Exploring the Pharmacological Mechanism of the modified Chinese medicine formulas “Shenfu-Linguizhugan decoction” treating Dilated cardiomyopathy Based on Network Pharmacology CURRENT

Background: Dilated cardiomyopathy (DCM) is a non-ischaemic cardiac muscle disease with structural and functional myocardial aberration can lead to extensive morbidity and mortality due to complications in particular heart failure and arrhythmia. Two classic Chinese medicine formulas, Shenfu decoction and Linguizhugan decoction, were both shown to exert therapeutic effects on heart disease. Thus, modified Shenfu and Linguizhugan decoction (SFLGZGD) is recommended for treatment DCM. However, its chemical and pharmacological characteristics remain to be elucidated. In the current study, a network pharmacology approach was applied to characterize the action mechanism and target genes of SFLGZGD on DCM. Methods: The gene expression of DCM was obtained from the Gene Expression Omnibus (GEO). All compounds were obtained from the correlative databases, and active mixture were selected according to their oral bioavailability (OB) and drug-likeness (DL) index. The potential targets of SFLGZGD were obtained from the traditional Chinese medicine systems pharmacology (TCMSP) database. The compound-target and target-pathway networks were constructed. The protein-protein interactive (PPI) network generated by R software was visualized by Cytoscape, and the topology scores, functional regions, and gene annotations were analyzed using plugins of Bisogenet and CytoNCA. The potential pathways related to target genes were determined by gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. Results: A total of 963 differentially expressed genes (DEGs), including 538 upregulated genes and 425 downregulated, were obtained from GSE19303. A total of 636 ingredients in SFLGZGD were obtained, among which, 93 were chosen as bioactive components. The compound-target network included 10 bioactive components and 18 potential targets and a total of 1939 genes obtained in the PPI network, among them, a total of 16 genes were screened out. Moreover,129 terms on the GO analysis and six pathways obtained. Among these potential targets, EGFR, CDKN1A, MMP1, COL1A1,

