compounds, which did not have 3D structures in the database. These compounds and targets were prepared for ligands and receptors by using AutoDockTools software, which included operations such as dehydration, hydrogenation, and deionization. These compound-target pairs were then docked, and the docking scores were made into a heat map. Finally, the lower binding energy docking results of each core target were visualized by LigPlot ([79]https://www.ebi.ac.uk/thornton-srv/software/LigPlus/) and PyMOL ([80]https://pymol.org/2/) software. 2.3. In vivo experiments 2.3.1. Experimental animals Male specific pathogen-free Sprague-Dawley rats (180–200 g) were purchased from the Beijing Vital River Laboratory Animal Technology Co Ltd (Beijing, China). All experimentals were conducted according to the Guide for the Care and Use of Laboratory Animals by the National Institute of Health (USA). All procedures were strictly approved by the Animal Experiment Ethics Committee of the Shaanxi University of Chinese Medicine (Ethics Approval No. SUCMDL20220310006). All rats were housed under the standard conditions (temperature, 25 ± 2 °C; constant humidity, 55 ± 5 %; 12 h dark/light cycle; freely available food and purified water). After one week of adaptive feeding, all rats were randomly selected for sham surgery or CHD model group. The rat CHD model was established by exhaustive swimming, restricted diet and ligation of the left anterior descending coronary artery. The rats in sham surgery group were fed normally. On the first day of the experiment, the rats in the CHD model group swam exhaustively once a day, with the standard of exhaustively swimming being that the head of the rats could not emerge from the water level for 10 s. At the same time, the food control was carried out, and the daily intake was half of the normal diet for 21 days. On day 22 of the experiment, a left thoracotomy was performed after rats were anesthetized with 2 % isoflurane. The rats were laid flat and disinfected, and incised from the front skin of the neck to extract tissue and muscle, exposing the windpipe, which was then inserted into the oral cavity and connected to a ventilator. The left third and fourth intercostal spaces of the rat were then propped open with hemostatic forceps, and the left anterior descending coronary artery (LAD) was then ligated with a 6-0 surgical sutures. After the ligation, the heart was placed back in the chest cavity, the air that entered the chest cavity was expelled, the chest cavity was closed, and the skin was sutured. 0.25 mL penicillin was injected immediately after chest closure. For the sham-operated group, rat LADs were only threaded and not ligated, and the remaining procedures were the same as those in the CHD group [[81]24]. All rats were fed normally after modeling. The 24 CHD-operated rats were divided into the CHD model group, the SXD low-dose group (4.33 g/kg), SXD high-dose group (12.99 g/kg), and Betaloc group (10 mg/kg, AstraZeneca, UK) according to a random number control table. Rats without ligation were placed in the sham-operated group. The rats in the sham and CHD model groups were given the same volume of normal saline via intragastric administration. 2.3.2. Myocardial infarct size detection After 28 days of treatment, the rats were given 12 h of fasting and water, 2 % isoflurane for anesthesia, all the hearts of the rats were separated, weighed after absorbing water, and the heart weight index (HWI) was calculated. Three rats in each group were randomly selected and their hearts were transferred at 0–4 °C PBS and frozen at −20 °C for 30 min. Slices of the heart were taken with a thickness of 2 mm and placed in 2 % red tetrazolium solution in a dark water bath at 37 °C for 30 min. The container was slightly shaken every 5 min to make it fully stained. The myocardial sections were removed and washed with PBS solution for 3–5 min. Next, the sections were soaked in 4 % paraformaldehyde solution for 24 h, and the myocardial infarction size was evaluated by 2,3,5-triphenyl-2H-tetrazolium chloride (TTC) staining. Image Pro Plus 6.0 was used to quantify the myocardial infarction size. Myocardial infarction area divided by total area was regarded as the percentage of myocardial infarction. 2.3.3. Hematoxylin and eosin (H&E) and Masson's trichrome staining The myocardial tissue was fixed in 4 % paraformaldehyde solution, gradient dehydrated with 70–100 % ethanol, embedded in paraffin, and sliced (4–5 μm). Hematoxylin-eosin (H&E) staining was used to observe myocardial structural changes under an optical microscope. The paraffin sections were dewaxed to water, followed by Masson staining of the sections. After dehydration and sealing, the changes of myocardial collagen fibers were observed under an optical microscope to evaluate the degree of myocardial fibrosis. 2.3.4. ELISA Blood was collected by intubation in the abdominal aorta of rats in each group, and serum was obtained by centrifugation at 3000 rpm for 15 min. The levels of serum IL-6 (MM-0190R1, MEIMIAN, China), IL-1β (MM-0047R1, MEIMIAN, China), Aldosterone (ALD, YJ002876, MEIMIAN, China), Angiotensin II (Ang II, YJ058803, MEIMIAN, China), N-terminal pro-brain natriuretic peptide (NT-proBNP, YJ003242, MEIMIAN, China) were assessed at 450 nm was detected by a microplate reader (RT-6100, Rayto, China) according to the ELISA kits. 2.4. In vitro experiments 2.4.1. Cell culture and the hypoxia/reoxygenation (H/R) cell model Rat cardiomyocytes (H9c2) were normally cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10 % FBS at a 37 °C normoxic atmosphere with 5 % CO[2]. The DMEM medium containing 10 % FBS was used as a complete medium. An H/R cell model was used to mimic the pathological process of cardiac injury. In brief, the H9c2 cells were in six groups: the control group, H9c2 cells were normally cultured; the H/R group, H9c2 cells were cultured in D-Hank's medium in a hypoxia atmosphere at 37 °C with 5 % CO[2] and 95 % N[2] for 8 h, and further cultured in complete medium at normoxic atmosphere at 37 °C with 5 % CO[2] and 95 % O[2] for 12 h; the H/R + treatment group, H9c2 cells were incubated in an anesthesia induction chamber with fresh gas (95 % O[2] and 5 % CO[2]) and the addition of drug (100 μM Diazoxide, 250 μg/mL SXD, 500 μg/mL SXD, 1 mg/mL SXD) for 24 h, and then treated with H/R. 2.4.2. Cell viability assay 3-(4,5)-dimethylthiahiazo (-z-y1)-3,5-di- phenytetrazoliumromide (MTT) assay was useded in the current study to evaluate cell viability after various treatments. After the collection of H9c2 cells, 10 μL MTT reagents (5 mg/mL) were added into the 96-well plate (1.0 × 10^5 cells/1 mL) and then incubated for 4 h. After the solution was discarded, 100 μL of dimethyl sulfoxide was added into the culture plate. Finally, an Imark microplate reader was used for measuring the cell absorbance at 570 nm. 2.4.3. Detection of intracellular reactive oxygen species (ROS) generation Assessment of intracellular ROS changes in cells during apoptosis was performed using Reactive Oxygen Species Assay Kit (S0033 M, Beyotime, China). A cell density of 10^5/mL was placed in each 35-mm culture plate. DCFH-DA was diluted in serum-free medium at 1:1000 to give a final concentration of 10 μM/L. Cells were collected and suspended in diluted DCFH-DA at a concentration of 1 million to 20 million cells/mL and incubated in a cell incubator at 37 °C for 20 min. The mixture was reversed every 5 min to make full contact between the probe and the cell. Cells were washed three times with serum-free cell culture medium and then analyzed by the Beckman CytoFlex Sflow cytometer system. 2.4.4. Quantification of cellular apoptosis The apoptosis of H9c2 cells was detected using Annexin V-FITC/PI stain kit (C1062 M, Beyotime, China). In brief, after the cell collection, the H9c2 cells were washed by PBS twice, centrifuged (500×g) for 5 min, and then mixed with 500 μL working buffer. The cells were further added with 5 μL Annexin V binding with FITC and 5 μL PI for 15 min staining. Lastly, cell signal was measured, and the apoptosis rate was analyzed by the Beckman CytoFlex Sflow cytometer system. 