Network Pharmacology and Molecular Docking Analyses of the Syner⁃gistic Mechanis

TAN Jingzhuang ,HE Yanbin ,HUANG Bin ,LIN Jiumao*

1 Academy of Integrative Medicine,Fujian University of Traditional Chinese Medicine,Fuzhou,Fujian 350122,China;

2 Fujian Key Laboratory of Integrative Medicine on Geriatrics,Fuzhou,Fujian 350122,China;

3 Key Laboratory of Integrative Medicine of Fujian Province University,Fuzhou,Fujian,350122,China

ABSTRACT Objective:To further explore the mechanism of Babao Dan (BBD) combined with oxaliplatin(L-OHP) in treating colorectal cancer (CRC) through a network pharmacology analysis.Methods:The analysis involved the following steps:screen the chemical components of BBD through literature review of Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP),Chemistry Database,Pub‐Chem,and other databases;obtain L-OHP-related targets through GeneCards database;and search CRC-related targets through OMIM,GeneMap,TTD,DisGeNET,CTD,GeneCards,and other databases.After the intersection and mapping of drugs and disease,the protein-protein interaction(PPI)network and core targets were obtained us‐ing STRING database and CytoScape software.MetaScape database was used to analyze the core targets to ob‐tain GO biological processes and KEGG pathways.Results:BBD contained 495 chemical components with 204 active components screened out through the Swiss ADME database and 770 targets were obtained through the Swiss Target Prediction database.After the intersection of BBD and 775 targets of L-OHP with the CRC targets,it resulted in 74 potential targets.Twenty-four core targets were determined from the 74 intersection targets,which were related to the positive regulation of kinase activity,the positive regulation of cell migration,and pepti‐dyl-serine modification in GO biological process.The KEGG pathway analysis showed that the core targets were related to pathway in cancer,proteoglycan in cancer,endocrine resistance,and microRNA in cancer,TNF signal‐ing pathway,platinum resistance,and other pathways.Molecular docking showed that the core targets could bind to the most examined compounds.Conclusion:Quercetin-7-olate,cyclo (L-tyrosyl-L-phenylalanyl),panaxadiol,and other compounds in BBD may play an anti-colorectal cancer effect in multiple pathways,including EGFR,AKT1,mTOR,and other targets in synergy with L-OHP.

KEY WORDS colorectal cancer;Babao Dan;oxaliplatin;network pharmacology;molecular docking

1 Introduction

Colorectal cancer (CRC) is one of the most com‐mon malignancies,with the third highest incidence rate and second highest mortality rate among cancer worldwide［1］.The 5-year overall survival (OS) rate of stage Ⅰand ⅡCRC patients is 99% and 70%,re‐spectively,while that of stage Ⅲpatients is only 55%to 65%［1-2］.CRC is mainly treated by surgery,but the surgery often needs to be combined with chemothera‐py,radiotherapy,and other therapies to improve the curative effect.Oxaliplatin (L-OHP) is a commonly used drug for CRC chemotherapy,but is shows ad‐verse effects,such as toxicity and drug resistance［3］.

Traditional Chinese medicine (TCM) has the char‐acteristics of multi-component,multi-target,and high safety in treating diseases.The current research re‐sults have shown that TCM offers significant antitu‐mor clinical efficacy.It can stimulate the body's im‐mune function and increase the serum immunoglobu‐lin level,prolonging the survival of cancer patients.Further,it can have a synergistic effect on chemother‐apy drugs and reduce side effects［4-5］.

Babao Dan (BBD) is a well-known traditional Chi‐nese patent medicine.Its formula has been derived from the secret recipe of the court in the JiaJing period of the Ming Dynasty,which includesCalculus Bovis,Fel Serpentis,Cornu Saigae Tataricae,Margarita,Moschus,andRadix Notogiseng,as well as two un‐published Chinese herbs.These herbs act by "clearing heat from liver and gallbladder by removing toxins","removing stasis and eliminating edema",and "acti‐vating blood and relieving pain".Recently,its clinical application was expanded from hepatobiliary diseases and surgical applications to other diseases such as can‐cer［6-9］.The treatment using BBD combined with plati‐num chemotherapy showed an antitumor effect,im‐proved chemotherapy sensitivity,and reduced drug re‐sistance in the lung and gastric cancers［10-11］.Some clinical studies also indicated that BBD could exert an antitumor effect in the CRC［12］.Although experimen‐tal studies combining BBD with chemotherapy in oth‐er cancers have been carried out［10-11］,the mechanisms of BBD actions in the CRC are yet to be elucidated.