suggested that network pharmacology prediction may provide a useful tool for describing the molecular mechanism of SFLGZGD on DCM.
Background Dilated cardiomyopathy (DCM) is a non-ischaemic cardiac muscle disease with structural and functional myocardial aberration by the presence of left ventricular ectasia and contractile dysfunction [1][2][3]. The WHO defines DCM as serious cardiac disarray that can lead to extensive morbidity and mortality just that complications in particular heart failure and arrhythmia [4].
Mutations in genes that encode sarcomere, cytoskeleton, and nuclear membrane proteins accounted for 35 percent of cases [1,3]. Because of DCM has a large number of genes and alleles, genetic diagnosis can help predict prognosis, especially the risk of arrhythmia in some subtypes [5].
Traditional Chinese Medicine (TCM) is one of the most valuable treasures of Chinese culture, and it has been used clinically in East and Southeast Asia for over 2,000 years. As one of the most popular models of complementary and alternative medicine in China, TCM is gradually accepted by non-Chinese people due to its apparent efficacy, abundant resources, and low toxicity, and is increasingly used in western countries [6,7]. In the clinical treatment of DCM, there are many proprietary Chinese medicines and TCM injections alone or as an adjuvant to conventional chemotherapy [8][9][10][11].
Shen fu and Lin Gui Zhu Gan decoction both are commonly used TCM formulas. Shen Fu decoction consists of only two herbs, Ginsen Radix Et Rhizoma Rubra (GRERR, Hong-shen in Chinese) and Aconiti Lateralis Radix Praeparata (ALRP, Fu-zi in Chinese), which was widely applied for the treatment chronic heart failure [12]. Lin Gui Zhu Gan decoction consists of four TCMs, Poria Cocos (PC, Fu-lin in Chinese), Cinnamomi Ramulus (CR, Gui-zhi in Chinese), Atractylodes Macrocephala Koidz (AMK, Bai-zhu in Chinese), and Licorice (Gan-cao in Chinese), which also can be used to treat chronic heart failure [13]. Hedysarum Multijugum Maxim (HMM, Huang-qi in Chinese), as a plant of the genus astragalus in the leguminous family, which was widely applied for the treatment of various diseases, also including DCM [14,15]. Based on the research, therefore, modified Shen fu decoction combination Lin gui zhu gan decoction (SFLGZGD) is recommended to be applied in the TCM treatment scheme of cardiomyopathy in China. However, the chemical and pharmacological 4 foundations of SFLGZGD in treating heart disease, especially DCM, was not comprehensively assessed by suitable approaches.
TCM is a complicated system with multiple targets and complex interactions among its components [6]. However, unlike chemical drugs found for specific proteins, the understanding of the molecular basis of TCM is still limited, and the study of modern TCM theories is lagging behind, which has slowed the development of novel TCM drugs [16]. With the rapid development in big data, bioinformatics, systems biology, and multi-pharmacology, network-based drug discovery is considered to be a promising low-cost approach to drug development. Based on the system biology method, the concept of network pharmacology of traditional Chinese medicine was first proposed in 2014 [17].
With drug discovery facing a significant bottleneck, drug research and development has gradually changed from the current "single target, single drug" model to the "network target, multi-component therapy" model [18]. It is helpful to evaluate the rationality and compatibility of TCM by providing a specific composite target and target-path network. In addition, network pharmacology methods have been used to study "complex protein/gene-disease" pathways that can describe the complexity between biological systems, drugs, and diseases from a network perspective, sharing a holistic philosophy similar to that of TCM. The application of systematic biological methods to determine the pharmacological action, mechanism, and safety of TCM are of great value to the research and development of modern TCM. It has been widely applied in the research on the mechanism of TCM treatment of complex diseases, such as ischemic stroke, cancer, chronic atrophic gastritis, type 2 diabetes, and cardiovascular disease [6,[19][20][21][22]].
In the present study, bioinformatics analysis was used to study the pharmacological network of SFLGZGD on DCM, and to predict its active components and potential target genes. In addition, we performed GO and KEGG pathway analyses to explore the functions and pathways involved. The detailed flow chart of the current study was shown in Fig. 1.

Differential gene expression analysis
The expression of gene extracted from the Gene expression dataset (GSE19303) [

Selecting procedure for Bioactive constituent in SFLGZGD
Oral TCM decoction must overcome the obstacles caused by the process of absorption, distribution, metabolism and excretion (ADME) to be active. During the ADME processes, oral bioavailability (OB) is one of the most considerable pharmacokinetic parameters [26]. High OB is usually a crucial barometer to determine the drug-likeness (DL) index of active ingredients. The ingredients with OB ≥ 30% were regarded to have high OB. As a qualitative concept for drug design, the DL index can be used to evaluate the pharmacist of molecules, which has great significance for the rapid screening of active substances. The average DL index is 0.18 in the Drug Bank database. Substances with a DL index ≥ 0.18 are considered to have high pharmacy ability. Consequently, the compounds in SFLGZGD with OB ≥30% and DL index ≥0.18 were selected as active substances in this study. The protein targets of the active ingredients in SFLGZGD were obtained from the TCMSP database.

Construction regulatory network of TCM compounds and PPI network
To further explore the molecular mechanism of SFLGZGD on DCM, the compound-target and target-pathway regulatory networks were generated using Cytoscape software (version 3.7.2, http://www.cytoscape.org), which is a JAVA-based network analysis and visualization tool [27]. We obtained the gene list, nodes, compound-target, and target-pathway interactions using the Perl Script. In these regulatory networks, the compounds, proteins, or pathways were performed as nodes, whereas the compound-target or target-pathway interactions were expressed as edges.
Protein-protein interactive (PPI) network is the basic frame for proteins to determine their function in the system biology. In the current work, the PPI network generated by Bisogenet plugin (version

Gene ontology and pathway enrichment analysis
Functional enrichment analysis of compound-target or target-pathway was based on the R/Bioconductor "clusterProfiler," "org.Hs.eg.db," "enrichplot," and "ggplot2" package, which identified Gene Ontology (GO) categories in the Biological Processes (BP), Cellular Components (CC), or Molecular Functions (MF) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to reveal bio-information of the identified genes, both up-and down-regulated. Count ≥ 2 and FDR 0.05 were set as a significance threshold to determine enrichment terms. Moreover, we constructed a relational network of KEGG by Perl Script and Cytoscape software.