2.4.5. ELISA H9c2 cells were inoculated in 6-well plates with a density of 1 × 10^5 cells per well. After the drug intervention, the supernatant was taken, and the optical density of tumor necrosis factor (TNF, FU-D1393, BIOFINE, USA), interleukin-6 (IL-6, FU-D1083, BIOFINE, USA) and interleukin-1 beta (IL-1β, FU-D1032, BIOFINE, USA) at 570 nm was detected by a microplate reader according to the ELISA kits. 2.4.6. Real time quantitative PCR (RT-qPCR) Total RNA was isolated from H9c2 cells with a RNeasy mini kit as described in the manufacturer's manual. These RNA samples were then reverse transcribed into cDNAs using Revert Ace kit (TOYOBO, Japan) and the mRNAs were measured through the SYBR-Green RT-qPCR. Cellular tumor antigen p53 (TP53), Caspase 3 and Cytochrome C expression was calculated relative to the β-actin expression ([82]Table 1). Table 1. Primers sequences used for RT-qPCR. Genes Forward primer Reverse primer Size (bp) TP53 AGATGTTCCGAGAGCTGAATGAG TTTTTTATGGCGGGACGTAGA 130 Caspase3 ACTGGAAAGCCGAAACTCTTCATCA GGAAGTCGGCCTCCACTGGTATC 127 Cytochrome C GCTGGATTCTCTTACACAGATGCC GGTCTGCCCTTTCTCCCTTCTT 151 β-actin AGAGGGAAATCGTGCGTGA CATTGCCGATAGTGATGACCT 144 [83]Open in a new tab 2.4.7. Western blot analysis H/R-induced H9c2 cells were treated with SXD, and then the proteins from each treatment group were isolated by radio immunoprecipitation assay buffer (RIPA, P1003B, Beyotime, China). The supernatant was centrifuged at 14,000 rpm for 15 min at 4 °C, and the total protein concentration was determined, followed by quantitation with the bicinchoninic acid assay (BCA) protein assay kit (P0010, Beyotime, China). After denaturation in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels, the equal amounts of proteins were transferred to nitrocellulose membrane (HATF00010, Millipore, USA). Blots were blocked with 5 % fat-free milk in TBS with Tween-20 (TBST) at room temperature (approximately 25 °C) for 1 h, and then Caspase3 (1:500, 19677-1-AP, Proteintech, USA), Caspase9 (1:500, bs-20773R, Bioss, Switzerland), Cytochrome C (1:500, bs-0013R, Bioss, Switzerland), β-actin (1:5000, 200068-8F10, ZEN-BIOSCIENCE, China) primary antibodies were added in 5 % blocking buffer at 4 °C overnight. After incubation with the corresponding HRP-conjugated secondary antibodies (1:2000, SA00001-1 and SA00001-2, Proteintech, USA), the enhanced chemiluminescence (ECL) kit (WBKLS0500, Millipore, USA) was used for detection. The grey scale values of protein bands were analyzed using Image J software (NIH, USA). 2.4.8. Caspase-3 enzymatic activities The caspase 3 can catalyze substrate acetyl-Asp-Glu-Val-Asp p-nitroanilide (Ac-DEVD-pNA) to produce yellow p-nitroaniline (pNA), so as to determine the activity of caspase 3 by using the caspase 3 Activity Assay Kit (C1115, Beyotime, China). The protein concentrations were determined by Bradford protein assay kit (P0006, Beyotime, China). Cellular extracts (30 μg) were incubated in a 96-well microtitre plate with 20 ng Ac-DEVD-pNA for 4 h at 37 °C, and OD[405nm] values were measured by using a microplate reader (BioTek Instruments, USA). 2.5. Statistical analysis The study data were analyzed using SPSS 22.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 8.3 software (GraphPad Software, USA), and comparisons between the groups were performed using Student's t-test (two-group comparison) and one-way analysis of variance (ANOVA, more than two groups). All data were based on at least three independent experiments. The statistical results are expressed as the mean ± standard deviation, and then statistical charts were drawn according to the statistical results. Value of p < 0.05 was considered to be statistically significant. 3. Results 3.1. Chemical fingerprint of SXD MassLynx V4.2 software was used to collect the data, and the obtained data were quickly matched to the database in UNIFI 1.8 (Waters) platform. Before conducting the matching process, we created a compound library in-house by gathering chemical formulas for SXD from various sources such as Chemical Book, PubChem, TCMSP, and relevant literature. The compounds in the SXD ethanol extraction solution produced the [M + H]^+, [M + Na]^+, [M + NH4]^+ and [M + K]^+ peaks in the positive ion mode, and produced the [M − H]^-, [M + CH3COO]^- and [M + HCOO]^- peaks in the negative ion mode. The chemical fingerprint of SXD was profiled by UPLC-Q-TOF/MS ([84]Fig. 1). Meanwhile, we identified 113 compounds in SXD, including flavonoids, organic acids, alkaloids, terpenoids, and saponins. [85]Table 2 lists the retention time, molecular formula, adduction ion and fragment ion of all identified compounds, and these compounds have possible biological activity. Seventeen of them were unambiguously identified by comparison with reference compounds ([86]Fig. S1). Fig. 1. [87]Fig. 1 [88]Open in a new tab Base peak ion (BPI) chromatogram of SXD based on UPLC-Q-TOF/MS. (A) Ethanol extraction solution; (B) Reference compounds solution. Table 2. Identification of chemical compounds in SXD by UPLC-Q-TOF/MS. No. Component name Formula Neutral mass (Da) Observed m/z Mass error (ppm) Adducts Observed RT (min) Fragments (m/z) Source 1 l(+)-Arginine C[6]H[14]N[4]O[2] 174.1117 173.1041 −1.7 [M − H]^- 0.78 131.0821 HQ 2 7-O-Methylmangiferin C[20]H[20]O[11] 436.1005 495.1097 −9.5 [M + CH[3]COO]^- 0.90 435.9157, 351.0805, 245.0365 ZM 3 Ononin C[22]H[22]O[9] 430.1264 429.1228 8.6 [M − H]^- 0.94 431.1304, 269.1013, 224.0434, 118.0476 HQ 4 Nicotinic acid C[6]H[5]NO[2] 123.0320 182.0448 −6.1 [M + CH[3]COO]^- 0.97 – HQ 5 Boc-D-Tyr-OH C[14]H[19]NO[5] 281.1263 299.1596 −1.7 [M + NH[4]]^+ 1.01 265.1544, 250.1418, 235.0709 HQ 6 N-Carbobenzoxy-dl-Leucine C[14]H[19]NO[4] 265.1314 283.1665 4.5 [M + NH[4]]^+ 1.35 230.1685, 212.1597 HQ 7 4-Hydroxybenzaldehyde C[7]H[6]O[2] 122.0368 181.0501 −2.7 [M + CH[3]COO]^- 1.46 – ZM 8 Guaiacol C[7]H[8]O[2] 124.0524 123.0451 −0.3 [M − H]^- 1.65 – SM 9 Neomangiferin[89]^a C[25]H[28]O[16] 584.1377 583.1161 – [M − H]^- 1.73 301.0249, 272.0245, 243.0229 ZM 10 4-Hydroxy-3,5-dimethoxycinnamic acid C[11]H[12]O[5] 224.0685 223.0615 1.4 [M − H]^- 1.98 223.0618, 135.0454 SM 11 1-Caffeoylquinic acid C[16]H[18]O[9] 354.0951 353.0878 0.1 [M − H]^- 2.04 353.0810, 85.0263 JG 12 Mangiferin[90]^a C[19]H[18]O[11] 422.3396 421.0559 – [M − H]^- 2.11 301.0249, 271.0113, 258.0078 ZM 13 Cimicifugamide or its isomer C[25]H[31]NO[10] 505.1948 506.2054 6.5 [M+H]^+ 2.18 344.1657, 177.0851, 145.0580 SM 14 Caffeic acid[91]^a C[9]H[8]O[4] 180.1574 179.0247 – [M − H]^- 2.4 165.0556, 135.0451, 134.0369 HQ, SM 15 Kaempferol-3-O-Rutinoside C[27]H[30]O[15] 594.1585 593.1511 −0.1 [M − H]^- 2.43 285.0423, 239.0477 HQ 16 Rutin[92]^a C[27]H[30]O[16] 610.5175 609.1237 – [M − H]^- 2.52 300.0175, 271.0113, 243.0161, 227.0225 CH 17 Calycosin 7-O-glucoside[93]^a C[22]H[22]O[10] 446.4041 491.1012 – [M + HCOO]^- 2.62 283.0540, 268.0344, 239.0275, 211.0333 HQ 18 Naringin C[27]H[32]O[14] 580.1792 639.1935 0.7 [M + CH[3]COO]^- 2.71 433.0741, 271.0401, 151.0402 HQ 19 Isorhamnetin 3,4′-diglucoside C[28]H[32]O[17] 624.1690 623.1613 −0.7 [M − H]^- 2.74 541.1690, 342.1341, 301.0348 HQ 20 Tuberosine A C[19]H[21]NO[5] 343.1420 342.1340 −2.1 [M − H]^- 2.76 342.1341, 193.0506 CH 21 Genistin C[21]H[20]O[10] 432.1057 477.1036 −0.6 [M + HCOO]^- 2.9 268.0371, 211.0101 HQ 22 Acacetin C[16]H[12]O[5] 284.0685 283.0608 −1.5 [M − H]^- 2.91 226.1017, 212.0438 JG 23 Cimicifugic acid D C[20]H[18]O[10] 418.0900 417.0825 −0.5 [M − H]^- 2.93 255.0475, 179.0306 SM 24 Timosaponin D C[45]H[74]O[19] 918.4824 963.4801 −0.6 [M + HCOO]^- 2.95 755.4228, 593.1825, 191.0169 ZM 25 Ferulic Acid[94]^a C[10]H[10]O[4] 194.1840 193.0453 – [M − H]^- 2.96 178.0206, 149.0545, 134.0313 SM 26 Timosaponin B II[95]^a C[45]H[76]O[19] 920.4981 965.4958 −0.5 [M + HCOO]^- 2.99 919.4790 ZM 27 Isoferulic acid[96]^a C[10]H[10]O[4] 194.1840 193.0393 – [M − H]^- 3.04 178.0306, 134.0363 SM 28 5-Methyl furfural C[6]H[6]O[2] 110.0368 109.0306 9.6 [M − H]^- 3.07 108.0217 SM 29 Paeoniflorin C[23]H[28]O[11] 480.1632 519.1229 −6.6 [M+K]^+ 3.09 449.1431, 327.1342, 165.0415, 121.0984 SM 30 Baicalin C[21]H[18]O[11] 446.0849 445.0781 1 [M − H]^- 3.15 269.0297, 113.5667 HQ 31 Ethyl 4-hydroxy-3-methoxycinnamate C[12]H[14]O[4] 222.0892 221.0818 −0.6 [M − H]^- 3.23 187.0971, 125.0970 SM 32 Cimilactone A C[33]H[50]O[9] 590.3455 591.3516 −1.9 [M+H]^+ 3.26 433.3441, 415.3356, 301.0968 SM 33 Desapioplatycodin D C[52]H[84]O[24] 1092.5353 1091.5288 0.7 [M − H]^- 3.27 1091.5263 JG 34 Puerarin C[21]H[20]O[9] 416.1107 475.1238 −1.8 [M + CH[3]COO]^- 3.31 267.0659, 252.0424, 223. 