Network Pharmacology-Based Prediction of the Active Ingredients and Potential Targets of Chinese Herbal Danyu Gukang Pills for Application to Osteonecrosis of the Femoral Head Disease

Ethnopharmacological relevance


Conclusion
The current study rstly researched the molecular mechanism of DGP on ONFH based on network pharmacology. The results indicated that DGP may exert the effect on ONFH targeting on HIF-1α and VEGFA via HIF-1 signaling pathway.

Background
Osteonecrosis of the femoral head (ONFH) is also de ned as avascular necrosis or aseptic necrosis, which is characterized by impaired blood ow to the bone due to trauma or non-traumatic blood ow, resulting in bone cell death [1,2]. The pathogenic factors of ONFH are complex so that it is di cult to diagnosis at the early stage. It has been reported that about 80% of patients will suffer from femoral head collapse in 1 to 4 years and nally leading to necrosis even disability without effective treatment [3]. Studies indicated that in China, there were about 8.12 million cases of ONFH in cohort over the age of 15 years; And there were about 11.76 cases of ONFH per 100,000 people in the general rural population, while 9.57 ONFH patients per 100,000 people in urban population [4,5]. Importantly, due to the high rate of disability and increasing incidence of ONFH among population aged from 30 to 50, the patients might lose the ability to work, which undoubtedly causes a huge burden to the patients' family and even the society.
Up to now, there is still a lack of a simple and effective treatment for ONFH, therefore ONFH has always been regarded as a major problem in the eld of orthopedics. With the development of ONFH to the late stage, patients will have to carry out total hip arthroplasty (THA), however, the trauma and economic pressure associated with the surgery will also become a burden for patients [6,7]. Therefore, it is crucial to research the methods of retain the femoral head of patients, especially the young patients. In conservative treatment, bisphosphonates, statins and anticoagulants were medicine which commonly use [8,9]. Although the medicine mentioned above can partially restore the function of bone cells through different pharmacological mechanisms, the obvious side effects such as gastrointestinal stimulation, renal toxicity or mandibular joint necrosis cannot be ignored [10]. Complementary medicine and alternative medicine (CAM) are commonly used for the treatment of ONFH due to the absence of speci c agents. In China, traditional Chinese medicine (TCM) has been used for hundreds of years to treat ONFH which named bone erosion in Huang Di Nei Jing. Zhang et,al performed a systematic review and metaanalysis of randomized controlled trials, their results indicated that TCM take essential part in the adjuvant treatment for ONFH [11].
Danyu Gukang Pills (DGP) as a new method in the treatment of ONFH developed by our research group, is possible to cure patients with ONFH in the early stage. This formular was consist mainly of the herb medicine which function as promoting blood circulation and removing blood stasis, such as Sanqi, Dahuang, Ruxiang, Moyao, Yanhusuo, Danshen; And Chuanxiong, Muxiang, Yujin to relieve pain. Since the development of DGP, it was considered as an effective way to treat the ONFH. However, its mechanism of action is still unclear, which may be related to the drug-target interaction network of "multicomponent, multi-target".
From the view of chemistry, TCM formula is a complicate system with multiple targets as well as interactions among their substances [12]. Different from western medicine of "one target, one drug", TCM emphasized the concept of the whole human body. Network pharmacology combines the systems biology with pharmacokinetic and pharmacodynamics to research medicines, targets and their pharmacological activities [13]. Systematically, network pharmacology is on the basis of the interactions of disease-gene-target-drug networks to observe the effects of drugs on diseases [14,15]. This research methods is in accordance with the theory that TCM emphasizes the diagnosis and treatment of diseases from a holistic perspective, and the synergistic effect of TCM and its compounds [16,17].
In the current study, we used the approaches of network pharmacology and molecular docking to investigate the potential molecular mechanisms of DGP on ONFH the interactions between activated compounds and protein targets. Additionally, in vitro experiments on human bone marrow mesenchymal stem cells (BMSCs) which played crucial role in the process of osteogenesis were also conducted to validate the potential underlying mechanism of DGP on ONFH. The detailed technical strategy of the current study was shown in Figure 1.