Therefore,this study aimed to analyze the activity of BBD combined with L-OHP in the treatment of CRC through a network pharmacology analysis to pre‐dict the target and possible signaling pathway,identi‐fy the mechanism of BBD action when combined with L-OHP–based chemotherapy in the CRC,and provide reference for the clinical application of BBD.The study workflow is shown in Figure 1.

Figure 1 Workflow for network pharmacology-based prediction

2 Materials and methods

2.1 Data preparation

Data for this study were obtained from the follow‐ing databases:TCMSP(Traditional Chinese Medicine Systems Pharmacology Database and Analysis Plat‐form;tcmspw.com/tcmsp.php),Chemistry Database(www.organchem.csdb.cn/scdb/default.htm),Pub‐Chem (pubchem.ncbi.nlm.nih.gov/),Swiss ADME(pharmacodynamic database of compounds;http://www.swissadme.ch/),Swiss Target Prediction (target prediction database;www.swisstargetprediction.ch/),OMIM (Online Mendelian Inheritance in Man;omim.org/about),TTD (Therapeutic Target Database;db.idrblab.net/ttd/),GeneCards (The Human Gene Data‐base;www.genecards.org/),DisGeNET (www.dis‐genet.org/home/),UniProt (Universal Protein Re‐source Database;www.uniprot.org/uniprot/),STRING(Protein– Protein Interaction Networks Database;string-db.org/),MetaScape (Gene Annotation and Analysis Database;https://me-tascape.org/gp/index.ht‐ml#/main/step1),RCSB PDB (https://www.rcsb.org/),VENNY2.1.0 (https://bioinfogp.cnb.csic.es/tools/ven‐ny/),CytoScape 3.8.1 software,Bioinformatics (On‐line drawing software;http://www.bioinformatics.com.cn/),PyMOL software,and AutoDock 4.2.6.

2.2 Screening of chemical components and target prediction of compounds

The L-OHP–related target (Target set A) was ob‐tained by searching"oxaliplatin"in the GeneCards da‐tabase (with a relevance score greater than 1).The chemical components of BBD were searched with Niu-Huang(Calculus Bovis),LingYangJiao(Cornu Saigae Tataricae),SheDan (Fel Serpentis liquidum),Zhen‐Zhu(Margarita),SheXiang(Moschus),and SanQi(Ra‐dix Notoginseng).CAS numbers and chemical SMILE structure formulas were obtained from the PubChem database［13］.

The chemical SMILE structure formulas of BBD chemical components were entered into the Swiss AD‐ME database.Lipinski,ghose,Veber,Egan,and mueg‐ge compounds were found to have high gastrointesti‐nal absorption and drug-likeness.The effective com‐pounds predicted by Swiss Target Prediction and the BBD-related target(Target set B)were identified usingP-value>0.1.

2.3 Screening of disease targets and mapping with drugs targets

The disease targets were obtained by searching"colorectal cancer","colorectal carcinoma",and "colorectal neoplasms"as keywords in OMIM genemap,TTD,DisGeNET,CTD,GeneCards.DisGeNET,based on aGDA score greater than 0.5,and the medium of the target from CTD and Gene Cards repeated to not less than 1 000 targets was used as screening criteria.All CRC target genes were deduplicated,and after Uni‐Prot correction,the species selected as "Homo sapi‐ens" and the verified genes (reviewed) were selected to obtain the CRC-related target(Target set C).Subse‐quently,using VENNY 2.1.0,Target sets A,B,and C were mapped to obtain the intersection targets.

2.4 Protein-protein interaction (PPI) network and topology analyses

The obtained intersection genes were imported into the "Multiple proteins" function of the STRING data‐base.The "Homo sapiens" species was selected to construct the PPI network diagram.The "string.inter‐actions.tsv" data of PPI were imported into the Cyto‐Scape 3.8.1 software,and three topological parame‐ters,including degree,betweenness,and closeness,were selected for screening core targets of BBD plus L-OHP-based chemotherapy used in CRC.

