Kun FANG, Zhengjie XU, Suxiao JIANG, Ping ZHANG, Lijuan DUAN

Surgical Department, Yinchuan Maternal and Child Care Hospital, Yinchuan 750001, China

Abstract [Objectives] To collect the main active components and targets of baicalein, and to explore the relationship between their targets and COVID-19 and the treatment mechanism of potential unknown targets. [Methods] The 2D and 3D structures of baicalein (CAS: 491-67-8) were obtained by searching Pubchem. The predicted target of baicalein was obtained through the Pharmapper website, and the species specific target was obtained by SEA search server. The COVID-19 related genes were obtained in Genecards and OMIM. The Draw Venny Diagram online program was used to establish the visualized network map of “drug-disease-target” with baicalein monomer target and disease common target and Cytoscape (3.7.2) software. The COVID-19 gene was constructed by using Cytoscape plug-in Bisogenet, and protein-protein interaction (PPI) network was constructed by using the baicalein COVID-19 gene. The CytoNCA topology was used to analyze the network topology and construct the core network. The core network genes were analyzed by GO and KEGG. [Results] Through screening, a total of 358 effective target genes of baicalein, 23 common targets of baicalein-COVID-19, and 80 genes of PPI core network were obtained. The total of 1 648 GO biological processes including DNA metabolism, positive regulation, intracellular receptor signaling pathway, transforming growth factor β receptor signaling pathway, regulation of apoptotic signaling pathway, RNA polymerase II transcription factor binding, and 115 KEGG-related signaling pathways involving cell cycle pathways, viral oncogenic pathways, Hepatitis B pathway, Hepatitis C pathway, PI3K-Akt pathway, and ubiquitin-mediated proteolysis, etc. [Conclusions] The effect of baicalein on COVID-19 reflects the characteristics of multiple targets, and it can also regulate the occurrence and development of COVID-19 through various channels. This study is expected to provide a certain basis for subsequent experiments.

Key words Network pharmacology, Baicalein, COVID-19, Action mechanism

1 Introduction

The acute novel coronavirus disease (COVID-19) has the characteristics of rapid and widespread transmission, strong contagion, and general susceptibility to all types of people. Mild patients have symptoms such as fever, fatigue, and dry cough. In severe cases, dyspnea, respiratory distress syndrome, or septic shock may occur[1-2]. 2019-nCoV is the seventh coronavirus discovered so far that can infect human beings[3-4], and it can be widely distributed in humans and other mammals[5-6]. 2019-nCoV is a β genus B subgroup coronavirus, and its genetic sequence is at least 70% similar to severe acute respiratory syndrome coronavirus (SARS-CoV)[7-8].

The COVID-19 has the characteristics of rapid and widespread transmission, strong contagion, and general susceptibility to all types of people, and it has a huge negative impact on human health, social and economic development and stability. By now, there is still no targeted effective drug for COVID-19, and the treatment is mainly symptomatic and supportive. How to form a quick and effective treatment plan is a major issue. Traditional Chinese medicine has a good effect on viral pneumonia, and has shown certain effects in the treatment of severe acute respiratory syndrome (SARS). In theChina’sNovelCoronavirusPneumoniaDiagnosisandTreatmentPlanissued and continuously updated by the National Health Commission, it is still recommended to take traditional Chinese medicine for the treatment of COVID-19[9].

Scutellariae Radix is the dried root ofScutellariabaicalensisGeorgi. It has a long history of medicinal use, and it has the functions of scavenging free radicals, anti-oxidation and anti-inflammatory. Baicalein is the flavonoid with the highest content in Scutellariae Radix, and it is also the main active component of Scutellariae Radix. Baicalein has a wide range of pharmacological effects, including anti-inflammatory[10], anti-virus[11], scavenging free radicals[12], anti-oxidation[13], and antipyretic effects. Besides, baicalein also has anti-gastric ulcer, anti-allergic, liver protection and acute kidney injury-preventing activities[14], inhibits the formation of hypertrophic scars[15], and protects the central nervous system from ischemia-reperfusion injury[16-17 ],etc.

Network pharmacology was first proposed by British pharmacologist Hopkins in 2007[18]. It is mainly used in drug discovery and mechanism research to solve the individual differences of single target molecule with poor efficacy, multiple side effects, strong drug resistance and treatment response. With the aid of the network pharmacology, it is possible to comprehensively analyze the components of traditional Chinese medicines, to explore the common targets of drugs and diseases, to analyze the biological functions and signal pathways involved in the targets, and to provide a theoretical basis for clinical trials[19]. In this study, using the network pharmacology methods, we explored the mechanism of baicalein in the treatment of COVID-19 at the molecular level. Through searching the database, we collected the targets contained in baicalein, constructed a drug-target-disease-network, identified important core genes in the network, and performed functional enrichment analysis on core genes to find their association with the COVID-19.