0388 SM 35 N-p-trans-Coumaroyltyramine C[17]H[17]NO[3] 283.1208 282.1129 −2.5 [M − H]^- 3.51 282.1137, 179.0347, 152.9960 ZM 36 Jaranol C[17]H[14]O[6] 314.0790 373.0926 −0.7 [M + CH[3]COO]^- 3.53 299.0926, 282.1137, 179.0347, 152.9960 HQ 37 Methylnissolin-3-O-glucoside C[23]H[26]O[10] 426.1526 507.1503 −1 [M + HCOO]^- 3.54 483.2362 HQ 38 Platycoside M[3] C[52]H[80]O[24] 1088.5040 1147.5211 2.9 [M + CH[3]COO]^- 3.59 933.4736, 653.3382 JG 39 7,2′-dihydroxy-3′,4′-dimethoxyisoflavane-7-O-glucoside or its isomer C[23]H[28]O[10] 464.1683 463.1603 −1.4 [M − H]^- 3.61 353.1018, 301.1079 HQ 40 Isochlorogenic acid C C[25]H[24]O[12] 516.1268 517.1384 8.4 [M+H]^+ 3.67 451.3312, 269.1081 CH 41 Timosaponin F C[39]H[64]O[15] 772.4245 817.4225 −0.3 [M + HCOO]^- 3.78 771.4069, 609.2312 ZM 42 Luteolin[97]^a C[15]H[10]O[6] 286.2363 285.0276 – [M − H]^- 3.84 145.9290 JG 43 Anemarsaponin B C[45]H[74]O[18] 902.4875 947.4865 0.8 [M + HCOO]^- 3.85 901.4807, 797.4692, 739.4276, 639.3372 ZM 44 Methylnissolin C[17]H[16]O[5] 300.0998 299.0925 −0.1 [M − H]^- 3.86 301.0325, 163.3195, 147.0660, 113.7024 HQ 45 Quercetin[98]^a C[15]H[10]O[7] 302.2357 301.0249 – [M − H]^- 3.91 285.0276, 268.0273 HQ, CH 46 Icariin I C[27]H[30]O[11] 530.1788 569.1412 −1.4 [M+K]^+ 3.93 451.3323, 258.1019, 270.0794 ZM 47 calycosin[99]^a C[16]H[12]O[5] 284.2635 283.0554 – [M − H]^- 3.95 268.0344, 211.0333 HQ 48 Nepasaikosaponin K C[48]H[80]O[18] 944.5345 989.5345 1.9 [M + HCOO]^- 4.06 943.5290, 829.4599, 783.4519, 725.4124, 199.8049 CH 49 Officinalisinin I C[45]H[76]O[19] 920.4981 959.4648 3.7 [M+K]^+ 4.09 741.8390, 417.1752, 273.2432 ZM 50 Isoflavanone C[15]H[12]O[2] 224.0837 269.0817 −0.7 [M + HCOO]^- 4.16 193.0141, 108.0236 HQ 51 6-Methylcoumarin C[10]H[8]O[2] 160.05243 161.0602 −0.19 [M+H]^+ 4.21 133.0875, 105.0690 HQ 52 Anemarrhenasaponin-Ia C[40]H[68]O[14] 772.4609 831.4751 0.4 [M + CH[3]COO]^- 4.49 839.4802, 785.4686, 767.4565 ZM 53 Saikosaponin F C[48]H[80]O[17] 928.5396 973.5380 0.2 [M + HCOO]^- 4.68 927.5326, 781.4735, 619.4216 CH 54 Oroxylin A-7-O-glucuronide C[22]H[20]O[11] 460.1006 483.0868 −6.3 [M+Na]^+ 4.69 423.3765, 405.3687 HQ 55 Kaempferol[100]^a C[15]H[10]O[6] 286.2363 285.0276 – [M − H]^- 4.7 – CH, ZM, CH 56 Isorhamnetin[101]^a C[16]H[12]O[7] 316.2623 315.0424 – [M − H]^- 4.84 300.0175, 285.0276 HQ, CH 57 Apigenin C[15]H[10]O[5] 270.0528 315.0510 −0.2 [M + HCOO]^- 4.85 251.8408, 241.0652, 225.1435, 151.0348 HQ 58 Astragaloside III C[41]H[68]O[14] 784.4609 829.4586 −0.6 [M + HCOO]^- 5.05 829.4602, 783.4532, 743.4594, 489.3580, 393.1592 HQ 59 Timosaponin I C[39]H[66]O[14] 758.4453 803.4425 −1.1 [M + HCOO]^- 5.13 757.4374, 665.3908, 595.3806, 529.3520 ZM 60 Cubebin C[20]H[20]O[6] 356.1260 355.1181 −1.7 [M − H]^- 5.16 161.0453 CH 61 Isoliquiritigenin C[15]H[12]O[4] 256.0736 255.0663 0.1 [M − H]^- 5.26 213.1294 ZM 62 Anemarrhenasaponin A[2] C[39]H[64]O[14] 756.4296 801.4273 −0.7 [M + HCOO]^- 5.35 635.3797, 593.3678, 375.1848 ZM 63 Sainfuran C[16]H[14]O[5] 286.0841 285.0762 −2.4 [M − H]^- 5.43 – CH 64 Pratensein C[16]H[12]O[6] 300.0634 299.0554 −2.4 [M − H]^- 5.5 223.0397, 196.0490 HQ 65 Formononetin C[16]H[12]O[4] 268.0736 267.0662 −0.4 [M − H]^- 5.54 252.0425, 223.0397, 195.0450, 167.0501 HQ 66 (−)-Catechin hydrate C[15]H[14]O[6] 290.0790 291.0890 9.1 [M+H]^+ 5.55 269.1081, 197.0906, 152.0943 CH 67 Astragaloside Ⅱ C[43]H[70]O[15] 826.4715 871.4731 3.9 [M + HCOO]^- 5.71 778.2931, 635.3801, 577.3384, 499.3055 HQ 68 Saikosaponin A[102]^a C[42]H[68]O[13] 780.9815 825.4397 – [M + HCOO]^- 5.79 779.4495 CH 69 Saikosaponin B[1] C[42]H[68]O[13] 780.4660 825.4650 1 [M + HCOO]^- 5.81 779.4588, 755.4225, 617.4057 CH 70 Saikosaponin B[4] C[43]H[72]O[14] 812.4922 811.4896 5.8 [M − H]^- 5.95 753.4062, 591.3549, 267.0649, 243.8989 CH 71 Cimicifugoside C[37]H[54]O[11] 674.3666 719.3640 −1.1 [M + HCOO]^- 6.1 655.3469, 520.3057, 295.2272 SM 72 Agroastragaloside Ⅰ C[45]H[74]O[16] 870.4977 915.4981 2.4 [M + HCOO]^- 6.2 915.4952, 419.9543, 329.2333 HQ 73 Malonylsaikosaponin A C[45]H[70]O[16] 866.4664 865.4593 0.2 [M − H]^- 6.24 631.3494, 299.0907, 193.0508 CH 74 Platycoside K C[42]H[68]O[17] 844.4457 867.4323 −3 [M+Na]^+ 6.25 787.4346, 478.3487 JG 75 Anemarrhenasaponin III C[39]H[64]O[14] 756.4296 801.4280 0.2 [M + HCOO]^- 6.37 755.4227, 635.3798, 577.3380 ZM 76 Cis-hinokiresinol C[17]H[16]O[2] 252.1150 251.1077 −0.2 [M − H]^- 6.72 235.0763, 162.8401, 117.0348 ZM 77 Hippeastrine C[17]H[17]NO[5] 315.1107 314.1019 −4.7 [M − H]^- 6.73 295.2269, 251.1077, 235.0763 ZM 78 Saikosaponin E C[42]H[68]O[12] 764.4711 809.4690 −0.4 [M + HCOO]^- 6.75 763.4639, 753.4090, 637.3955, 619.3831 CH 79 Anemarsaponin F C[50]H[82]O[23] 1050.5247 1095.5236 0.6 [M + HCOO]^- 6.83 1049.5169, 915.4600, 773.4369, 721.3797, 617.3706 ZM 80 6″-O-Acetylsaikosaponin b3 C[45]H[74]O[15] 854.5028 913.5159 −0.8 [M + CH[3]COO]^- 7.02 885.4846, 677.3909, 617.3681, 559.3280 CH 81 7-O-methylisomucronulatol C[18]H[20]O[5] 316.1311 315.1227 −3.6 [M − H]^- 7.03 283.0609 HQ 82 Astragaloside Ⅳ[103]^a C[41]H[68]O[14] 784.9702 783.3947 – [M − H]^- 7.35 739.4159, 677.3754, 449.2163 HQ 83 25-O-Acetylcimigenol xyloside C[37]H[58]O[10] 662.4030 661.3932 −3.8 [M − H]^- 7.41 659.3799, 513.2282, 389.1742 SM 84 Saikosaponin D[104]^a C[42]H[68]O[13] 780.9815 825.4397 – [M + HCOO]^- 7.49 779.4495 CH 85 Timosaponin AIII[105]^a C[39]H[64]O[13] 740.4347 785.4329 0 [M + HCOO]^- 7.87 739.4159 ZM 86 Ethyl caffeate C[11]H[12]O[4] 208.0736 226.1073 −0.5 [M + NH[4]]^+ 7.96 207.1474, 197.1642 CH 87 Diosgenin C[27]H[42]O[3] 414.3134 415.3168 −9.3 [M+H]^+ 7.97 415.3153, 378.2821, 361.2735 ZM 88 Stigmasterol C[29]H[48]O 412.3705 451.3329 −1.8 [M+K]^+ 8.24 397.2922, 335.2605, 259.2340, 97.0604 ZM, CH, SM 89 6″-O-Acetylsaikosaponin D C[44]H[70]O[14] 822.4766 867.4751 0.4 [M + HCOO]- 8.27 821.4694, 779.4586, 761.4480, 617.4050 CH 90 (+)-Anomalin C[24]H[26]O[7] 426.1679 427.1780 6.6 [M+H]+ 8.45 317.2493, 171.1482 CH 91 27-Dexyactein C[37]H[56]O[10] 660.3874 659.3796 −0.7 [M − H]- 8.51 659.3798, 559.3257, 523.3100, 485.3272 SM 92 Cimigenol 3-O-β-d-Xylopyranoside C[35]H[56]O[9] 620.3924 665.3903 −0.6 [M + HCOO]- 8.75 619.3835, 595.2781, 315.2512 SM 93 Platycoside M[1] C[36]H[54]O[12] 678.3615 696.3953 −0.1 [M + NH[4]]+ 8.99 263.2641, 245.2559 JG 94 Phenprobamate C[10]H[13]NO[2] 179.0946 197.1271 −7 [M + NH[4]]+ 9.05 149.0562 HQ 95 Acetytastragaloside C[47]H[74]O[17] 910.4926 955.4914 0.7 [M + HCOO]^- 9.63 909.4861, 821.4706, 603.3378, 279.2325 HQ 96 Coronaric acid C[18]H[32]O[3] 296.2351 295.2279 0.2 [M − H]^- 10.23 277.2175, 195.1389, 116.9286 JG 97 β-Sitosterol C[29]H[50]O 414.3862 453.3479 −3.1 [M+K]^+ 10.82 277.2434, 149.0556 SM 98 7,8-Didehydrocimigenol C[30]H[46]O[5] 486.3345 545.3481 −0.4 [M + CH[3]COO]^- 10.96 485.3272, 445.2909, 393.1357, 185.1178 SM 99 Cimigenol C[30]H[48]O[5] 488.3502 547.3637 −0.5 [M + CH[3]COO]^- 11.26 487.3410, 383.2895, 279.2326, 243.8991 SM 100 Pentadecanoic acid C[15]H[30]O[2] 242.2246 287.2247 6.5 [M + HCOO]^- 11.32 – SM 101 Scoparone C[11]H[10]O[4] 206.0579 265.0732 5.3 [M + CH[3]COO]^- 12.55 265.0752 CH 102 Anhydroicaritin C[21]H[20]O[6] 368.1260 427.1440 9.8 [M + CH[3]COO]^- 13.47 277.2170, 251.1648, 116.9281 ZM 103 Prosaikogenin A C[36]H[58]O[8] 618.4132 641.4019 −0.7 [M+Na]^+ 13.68 457.2680, 177.1953, 93.1013 CH 104 Linolenic acid C[18]H[30]O[2] 278.2246 323.2230 0.6 [M + HCOO]^- 14.17 323.2208, 305.2109 HQ 105 Linoleic acid C[18]H[32]O[2] 280.2402 279.2329 0 [M − H]^- 14.62 279.2329, 211.1338, 116.9280 HQ 106 Bifendate C[20]H[18]O[10] 418.0900 419.0984 2.6 [M+H]^+ 14.63 338.3643, 270.3067, 245.2555, 225.0726 HQ 107 Palmitic acid C[16]H[32]O[2] 256.2402 255.2326 −1.5 [M − H]^- 15.56 261.1982, 205.1383, 165.0720, 149.0771, 125.0043 HQ 108 Linoleyl acetate C[20]H[36]O[2] 308.2715 307.2642 0 [M − H]^- 16.16 281.2481, 234.8201, 116.9279 CH 109 Anemarsaponin E C[46]H[78]O[19] 934.5137 973.4805 3.7 [M+K]^+ 16.29 905.4240, 849.4228, 734.4902 ZM 110 Chrysanthemaxanthin C[40]H[56]O[3] 584.4230 607.4104 −2.9 [M+Na]^+ 17.14 485.3876, 312.3499, 256.2919 ZM 111 Baohuoside I C[27]H[30]O[10] 514.1839 532.2143 −6.4 [M + NH[4]]^+ 17.8 413.2813, 149.0554 ZM 112 Hederagenin C[30]H[48]O[4] 472.3553 531.3695 0.8 [M + CH[3]COO]^- 18.02 416.7036, 316.7786, 216.8533 HQ 113 Mairin C[30]H[48]O[3] 456.3604 455.3531 0.1 [M − H]^- 18.22 455.3526, 400.7358, 339.7611, 222.8403 184.9164 HQ [106]Open in a new tab ^a Compounds identified by reference compounds. 3.2. Collecting and screening of effective compounds and the targets in SXD In the present study, based on UPLC-Q-TOF/MS analysis and the set conditions in the database, a total of 57 potentially effective compounds were retrieved and screened from chemical fingerprint, TCMSP and literature mining, including 17, 11, 12, 8, 5 and 4 form HQ, ZM, CH, SM, JG and common compounds in SXD ([107]Table 3). Among them, isorhamnetin (M54) was shared in HQ and CH, kaempferol (M55) was shared in HQ, ZM and CH, quercetin (M56) was shared in HQ and CH, and stigmasterol (M57) was shared in ZM, CH and SM. Some compounds that did not meet both the OB and DL criteria were also selected in the cases of high bioactivities and huge amounts, including astragaloside IV (M15), ononin (M16), calycosin 7-O-glucoside (M17), timosaponin B II (M28), saikosaponin A (M38), saikosaponin D (M39), rutin (M40), caffeic acid (M46), isoferulic acid (M47), ferulic acid (M48) and platycodin D (M53). Specifically, astragaloside IV, calycosin 7-O-glucoside, saikosaponin A, saikosaponin D, timosaponin B II, isoferulic acid and platycodin D have been chosen as the marker compounds for quality control of corresponding herbs in Chinese Pharmacopoeia [[108]25]. Similarly, although ononin holds low OB value, it also exhibits remarkable pharmacological effects by alleviating H[2]O[2]-induced cardiomyocyte