Experimental evidence and network pharmacology-based analysis reveal the molecular mechanism of Tongxinluo capsule administered in coronary heart diseases

Abstract Background: Tongxinluo (TXL) capsule, a polypharmacy derived from traditional Chinese medicine (TCM), has been widely used in coronary heart disease (CHD), while the underlying mechanism of TXL capsule is still unclear. The present study aimed at investigating the underlying mechanism of TXL acting on CHD patients and providing substantial evidence in molecular evidence by means of a network pharmacological analysis. Method: Active compounds and targeted genes of TXL were retrieved from TCM systems pharmacology (TCMSP) and TCM integrative database (TCMID). CHD and coronary artery disease were treated as search queries in GeneCards and Online Mendelian Inheritance in Man (OMIM) databases to obtain disease-related genes. Visualization of disease–targets network was performed under administration of Cytoscape software. Besides, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were administered. H9c2 cells were used to validate the predicted results in cardiomyocytes/reoxygenation model, and anti-inflammatory ability was examined. Results: A network of a total of 212 nodes and 1016 edges was obtained. Peptide and ubiquitin-like protein ligase binding occupied a leading position of GO enrichment. For KEGG analysis, fluid shear stress and atherosclerosis, as well as inflammation-related pathways were enriched. Cellular validation revealed the anti-inflammatory effect of β-sitosterol, eriodictyol, odoricarpin, and tirucallol as active compounds of TXL. Conclusion: Our study provided substantial molecular evidence that TXL capsule possessed the characteristics of multitargets with safe profile, and the main component is capable of regulating cytokine level in CHD patients.


Introduction
Coronary heart disease (CHD), one of the most common cardiovascular diseases is caused by reduction in blood flow to cardiomyocyte owing to build-up of plaque in arteries of heart [1,2]. CHD has become a leading cause of death and the mortality increased from 5.2 million to over 7 million between 1990 and 2010 [3]. It affects individuals at any age while becomes approximately triple in progressively elder populations compared with other age groups, and the morbidity in males is larger than that in female population [4]. Statin, as the cornerstone in anti-atherosclerotic regimen, has demonstrated the substantial efficacy at reducing cardiovascular events. However, even with intensive statin therapy, many patients still suffered from high residual risks in cardiovascular events [5]. Thus, exploration of alternative anti-atherosclerotic medications with high efficacy as well as low side-effect is needed.

Targets of active compounds
We comprehensively searched the direct targeted receptors of each active compound via DrugBank database, a specific bioinformatics and cheminformatics resource with detailed drug data, as well as targeted receptors (https://www. drugbank.ca). Full names of targeted protein receptors were obtained and converted into gene symbol on the basis of UniProt ID (https://www.uniprot.org/) for following analysis.

Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analysis
Overlapped genes were retrieved for GO and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis with the criterion of P-value <0.05. Bar plots of GO and KEGG were exported and signal pathways involved in this network analysis were visualized in forms of diagram.

Cells
H9c2 cells were purchased from Tongpai Technology Company (Shanghai, China) and cultured in Dulbecco's modified Eagle's medium (DMEM) bought from Thermo Fisher Scientific (Guangzhou, China), with the supplement of 10% v/v FBS and 1% v/v penicillin/streptomycin in CO 2 incubator at 37 • C and 95% relative humidity.

Cell models
Regarding the investigation of the protective effect of TXL, hypoxia/reoxygenation (H/R) model was administered. Cells were put into an incubator with Krebs-Ringer bicarbonate buffer medium saturated with 99.99% N 2 for 140 min [15]. Cells were reoxygenated through changing the DMEM back and cultured under normal oxygen level (21%) for 1 h. The molecules were applied for 48 h before hypoxia until the end of oxygenation.

Cell viability test
Cell viability test was performed under the assistance of cell counting kit-8 (CCK-8) after the administration of abovementioned active components of TXL. Cells with different molecules were seeded in a 96-well plate at a density of 1 × 10 4 cells for 24 h. Then, 10% CCK-8 was added and OD value was read at 450 nm after 1 h. In addition, optimal concentration of each molecule was explored ranging from 5 to 100 μM [16][17][18][19]. Each cell viability test with different molecules was repeated five times and measurement of relative cell viability was recorded.

