Network Pharmacological Analysis of Huanglian Jiedu Decoction for Anti- Atherosclerosis


 Objective: We aimed to predict the possible active components,key targets and pathways of Huanglian Jiedu Decoction(HLJDD) for anti-atherosclerosis. Methods: The TCMSP database was searched to obtain the active components and targets of HLJDD, the GeneCards and OMIM databases were searched to obtain related targets of atherosclerosis, and we obtain the intersection targets of them, which are the potential targets of HLJDD for anti-atherosclerosis.Application of Cytoscape 3.6.0 software to build a herbal-active ingredient-potential target regulation network.We perform protein-protein interaction(PPI) network analysis of potential targets through STRING 11.0 database and obtain the key targets,and the results of PPI network of key targets were visualized by Cytoscape3.6.0 software. Gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the key targets were performed using STRING11.0 database, and we finally constructed the possible pharmacological network of HLJDD for anti-atherosclerosis .Results: We finally obtained 14 key active ingredients of HLJDD, 65 key targets of anti-atherosclerosis, and 14 key active ingredients corresponded to 52 of these targets. These targets are mainly involved in biological processes such as reaction to organic substance, reaction to chemical stimulation,etc.They mainly involved in biological signaling pathways such as pathways in cancer,IL-17 signaling pathway,etc. Conclusion: HLJDD may act on 52 key targets such as PTGS2, HSP90AA1 and RELA through 14 key active ingredients, and influence the signaling pathways including fluid shear stress and atherosclerosis,PI3K-Akt signaling pathway,IL-17 signaling pathway,AGE-RAGE signaling pathway in diabetic complications,TNF signaling pathway,etc.Thus, it may play an anti-atherosclerosis role by inhibiting inflammatory reaction, oxidative stress and improving hemodynamics,etc.

targets,performed GO analysis and KEGG pathway enrichment analysis of key targets,and constructed the pharmacological network of "HLJDD-key target-pathway".We aim to find out possible key active ingredients, key targets and action pathways of HLJDD for anti-atherosclerosis, so as to provide some references for further research on the effect of HLJDD for atherosclerosis.

Materials and Methods 2.1 Screening of Active Ingredients and Targets of HLJDD
The chemical components of four Chinese herbal medicines including CR, SR, PCC, and GF were obtained by searching Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform(TCMSP, http://tcmspw.com/tcmsp.php) [20] . Based on ADME(absorption, distribution, metabolism and excretion) prediction model, the active ingredients of HLJDD were screened with oral bioavailability (OB) ≥ 30% and drug similarity (DL) ≥0. 18. At the same time, we searched TCMSP to obtain all the reported targets of the four Chinese herbal medicines, and a database of targets was constructed. Perl software (https://www.perl.org/get.html)was used to screen out the targets corresponding to the active ingredients, and we construct the file of "HLJDD -Active Ingredients -Targets ".

Obtaining Gene Name of The Targets
The protein name of the target was input into the online database STRING11.0(https://string-db.org/cgi/input.pl) [21] ,and we converted the protein names into corresponding gene names in the context of setting the species as homo sapiens. Some of the targets that could not get the corresponding gene names were deleted, and finally we constructed the file of "HLJDD-Active Ingredient -Target -Gene".

Screening of The Potential Targets of HLJDD for Anti-atherosclerosis
Based on the targets of the active ingredients of HLJDD and the related targets of atherosclerosis, R 3.6.1 software (https://www.r-project.org/) was used to obtain the common targets of the active ingredients of HLJDD and atherosclerosis, which was considered as a potential target of HLJDD for anti-atherosclerosis.

Construction of "HLJDD-Active Ingredient-Potential Targets " regulation network
Perl software was applied to identify the Chinese herbal medicine and its active components corresponding to the potential targets of HLJDD for anti-atherosclerosis, and the file of "HLJDD-active ingredient-potential targets" was constructed. Cytoscape3.6.0 software was used to construct the regulation network of "HLJDD-Active Ingredient-Potential Targets". Through topological analysis of the network, the active components with node connection degree greater than the average node connection degree of the network were found out and identified as the key active components of HLJDD for anti-atherosclerosis.

Construction of Protein-Protein Interaction(PPI) Network of The Potential Targets
Enter the gene name of the potential targets into the online database STRING11.0, set the species as homo sapiens, confidence score ≥0.4, do not display free targets, we finally obtain PPI network of the potential targets and corresponding result files. The greater the node connection degree of each target in the network, it indicates that it has a more important position in the network and has a higher probability to become a key target.Analyzing the network, the protein that the node connection degree is greater than the average node connection degree is obtained, which is considered as the key node of the network, that is, the key targets. R 3.6.1 software was used to visualize the key targets and their corresponding node connection degree. The gene names of the key targets were imported into the online database STRING11.0 again, and the PPI network analysis of the key targets was performed in the same operation method.The results were exported in text format, and the PPI network was visualized by Cytoscape3.6.0 software.