2 Methods

2.1 Single structure search and target acquisition

2.1.1Single structure search. Through searching Pubchem[16](https://pubchem.ncbi.nlm.nih.gov/), we obtained the sdf file, pattern diagram and PubChem_CID of baicalein (CAS: 491-67-8) 2D and 3D structures and SMILES files.

2.1.2Target acquisition. Pharmmapper[17](http://www.lilab-ecust.cn/pharmmapper/): through submitting the compound structure sdf file (the maximum return result was set to 1 000, and the rest of the parameters were set at the default), we obtained compound prediction targets. SEA Search Server (http://sea.bkslab.org/): through submitting SMILES files, and selected targets of human species, we obtained compound prediction targets. Swiss Target Prediction (http://www. swisstarget-prediction.ch/): through submitting SMILES files, we obtained compound prediction targets.

2.2 Acquisition of disease-related genesWe searched for disease-related genes in two databases: Genecards (https://www.genecards.org/) and OMIM (https://omim.org/). The search term was novel coronavirus pneumonia (COVID-19).

2.3 Acquisition of possible targets for single drugs to treat diseasesUsing the Draw Venny Diagram online program (http://bioinformatics.psb.ugent.be/webtools/Venn/), we intersected the Disease and Drug target genes, and obtained the common target of the monomer and the disease. This set was the possible targets for single drugs to treat diseases.

2.4 Drawing of network diagrams of monomers, diseases, and target genesBased on the PPI network data, we constructed the monomer-gene-gene-disease network diagram. With the aid of Cytoscape 3.7.2, we plotted the monomer, disease, and target gene network diagram. The graphics drawing file preparation was shown in 05.PrepareCyto folder, and the network drawing was shown in Cytoscape folder.

2.5 Core network analysisUsing Cytoscape plug-in Bisogenet, we constructed a protein-protein interaction (PPI) network with intersection genes; using CytoNCA plug-in, we performed network topology analysis and core network construction, and first-level sub-network screening: DC>61, second-level sub-network (core network) screening: BC>600, we obtained a total of 80 core network genes. We analyzed these 80 core network genes.

2.5.1Gene Ontology (GO) analysis: using R3.6.0 software, we installed colorspace, stringi, DOSE, clusterProfiler, pathview and ggplot2 and other program packages, and performed GO full classification [Biological Process (BP), Molecular Function (MF) and Cellular Component (CC)] enrichment analysis according to the id.txt file, and obtained bar graphs and bubble chart (each category only displayed the first 10 statistically significant results).

2.5.2Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis: using R3.6.0 software, we installed colorspace, stringi, DOSE, clusterProfiler, pathview and ggplot2 packages, and performed KEGG enrichment analysis based on the id.txt file, andobtained bar graphs, bubble chart and key signal pathway graphs (only displayed the first 20 statistically significant results).

3 Results

3.1 Collection and sorting of target genesThrough Pharmmapper, we obtained 231 compound prediction targets, and through SEA Search Server (http://sea.bkslab.org/), we obtained 67 compound prediction targets; through SwissTarget-Prediction (http://www.swisstargetprediction.ch/), we obtained 99 compound prediction targets. After de-duplication, we obtained 357 effective targets of baicalein. Through the Genecards (https://www.genecards.org/) database, we obtained 252 disease-related genes of the novel coronavirus pneumonia (COVID-19); through the OMIM (https://omim.org/) database, we obtained 2 disease-related genes. After de-duplication and sorting, we obtained 251 target genes of the COVID-19.

3.2 Main active components of baicalein: the common target of COVID-19Using the Draw Venn diagram online program (http://bioinformatics.psb.ugent.be/webtools/Venn/), we selected the intersection of the obtained baicalein active component target gene set and COVID-19 related target gene set, we obtained a total of 23 potential targets of baicalein for the treatment of COVID-19, as shown in Fig.1. In the figure, the left circle is the target of baicalein’s active components, the right is the target gene related to COVID-19, and the middle intersection is the potential target gene of baicalein acting on the COVID-19.

Fig.1 Venn diagram

3.3 Construction of drug - key chemical component-disease networkWith the aid of Cytoscape 3.7.2, we constructed a visual processing diagram of drug-key compound-disease-target gene network (Fig.2).