2.5 Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes(KEGG)path⁃way enrichment analyses

Using the MetaScape database set as "Homo sapi‐ens" as the standard for screening,the enriched core targets from biological process (BP),cellular compo‐nent (CC),and molecular function (MF) were ana‐lyzed further via the KEGG pathway analysis.Finally,GO analysis results were presented as a bar graph and KEGG analysis results as a bubble graph to construct a compound–target–pathway network.

2.6 Molecular docking

The protein target for molecular docking was se‐lected based on CytoScape topology parameters and the number of source compounds.The PDB format file of the 3D structure of these targets was download‐ed from the PDB database.The SDF format of the 2D structure of the compounds was downloaded from the PubChem database,and the unique ligands selected by PyMOL were saved in the PDB format.The molec‐ular docking between the target and the selected com‐pound was conducted using the “local search parame‐ters” method with the AutoDock software.The binding energy(BE)was used to evaluate the binding ability.Finally,PyMOL was used to visualize the results.

3 Results

3.1 Intersection target of BBD and L-OHP for synergistic intervention of CRC

The six TCMs,including NiuHuang,LingYangJiao,SheDan,ZhenZhu,SheXiang,and SanQi,selected through database and literature searching,contained 495 chemical components.The chemical SMILE struc‐ture formula of chemical components predicted by the Swiss ADME database and 204 active components se‐lected according to the standard are summarized in Ta‐ble 1.The chemical SMILE structure formula of 204 active components was analyzed by the Swiss Target Prediction database,and 770 targets(target set B)with theP-value exceeding 0.1 as the standard were ob‐tained.The 775 targets of L-OHP (Target set A) ob‐tained from the Gene Cards database were entered in‐to VENNY software to draw a Venn diagram with Tar‐get set B of BBD and Target set C of CRC,as shown in Figure 2.The 74 targets intersected by the three tar‐get sets were considered potential targets for the BBD plus L-OHP-based chemotherapy in the CRC treat‐ment.

3.2 Protein interaction network and core genes

To further explore the interaction mechanism of in‐tersection targets,74 intersection targets were input‐ted into the STRING database through the "multiple proteins" function.Setting the species as "Homo sapi‐ens",a protein interaction network (PPI) composed of 74 nodes and 1,115 edges was obtained (Figure 3).The network data were saved in the"TSV"format and inputted into the CytoScape software to analyze for core genes.

Figure 2 Venn diagram of the intersection targets of BBD,L-OHP,and CRC

The"Analyze Network"function of CytoScape soft‐ware was used to calculate three topological parame‐ters:degree,betweenness,and closeness.With the de‐gree of 30.135 14,betweenness of 0.008 355 959,and closeness of 0.637 690 391 as the screening criteria,24-core targets were obtained:EGFR,VEGFA,HSP90-AA1,CASP3,AKT1,CCND1,HRAS,STAT3,SRC,mTOR,ERBB2,IL-6,MAPK1,Bcl-2L1,MDM2,PTGS2,KDR,PIK3CA,MMP9,MAPK8,SIRT1,IGF1R,EZH2,and PARP1,listed from large to small sequence,are summarized in Table 2.These 24-core targets,using CytoScape software combined with the compound of intersection target,were constructed in‐to a "Babao Dan-Chinese medicine-compound-targetcolorectal cancer" network diagram,as shown in Figure 4.

Table 1 Basic information of selected compounds

Figure 4 "BBD-Chinese medicine-compound-target-colorectal cancer"network diagram

Table 2 Frequency of association between compounds with core targets and Chinese medicine

Figure 3 PPI network.

The red hexagons stand for Babao Dan,the orange triangles for Chinese medicine,the green squares for compounds,the yellow squares for common com‐pounds,the blue ellipses for the intersection targets,the indigo diamonds for the core target,and the pur‐ple hexagons for colorectal cancer.