Investigation of anti-inflammatory effect
For anti-inflammatory effect, cells were seeded in a 96-well plate incubated for 24 h, and treated with 0.01 μg/ml LPS 30 min after incubation with optimal concentration of abovementioned molecules was obtained. Then, supernatant was collected by adding 150 μl dimethyl sulfoxide (DMSO) and stored at −80 • C for downstream analysis. Concentration of cytokine was measured by enzyme-linked immunosorbent assay (ELISA) under corresponding protocol and IL-6 (K4144-100, Biovision) and IL-8 (K4169-100, Biovision) ELISA kits were administered in the presentstudy. Each test with different molecules was repeated five times and average concentration of corresponding results was recorded.

Identification of putative ingredient targets
With the mentioned search queries of Panax Ginseng C. A. Mey., Radix Paeoniae Rubra, Ziziphi Spinosae Semen, Dalbergiae Odoriferae lignum, Santalum Album L., Olibanun, Cicadae Periostracum, Borneolum Syntheticum, hirudo, Scorpio, Scolopendra, Cicadae periostracum and criteria of OB ≥ 30% as well as DL ≥ 0.18, a total of 111 chemical ingredients were collected within TXL prescription from TCMSP and TCMID databases. Besides, the targeted genes of each retrieved chemical ingredients were explored and a total of 1205 targeted genes were obtained. The names of targeted genes were converted into gene ID on basis of UniProt database, and eventually 861 eligible targeted genes with molecular names and symbol ID were acquired. The active compounds involved in the present study with the amount as well as ratio of each component [20] were shown in Table 1 and detailed information of putative ingredients with targeted genes were documented in Supplementary Table S1.

Identification of disease-related genes
Since the application of TXL is to lower serum lipid level, anti-oxidation and anti-inflammation, which are standard management in CHD [21], the CHD and coronary artery disease were treated as keywords to acquire relevant genes. After the administration of search queries in GeneCards and OMIM databases, a total of 7389 CHD-relevant genes were obtained. Furthermore, intersection between ingredients-targeted and CHD-relevant genes were performed and 138 overlapped genes were obtained eventually. The Venn diagram of overlapped genes were displayed in Supplementary Figure S1.

Network visualization
A complete ingredient-target network consisting of a total of 212 nodes and 1016 edges (138 target nodes, 72 putative ingredients nodes, 1 disease node and 1 TXL node) was obtained after administration of Cytoscape software as shown in Figure 2. For detailed information, each node included in this ingredient-target network was documented in Supplementary Table S2. Overlapped genes were processed by STRING to produce a PPI network with confidence > 0.4 and shown in Figure  3A. PPIs were displayed by a total of 138 nodes and 1939 edges with average node degree of 28.1. Within the PPI net-   Figure 3B and the detailed information of PPI is documented in Suplementary Table S3.

GO and KEGG enrichment analyses
Overlapped genes' names were converted into symbol ID via UniProt database for GO and KEGG enrichment analyses. Regarding GO enrichment analysis, function of peptide binding and ubiquitin-like protein ligase binding occupied the leading position among all relevant genes with adjusted P-value of 6.35e −7 and 1.00e −6 , respectively. Heme binding and tetrapyrrole binding function were at second place of overlapped genes enrichment analysis with adjusted P-value of 3.49e −8 and 6.83e −8 , respectively. Top 20 categories of GO enrichment analysis are shown in Figure  4A,B.
When it comes to KEGG enrichment analysis, AGE-RAGE signaling pathway, and fluid shear stress and atherosclerosis pathway occupied the predominant position with adjusted P-value of 5.60e −19 and 3.88e −17 , respectively. Moreover, inflammation-related pathways, such as IL-17, TNF and T-cell receptors signaling pathways, were principal pathways within the TXL-CHD overlapped genes enrichment, with the adjusted P-value of 3.19e −17 , 1.13e −14 ,

Active ingredients protect H9c2 cells from H/R injury
Six potential ingredients, β-sitosterol, ellagic acid, formononetin, eriodictyol, odoricarpin, tirucallol (detailed information shown in Table 2), were obtained and used for validation. Regarding to cell viability tests, β-sitosterol, eriodictyol, odoricarpin and tirucallol revealed positive improvement effect, while ellagic acid and formononetin were found to be cytotoxic to H9c2 cells in H/R model ( Figure 5A). Improvement rate at different concentrations was investigated to obtain optimal dosage. From the results, the optimal dosage of β-sitosterol, eriodictyol, odoricarpin, tirucallol were 40, 20, 20 and 40 μM in this model, respectively, and decreased relative cell viability was observed in each test when concentration exceeded 50 μM ( Figure 5B).