GO Analysis And KEGG Pathway Enrichment Analysis of The Key Targets
Application of the analysis function in the online database STRING11.0,the GO analysis and KEGG pathway enrichment analysis of key targets were performed, and the false discovery rate(FDR)≤0.05 was considered to be of statistical significance. The "ggplot2" package of R 3.6.1 software was used to visualize the GO analysis and KEGG pathway enrichment analysis results.

Construction of "HLJDD-Key Active Ingredient-Key Targets-Pathway" network
Based on the results of key active ingredients, key targets and KEGG pathway enrichment analysis, the network of "HLJDD-Key Active Ingredient-Key Targets-Pathway" was constructed by using Cytoscape3.6.0 software.

Result 3.1 Active Ingredient and Targets
Screening of HLJDD By screening the TCMSP database, a total of 429 ingredients were retrieved, which are as follow:CR,48;SR,143;PCC,140;GF,98. According to OB and DL, 102 active ingredients were obtained, including 14 of CR, 36 of SR, 37 of PCC, and 15 of GF(supplementary Table1). In addition, a total of 248 corresponding targets were obtained.Finally,a total of 222 gene names corresponding to the targets were found in the STRING database(supplementary Table2).

The Related Target of Atherosclerosis
Through screening the GeneCards and OMIM databases, we obtained in total of 3812 related targets of atherosclerosis(supplementary Table 3).

The Potential Targets Screening HLJDD for Anti-atherosclerosis
The R 3.6.1 software was used to obtain the intersection of the targets of active ingredient of HLJDD and the related targets of atherosclerosis, and finally 175 common targets were screened, as shown in Figure 1. The blue part of picture stands for the targets of active ingredient of HLJDD, and the yellow one represents the related targets of atherosclerosis, the purple one stands for the common targets, which means the potential targets of HLJDD for anti-atherosclerosis.

Figure1. Vene figure of Screening Potential Targets of HLJDD for Anti-atherosclerosis 3.4 The regulatory network of "HLJDD-active ingredient-potential target"
Based on the established file of "HLJDD-active ingredient-potential target", we constructed a regulatory network, as shown in Figure 2. The regulatory network contains 4 Chinese herbal medicine nodes (yellow nodes), 65 active ingredient nodes (green nodes), 175 target nodes (blue nodes), 78 associations of "Chinese herbal medicine-active ingredient" and 1262 associations of "active ingredient-potential target". The 65 active ingredients may play an anti-atherosclerotic role by acting on 175 targets. A topological analysis of the regulatory network shows that the average degree value of the active ingredients was 19.42, and there were 14 active ingredients with degree value ≥ 20, as shown in Table 1. These active ingredients play an important role in the network, suggesting that they may be the key active ingredients of HLJDD for anti-atherosclerosis.

PPI network of the key targets of HLJDD for anti-atherosclerosis
The gene names of 175 potential targets were entered into the online database STRING11.0 for PPI network analysis, and build a PPI network,as shown in supplementary Figure1. The analysis of the network shows that the network consists of 175 target nodes and 3369 associations and the average node degree of network is 38.05. The results are exported in text format. R 3.6.0 software is used for statistics, and targets with node degree > 38.5 in the network are screened out. Finally, 65 targets were obtained, and the node connection degrees of each target node were shown in Figure 3. These targets play a key role in the PPI network, indicating that they may be the key targets of HLJDD for anti-atherosclerosis. Then we performed the same operation for the PPI network analysis of 65 key targets, using the Cytoscape software to visualize the results of PPI network,as shown in Figure 4. The larger the node in the figure, the greater the node connection degree, and the more targets associated with it.

Figure4.PPI Network of Key Targets of HLJDD for Anti-atherosclerosis 3.6 GO Analysis And KEGG Pathway Enrichment Analysis
The gene names of 65 key targets were entered into the database STRING11.0, and the analysis function provided by it was used for GO analysis, mainly involving biological process (BP).The results indicated that 65 key targets mainly involved 1451 BP, such as response to organic substance, response to chemical stimulus, response to cytokines, response to oxygen-containing compound, positive regulation of cellular process, regulation of cell population proliferation, and response to lipids.KEGG pathway enrichment analysis indicated that HLJDD may act on 157 signaling pathways including pathways in cancer,IL-17 signaling pathway,AGE-RAGE signaling pathway in diabetic complications, fluid shear stress and atherosclerosis, and TNF signaling pathway,ect. R 3.6.0 software was applied to visualize the first 20 GO terms and KEGG pathways, and the results were shown in figure 5 and figure 6. The results of enrichment analysis indicate that HLJDD may play a synergistic role in anti-atherosclerosis by regulating multiple biological processes and signaling pathways.

