System Pharmacology-Based Strategy to Decode the Synergistic Mechanism of GanDouLing for Wilson's Disease

Results Firstly, 324 active compounds have been identified in the GDL formula. Meanwhile, we identified 1496 human genes which are related to WD or liver cirrhosis. Functional and pathway enrichment analysis indicated that NOD-like receptor signaling pathway, bile secretion, calcium signaling pathway, steroid hormone biosynthesis, T cell receptor signaling pathway, apoptosis, MAPK signaling pathway, and so forth can be obviously regulated by GDL. Further, in a mouse model of WD, in vivo experiments showed that GDL treatment can not only reduce the pathological symptoms of the liver but also reduce the apoptosis of hepatocytes. Conclusions In this study, systemic pharmacological methods were proposed and the mechanism of GDL combined therapy for WD was explored. This method can be used as a reference for the study of other mechanisms of traditional Chinese medicine.


Introduction
Wilson's disease (WD) is a hereditary disease which was first proposed by Kinnear Wilson in 1912 and caused by mutations in the ATP7B copper transporter gene that causes copper to accumulate in the liver and brain [1]. e phenotypic manifestations of WD are often highly variable, involving varying degrees of liver and/or neurological symptoms [2][3][4]. ere is increasing evidence showing that the clinical manifestations of WD are affected by genetic and epigenetic factors [5][6][7], which may be responsible for the high phenotypic variability of WD. WD's liver damage ranges from mildly symptomatic to steatosis, cirrhosis, or acute liver failure. e involvement of the nervous system is characterized by dyskinesia, manifested as Huntington's disease-like tonic and tremor.
According to its pathogenesis, WD is treated by blocking intestinal copper uptake by chelating agents, such as triene amine, tetrathiomolybdate, or zinc salts. However, their high cost or side effects may hinder clinical use [8]. At present, penicillamine (d-penicillamine) is widely used, because of its low cost and remarkable curative effect [9,10]. But, with many adverse reactions, such as fever, rash, gastrointestinal reaction, leukopenia, systemic lupus erythematosus, and nephritis, which limit its clinical application [11], a large number of studies have shown that Penicillium has therapeutic effects on liver diseases [12,13], especially WD [9,13]. A large number of clinical practice and theoretical studies have proved that traditional Chinese medicine is an effective drug for the treatment of WD. Traditional Chinese medicine can eliminate the abnormal deposition of copper ions in the liver, brain, and other tissues through not only kidney metabolic pathway but also bile secretion, so as to improve the curative effect and reduce side effects. More importantly, the combination of traditional Chinese and Western medicine can reduce side effects. GanDouLing is a TCM, and as an in-hospital preparation, it was prepared by the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine [14,15], which has proved that there can be effective improvement of liver and neuropathy in clinical, animal, and cell models [16][17][18][19]. Specifically, it can effectively play an important role in removing blood stasis from the liver and gallbladder, promoting blood circulation and removing copper. GanDouLing is mainly composed of  [20,21].
Network pharmacology is an emerging discipline, which is based on the characteristics of biomolecules and a number of authoritative databases showing us a preliminary understanding of mechanisms of medicine and disease. e previous research study used the method of network pharmacology to provide a promising method for understanding the pharmacological mechanism of TCM. In the article, we developed a network pharmacology analysis to determine the anti-WD mechanism of GDL and the experiments in vitro were then performed given the results of network pharmacology analysis.

Chemical Components
Database, ADME Screening, and Targets Identification. GanDouLing is a TCM, and as an inhospital preparation, it was prepared by the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine and the quality control of chemical fingerprint has been performed [14,15], which is mainly composed of BJT, BWWZ, DH, EZ, HL, JH, JQC, and SDH [20,21]. Following with a strategy of system pharmacology, the pharmacology mechanisms of WD by GDL were encoded. Firstly, all compounds of GDL containing eight herbs were gathered from the TCMSP database (http://lsp.nwu.edu.cn/index. php) [22] with the parameter of oral bioavailability (OB) [23], drug-likeness (DL) [24], and so forth. Next, the active components were filtered out by the ADME method. Targets of the active components in GDL were predicted by BAT-MAN-TCM [25]; next, the intersection genes of GDL targets and WD-related genes including liver cirrhosis collected from the Comparative Toxicogenomics Database (CTD) [26] and DisGeNET [24,27] were identified. en, GO [28] and KEGG [29] pathway analysis was performed by clus-terProfiler [30]. Last, the PPI network was constructed by STRING [31] and visualized by Cytoscape [32], and the modules of PPI were identified by MCODE [33].