apoptosis and improving cardiac function [[109]26]. In addition, cardioprotective effects of caffeic acid and ferulic acid are also good [[110]27,[111]28]. Based on the above considerations, it was reasonable to believe that 57 compounds could be listed as potentially effective compounds for SXD. Subsequently, the effective compounds of SXD were entered into the database to obtain the corresponding targets. A total of 502 SXD-related targets were screened by correcting and deleting duplicate values, which were included in the subsequent analysis. Table 3. The effective compounds of SXD. No. Name PubChem CID 2D structure OB (%) DL Herb M1 Mairin[112]^a 64971 Image 1 55.38 0.78 HQ M2 Jaranol[113]^a 5318869 Image 2 50.83 0.29 HQ M3 Hederagenin[114]^a 73299 Image 3 36.91 0.75 HQ M4 3,9-di-O-methylnissolin 15689655 Image 4 53.74 0.48 HQ M5 7-O-methylisomucronulatol[115]^a 15689652 Image 5 74.69 0.30 HQ M6 Methylnissolin-3-O-glucoside[116]^a 74977390 Image 6 36.74 0.92 HQ M7 Methylnissolin[117]^a 5319733 Image 7 64.26 0.42 HQ M8 Bifendate[118]^a 108213 Image 8 31.10 0.67 HQ M9 Formononetin[119]^a 5280378 Image 9 69.67 0.21 HQ M10 Isoflavanone[120]^a 160767 Image 10 109.99 0.30 HQ M11 Calycosin[121]^a 5280448 Image 11 47.75 0.24 HQ M12 Folic acid 135398658 Image 12 68.96 0.71 HQ M13 Isomucronulatol-7,2′-di-O-glucosiole 15689653 Image 13 49.28 0.62 HQ M14 1,7-Dihydroxy-3,9-dimethoxy pterocarpene 5316760 Image 14 39.05 0.48 HQ M15 Astragaloside IV[122]^a 13943297 Image 15 22.50 0.15 HQ M16 Ononin[123]^a 442813 Image 16 11.52 0.78 HQ M17 Calycosin 7-O-glucoside[124]^a 5318267 Image 17 5.49 0.81 HQ M18 Asperglaucide[125]^a 10026486 Image 18 58.02 0.52 ZM M19 Mangiferolic acid 45270099 Image 19 36.16 0.84 ZM M20 Anhydroicaritin[126]^a 5318980 Image 20 45.41 0.44 ZM M21 Anemarsaponin F_qt[127]^a / Image 21 60.06 0.79 ZM M22 Chrysanthemaxanthin[128]^a 21160900 Image 22 38.72 0.58 ZM M23 Hippeastrine[129]^a 441594 Image 23 51.65 0.62 ZM M24 Icariin I[130]^a / Image 24 41.58 0.61 ZM M25 Anemarsaponin E_qt[131]^a / Image 25 30.67 0.86 ZM M26 Diosgenin[132]^a 99474 Image 26 80.88 0.81 ZM M27 Coumaroyltyramine[133]^a / Image 27 112.9 0.20 ZM M28 Timosaponin B Ⅱ[134]^a 44575945 Image 28 13.87 0.04 ZM M29 Linoleyl acetate[135]^a 21159087 Image 29 42.10 0.20 CH M30 3′,4′,5′,3,5,6,7-Heptamethoxyflavone[136]^a 389001 Image 30 31.97 0.59 CH M31 Areapillin 158311 Image 31 48.96 0.41 CH M32 Cubebin[137]^a 117443 Image 32 57.13 0.64 CH M33 Sainfuran[138]^a 185034 Image 33 79.91 0.23 CH M34 Troxerutin 252216528 Image 34 31.6 0.28 CH M35 (+)-Anomalin[139]^a 6450453 Image 35 46.06 0.66 CH M36 Saikosaponin C_qt / Image 36 30.50 0.63 CH M37 Petunidin 441774 Image 37 30.05 0.31 CH M38 Saikosaponin A[140]^a 167928 Image 38 32.39 0.09 CH M39 Saikosaponin D[141]^a 107793 Image 39 34.39 0.09 CH M40 Rutin[142]^a 5280805 Image 40 3.20 0.68 CH M41 Tuberosine A[143]^a 5322166 Image 41 102.67 0.34 SM M42 Cimicifugic acid 100913813 Image 42 83.02 0.45 SM M43 Visamminol 5315249 Image 43 50.01 0.23 SM M44 (20r,24r)-24,25-epoxy-3-beta-(beta-d-xylopyranosyloxy)-9,19-cyclolanost -7-ene-16,23-dione_qt / Image 44 40.10 0.76 SM M45 Paeoniflorin[144]^a 442534 Image 45 53.87 0.79 SM M46 Caffeic acid[145]^a 689043 Image 46 25.76 0.05 SM M47 Isoferulic acid[146]^a 736186 Image 47 58.83 0.06 SM M48 Ferulic acid[147]^a 445858 Image 48 54.79 0.06 SM M49 Acacetin[148]^a 5280442 Image 49 34.97 0.24 JG M50 Cis-Dihydroquercetin 443758 Image 50 66.44 0.27 JG M51 Luteolin[149]^a 5280445 Image 51 36.16 0.25 JG M52 Robinin 5281693 Image 52 39.84 0.71 JG M53 Platycodin D 162859 Image 53 7.60 0.01 JG M54 Isorhamnetin[150]^a 5281654 Image 54 49.6 0.31 HQ, CH M55 Kaempferol[151]^a 5280863 Image 55 41.88 0.24 HQ, ZM, CH M56 Quercetin[152]^a 5280343 Image 56 46.43 0.28 HQ, CH M57 Stigmasterol[153]^a 5280794 Image 57 43.83 0.76 ZM, CH, SM [154]Open in a new tab ^a Compounds identified by UPLC-Q-TOF/MS. 3.3. Screening of potential targets for SXD treatment of CHD and construction of PPI network The GeneCards, OMIM, TTD and DisGeNET databases were used to search for CHD-related targets, and 610 corresponding targets were screened and sorted out. Subsequently, the 96 potential targets of SXD for the treatment of CHD were screened by getting their intersection targets ([155]Fig. 2A). The 96 potential targets in herbs and diseases accounted for 19.12 % and 15.74 %, respectively. Among the 96 potential targets, the number of potential targets of each herb in SXD was shown in [156]Fig. 2B. HQ, ZM, CH, SM and JG accounted for 90.62 %, 44.79 %, 82.29 %, 46.88 % and 27.08 % of the potential targets, respectively, which implied that the herbs contained potential targets with the same effect. HQ, as a monarch herb in SXD, contained more potential targets than other herbs, and had a synergistic effect with the 72 and 21 overlapping targets of CH and SM, respectively. Similarly, ZM and JG, as assistant and guide herbs in the prescription, could also assist the monarch herb to exert a therapeutic effect. This was also in line with the characteristics of TCM prescriptions for the synergistic treatment of diseases. Fig. 2. [157]Fig. 2 [158]Open in a new tab Analysis of potential targets for SXD against CHD. (A) Venn diagram of SXD and CHD related targets; (B) Distribution of potential targets of each herb in SXD; (C) PPI network analysis of potential targets for SXD against CHD (The red hexagon nodes represent the 15 key targets with the average of degree and betweenness in the topology analysis. MCODE was used to reclassify existing potential targets). The 96 potential targets were imported into the STRING database, and the medium confidence (0.4) between the targets was set to obtain a PPI network file, which was imported into Cytoscape 3.8.2 software for visualization. The PPI network involves 96 nodes and 1671 edges, and the average degree and betweenness centrality of all nodes are 34.8125 and 0.0132, respectively ([159]Fig. 2C). In order to further understand the biological role of the PPI network, the constructed PPI network was used as input to conduct network clustering analysis based on MCODE cluster, and a total of 3 network clusters were identified. According to the MCODE score, the largest cluster contained 46 nodes and 885 edges, with a score of 39.333, and the smallest cluster contained 8 nodes and 13 edges, with a score of 3.714. The targets in the PPI network are greater than the average of degree and betweenness centrality and included in the core sub-networks cluster as key targets, which are TNF, IL-6, peroxisome proliferator-activated receptor alpha (PPARG), TP53, RAC-alpha serine/threonine-protein kinase (AKT1), angiotensin-converting enzyme 2 (ACE), IL-1β, nitric oxide synthase, endothelial (NOS3), estrogen receptor (ESR1), epidermal growth factor receptor (EGFR), vascular endothelial growth factor A (VEGFA), prostaglandin G/H synthase 2 (PTGS2) and catenin beta-1 (CTNNB1). The key targets possess a higher degree value and are more likely to play a critical role in the network of SXD acting on CHD. Meanwhile, the S-value of each key target was the maximum value of 4 by CytoHubba plugin analysis excluding ESR1, EGFR and CTNNB1, and then the S-value of each target in the PPI network was analyzed ([160]Table 4). Table 4. Potential targets of SXD against CHD and their weight analysis. No. Target UniProt ID Gene name Degree S-value T1 RAC-alpha serine/threonine-protein kinase [161]P31749 AKT1 73 4 T2 Tumor necrosis factor [162]P01375 TNF 80 4 T3 Interleukin-8 [163]P10145 IL-8 63 4 T4 Interleukin-6 [164]P05231 IL-6 80 4 T5 Interleukin-1 beta [165]P01584 IL-1β 70 4 T6 Vascular endothelial growth factor A [166]P15692 VEGFA 70 4 T7 Cellular tumor antigen p53 [167]P04637 TP53 67 4 T8 Matrix metalloproteinase-9 [168]P14780 MMP9 64 3 T9 C–C motif chemokine 2 [169]P13500 CCL2 62 4 T10 Signal transducer and activator of transcription 3 [170]P40763 STAT3 56 3 T11 Epidermal growth factor receptor [171]P00533 EGFR 62 3 T12 Vascular cell adhesion protein 1 [172]P19320 VCAM1 54 3 T13 Caspase-3 [173]P42574 CASP3 59 4 T14 Toll-like receptor 4 [174]O00206 TLR4 58 4 T15 Interleukin-10 [175]P22301 IL10 60 3 T16 Fibroblast growth factor 2 [176]P09038 FGF2 55 4 T17 Prostaglandin G/H synthase 2 [177]P35354 PTGS2 60 4 T18 Catenin beta-1 [178]P35222 CTNNB1 58 3 T19 Nitric oxide synthase, endothelial [179]P29474 NOS3 56 4 T20 Intercellular adhesion molecule 1 [180]P05362 ICAM1 57 4 T21 Peroxisome proliferator-activated receptor gamma [181]P37231 PPARG 59 4 T22 Hypoxia-inducible factor 1-alpha [182]Q16665 HIF1A 59 4 T23 72 kDa type IV collagenase [183]P08253 MMP2 51 3 T24 Plasminogen activator inhibitor 1 [184]P05121 SERPINE1 52 4 T25 Interleukin-4 [185]P05112 IL-4 49 3 T26 Interferon gamma [186]P01579 IFNG 53 4 T27 Interleukin-2 [187]P60568 IL-2 45 4 T28 E-selectin [188]P16581 SELE 43 