Results
A total of 963 DEGs, including 538 upregulated genes and 425 downregulated, were obtained from DCM-expression microarray dataset GSE19303, which was statistical significance between normal controls and DCM patients. The TOP 20 genes, both in the upregulated and downregulated groups, 7 are shown in the heatmap and all the DEGs through volcano plot filtering (Additional file 1: Fig. 1). A detailed list of the 936 genes and the corresponding statistics is shown in Additional file 1.

Identification of Bioactive Components in the SFLGZGD
A total of 636 ingredients in SFLGZGD were obtained from the TCMSP database. Among the 636 components in SFLGZGD, 324 (50.9%) met the requirement of OB ≥30% and 93 (14.6%) met the requirements of OB ≥30%, and DL index ≥0.18. The whole components list manifest in Table 1, and the detailed list of the 93 candidate Bioactive Components as shown in Additional file 2.

Herb-Compound-Compound Target Network
Exploring the molecular basis of TCM is very crucial for the modernization of TCM, and understanding the targets of TCM is momentous. In the present study, the compound-target network of SFLGZGD on DCM was constructed (Fig.2), which was composed of 28 nodes (10 for bioactive components and 18 for potential targets). These potential targets, including ACHE, MMP3, EGFR, CDKN1A, MMP1, ICAM1, PTGER3, HSPB1, MGAM, COL1A1, ABCG2, PSMD3, COL3A1, CLDN4, CTSD, IGFBP3, MTTP, and ND6, associated with the 10 bioactive components. Except for the MOL000098 derived from PS and HMM, and the rest of bioactive ingredients comes from HMM, CR, ALRP, AMK, and PS, respectively, and the detailed information was shown in Table 2, which signify that these five TCMs and these targets in the network play a significant effect in the process of SFLGZGD treating DCM.

PPI Network Analysis
The PPI network was constructed by the Bisogenet plugin. A total of 1939 genes obtained, the degree ranged from 1 to 933 (Additional file: Fig. 2). The topologically essential genes screened by CytoNCA, a total of 249 genes selected by the parameter "Without weight" and DC min value greater than 81, then, we generated the sub-network based on the previously filtered data, which was shown in the Fig.3A. Besides, we constructed the network by the parameter "Without weight" and BC min value greater than 600 based on the sub-network of previous 249 genes, a total of 16 genes were screened out (Fig.3B). Finally, we also generated the sub-network, these genes in the network may account for 8 the significantly essential therapeutic effects of SFLGZGD on DMC, especially the high-degree protein targets, such as EGFR (degree=933), NTRK1 (degree=674), and HSPB1(degree=415). The detailed PPI information was shown in Additional file 3.

Gene ontology enrichment and KEGG pathway analysis
To identify the biological characteristics of presumptive targets of SFLGZGD on DCM in detail, the GO and KEGG pathway enrichment analyses of involved targets were conducted using several Rpackages by R software. We obtained 129 terms on the GO analysis, including 94 BP, 25 CC, and 10 MF, respectively, Count ≥ 2 and FDR 0.05 were as cutoffs (Fig. 4). The detailed GO information was The KEGG enrichment analysis revealed that 6 pathways were significantly associated with targets of SFLGZGD on DCM as shown in Fig. 6 and Table 3. Among these potential pathways, "Relaxin signaling pathway" was considered the most significant one with the highest degree value (Fig. 6A). Among these potential targets, EGFR, CDKN1A, MMP1, COL1A1, COL3A1, MMP3, ICAM1, and HSPB1 were identified as relatively high-degree targets, which played a crucial role in the development of DCM and were considered as the key markers of SFLGZGD treatment on DCM (Fig. 6B). From the incorporated drug target prediction, GO, and pathway enrichment as well as network analyses, we speculated that the effects of SFLGZGD on DCM might be associated with the roles of its key targets 9 including EGFR, MMP1, COL1A1, COL3A1, CDKN1A, MMP3, ICAM1, and HSPB1 in regulating myocardial cell function.