3.3 GO enrichment and KEGG pathway enrich⁃ment analyses

To explore the enrichment of 24-core genes in BP,CC,MF,and KEGG pathways,the study data were ana-lyzed using MetaScape database.As shown in Figure 5,the BP was enriched in positive regulation of kinase activity,response to inorganic substances,positive regulation of cell migration,peptidyl-serine modification,response to radiation,cellular response to organonitrogen compound,regulation of cellular re‐sponse to stress,wound healing,and other functions.The CC was distributed in the cytoplasm's nuclear en‐velope,membrane raft,and perinuclear region.The MF was related to functions such as protein binding and activity.

Figure 5 GO enrichment analysis bar chart

The KEGG enrichment analysis is shown in Figure 6.The results showed that the KEGG pathway was in‐volved in cancer pathways,proteoglycans,microR‐NAs,endocrine resistance,platinum drug resistance,TNF signaling,and other pathways involving the blad‐der,colorectal,gastric,lungs,breast,pancreatic,and other tumors.

Figure 6 KEGG enrichment analysis bubble chart

3.4 Construction of "Compound-Core Target-Pathway"network

Based on the KEGG enrichment analysis results,"pathways in cancer" were selected that had the greatest correlation with the core targets to analyze the re‐lationship between the core targets and the signaling pathways in cancer,as shown in Figure 7.The green box marks the core target of the compound in the can‐cer pathway.BBD may regulate 19 targets through the MAPK,Jak/STAT,VEGF,P13K/AKT,and mTOR signaling pathways and thus participate in tumor pro‐liferation,evading apoptosis,sustained angiogenesis,and other processes.According to the "compoundcore target" and the tumor-related KEGG pathways,a"Compound-Target-Pathway" network diagram con‐taining 109 compounds,24-core targets,and 12 path‐ways was drawn(Figure 8).

Figure 7 Core target marking map in"pathways in cancer"

Figure 8 The"Compound-Target-Pathway"network diagram

3.5 Molecular docking of BBD compound with a core target

To explore the binding ability of the core targets and their source compounds,the three core targets,EGFR,AKT1,and mTOR,with the largest quantity of compounds among the top 10 core targets were ranked by degree value and were used for docking tar‐gets.For the 3 core targets derived from 25 com‐pounds,36 groups for molecular docking and their binding energies are tabulated in Table 3,which indi‐cates that the core targets had a certain binding ability with the compound,and selective binding models are shown in Figure 9.

Figure 9 Docking diagram of selected compounds and target-protein molecules

Table 3 Selected BBD compounds and core-target molecules docking binding energy（kJ/mol）

4 Discussion

CRC as the third most common cancer worldwide has significant adverse effects on health and function.According to the 2020 cancer survey statistics,1 931 590 cases of CRC were diagnosed globally,of whom 935 173 died;this is related to the rapid population growth and changes in dietary structure［14-15］.The mechanism of tumorigenesis is complex,involving a dynamic process of multiple factors,stages,and genes.Chemotherapy remains one of the three con‐ventional methods for CRC treatment to increase the tumor resection rate and reducethe risk of recurrence of high-risk stages Ⅱand ⅢCRC.However,chemo‐therapy can cause more than 500 adverse reactions,in‐cluding bone marrow suppression,digestive system reactions,cardiotoxicity,stomatitis,and drug extrava‐sation-caused phlebitis or severe tissue necrosis［16］.

In TCM,CRC belongs to the categories of "chronic dysentery","intestinal regurgitation",and "intestinal seclusion".The TCM theory believes that cancer is mostly due to Qi deficiency,and external pathogenic invasion that cause functional disorders of the Zang-Fu organs,imbalance of Qi,blood,and body fluid,which result in pathological changes such as Qi stag‐nation,blood stasis,phlegm-turbid retention,and heatpoison［17］.NiuHuang,LingYangJiao,SheDan,ZhenZhu,SheXiang,and SanQi in BBD have the effects of clearing heat and removing toxins,activating blood,and relieving pain.The research results has shown that BBD protects liver and kidney functions,reduces transaminase,has anti-inflammatory properties,clears heat from the gallbladder,relieves icterus,and thus promotes bilirubin metabolism and antitumor activi‐ty［8］.Moreover,experimental studies have shown that BBD could inhibit the proliferation of tumor cells and promote apoptosis to some extent.However,a rele‐vant research basis for BBD plus L-OHP to reduce the toxic and side effects of CRC is still missing.