3 T29 GTPase HRas [189]P01112 HRAS 45 3 T30 Estrogen receptor [190]P03372 ESR1 51 3 T31 Osteopontin [191]P10451 SPP1 45 4 T32 Angiotensin-converting enzyme 2 [192]Q9BYF1 ACE 54 4 T33 Heme oxygenase 1 [193]P09601 HMOX1 46 4 T34 Caveolin-1 [194]Q03135 CAV1 46 3 T35 Vascular endothelial growth factor receptor 2 [195]P35968 KDR 42 4 T36 Receptor tyrosine-protein kinase erbB-2 [196]P04626 ERBB2 47 3 T37 Renin [197]P00797 REN 43 4 T38 Myeloperoxidase [198]P05164 MPO 43 3 T39 Interleukin-1 alpha [199]P01583 IL1A 37 4 T40 Peroxisome proliferator-activated receptor alpha [200]Q07869 PPARA 39 3 T41 Mast/stem cell growth factor receptor Kit [201]P10721 KIT 40 3 T42 Stromelysin-1 [202]P08254 MMP3 38 4 T43 Signal transducer and activator of transcription 1-alpha/beta [203]P42224 STAT1 41 3 T44 Interstitial collagenase [204]P03956 MMP1 36 4 T45 NAD-dependent protein deacetylase sirtuin-1 [205]Q96EB6 SIRT1 41 4 T46 Tissue factor [206]P13726 F3 33 1 T47 Nitric oxide synthase, inducible [207]P35228 NOS2 31 1 T48 Urokinase-type plasminogen activator [208]P00749 PLAU 32 1 T49 Mitogen-activated protein kinase 1 [209]P28482 MAPK1 40 2 T50 Prothrombin [210]P00734 F2 34 2 T51 Tyrosine-protein phosphatase non-receptor type 11 [211]Q06124 PTPN11 33 0 T52 Amyloid-beta precursor protein [212]P05067 APP 40 3 T53 Type-1 angiotensin II receptor [213]P30556 AGTR1 30 0 T54 CD40 ligand [214]P29965 CD40LG 29 0 T55 Neutrophil cytosol factor 1 [215]P14598 NCF1 27 1 T56 Phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit alpha isoform [216]P42336 PIK3CA 33 0 T57 Heat shock protein beta-1 [217]P04792 HSPB1 24 1 T58 Gap junction alpha-1 protein [218]P17302 GJA1 27 1 T59 Androgen receptor [219]P10275 AR 28 1 T60 Estrogen receptor beta [220]Q92731 ESR2 24 0 T61 Telomerase reverse transcriptase [221]O14746 TERT 24 0 T62 Tissue-type plasminogen activator [222]P00750 PLAT 24 0 T63 Superoxide dismutase [Cu–Zn] [223]P00441 SOD1 24 1 T64 Macrophage migration inhibitory factor [224]P14174 MIF 23 1 T65 Polyunsaturated fatty acid 5-lipoxygenase [225]P09917 ALOX5 24 0 T66 Thrombomodulin [226]P07204 THBD 23 1 T67 Fibroblast growth factor 1 [227]P05230 FGF1 24 0 T68 Vitamin D3 receptor [228]P11473 VDR 20 1 T69 Tyrosine-protein kinase ABL1 [229]P00519 ABL1 24 0 T70 RAF proto-oncogene serine/threonine-protein kinase [230]P04049 RAF1 22 0 T71 Oxidized low-density lipoprotein receptor 1 [231]P78380 OLR1 19 1 T72 Phospholipase A2 [232]P04054 PLA2G1B 17 1 T73 Neprilysin [233]P08473 MME 17 0 T74 Beta-2 adrenergic receptor [234]P07550 ADRB2 18 0 T75 Monocyte differentiation antigen CD14 [235]P08571 CD14 14 0 T76 Serum paraoxonase/arylesterase 1 [236]P27169 PON1 15 1 T77 Serine/threonine-protein kinase B-raf [237]P15056 BRAF 12 1 T78 HLA class II histocompatibility antigen, DRB1 beta chain [238]P01911 HLA-DRB1 10 0 T79 Tyrosine-protein phosphatase non-receptor type 22 [239]Q9Y2R2 PTPN22 9 0 T80 Oxysterols receptor LXR-alpha [240]Q13133 NR1H3 8 0 T81 Coagulation factor VII [241]P08709 F7 10 1 T82 3-hydroxy-3-methylglutaryl-coenzyme A reductase [242]P04035 HMGCR 9 1 T83 Glutathione S-transferase Mu 1 [243]P09488 GSTM1 10 0 T84 Beta-1 adrenergic receptor [244]P08588 ADRB1 8 0 T85 Sex hormone-binding globulin [245]P04278 SHBG 10 0 T86 Cyclin-dependent kinase 8 [246]P49336 CDK8 6 1 T87 Cytochrome P450 2C19 [247]P33261 CYP2C19 10 1 T88 Transthyretin [248]P02766 TTR 8 1 T89 Steroid 17-alpha-hydroxylase/17,20 lyase [249]P05093 CYP17A1 7 1 T90 Lysosomal acid glucosylceramidase [250]P04062 GBA 3 0 T91 Potassium voltage-gated channel subfamily H member 2 [251]Q12809 KCNH2 4 1 T92 Endothelin-converting enzyme 1 [252]P42892 ECE1 3 0 T93 Sodium channel protein type 5 subunit alpha [253]Q14524 SCN5A 3 1 T94 Gamma-aminobutyric acid type B receptor subunit 1 [254]Q9UBS5 GABBR1 2 0 T95 Aldehyde dehydrogenase, mitochondrial [255]P05091 ALDH2 3 0 T96 Alpha-ketoglutarate-dependent dioxygenase FTO [256]Q9C0B1 FTO 1 0 [257]Open in a new tab 3.4. C-T-D network construction and CI analysis A C-T-D network was constructed by the Cytoscape 3.8.2 software to facilitate the visualization and interpretation of the complex relationships according to the corresponding information, including effective compounds, potential targets and related diseases of SXD, which consisted of 159 nodes (96 target nodes, 57 compound nodes, 5 herb nodes and 1 disease node) and 523 edges ([258]Fig. 3A). Most targets were shared by candidate effective compounds of each herb in SXD. The high interconnectedness of the C-T-P network was due to the high interconnection degrees of these candidate effective compounds, including quercetin (M56, degree = 51), luteolin (M51, degree = 23), ferulic acid (M48, degree = 19), isoferulic acid (M47, degree = 19), caffeic acid (M46, degree = 17), kaempferol (M55, degree = 13) and anhydroicaritin (M20, degree = 13). Subsequently, we calculated the parameters of the C-T-D network by using NetworkAnalyzer function and analyze the RSR of parameters by the SPSS AU platform ([259]Table S1). Quercetin, the compound with the largest degree in the C-T-D network, only ranked 12 according to RSR. However, the compound anhydroicaritin, the seventh-ranked degree in the network, ranked the highest RSR value. So, a more systematic and accurate index was needed to screen effective compounds in SXD. Based on the network parameters, the content and pharmacokinetic parameter of the compounds in SXD, CI was proposed to screen the core compounds ([260]Tables S2 and S3). Isoferulic acid (M47), quercetin (M56), calycosin (M11), ferulic acid (M48), kaempferol (M55), calycosin 7-O-glucoside (M17), formononetin (M9), astragaloside IV (M15) and saikosaponin D (M39), were ranked as the top 9 compounds according to the accumulative CI value more than 90 %, which could be the core compounds of SXD in the treatment of CHD ([261]Fig. 3B). In addition, among the top 15 effective compounds in the CI value ranking, the emperor herb HQ contained 7 effective compounds, and CH and SM as the minister herbs contained 4 and 3 effective compounds, respectively. Fig. 3. [262]Fig. 3 [263]Open in a new tab Screening of core compounds for SXD against CHD. (A) C-T-D network for SXD against CHD (The blue diamond represents a compound shared by two or more herbs); (B) Contribution index of each effective compound (top 15) for SXD against CHD. 3.5. Enrichment analyses and construction of T-P network The 96 potential targets were entered into the DAVID database for GO and KEGG enrichment analysis. The GO enrichment analysis results were reflected in three aspects: BP, CC and MF. Under the condition of p < 0.05, 640, 63 and 101 items were obtained from these three aspects, respectively. The top of 10 items were taken to make a visual bar chart using the bioinformatics platform, as shown in [264]Fig. 4A, which indicated that SXD may regulate ERK1 and ERK2 cascade, transcription from RNA polymerase II promoter, MAPK cascade, RNA polymerase II transcription factor activity, protein binding, and perinuclear region of cytoplasm for treating CHD. Moreover, a total of 155 pathways were identified by KEGG analysis. All the pathways involved in the S-value of potential targets were ranked and analyzed. Based on the p < 0.05, the top 20 pathways enriched by KEGG were selected according to the P-weight, as shown in [265]Fig. 4B and C and [266]Table S4. Among them, fluid shear stress and atherosclerosis, lipid and atherosclerosis, PI3K-Akt signaling pathway, MAPK signaling pathway, HIF-1 signaling pathway, and EGFR tyrosine kinase inhibitor resistance were probably associated with SXD against CHD. Fig. 4. [267]Fig. 4 [268]Open in a new tab Potential targets enrichment analysis and T-P network analysis of SXD treatment for CHD. (A, B) GO and KEGG enrichment analysis for SXD against CHD; (C) T-P network of SXD against CHD where the size of pathway node is based on the P-weight, and the color of the target node represents the degree value. 3.6. Core targets expression analysis In [269]GSE66360, we analyzed the expression of key targets in MI samples (n = 49) and control samples (n = 50). As shown in [270]Fig. 5, the expression levels of TNF, IL-6, IL-1β, PTGS2 and VEGFA were up-regulated than those of the control group (p < 0.05), while the expression level of TP53 was down-regulated (p < 0.05). There was no significant difference in the expression levels of other key targets (p > 0.05). Therefore, we considered the targets with significant expression differences as core targets. Subsequently, the core targets were entered into the VaeElect database to obtain the correlation score between the core targets and the disease ([271]Table S5). Obviously, we found that the average disease's causing likelihood of IL-1β was as high as 81.8 %, but that of IL-6 was relatively low, which was only 56.1 %. Therefore, these core targets were used as the main targets of SXD against CHD and included in subsequent analysis. Fig. 5. [272]Fig. 5 [273]Open in a new tab Analysis of core targets expression based on GEO dataset (The values of p < 0.05 was considered to be significant expression differences). 