Figure6.KEGG Pathway Enrichment Analysis of Key Targets of HLJDD for
Anti-atherosclerosis 3.7 HLJDD-Key Active Ingredients-Key Targets-Pathway Network 4 Chinese herbal medicines, 14 key active ingredients, 65 key targets, and the pathways of the top 20 KEGG enrichment pathways with enrichment targets > 20 were included in the network construction, and the targets unassociated with the key active ingredients and pathways were deleted, as shown in Figure 7. The network consists of 4 Chinese herbal medicine nodes(yellow prismatic nodes), 14 key active ingredient nodes (green triangle nodes), 52 key target nodes (pink oval nodes), and 10 pathway nodes (blue hexagon nodes). The analysis of the network shows that among the 14 key active ingredients, quercetin is the node with the highest connection degree, with 47 targets connected, and Isocorypalmine and (S) -Canadine are the nodes with the lowest connection degree, with 2 targets respectively. Among the 52 targets, the top 3 with the highest node connection degree are PTGS2, HSP90AA1, and RELA. PTGS2 is connected to 14 ingredient nodes and 4 pathway nodes, and HSP90AA1 is connected to 11 ingredient nodes and 4 pathway nodes, while RELA is connected to 5 ingredient nodes and 9 pathway nodes.

Figure7.HLJDD-Key Active Ingredients-Key Targets-Pathway Network of HLJDD for Anti-atherosclerosis 4 Discussion
In the field of TCM, HLJDD is a typical representative of heat-clearing and detoxifying prescription, which is often used to treat diseases caused by fire evil (TCM divides the pathogenic factors into six types, namely wind, cold, heat, wet, dry and fire). From the perspective of TCM, the incidence of atherosclerosis is related to the fire evil [24] . Therefore, HLJDD may have an anti-atherosclerotic effect, which has been confirmed in previous studies. In this study, we finally identified 14 key active ingredients, 65 key targets and 20 signaling pathways of HLJDD for anti-atherosclerosis, and constructed the pharmacological network of HLJDD for anti-atherosclerosis.The 14 key active ingredients are quercetin, beta-sitosterol, stigmasterol, kaempferol, wogonin, palmatine, baicalein,berberine, coptisine, isocorypalmine, acacetin, (S) -Canadine, berberrubine, oroxylin a, and they have corresponding relationship with 52 of 65 key targets, and may play an anti-atherosclerotic role by affecting the main 10 related signal pathways,which include Pathways in cancer, Kaposi's sarcoma-associated herpesvirus infection, Fluid shear stress and atherosclerosis, PI3K-Akt signaling pathway, IL-17 signaling pathway, AGE-RAGE signaling pathway in diabetic complications, Hepatitis B, Chagas disease (American trypanosomiasis), TNF signaling pathway, Proteoglycans in cancer.
Of the 14 components, some have been shown to have anti-atherosclerosis effects in previous studies.With the most targets, it can be found from the literature that the research on the anti-atherosclerosis effect of quercetin is also the most extensive.Quercetin can improves blood vessel structure by reducing serum lipids, oxidative stress factors, aortic contraction, weight gain and blood pressure, prevents the formation of arterial plaques through antioxidant activity [25] ,and fight atherosclerosis by enhancing autophagy in ApoE-/-mice [26] .Furthermore, in a high fructose-induced atherosclerosis mouse model, quercetin can attenuate atherosclerosis by inhibiting inflammation and apoptosis through the PI3K / AKT signaling pathway regulated by reactive oxygen species [27] .Both beta-sitosterol and stigmasterol belong to Phytosterols, and moderate intake of Phytosterols is considered to reduce plasma cholesterol and prevent atherosclerosis [28] .In addition, Lifsey HC et al reported that stigmasterol can promote the excretion of cholesterol in the small intestine [29] , while Chien MH et al reported that beta-sitosterol can inhibit the proliferation of vascular smooth muscle cells and has a potential role in anti-atherosclerosis [30] .In the rabbit model of atherosclerosis induced by high cholesterol, kaempferol can regulate the expression of inflammatory factors, showing an anti-atherosclerosis effect [31] .Ku SK et al reported that wogonin and baicalein blocked hyperglycaemic vascular inflammation by inhibiting NF-κB in human primary endothelial cell models, suggesting that they had a therapeutic effect on atherosclerosis [32] .Shi Y et al reported that berberine can delay the progression of atherosclerosis in ApoE-/-mice induced by high-fat diet and reduce the expression of inflammatory factors, which may be related to the regulated intestinal flora [33] .Other studies have shown that berberine down-regulates visfatin expression to inhibit endothelial dysfunction and improve diet-induced