Fig.2 A network diagram for baicalein extract-key chemical components and COVID-19 targets

3.4 Screening of hub (core) gene network through PPI interaction network analysisWith the aid of Cytoscape plug-in Bisogenet, we constructed a PPI network diagram with intersection genes (Fig.3); with the aid of CytoNCA plug-in, we performed the network topology analysis and core network construction, and obtained a total of 80 core network genes (Fig.4).

Fig.3 PPI network diagram

Note: A: Interactive PPI network of potential target genes of baicalein and COVID-19 related targets; B: Important protein extracted from A by PPI network; C: PPI network for baicalein extracted from B to treat COVID-19 target.Fig.4 Determination of candidate target genes of baicalein for COVID-19

3.5 Biological function and pathway enrichment analysisWe imported the core gene targets of baicalein that may play a role in the COVID-19 into the software R3.6.0 and script calculations, and performed gene ontology (GO) function enrichment analysis (Fig.5) and KEGG pathway enrichment analysis (Fig.6).

Fig.5 GO analysis of baicalein in the treatment of COVID-19

Fig.6 KEGG analysis of baicalein in the treatment of COVID-19

After screening through functional enrichment analysis, we obtained the top 20 pathway results. According to the results, the positive regulation of DNA metabolism, the intracellular receptor signaling pathway, and the transforming growth factor β receptor signaling pathwaywere the most obvious biological processes. These processes involve 17 targets, which may be the most important biological process of baicalein in the treatment of the COVID-19. KEGG pathway enrichment analysis results show that viral carcinogenic pathways, PI3K-Akt pathways, cell cycle pathways, hepatitis B pathways,etc. are the most significant signaling pathways, and may be the most important pathway of Scutellariae Radix for treating depression. Other pathways include prostate cancer pathway, hepatitis C pathway, and ubiquitin-mediated proteolysis pathway.

4 Discussion

Baicalein has an excellent prevention and treatment effect on infectious diseases, and can exert antibacterial[18], antiviral[19], anti-inflammatory[20]and other pharmacological activities. The main mechanism of its antiviral effect[21]includes: (i) directly killing the virus or reducing the virulence of the virus; (ii) preventing the adsorption and invasion of the virus; (iii) inhibiting the transcription and replication of the virus’s DNA or mRNA; (iv) antioxidant effect; (v) regulating the cytokine production. According to the prediction results of this study, ALB, MAPK3, EFGR, and IL12 are the potential main targets of baicalein to resist the COVID-19, which suggests that baicalein can play a protective role in the middle and late stages of the COVID-19.

In order to study the role of targets in gene function and signaling pathways, we performed GO biological function enrichment analysis. It is found that the above-mentioned PPI core genes can play a role in antagonizing the intracellular reverse transcription process of the COVID-19 and resisting viral replication through the positive regulation of the DNA metabolism process, the intracellular receptor signaling pathway, and the transforming growth factor β receptor signaling pathway. KEGG pathway enrichment analysis results show that the above-mentioned PPI core genes are mainly significantly enriched in pathways involving cell cycle pathways, viral carcinogenic pathways, hepatitis B pathways, hepatitis C pathways, PI3K-Akt pathway, and ubiquitin-mediated proteolysis pathways. Among these pathways, the PI3K/Akt pathway regulates a variety of cellular processes, including cell proliferation, RNA processing, protein translation, autophagy and apoptosis[22]. In order to replicate in host cells, the virus has developed a strategy to stimulate the basic PI3K activity[23]. As an adaptive strategy, PI3K can activate IFN-I response and autophagy to counteract virus invasion. However, autophagy may be hijacked by viruses to achieve efficient replication, possibly in a cell type and virus-specific manner[24]. According to our analysis and research, the potential pathways of baicalein acting on the COVID-19 also have the PI3K/Akt pathway, which seems to provide an idea that the PI3K/AKT pathway of baicalein acting on the COVID-19 can activate the IFN-I response and autophagy to counteract the virus, and accordingly using baicalein to intervene in the early stage of virus replication. The binding energy of baicalein to ACEII, a currently known potential target of COVID-19[25], is 35.56 kj/mol. Baicalein has the same binding energy of 2019-nCoV3CL hydrolase as lopinavir and remdesivir[26]. This also provides a basis for the potential feasibility of baicalein against the COVID-19. According to our analysis, the effective targets of baicalein include env integrase and pol packaging protein and other HIV virus originals, indicating that baicalein can inhibit the integration of HIV virus into the human genome to exert its ability to inhibit HIV virus replication. The analysis results are consistent with those of previous studies[27-30].

In summary, we used the method of network pharmacology to explore the potential targets of baicalein on the COVID-19. It is found that baicalein may have the potential value of treating the COVID-19.