3.7. Analysis of molecular docking results Molecular docking is the docking of ligands and receptors in the active pocket through one or more hydrogen bonds, which is a process accompanied by a change in binding energy. In order to verify the reliability of the network pharmacology research results, these core targets, TNF (PDB ID: [274]7KP9), IL-6 (PDB ID: [275]4O9H), IL-1β (PDB ID: [276]5I1B), VEGFA (PDB ID: [277]5DN2), TP53 (PDB ID: [278]5AOI) and PTGS2 (PDB ID: [279]5F19) were selected as receptors, and they were docked with the core compounds in [280]Fig. 6A. In general, when the ligand-receptor binding energy is lower than 0, the binding energy is about the greater the possibility of interaction. When the binding energy is lower than −5.0 kcal/mol, it contains good affinity. For these effective compounds, quercetin, calycosin, kaempferol, calycosin 7-O-glucoside, formononetin, astragaloside IV and saikosaponin D could produce lower binding activity with most targets, especially with TNF and PTGS2. Among them, calycosin 7-O-glucoside have the lowest docking binding energy to PTGS2 (−11.3 kcal/mol), and isoferulic acid has the highest binding energy to IL-6 (−5.2 kcal/mol). Subsequently, six results with relatively low binding energy of core targets were selected for visualization. As shown in [281]Fig. 6B, calycosin7-O-glucoside could interact with Lys11 (B), Leu157 (A) and Leu (B) via one hydrogen bonds in TNF, respectively. Astragaloside IV was more capable of interacting with Gly42 (H) via two hydrogen bonds and with Ley175 (H), Ala173 (H), Lys169 (L), Gly43 (L) and Gln39 (H) via one hydrogen bond in IL-6. Calycosin 7-O-glucoside could interact with Tyr90 (A), Pro87 (A), Glu64 (A), Ser5 (A), Lys65 (A) and Tyr68 (A) via one or two hydrogen bonds in IL-1β. The results of the interaction of VEGFA, TP53 and PTGS2 with the core compounds were also seen in [282]Fig. 6B. Therefore, these results suggested a strong binding between core compound-target pairs and demonstrated that the target proteins were in a favorable conformation. Fig. 6. [283]Fig. 6 [284]Open in a new tab Molecular docking of the core compound-target pair of SXD treatment for CHD. (A) Heat map of estimated binding energy where colour change from blue to red indicates binding energy from high to low. (B) Molecular docking of core compounds of SXD and receptor proteins. 3.8. Experimental verification 3.8.1. SXD ameliorates the myocardial injury and inflammatory response in CHD model rats As shown in [285]Fig. 7A, HWI in the CHD model group was significantly higher than that in the sham-operated group. Compared with the CHD model group, the HWI levels of Bet and SXD groups were reduced to different degrees. TTC staining was used to evaluate myocardial infarct size. Compared with sham-operated group, myocardial infarction size was significantly increased in CHD model group, while myocardial infarction size decreased after SXD treatment ([286]Fig. 7B). In the CHD model group, H&E and Masson staining showed severe necrosis of myocardial fibers, inflammatory cell infiltration, and tissue structure disorder. The degree of tissue fibrosis and cell damage were significantly improved in Bet and SXD groups ([287]Fig. 7C). In addition, serum IL-1β and IL-6 levels were increased in the CHD model group compared to the Sham group (p < 0.01). Compared with the model group, the levels of serum IL-1β and IL-6 in each dosing group were decreased to varying degrees ([288]Fig. 7D). In addition, SXD could also change the levels of serum ALD, Ang Ⅱ and NT-proBNP in CHD rats. The results showed that SXD significantly reduces HWI, alleviates myocardial tissue injury and reduces the level of serum inflammation in CHD rats. Fig. 7. [289]Fig. 7 [290]Open in a new tab SXD significantly alleviates myocardial injury and inflammatory response in CHD rats. Sham: Sham operation group; Mod: CHD model group; Bet: Betaloc group (10 mg/kg); SXDL: SXD low-dose group (4.33 g/kg); SXDH: SXD high-dose group (12.99 g/kg). (A) Heart weight index (n = 6); (B) Infarct size of heart tissue (n = 3); (C) H&E and Masson staining of the heart tissue; (D) ELISA was performed to examine the contents of IL-6, IL-1β, ALD, Ang Ⅱ and NT-proBNP in the rat serum (n = 6). *p < 0.05, **p < 0.01, vs. Control; ^#p < 0.05, ^##p < 0.01, vs. H/R. 3.8.2. SXD effectively enhances cell viability and decreases the production of ROS To identify the most effective concentration of SXD for myocardial protection, the MTT method was used. Compared with the control group, the H/R-induced cardiomyocytes showed lower cell viability ([291]Fig. 8A). In the H/R + Diazoxide groups comparing with the H/R group, Diazoxide increased the viability. Further, a 24 h pretreatment with SXD at various concentrations (250 μg/mL, 500 μg/mL and 1 mg/mL) significantly increased cell viability, compared with the H/R group (p < 0.05). Thus, the concentrations of 250 μg/mL, 500 μg/mL and 1 mg/mL of SXD were selected to explore the possible mechanism. To confirm that SXD could protect H9c2 cells through anti-oxidative activity, we analyzed the ROS production ([292]Fig. 8B and [293]S2). As shown in [294]Fig. 8B, the ROS level was higher in the H/R group than control, while SXD markedly decreased the production of ROS in H/R stimulated H9c2 cells. Fig. 8. [295]Fig. 8 [296]Open in a new tab SXD increases cell viability and attenuates H/R-induced H9c2 cell injury. (A)The viability of cardiomyocyte cell line H9c2 after treated with H/R or SXD was determined by MTT assays (n = 5); (B) ROS levels were detected by flow cytometry (n = 3); (C) Effects of SXD on the apoptosis ratio of H9c2 cells (n = 3); (D) ELISA was performed to examine the contents of TNF, IL-6 and IL-1β in H9c2 cells culture supernatant (n = 3). *p < 0.05, **p < 0.01, vs. Control; ^#p < 0.05, ^##p < 0.01, vs. H/R. 3.8.3. SXD protects H9c2 cells from H/R-induced myocardial damage Meanwhile, to investigate the anti-apoptotic effect of SXD, flow cytometry was used. As displayed in [297]Fig. 8C, compared with the control group, the number of apoptotic cells increased in the H/R group, whereas pretreatment with SXD reduced the apoptotic ratio (p < 0.01). Collectively, these data show that SXD can reduce the cellular apoptosis induced by H/R. Moreover, three different proinflammatory cytokines (TNF, IL-6 and IL-1β) were detected, and as shown in [298]Fig. 8D, ELISA proved that the contents of TNF, IL-6 and IL-1β in H9c2 cells culture supernatant were reduced in SXD-treated H9c2 cells, as compared to H/R-induced H9c2 cells (p < 0.01). The results showed that H/R-stimulated production of TNF, IL-6 and IL-1β in H9c2 cells were abolished by SXD. 3.8.4. The expression of mRNA and protein during the H/R-induced inhibition of apoptosis It was found that ROS production and DNA damage activated the TP53 signaling pathway, and TP53 further up-regulated cytochrome C-mediated caspase-3 activation, leading to apoptosis [[299]29]. TP53, caspase3 and cytochrome C mRNA expression levels were examined in order to further clarify the molecular mechanism of the H/R-induced inhibition of apoptosis. RT-qPCR results in [300]Fig. 9A showed that TP53 expression was significantly up-regulated in the H/R group (p < 0.01), and this expression was dramatically reversed by SXD treatment at a concentration of 500 μg/mL and 1 mg/mL for 24 h (p < 0.01). However, no obvious difference in TP53 expression was observed between the groups treated with SXD at 250 μg/mL, diazoxide and the H/R group, separately (p > 0.05). RT-qPCR results showed that caspase3 expression was significantly up-regulated in the H/R group (p < 0.01) and this expression was dramatically reversed by treatment with SXD at 250 μg/mL, 500 μg/mL and 1 mg/mL for 24 h (p < 0.01). RT-qPCR results showed that cytochrome C expression was significantly up-regulated in the H/R group (p < 0.01) and this expression was dramatically reversed by treatment with SXD at 250 μg/mL, 500 μg/mL and 1 mg/mL for 24 h (p < 0.01). However, no significant difference in caspase3 and cytochrome C expression was observed between the groups treated with diazoxide and the H/R group. Fig. 9. [301]Fig. 9 [302]Open in a new tab The expression of mRNA and protein during the H/R-mediated inhibition of apoptosis. (A) The mRNA expressions of TP53, caspase3 and cytochrome C in H/R-induced cardiomyocyte cell line H9c2 after treated with SXD were determined (n = 3); (B) Western blotting analysis detected the caspase3, caspase9 and cytochrome C protein expression levels in H/R-induced H9c2 cells after treated with SXD (n = 3); (C) Activity Assay Kit was performed to examine the activity of caspase3 in H/R-induced cardiomyocyte cell line H9c2 after treated with SXD (n = 3). *p < 0.05, **p < 0.01, vs. Control; ^#p < 0.05, ^##p < 0.01, vs. H/R. To further elucidate the mechanism by which SXD protect the H/R-induced inhibition of apoptosis, we analyzed the expression of caspase3, caspase9 and cytochrome C that regulate cellular apoptosis using western blotting analysis. The results revealed that the expression of caspase3, caspase9 and cytochrome C were significantly decreased by SXD treatment for 24 h compared with the H/R-induced H9c2 cells model group, especially when the concentration was 1 mg/mL in [303]Fig. 9B (The uncropped versions of [304]Fig. 9B was provide as [305]Supplement Fig. S3). Additionally, the activity of caspase3 in H/R-induced H9c2 cells culture supernatant were reduced in SXD-treated H9c2 cells, as compared to H/R-induced H9c2 cells in [306]Fig. 9C (p < 0.05). In summary, these data indicate that SXD significantly relieve myocardial injury from the perspective of anti-apoptosis via down-regulating the caspase3, caspase9 and cytochrome C protein expression levels. 