In this study,we screened chemical components and predicted targets of the six herbs in BBD using network pharmacology and intersection mapping with the targets of L-OHP and CRC.We then constructed a BBD compound–target–CRC network diagram and conducted the core target enrichment analysis to ex‐plore the mechanism of BBD plus L-OHP in the CRC treatment.From the Venn diagram of BBD and LOHP intervention for CRC,we identified 74 targets of intersection.The network topology analysis and en‐richment analysis showed 24-core targets of the 74 in‐tersection targets;24 CRC-related pathways had veri‐fied the features of the "multi-component","multi-tar‐get",and"multi-pathway"effects of Chinese medicine.According to the degree of the intersection target and the frequency of association between the compound and Chinese medicine,EGFR was the most important target,followed by AKT1 and mTOR.

Epidermal growth factor receptor (EGFR) is a member of the receptor tyrosine kinase family and is the activation target of essential signal pathways such as MAPK,Jak/STAT,and PI3K/AKT to promote tu‐mor cell growth.Proliferation,differentiation,inva‐sion,metastasis,inhibition of apoptosis,angiogenesis,and internal environment stability are critical factors in the occurrence and progression of tumors［18］.The EGFR is derived from 18 compounds that play impor‐tant roles in BBD,such as quercetin,ginsengtriol,nar‐cotine,morin,tyrosine,phenylalanine,and other com‐pounds.Studies have shown that as the degree of dif‐ferentiation of colon cancer decreases,the positive ex‐pression rate of EGFR gradually increases.Moreover,the positive expression rate of EGFR with lymph node metastasis was higher than that without lymph node metastasis.It is believed that EGFR is highly ex‐pressed in CRC tissues;its expression level was con‐sistent with and closely related to the clinicopathologi‐cal characteristics of patients［19］.AKT (RAC-alpha serine/threonine-protein kinase) is a serine/threonineprotein kinase,one of the three subtypes of protein ki‐nase B,the main target downstream of the PI3K/AKT signaling pathway［20］.It could promote the growth,proliferation,and differentiation of tumor cells and participate in the regulation of tumor angiogenesis.The expression of AKT protein is closely related to the depth of tumor invasion,lymph node metastasis,vascular invasion,and clinicopathological stage［21］.The mammalian target of rapamycin (mTOR),an atypical serine/threonine-protein kinase,belonging to the phosphatidylinositol 3-kinase (PI3K)–related ki‐nase family is also one of the downstream targets of PI3K/AKT signaling pathway.Its activation could af‐fect the growth,apoptosis,and metabolism of tumor cells and is closely related to the occurrence and de‐velopment of CRC［22］.

GO enrichment analysis showed that the target's BP was related to the function of kinase activity,cell mi‐gration,peptidyl-serine modification,and cellular re‐sponse to stress.KEGG pathway enrichment analysis showed that the core targets in the "Pathways in Can‐cer"were mostly related to tumors associated with the MAPK,Jak/STAT,and PI3K/AKT signaling path‐ways,with EGFR,ERBB2,IGF1R,HRAS,MAPK1,CCND1,and mTOR being during the process of tu‐mor cell proliferation in the MAPK signaling path‐way.EGFR,IGF1R,HRAS,PIK3CA,AKT1,mTOR,Bcl-2L1,and MDM2 were enriched during the evad‐ing apoptosis in the P13K/AKT signaling pathway.The continuous angiogenesis enriched EGFR,IL-6,STAT3,and VEGFA in the Jak/STAT signaling path‐way.The molecular docking results of EGFR,AKT1,and mTOR showed that most compounds could bind the target,indicating that BBD combined with L-OHP had a synergistic effect in the treatment of CRC.

In summary,through intersection mapping,network topology analysis,and enrichment analysis,our study indicated that quercetin-7-olate,cyclo (L-tyrosyl-Lphenylalanyl),panaxadiol,and other compounds in BBD exert an anti-CRC effect via EGFR,AKT1,mTOR,and other multiple pathways and targets syn‐ergic with L-OHP.However,since network pharma‐cology does not include factors such as compound content or target interaction,it is necessary to conduct experimental studies to verify the active ingredients and targets of BBD synergistic with L-OHP［23-24］to further clarify the mechanism of action of BBD plus L-OHP in CRC treatment.