atherosclerosis in ApoE-/-mice [34] .Feng M et al reported that in the atherosclerosis model of ApoE-/-mice, the intervention of coptisine can reduce the level of blood lipid and down-regulate the expression of relevant inflammatory cytokines, thus reducing the area of atherosclerotic plaque [35] .Although no previous studies have directly shown that palmatine, berberrubine, acetacetin, oroxylin a paly a role of anti-atherosclerosis, studies have shown that they all have different degrees of effects of anti-inflammatory, lipid-lowering, and anti-oxidation,et al.Berberrubine is the main metabolite of berberine in vivo. Studies have shown that it can reduce blood lipid ,inhibit the activity of intracellular xanthine oxidase, and has a potential role in treating endothelial dysfunction [36][37] .Palmatine and oroxylin a both have anti-inflammatory effects,and the former can reduce blood lipid levels, while the latter can also inhibit intracellular lipid accumulation [38][39] .
As a flavonoid,Liou CJ et al reported that acacetin can reduce the expression of transcription factors related to adipogenesis, inhibit lipid accumulation in adipocytes, and inhibit the inflammatory response of macrophages stimulated by 3T3-L1 cell differentiation medium [40] ,while Wu WY et al show that it can participate in anti-oxidation, anti-inflammatory and anti-apoptosis by activating AMPK-mediated Nrf2 pathway, which will play its protective effect on myocardial cell hypoxia / reoxygenation injury [41] .Based on the above research status, we speculate that the above chemical components are most likely related medicinal substances of HLJDD against atherosclerosis.
In terms of targets, in the final constructed pharmacological network, the top 3 nodes with the highest node degrees are PTGS2, HSP90AA1, and RELA.PTGS2 is connected to 14 component nodes and 4 pathway nodes, HSP90AA1 is connected to 11 component nodes and 4 pathway nodes, and RELA is connected to 5 component nodes and 9 pathway nodes.The PTGS2 gene encodes cyclooxygenase-2 (COX-2), and previous studies have shown that COX-2 is widely expressed in human atherosclerosis by monocytes, macrophages, medial smooth muscle cells, and endothelial cells [42][43] , which suggests that COX-2 may be involved in the formation and development of atherosclerosis.Atherosclerosis is a chronic inflammatory response process. In inflammatory response,COX-2 expression can be rapidly induced by a variety of factors, such as IL-1 and TGF-β, while cox-2-induced prostaglandin production can promote inflammatory response of atherosclerosis.The activation of macrophages is one of the key mechanisms for the formation and development of atherosclerosis, and this process is related to the high expression of COX-2. The expression of COX-2 can promote the release and activation of matrix metalloproteinases,which play an important role in the migration of macrophages.Low density lipoprotein stimulates monocyte chemotaxis through the cox-dependent pathway.In general, Cox-2 expression induces prostaglandin production through activation of chemotaxis, induction of vascular permeability, cascade propagation of inflammatory cytokines, and stimulation of smooth muscle cell migration and proliferation,et al, thus promoting the formation and development of atherosclerosis [44] .HSP90AA1 is a subtype of the heat shock protein HSP90, and the overexpression of HSP90 in atherosclerotic plaques plays an important role in the maintenance of inflammation, which is related to the activation of nuclear factor-κB (NF-κB) pathway. HSP90 folds and activates IκB kinase,which phosphorylates and induces the degradation of the proteasome inhibitor protein IκBα, so that the nuclear factor-κB p65 subunit enters the nucleus, promoting the inflammatory response and promoting the transcription of oxidative genes [45] .The RELA gene encodes Transcription factor p65, a subunit of the transcription factor family NF-κB, which is a key regulator of eukaryotic inflammatory responses.Previous studies have shown that NF-κB is closely related to the formation and development of atherosclerosis,which induces the expression of adhesion proteins, monocyte chemotactic protein, IL-1βand other proinflammatory molecules in vascular endothelial cells to promote vascular inflammation [46] .The post-translational modification of RelA / p65 can finely regulate the transcriptional activation of NF-κB and play an important role in the occurrence and development of inflammatory response and diseases related to inflammatory response [47] .Furthermore, many previous studies have shown that other targets are closely related to the pathogenesis of atherosclerosis. For example, Interleukin-6, Interleukin-1βand other pro-inflammatory factors are involved in the formation and development of atherosclerosis [48][49] ,and the effects of matrix metalloproteinases such as MMP-1, MMP-3, and MMP-9 are closely related to the instability of atherosclerotic plaques [50] .All these previous studies provided certain theoretical support for our study. Therefore, we speculated that HLJDD might play an anti-atherosclerosis role by acting on 52 targets including PTGS2, HSP90AA1, RELA, IL-6, IL-1, MMP-1, MMP-3, and MMP-9.