Establishment of Animal Models.
Twenty female TX mice (20 ± 2 g) and ten female DL mice were at the ages of 8-10 weeks [34]. Original mice were obtained from the Jackson laboratory with the help of Beijing Vital River Laboratory Animal Technology, Ltd. is study was carried out in strict accordance with the "National Institutes of Health Guidelines for the Care and Use of Experimental Animals." It was approved by the Institutional Animal Care and Use Committee of the Anhui Hospital of TCM. e twenty female TX mice were divided into two groups, 10 mice in each group: the Wilson group and GanDouLing group (0.486 g/kg/day, GanDouLing was produced by the Anhui Hospital of TCM: the active ingredients of each herb were extracted with 65% ethanol and then combined with extracted filtrate; the filtrate combinations were baked into dry paste at the right temperature and starch was added and packed into GanDouLing troche) [16]. e ten female DL mice (20 ± 2 g) were treated as the control group. e mice in the GanDouLing groups were treated by intragastric administration with GanDouLing for 8 weeks. e Wilson group and control group were treated by intragastric administration with an equivalent volume of 0.9% saline every day. All the mice were housed in a controlled humidity (50-70%) and room temperature (18-22°C), fed in the isolation cages with independent air supply, and were given free access to food and water ad libitum in an alternating 12 h light/dark cycle over a period of 8 weeks.

Hematoxylin and Eosin Stain.
e mice liver isolated tissues were fixed with 4% paraformaldehyde for 3 h and then dehydrated with ethanol and xylene. According to the previous literature, staining was carried out for liver samples [16]. e result of staining was that the cell nucleus was dyed blue by Hematoxylin and the cytoplasm was dyed red by Eosin.
2.6. Statistical Analysis. GraphPad Prism 7 was employed for comparing the expression of Bcl-2 and Bax and the activity of caspase-3 and caspase-8. And the method of Student's t-test was used for analyzing the comparison. When P < 0.05, it was considered statistically significant.

Results
Following a system pharmacology model, the pharmacology mechanisms of WD by GDL were encoded. Firstly, all compounds of GDL were gathered from a database. Next, the active components were filtered out by the ADME method. en, the targets, pathway, and PPI network were identified from integrated predictive models. Finally, we employed the TX-j mouse model for proving that GDL could improve WD liver injury by regulating apoptosis.

Active Components Target for Anti-WD in GDL.
In total, 802 compounds were obtained for GDL, including 174 compounds for Bajitian, 130 compounds for Beiwuweizi, 92 compounds for Dahuang, 202 compounds for Danshen, 81 compounds for Ezhu, 48 compounds for Huanglian, 52 compounds for Jianghuang, 61 compounds for Jinqiancao, and 76 compounds for Shudihuang ( Figure 1). Up to 65 compounds are contributed to more than one herb. By contrast, the other compounds belong to only one of the nine herbs. And then, two ADME properties of these components including OB and DL were focused to further state the components of GDL. Generally speaking, any TCM formulation consists of many components, but most of the components do not possess satisfactory properties of pharmacodynamic and pharmacokinetic. In the present study, OB and DL were adopted to filter for active components. After screening, 324 active components were screened out of the 802 components of GDL. And the targets for active components were predicted by BATMAN-TCM. In total, 1,496 genes were identified which are related to WD and liver cirrhosis, and 287 genes could be a target with 151 active components. And the components-target network was performed by Cytoscape ( Figure 2).

GO and Pathway Analysis to Encode the Pharmacological Mechanisms of GDL.
To explain the molecular biology of the therapy mechanism of GDL for WD, all target proteins were used to perform the enrichment of GO and KEGG pathways by clusterProfiler.

PPI Network.
e PPI network for the target genes was constructed based on the STRING database. As shown in Figure 4, the PPI network consists of 283 target genes and 4,332 edges, which means some target genes could interact with numerous genes. In fact, we found that dozens of target genes exhibited high degree characteristics, such as INS (insulin, degree � 160), ALB (albumin, degree � 152), AKT1 (AKT serine/threonine kinase 1, degree � 152), IL6 (interleukin 6, degree � 150), TP53 (tumor protein p53, degree � 128), TNF (tumor necrosis factor, degree � 125), and EGFR (epidermal growth factor receptor, degree � 111). High connectivity nodes with degree ≥70 are listed in Table 1.
e plug-in of MCODE was used to identify the modules with the PPI, and a total of 13 modules were gained. By sorting with MCODE score, two modules with MCODE score ≥3 and nodes ≥10 were screened out and named module 1 and module 2. And then, the subnetwork visualization was performed for module 1 and module 2 ( Figure 5).

Effects of GDL on the Mouse Model of WD.
In order to observe the protection effects of GDL with WD, TX mice were treated by intragastric administration with Gan-DouLing in 0.486 g/kg every day for 8 weeks. e results indicated that WD induced liver injury in the mouse model. HE staining results shown in Figure 6 suggested that the control group presented healthy tissues, whereas the WD models presented a diffuse lesion in the liver. A large number of hepatic parenchymal cells were observed to be necrotic in TX compared with the control. And GDL could effectively improve these lesions.
Collagen was stained blue in liver sections with Masson staining. e TX mice showed obvious collagen accumulation compared with the control group ( Figure 6(b)), and administration GanDouLing can reverse.