Discussion
DCM is a non-ischaemic cardiac muscle disease with structural and functional myocardial malformations. Gene mutation can induce DCM, including genes encoding structural elements of the sarcomere and desmosome [2]. In the present study, we obtained the DCM gene data from the GSE19303 dataset using the bioinformatics analysis method and identified 538 genes that might involve in this disease. TCM uses therapeutic herbs to treat diseases based on a patient's syndrome and to recover from a balance of life and body functions [30] and may manifest extensive pharmacological activities with multiple targets and pathways [31], which may benefit the therapy of DCM. However, this characteristic may bring difficulty to lucubrate the internal mechanism of TCM.
The Network pharmacology approach, which based on the rapid progress in bioinformatics, systems biology, and polypharmacology, may offer a promising approach for the mechanistic study of intricated TCM. In the traditional Chinese medicine system, compounds lacking appropriate pharmacokinetic properties cannot reach the target organ to convey biological activities. In the present study, the SFLGZGD mixture with OB ≥ 30% and DL index ≥0.18 were considered pharmacokinetically active, which may be likelihood absorbed and distributed in the human body. In the ingredients-target network, ingredients with high-degree may account for the major therapeutic effects of SFLGZGD on DCM.
In the present study, the compound-target network of SFLGZGD on DCM was constructed, which was composed of 10 candidate bioactive components and 18 potential targets. Among the network, only Finally, the GO and KEGG pathway enrichment analyses of involved targets were conducted using several R-packages by R software to identify the biological characteristics of presumptive targets of SFLGZGD on DCM in detail. 129 enriched GO terms revealed fundamental pathophysiological alterations of DCM. These target genes mainly rich in the terms of extracellular matrix and its related, which demonstrated that SFLGZGD treats DCM patients may be via inhibit cardiomyocyte fibrosis and cardiomyocyte hypertrophy. Adverse reconstructing of the extracellular matrix (ECM) is a significant feature of heart failure. ECM reconstructing is extensive in several heart diseases and hinders cardiac filling, often leading to heart failure [47][48][49]. Response to oxidative stress, steroid hormone are also leading causes of adverse remodel reconstructing during DCM development [50,51]. Furthermore, 8 genes (EGFR, CDKN1A, MMP1, COL1A1, COL3A1, MMP3, ICAM1, and HSPB1) identified by KEGG analysis were significantly enriched pathway terms, including Relaxin signaling pathway, bladder cancer, rheumatoid arthritis, prostate cancer, AGE-RAGE signaling pathway in diabetic complications, and amoebiasis. EGFR is a transmembrane glucoprotein that is a member of the protein kinase superfamily, and a receptor of the epidermal growth factor family, binding of the protein to a ligand induces receptor dimerization and leads to cell proliferation. Modulating EGFR transactivation and signal is a fundamental mechanism during DCM development [52,53]. Specifically, ErbB2 signaling in cardiomyocytes is, therefore, essential for the prevention of dilated cardiomyopathy. CDKN1A encodes a potent cyclin-dependent kinase inhibitor to inhibit the activity of cyclin-cyclin-dependent

Conclusion
In conclusion, the pharmacological mechanism by which SFLGZGD treating DCM was investigated with the combination of TCM-network pharmacology prediction. We demonstrated that SFLGZGD might inhibit cardiomyocyte fibrosis and cardiomyocyte hypertrophy to treat DCM patients, which mainly via the regulation of the extracellular matrix, extracellular structure organization, and Relaxin signaling pathway. Our study further suggested that TCM-network pharmacology prediction may offer a useful tool to characterize the action mechanism of TCM in detail. We underline a central role of some hub genes such as EGFR, CDKN1A, and MMP1 in the process of SFLGZGD treating on DCM, which may     Figure 1 The flow chart of the current study.