Searching Strategy for Bioactive Compounds in DGP
To be effective, oral TCM must overcome the obstacles in the process of absorption, distribution, metabolism and excretion (ADME). Oral bioavailability (OB) is one of the most important pharmacokinetic parameters in ADME. High OB is commonly necessary to determine the drug-likeness (DL) index of bioactive substances. The components with OB ≥ 30% were considered to have high OB.
The DL index is crucial for rapid screening of active compounds as a qualitative concept when it was applied in drug design to estimate the druggability of a molecule. In the DrugBank database, the components with DL ≥ 0.18 were considered to have high druggability. Therefore, the components in DGP with OB ≥ 30% and DL ≥ 0.18 were selected as active substances in our study.

Prediction of Herbs' Targets for DGP
The protein targets of the active compounds in DGP were obtained from the TCMSP databases (http://tcmspw.com/).

GO and KEGG Pathway Enrichment Analysis for ONFH-Related Targets of DGP
The gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were conducted using the R Project for Statistical Computing. (Version 3.6.3)

Interaction Network Construction and Analysis
To investigate the in-depth molecular mechanism of DGP on ONFH, the compounds-targets and targetspathways networks were constructed by Cytoscape software (Version. 3.6.1). In these networks, the components, proteins, or pathways were represented as nodes, whereas the interactions between nodes were represented as edges.

Molecular Docking
In order to study the association between quercetin and its crucial targets in the network, molecular docking approach was applied to analyzed the strength and mode of interactions between quercetin and HIF-1α or VEGFA. The protein crystal structure of HIF-1α or VEGFA were obtained from Protein Data Bank (PDB ID: 4H6J for HIF-1α and 5DN2 for VEGFA). The process is simply described as follows: ChemDraw and Chem3D were applied to prepare the chemical structure of quercetin. Autodock vina software (version: 1.1.2) [20] was applied for molecular docking. Discovery Studio was used to analyze the results of docking and presented as nal gures.

Cell Counting Kit-8 (CCK8) Assay
Adjust the density of the cells in the control and the DP groups to be tested to 5 × 10 3 , and inoculated in 96-well plates. Cell proliferation was assessed by CCK8 method according manufacturer's instructions. CCK8 ( nal concentration 5 mg/ml) was added to each well incubated at 37 °C 4 h, discarding the supernatant and adding DMSO. Absorbance value (OD) at A570 nm was measured using a microplate reader. The experimental results were repeated three times independently.

Data analysis
Data are presented as mean ± standard deviation (SD). Statistical differences were analyzed with SPSS version 20.0 (SPSS Inc., Chicago, IL, USA) using one-way ANOVA followed by Bonferroni post hoc tests.

Results
Identi cation of Bioactive Substances in DGP 2305 components in DGP were obtained from TCMSP and BATMAN-TCM databases, the remaining components were 370 after threshold screening (OB ≥ 30% and DL ≥ 0.18) ( Table 1).
Target identi cation of DGP on ONFH For the 370 candidate bio-active substances, 4990 protein targets were collected from TCMSP databases. And then, 1413 protein targets were remained for the next analyses after eliminating the overlaps. 124 ONFH-related gene of homo-sapiens were collected from Pubmed Gene and Gene card databases. As shown in Figure 2, all herb protein targets for herb-related and ONFH-related proteins were listed as two separate sets. A Venn diagram is obtained by using the closed loop form of xed position to represent the set and its relations. As results, 18 targets of 44 components in DGP were associated with ONFH. The information of 18 targets and the interaction degrees between the targets were shown in Table 2.

Compounds-Targets Network and Analysis
Traditional Chinese medicine formulas were characteristic for their multi-target and multipharmacological effects. It is of great signi cance to research the internal mechanism of Chinese medicine formulas in the treatment of complicate diseases through the analysis of networks. In our current study, we contrasted the components-targets network of DGP on ONFH by R/Bioconductor platform (Figure 3), which included 62 nodes (44 for bio-active components and 18 for gene targets).
Among these compounds, there were several high-degree substances associated with multiple ONFH targets, mostly including quercetin (MOL000098), luteolin (MOL000006,). These targets in the network were potentially represent the essential therapeutic effects of DGP on ONFH.