Effects of GDL on Protein Expressions and Caspase Activity
Associated with Apoptosis. In the analysis of network pharmacology, we found that there are more than 102 targets for anti-WD with GDL enriched to cell death, and up to 10 targets for anti-WD with GDL were enriched to the apoptosis signaling pathway, including NFKB1, PPP3CA, APAF1, RELA, AKT1, PPP3CB, IL1B, NFKBIA, TNF, and BCL2.
In order to explore the possible mechanism for WD, effects of the GDL formula on apoptosis relative proteins including Bcl-2 and Bax were observed using the western blotting method. In comparison with the control group, Bax was obviously overexpressed and Bcl-2 was suppressed in    Figures 7(a) and 7(b)). As shown in Figure 7(c), the TX group led to a 2.8-fold increase in caspase-3 and 1.9-fold increase in caspase-8 expression levels compared with controls. e above results indicated that GanDouLing can inhibit apoptosis of liver cells.

Discussion
At present, systemic pharmacology provides a powerful tool for encoding the compatibility and mechanism of traditional Chinese medicine [35][36][37]; there are many studies focusing on liver disease or injury based on the theory of systems pharmacology. On the basis of network pharmacology, the antioxidative effect of Zhizi Dahuang Decoction on alcoholic liver disease was expounded [38]. Wei et al. successfully revealed that 16 targets related to 11 compounds of SCG are closely related to the treatment of liver fibrosis, and TGF-β 1/ Smad signal pathway is relatively important. ey also demonstrated through animal experiments that SCG can significantly improve liver fibrosis by inhibiting TGF-β 1/ Smad pathway [39]. In order to explain the pharmacological mechanism of Yinchenhao decoction (YCHD) which is a classical TCM formula that has been widely used in the treatment of liver fibrosis caused by chronic hepatitis B and jaundice for more than 1800 years, Cai et al. have employed the methods of network pharmacology and transcriptomic analysis to systematically describe the pharmacological mechanism [40]. At the same time, a large number of studies on the treatment of hepatocellular carcinoma with TCM have been carried out [41][42][43].
In this study, we performed the system pharmacology to construct a strategy for decoding the TCM pharmacologic molecular mechanism of GanDouLing which has proved   Evidence-Based Complementary and Alternative Medicine that it can be an effective improvement of liver and neuropathy in clinical, animal, and cell models [16][17][18][19] With the integrated application of physical and chemical properties, network topological features, functional analysis, and path analysis provide a reference for the new method. However, its discovery mainly depends on theoretical analysis, so more experiments are needed to verify our findings and potential clinical significance. We find that some recent studies have introduced experiments such as transcriptome, proteomics, qPCR, and WB. [39,40,44]. In our study, with regular system pharmacology, we found that hundreds of GO-terms could be affected by GDL, such as cell proliferation, cell death [45], inflammatory response [46], response to oxidative stress [47], and response to lipid [48]. Meanwhile, dozens of pathways could be affected by GDL including adipocytokine signaling pathway, NOD-like receptor signaling pathway, bile secretion, retinol metabolism, calcium signaling pathway, steroid hormone biosynthesis, T cell receptor signaling pathway, apoptosis [45], MAPK signaling pathway, cytokine-cytokine receptor interaction, PPAR signaling pathway, Toll-like receptor signaling pathway, B cell receptor signaling pathway, Jak-STAT signaling pathway, and focal adhesion. e TX-j mouse model was employed for proving that GDL could improve WD liver injury, by regulating apoptosis. Firstly, HE and Masson staining showed that GanDouLing could improve the necrosis and fibrosis of the TX-j liver tissue. Furthermore, targeting the apoptosis signaling pathway identified by network pharmacology, we found that GanDouLing significantly inhibited the expression of proapoptosis Bax protein and promoted the expression of antiapoptosis BCL2. In addition, the activity of caspase-3 and caspase-8 was tested and found to be significantly increased in TX-j mice, which was consistent with the previous research results: the activity of caspase-3 increased in WD patients or copper-induced hepatocytes [49,50]. Meanwhile, GanDouLing can reduce its activity to a certain extent.

Conclusion
In conclusion, the mechanism of GDL in the treatment of WD involves a variety of active components, targets, and pathways. In this study, 151 active components, 287 potential targets, and 43 related signaling pathways were predicted. Animal validation tests showed that GDL could effectively improve WD liver lesions, especially by inhibiting the apoptosis of liver tissues.
Data Availability e datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
e authors declare no conflicts of interest.

Authors' Contributions
Juan Zhang, Daojun Xie, and Wenming Yang conceived and designed the study. Juan Zhang and Hong Chen performed the system pharmacology analysis. Juan Zhang and Hong Chen performed the animal experiments, HE analysis, and WB. Yuancheng Bao, Jingjing Xu, Huaizhou Jiang, Ting Dong, and Hui Han analyzed the data and previewed the manuscript. All authors read and approved the final manuscript. Juan Zhang and Hong Chen contributed equally to this work.