4. Discussion CHD is a complex cardiovascular disease that seriously threatens human health. The pathological mechanism of CHD is in line with the pathogenic characteristics of the “qi” stagnation. The Chinese medicine understanding of the “qi” stagnation will cause the weak and slow blood flow and abnormal hemorheology. Therefore, treating this disease from the “qi” stagnation and blood stasis has important practical significance for the diagnosis and treatment of modern CHD. Treating this disease from the “qi” stagnation and blood stasis has important practical significance for the diagnosis and treatment of modern CHD. SXD is a classic TCM prescription and possesses the effect of nourishing the “qi” and blood for the body. SXD, which is composed of Astragalus membranaceus var. mongholicus (Bunge) Hsiao (HQ), Anemarrhena asphodeloides Bge. (ZM), Bupleurum chinense DC. (CH), Platycodon grandiflorum (Jacq.) A. DC. (JG) and Cimicifuga foetida L. (SM), has been successfully applied in clinical treatment of CHD [[307]30]. The combination of each herb in SXD has a synergistic effect on the treatment. Research has indicated that JG can enhance the distribution or absorption of SXD, which reflects the courier role of PG [[308]10]. SXD has not been found to be used in other countries. But it should be noted that those plants used for the preparation of SXD was not unique for TCM. They also have been used in other countries, including Japan, Korea, Russia and other regions [[309][31], [310][32], [311][33], [312][34]]. For example, HQ is included in the Japanese Pharmacopoeia, and the team from Japan investigated the different protective effects of astragalus [[313]31,[314]32]. In addition, CH, which is widespread in the eastern regions of Russia as well as in Mongolia and Korea, have been widely applied as a choleretic and hepatoprotective remedy in traditional Russian medicine [[315]33]. The standardized DA-9805 consists of CH from Dong-A ST (Yongin, Korea) can protect dopaminergic neurons against 6-hydroxydopamine in a Parkinson's model of neurotoxic disease [[316]34]. At present, SXD is widely used in the treatment of CHD, but its therapeutic mechanism is not clear. The purpose of this study is to provide scientific basis for the treatment of SXD with CHD. In this study, through the combination of UPLC-Q-TOF-MS/MS technology and an integrated network pharmacology, we directly identified the potential active ingredients and targets of SXD in the treatment of CHD from a large number of data, and understood the molecular mechanism and important pathways of TCM. 57 effective compounds and 96 corresponding potential targets were selected, which were largely involved in multiple biological processes and pathways associated with CHD. To better understand the contribution of compounds in SXD, we introduced a parameter CI that simulated compatible combinations of effective compounds in SXD in terms of intrinsic properties and network importance. From the C-T-D network and CI calculation, it was revealed that the core compounds, including isoferulic acid, quercetin, calycosin, ferulic acid, kaempferol, calycosin 7-O-glycoside, formononetin, astragaloside IV and saikosaponin D, were ranked as the top 9 compounds according to the sum of their CI value more than 90 %. The core compounds were the main compounds of SXD in the treatment of CHD. Pharmacological studies have shown that the emperor herb HQ in SXD and its compounds have various biological activities, such as regulating immunity, anti-oxidation, anti-inflammatory, anti-apoptosis and preventing cardiovascular diseases [[317][35], [318][36], [319][37], [320][38]]. Through compounds collecting and targets screening, it turned out that HQ contained the vast majority of 20 compounds and more targets than other herbs. The analysis of the top 15 compounds with CI calculation found that HQ involved 7 effective compounds. Therefore, we could conclude that the effective compounds of SXD in the treatment of CHD were mainly contained in HQ. HQ contains astragaloside IV with significant anti-inflammatory and cardiovascular protective effects [[321]39,[322]40]. For the quercetin with a high CI, it has good antithrombotic and cardiac function improvement effects, which is of great significance for the treatment of cardiovascular diseases [[323]41]. And quercetin can reduce the levels of TNF and IL-1β to show anti-inflammatory properties in the serum of patients with coronary artery disease [[324]42]. Meanwhile, kaempferol can upregulate miR-26a-5p by inhibiting the TLR4/NF-κB signaling pathway, reduce the apoptosis of endothelial cells induced by oxidized low-density lipoprotein, and play an important role in anti-atherosclerosis [[325]43]. Formononetin, a methoxyisoflavone abundant in many herbs, has been found to alleviate atherosclerosis by modulating the interaction between KLF4 and SRA in Apo E-deficient mice [[326]44]. Research has shown that calycosin exhibits anti-apoptotic effects by activating ERα/β and enhancing Akt phosphorylation in cardiomyocytes [[327]45]. SM is used as the minister in SXD, and the compounds including isoferulic acid, ferulic acid and caffeic acid were the main effective compounds that can play an antioxidant role [[328]46]. In addition to this, some other compounds have good activity in the prevention and treatment of CHD. A study by Zhang et al. luteolin has a strong cardioprotective effect, and its mechanism may be related to the down-regulation of TLR4-mediated NF-κB/NLRP3 inflammasome in vivo and in vitro [[329]47]. Therefore, it is speculated that SXD can treat CHD through the synergistic effect of multiple compounds. PPI network and the published microarray data result showed that TNF, IL-6, IL-1β, VEGFA, TP53 and PTGS2, were the core target proteins. Through cluster analysis of protein interaction network MCODE, it was found that the core target proteins were mainly concentrated in cluster 1. Cluster 1 mainly involved signaling pathways such as fluid shear stress and atherosclerosis, lipid and atherosclerosis, HIF-1 signaling pathway, IL-17 signaling pathway, TNF signaling pathway, PI3K-Akt signaling pathway, MAPK signaling pathway, etc. Cluster 2 mainly involved AGE-RAGE signaling pathway in diabetic complications, phospholipase D signaling pathway, etc. And cluster 3 mainly involved Ras signaling pathway, VEGF signaling pathway, etc. Meanwhile, these core target proteins are involved in various signaling pathways related to inflammation, apoptosis, immunity, metabolism, cell apoptosis and proliferation in the treatment of CHD. For the target with the highest degree value, TNF is an important inflammatory mediator in post-ischemic myocardial dysfunction, and the level of TNF in plasma can reflect the degree of cardiac damage in patients with CHD [[330]48]. Similarly, IL-6 is a key cytokine, which can mediate a variety of inflammatory responses and immune regulation pathways, and it plays an indispensable role in the occurrence and development of cardiovascular diseases [[331]49]. Besides, IL-6 is an inflammatory marker of coronary microvascular disease in women [[332]50]. Compared with normal subjects, IL-1β is significantly increased in patients with atherosclerosis, which plays an important role in atherosclerosis [[333]51]. VEGFA is a major factor for promoting angiogenesis, and increased spontaneous production of VEGFA may induce angiogenesis after acute myocardial infarction, which is due to initiating ROS-ER stress-autophagy axis in the vascular endothelial cells [[334]52,[335]53]. TP53 is an important tumor suppressor gene and plays