In terms of pathways, KEGG pathway enrichment analysis results show that 65 key targets are also significantly enriched in diseases-related pathways such as herpes virus infection, hepatitis B, cancer, diabetes, and Chagas disease, suggesting that there may be correlations and effects between different diseases, Which has also been verified in previous research.Studies have shown that the hepatitis B core antibody (anti-HBc) positive / hepatitis B surface antigen (HBsAg) negative pattern is associated with intracranial atherosclerotic stenosis in hepatitis B virus infection [51] .In virologically suppressed HIV patients, human herpesvirus 8 (HHV-8) infection increases inflammation and activates immunity, increasing the risk of atherosclerosis [52] .With common risk factors, cancer and atherosclerosis have the same basis as metabolic or vascular abnormalities,and the theory of inflammation is the common pathological basis of both them [53] .Chronic inflammation caused by Chagas disease may trigger endothelial damage and platelet activation, accelerating atherosclerosis [54] .In the progression of atherosclerosis, the PI3K-Akt signaling pathway plays an important role in the survival, proliferation and migration of macrophages [55] ,which suggesting that HLJDD may prevent atherosclerosis by regulating PI3K-Akt signaling pathway and inhibiting the related effects of macrophages.Signaling pathways such as IL-17 signaling pathway and TNF signaling pathway are closely related to inflammatory response. Activation of IL-17 signaling pathway and TNF signaling pathway can activate downstream NF-κ B signaling pathways, leading to the production of various mediators associated with atherosclerosis, including inflammatory factors ,chemokines and others [56][57] . It is suggested that HLJDD may inhibit the production of inflammatory mediators by affecting the activation of il-17 signaling pathway, TNF signaling pathway and its downstream pathways, thus playing an anti-atherosclerosis role.Atherosclerosis is a common complication of diabetes. Activation of AGE-RAGE signaling pathway in diabetic complications participates in the formation and development of atherosclerosis. Advanced glycation endproducts (AGEs) binds to its receptor (RAGE) to activate a variety of intracellular signaling pathways such as NADPH oxidase, which eventually activates NF-κB and leads to the expression of pro-inflammatory factors such as IL-6 and TNF-α, leading to atherosclerosis [58] , suggesting that HLJDD may prevent atherosclerosis by affecting the activation of AGE-RAGE signaling pathway in diabetic or its' downstream pathways and inhibiting the production of inflammatory factors.Hemodynamics is an important factor affecting atherosclerosis. Stable blood flow can prevent atherosclerosis by inhibiting oxidative stress and inflammation in the arterial wall,while disordered blood flow can promote oxidative stress and inflammatory response in the arterial wall to induce atherosclerosis, of which the mechanism is related to activation of endothelial cell activator protein 1 (AP-1) and nuclear factor kappaB (NF-κB) [59] , suggesting that HLJDD may inhibit the oxidation and inflammation of the arterial wall by improving hemodynamics and regulating the fluid shear stress and atherosclerosis signal pathway.
In summary,through the application of network pharmacology, we finally predicted 14 key active ingredients, 65 key targets and 10 pathways that may be related to the anti-atherosclerosis activity of HLJDD.HLJDD may affect 52 key targets such as PTGS2, HSP90AA1, and RELA through 14 key active ingredients such as quercetin, wogonin, and baicalein, and regulate Fluid shear stress and atherosclerosis, PI3K-Akt signaling pathway, IL-17 signaling pathway, AGE-RAGE signaling pathway in diabetic complications, TNF signaling pathway and other related signaling pathways, and then by inhibiting inflammatory response and oxidative stress, improving hemodynamics and other effects to prevent atherosclerosis.However, our results are only a prediction based on the results of previous studies, which need to be further verified by experimental research.

Data Availability
The data used to support the findings are included in the article and the supplementary files.