Anti-inflammatory effect of TXL
Due to the significance of anti-inflammatory regulation in CHD management, the anti-inflammatory effect of TXL was investigated. Since the enriched pathways in anti-inflammatory regulation (Supplementary Figure S2), concentrations of IL-6 ( Figure 6A) and IL-8 ( Figure 6B) were investigated with the abovementioned optimal concentration of four compounds. β-sitosterol, eridictyol, odoricarpin and tirucallol indicated significant inhibition on concentration of IL-6 as well as IL-8 (P<0.05). Moreover, tirucallol revealed to have a significant anti-inflammation effect compared with DXM group (P<0.05). Collectively, active compounds of TXL is capable of regulating anti-inflammation.

Discussion
In previous study, resistance to statin regimen led to rapid progression of atheroma, indicating warranted alternative to lipid-lowering medication [22]. As indicators of plaque progression, IMT and maximal plaque area are favored indicators for CHD assessment. In CAPITAL trial, as the additional anti-atherosclerotic regimen to routine CHD therapy, TXL revealed superiority compared with control group in slowing down the progression of CHD significantly [10]. However, the underlying anti-atherosclerotic effects of TXL were unclear. After this research, substantial evidences might be provided at the molecular level. Network pharmacology was designed for investigating single-medication targeting on multiple targets so as to enhance efficacy as well as reducing toxicity to patients [23]. Besides, TXL capsule was a mixture of 12 plant and animal products with multiple ingredients and targets, which conformed to the abovementioned perspective and was proved to be effective in cellular level in the present study.
Regarding enrichment analysis, several pathways revealed the potential mechanism of TXL capsule acting on anti-atherosclerotic events. Peptide and ubiquitin-like protein ligase binding occupied the predominant position among GO enrichment analysis, in which rising ubiquitin was reported as positively correlated indicators with the severity of pathologies such as trauma, burn, and especially in CHD and acute myocardial infarction (AMI) patients [24][25][26]. Also, extracellular ubiquitin was shown to be elevated in CHD patients, especially in patients with acute coronary syndrome (ACS) attack, and it was positively related to Gensini score reflecting the degree of atherosclerosis in CHD [27]. Moreover, ubiquitin was suggested to be positively related to inflammatory markers CRP, CK-MB and cTnl, which were associated with progression of atherosclerosis as well as AMI [28]. To sum up, ubiquitin is an  alternative biomarker to predict the severity of CHD. Predominant function of targeted genes on ubiquitin-like protein ligase binding might hint that TXL capsule had the capacity on regulating extracellular ubiquitin level to prevent the progression of atherosclerosis.
Fluid shear stress and atherosclerosis pathway was enriched in KEGG analysis, and it was found to be associated with microvascular and epicardial endothelial dysfunction in CHD patients. Coronary arteries exposed to abnormal microvascular endothelial function exhibited significantly lower shear stress compared with normal coronary arteries [29]. Apart from systemic risk factors, local factors as low shear stress might contribute to promotion of early focal epicardial endothelial dysfunction and potential plaque progression [30,31]. A fall in shear stress might be triggered by microvascular endothelial dysfunction which induced by established systemic risk factors like inflammation and oxidative stress at early stage of disease, further provoking as well as exacerbating inflammatory processes of coronary endothelium. Moreover, inflammation plays an indispensable role in the progression of atherosclerosis [32,33], and inflammation-related pathways such as IL-17, TNF, toll-like receptor, T-cell receptor signaling pathways, were enriched among KEGG analysis. Targeted anti-inflammatory regimen and reduction in CRP have been shown to reduce major adverse cardiovascular events in established CHD patients [34,35]. As discussed above, TXL was also capable of regulating ubiquitin to adjust CRP level, and the active compounds of TXL were validated to be effective in regulating inflammation-related pathway, which further confirmed the theory of anti-inflammatory effects of TXL capsule on CHD patients.
However, several limitations should be considered in the present study. First, retrieved active ingredients might be inconsistent with the exact compounds absorbed by patients. Second, only targeted genes of active ingredients could be found but the exploration of predominantly targeted genes by active compounds is difficult. Third, errors might occur in GO and KEGG enrichment analyses due to the complex formula of TXL capsule and enriched pathway might be confused. Last but not the least, validation is performed at cellular level and the verification in animal model to investigate more indicators is still necessary in future research.

Conclusion
Our study provided substantial molecular evidence that TXL capsule possessed the characteristics of multitargets with safe profile, and its main component is effective in regulating cytokine level as well as improving hypoxia to protect myocardial cells on CHD patients.