an important role in apoptosis, genome stability, inhibition of angiogenesis, etc [[336]54]. The above studies showed that SXD had multi-target effects on the regulation of CHD, and these targets involved cell proliferation and apoptosis, signal transduction, inflammatory response, oxidative stress, gene expression, angiogenesis, etc. These targets were closely related to the occurrence and development of CHD, which confirmed that SXD has the effect of treating CHD. In the process of GO enrichment analysis, BP mainly included cytokine-mediated signaling pathway, positive regulation of gene expression, negative regulation of gene expression; CC mainly involved extracellular space, and extracellular region, cell surface; MF mainly involved enzyme binding, identity protein binding, and RNA polymerase II transcription factor activity, ligand-activated sequence-specific DNA binding. From the analysis of KEGG enriched pathways, it could be seen that AGE-RAGE signaling pathway in diabetic complications, pathways in cancer, fluid shear stress and atherosclerosis, lipid and atherosclerosis, PI3K-Akt signaling pathway, MAPK signaling pathway, HIF-1 signaling pathway, EGFR tyrosine kinase inhibitor resistance. The PI3K-Akt signaling pathway is an important signaling pathway in organisms. Phosphorylated Akt can activate endothelial nitric oxide synthase and heat shock proteins to protect the myocardium [[337]55]. Studies have found that ischemic preconditioning can activate the PI3K-Akt signaling pathway in cardiomyocytes, thereby reducing apoptosis, eliminating intracellular reactive oxygen species, and protecting mitochondrial function [[338]56]. MAPK signaling pathway is an important intracellular transduction pathway. Studies have shown that inhibiting MAPK signaling pathway can reduce inflammation, improve intimal thickening and inhibit plaque formation [[339]57]. HIF-1 is a heterodimeric transcription factor composed of two subunits, HIF-1α and HIF-1β. HIF-1α is highly expressed in atherosclerotic sites, and the HIF-1 signaling pathway obstruction can exert myocardial protective effect, reduce ischemia-reperfusion injury, improve cardiac function and reduce infarct size [[340]58,[341]59]. In addition, by regulating AGE-RAGE pathway, it can affect oxidative stress and inflammatory response in IR-induced myocardial injury in diabetic rats [[342]60]. This is a complex process involving multiple compounds, multiple targets and multiple pathways, which reflects the holistic and systematic nature of SXD treatment of the CHD. The results of molecular docking further showed that the core targets can be stably combined with the core effective compounds in SXD, which further verified the reliability of the prediction of the network pharmacology method. For the core targets TNF, IL-6 and other core targets mentioned in this paper, we found that the gene expression of the core genes was significantly difference in circulating endothelial cells of patients with myocardial infarction [[343]61]. Therefore, the network pharmacological results indicate that SXD may improve myocardial ischemia and reperfusion injury through inflammation and anti-apoptosis. The results of in vivo experiments in rats showed that SXD can significantly reduce HWI, myocardial tissue damage, serum inflammation, ALD, Ang Ⅱ and NT-proBNP in rats with CHD. Therefore, the results indicated that SXD can effectively treat CHD. Based on what mentioned above, the rat cardiomyocytes H9c2 were oxygenated for 12 h after 8 h of hypoxia to establish the cell injury model of H/R, and screen whether SXD has the effect of treating myocardial damage and other injuries. The results showed that H/R could significantly inhibit the cell viability of H9c2, and different concentrations of SXD and positive drugs could improve the cell damage caused by H/R and increase the cell viability. In addition, we further explored the possible mechanism of SXD therapy. H/R can induce apoptosis of H9c2 cells, and this drug can inhibit apoptosis to a certain extent, which may be because SXD can improve cell vitality to a certain extent. Similarly, it has been reported that H/R can lead to inflammatory responses resulting from myocardial cell damage [[344]62]. On the basis of this study, we found that SXD can inhibit the secretion of inflammatory factors in H9c2 cells and reduce the expression of inflammatory factors, and then SXD has a certain anti-inflammatory effect, so as to alleviate the secondary damage caused by the accumulation of inflammatory factors in cardiomyocytes. As shown in the review, through the predicted results of network pharmacology and experimental verification, we revealed that SXD can relieve myocardial ischemia reperfusion injury or myocardial injury caused by myocardial infarction from the perspective of anti-inflammation and anti-apoptosis. The compounds of SXD are expected to become potential drugs for the treatment of CHD, and may become the focus of future research. 5. Conclusions In conclusion, we performed the UPLC-Q-TOF-MS/MS technology, an integrated network pharmacology and experimental verification to reveal the pharmacological mechanism of SXD against CHD. Among all effective compounds, isoferulic acid, quercetin, calycosin, ferulic acid, kaempferol, calycosin 7-O-glycoside, formononetin, astragaloside IV and saikosaponin D in SXD have high CI and good affinity with potential targets, which may play an important role in the treatment of CHD. Moreover, a further in vivo validation experiment that SXD alleviates myocardial tissue injury and reduce the level of serum IL-6, IL-1β, ALD, Ang Ⅱ and NT-proBNP in vitro validation experiment illustrated that SXD may protect myocardial cell via reducing the expression of the inflammatory factors and pro-apoptotic protein, thus playing a protective role in myocardial injury caused by CHD. Although the verification of the mechanism is not completely in depth, our results provide a partial research basis for the application of SXD and a new idea for the clinical prevention and treatment of myocardial injury caused by CHD. Ethics approval and consent to participate All procedures were strictly approved by the Animal Experiment Ethics Committee of the Shaanxi University of Chinese Medicine (Ethics Approval No. SUCMDL20220310006). Funding This research was funded by the National Natural Science Foundation of China (82004011 and 81903786), the Natural Science Foundation of Shaanxi Province (2022SF-221), and Subject Innovation Team of Shaanxi University of Chinese Medicine (2019-YL10). Data availability statement Data will be made available on request. CRediT authorship contribution statement Hao-ming Zhou: Writing – original draft, Software, Data curation, Conceptualization. Shi-jun Yue: Writing – review & editing, Funding acquisition, Conceptualization. Wen-xiao Wang: Methodology, Funding acquisition. Qiao Zhang: Investigation, Formal analysis. Ding-qiao Xu: Resources, Formal analysis. Jia-jia Li: Investigation. Yu-ping Tang: Funding acquisition, Conceptualization. Xin-yu Yang: Writing – review & editing, Funding acquisition. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Footnotes ^Appendix A Supplementary data to this article can be found online at [345]https://doi.org/10.1016/j.heliyon.2024.e29558. Contributor Information Shi-jun Yue, Email: shijun_yue@163.com. Yu-ping Tang, Email: yupingtang@sntcm.edu.cn. Xin-yu Yang, Email: yangxinyu@bjsjth.cn. Abbreviation ALD Aldosterone Ang Ⅱ Angiotensin Ⅱ BP biological process CC cellular component CH roots of Bupleurum chinense DC CHD coronary heart disease CI contribution index C-T-D: compound-target-disease DL: drug-likeness ECL: chemiluminescence GO Gene ontology HQ roots of Astragalus membranaceus var. mongholicus (Bunge) Hsiao H/R hypoxia/reoxygenation HWI heart weight index JG roots of Platycodon grandiflorum (Jacq.) A. DC KEGG Kyoto Encyclopedia of Genes and Genomes LAD left anterior descending coronary artery MF molecular function MI myocardial infarction MTT 3-(4,5)-dimethylthiahiazo (-z-y1)-3,5-di- phenytetrazoliumromide assay NT-proBNP N-terminal pro-brain natriuretic peptide; OB oral bioavailability PPI protein-protein interaction PVDF polyvinylidene difluoride; RIPA radio immunoprecipitation assay ROS reactive oxygen species RSR rank-sum ratio RT-qPCR real time quantitative PCR SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis SM rhizomes of Cimicifuga foetida L SXD Shengxian Decoction TCM traditional Chinese medicine; T-P: target-pathway UPLC-Q-TOF-MS/MS ultra-performance liquid chromatography-quadrupole-time of flight-mass spectrometry TTC 2,3,5-triphenyl-2H-tetrazolium chloride; ZM rhizomes of Anemarrhena asphodeloides Bge Appendix A. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 [346]mmc1.docx^ (3.4MB, docx) References