Gene Functions and Pathway Enrichment Prediction
In order to clarify the biological characteristics of the potential targets of DGP on ONFH, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed by R/Bio-conductor platform. As shown in Figure 4, there were listed the top 20 (Count ≥ 2 and p < 0.05) signi cantly enriched terms in functions of these targets, which revealed that DGP may regulate the development of ONFH via cytokine activity, receptor ligand activity, growth factor activity, cytokine receptor linking to their treatment effect on ONFH. The involved pathways of DGP on ONFH were detected using KEGG analysis. As results, the signi cant pathways of DGP on ONFH were shown in Figure 5.

Targets-Pathways Network Analysis
In order to further clarify the molecular mechanism by which DGP alleviates ONFH, we constructed a targets-pathways network according to the relevant proteins and their relevant signaling pathways. As shown in Figure 6, the network included 33 nodes (13 for proteins and 20 for pathways). From these potential pathways, Proteoglycans in cancer, AGE-RAGE signaling pathway in diabetic complications, HIF-1 signaling pathway have been the top three degrees value. However, Proteoglycans in cancer and AGE-RAGE signaling pathway in diabetic complications were closely related to the cancer and diabetic complications respectively. HIF-1 signaling pathway was most associated with the effect of DGP on ONFH. Among these potential targets, VEGF, TP53 and TGFB1 were recognized as relatively high-degree targets, which played essential role in development of ONFH. From the combined drug target prediction, GO analysis, KEGG pathway enrichment analysis as well as targets-pathway network, we found that the pharmacology effects of DGP on ONFH might to be the regulation of VEGFA through HIF-1 signaling pathway. The detailed of HIF-1 signaling pathway was shown in Figure 7.

Molecular Docking Analysis
To investigate the reliability of molecular-target interaction and research accurate binding patterns, we chose quercetin as crucial molecular and HIF-1α or VEGFA as key targets according to the results of network analysis above. The results indicated that there was stronger interaction between quercetin and HIF-1α or VEGFA. As shown in Figure 8A, quercetin combined to HIF-1α mainly interacted with amino acid residues of SER by conventional hydrogen bond, TYR by pi-pi stacked and pi-donor hydrogen bond, ARG by unfavorable donor-donor. And in Figure 8B, quercetin combined to VEGFA mainly interacted with amino acid resides of THR, ILE, HIS by conventional hydrogen bond, TYR by conventional hydrogen bond and pi-pi stacked or pi-pi T-shaped, ASP by pi-anion.

DGP could improve the viability of BMSCs
The viability of BMSCs increased signi cantly after DGP treating for 24 H ( Figure 9A). By using CCK-8 assay, the results indicated that 0.1, 0.25 and 0.5 mg/ml DGP treatment for 12, 24, 36 and 48 H could promote the proliferation of BMSCs compared with control group. However, 0.25 mg/ml DGP treatment obtained the best e cacy than the other two ( Figure 9B).

DGP improve the mRNA expression of HIF-1α and VEGFA
In BMSCs, the consequence of the RT-PCR test demonstrated that 0.1 and 0.25 mg/ml DGP treating could signi cantly increase the expression of HIF-1α and VEGFA mRNA, while 0.5 mg/ml DGP treating had a reverse effect (Figure 10).

Discussion
Femoral head necrosis (ONFH) is a refractory disease characterized by damaged subchondral microcirculation, skeletal necrosis and microfracture accumulation without continuous remodeling [21]. The pathological mechanism of ONFH is very complicate, which are associated with multiple proteins and pathways during the development and procession [22]. TCM is commonly composed of a variety of compounds, with a wide range of pharmacological activities and a variety of targets and pathways [23].
Accumulating evidence indicated that TCM may bene t the treatment of ONFH [11,24,25]. Nevertheless, this characteristic of TCM may make it di cult to further study of the underlying mechanisms. Network pharmacology is an organic combination of system biology and omics, which can provide a direction for the mechanism research of complex TCM [26]. In this research, we used these methods to elucidate the pharmacological mechanisms by which DGP alleviates ONFH.
Among the TCM, components lacking appropriate pharmacokinetic properties cannot arrive to the target organ and then transmit biological activity. Thus, we screen the active compounds in the DGP with OB ≥ 30%, DL ≥ 0.18, which are considered pharmacokinetic active [18]. Importantly, it has been realized that components with high-degree may represent the crucial treating effect of DGP on ONFH. In our research, quercetin was the most signi cant components, and then, the luteolin, sesamin and kaempferol were followed. Qucercetin, a natural avonoid, was con rmed to promote BMSCs proliferation and osteogenic differentiation [27,28]. BMSCs have powerful self-proliferation ability and multi-potential differentiation capacity and can undergo osteogenesis through induction. Recently, BMSCs have been researched and a new technology has been developed and applied for regenerative medicine. Numerous reports have indicated successful results of the treatment of ONFH by using stem cell transplantation. [29][30][31].
According to the results above, DGP may exert its pharmacological functions on the treatment of ONFH through regulation of cytokine activity and growth factor activity, which was considered as the important mechanism of ONFH progression [32,33]. As predicted by network pharmacology, DGP may play therapeutic role on ONFH mainly through hypoxia-inducible factor-1 (HIF-1) signaling pathway and targeting to the HIF-1α and vascular endothelial growth factor A (VEGFA). The transcription factor HIF-1 is a key regulatory factor responsible for the induction of genes that promote cell adaptation and survival and the entire organism from normal oxygen to hypoxia. HIF-1α is a subtype discovered by the identi cation of a hypoxia response element. The functional HIF-1 is composed of two subunits, HIF-1α and HIF-1β, with the HIF-1α subunit being responsive to hypoxia [34,35]. It has been reported that HIF-1α can improve the expression of RUNX-2, OCN, and ALP in BMSCs [36]. Meanwhile, the over-expression of HIF-1α can promote the differentiation from BMSCs to osteoblasts after the osteogenesis induction [37].
Moreover, VEGF is regulated by HIF-1α. The over-expression of HIF-1α enhanced the secretion of VEGF in BMSCs [38,39]. VEGF is the currently recognized as the growth factor with the strongest ability to promote angiogenesis, which could improve the proliferation and differentiation of endothelial cells and induce the formation of new vessels [40].
To further validate the prediction of the targets of DGP, we used molecular docking. The results demonstrated that both HIF-1α and VEGFA had a high interaction with quercetin by combined with amino acid residues. Furthermore, we designed an in vitro experiment to validate the postulation, which DGP may alleviated ONFH via HIF-1 signaling pathway. The results showed that DGP treatment signi cantly increased BMSCs viability and RT-PCR assay indicated that DGP treatment could improve the mRNA expression of HIF-1α and VEGFA. These results revealed that DGP might exert the effect on ONFH though HIF-1 signaling pathway. The in-depth molecular mechanism needs further clari ed.

Conclusion
To summary, our current research rstly gives a systematic view of potential targets and pathways relevant to the treating effect of DGP on ONFH. Importantly, for the further con rmation, we still need more in-depth molecular experiments to verify the predication from the network pharmacology. All compounds of the 19 Chinese medicine herbs in DGP were obtained from the traditional Chinese medicine systems pharmacology (TCMSP: http://tcmspw.com/) [18] and Bioinformatics analysis tool for molecular mechanism of Traditional Chinese Medicine (BATMAN-TCM: http://bionet.ncpsb.org/batman-tcm/index.php) [19] databases.All participants signed informed consent forms before participating in this study.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests  The technical strategy of the current study.   The compounds-target network for DGP on ONFH. The purple nodes represent candidate active compounds and the red nodes represent potential protein targets. The edges represent the interactions between them and shade of color is proportional to their degree.  The 20 most signi cance of GO analysis of therapy target gene of DGP on ONFH.

Figure 5
The 20 most signi cance of KEGG pathway enrich analysis f therapy target gene of DGP on ONFH.

Figure 6
The target-pathway network for DGP on ONFH. The purple nodes represent targets and the green nodes represent pathways. The edges represent the interactions between them and node size is proportional to their degree.   The docking patterns shown as 2D presentation of quercetin with (A) HIF-1α and (B) VEGFA. The red arrows refer to the amino acid residues that interact with quercetin in binding site of HIF-1α or VEGFA.