key: cord-1045981-exfoeh2k
authors: Li, Xinrui; Wen, Zishuai; Si, Mingdong; Jia, Yuxin; Liu, Huixian; Zheng, Yuguang; Ma, Donglai
title: Exploration of hanshi zufei prescription for treatment of COVID-19 based on network pharmacology
date: 2022-03-31
journal: Chin Herb Med
DOI: 10.1016/j.chmed.2021.06.006
sha: 99874418b02bd28c1e56d1618ef48693d4018d53
doc_id: 1045981
cord_uid: exfoeh2k

OBJECTIVE: Network pharmacology combines drug and disease targets with biological information networks based on the integrity and systematicness of the interactions between drugs and disease targets. This study aims to explore the molecular basis of Hanshi Zufei formula for treatment of COVID-19 based on network pharmacology and molecular docking techniques. METHODS: Using TCMSP, the chemical constituents and molecular targets of Atractylodis Rhizoma, Citri Reticulatae Pericarpium, Magnoliae Officinalis Cortex, Pogostemonis Herba, Tsaoko Fructus, Ephedrae Herba, Notopterygii Rhizoma et Radix, Zingiberis Rhizoma Recens, and Arecae Semen were investigated. The predicted targets of novel coronavirus were screened using the NCBI and GeneCards databases. To further screen the drug-disease core targets network, the corresponding target proteins were queried using multiple databases (Biogrid, DIP, and HPRD), a protein interaction network graph was constructed, and the network topology was analyzed. The molecular docking studies were also performed between the network’s top 15 compounds and the coronavirus (SARS-CoV-2) 3CL hydrolytic enzyme and angiotensin conversion enzyme II (ACE2). RESULTS: The herb-active ingredient-target network contained nine drugs, 86 compounds, and 49 drug-disease targets. Gene ontology (GO) enrichment analysis resulted in 1566 GO items (P < 0.05), among which 1438 were biological process items, 35 were cell composition items, and 93 were molecular function items. Fourteen signal pathways were obtained by enrichment screening of the KEGG pathway database (P < 0.05). The molecular docking results showed that the affinity of the core active compounds with the SARS-CoV-2 3CL hydrolase was better than for the other compounds. CONCLUSION: Several core compounds can regulate multiple signaling pathways by binding with 3CL hydrolase and ACE2, which might contribute to the treatment of COVID-19.

The 2019 coronavirus disease is caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), which can be transmitted by respiratory droplets, close contact, and aerosols, and is characterized by strong infectiousness, rapid transmission, and rapid disease progression (Qin et al., 2020 ; Chinese preventive medicine association COVID-19 prevention and control expert group, 2020). The National Health Commission lists SARS-CoV-2 as a class B infectious disease and adopts the prevention and control measures for class A infectious diseases (Qin et al., 2020) . In late 2019, COVID-19 broke out around the world. By April 20, 2020, 2 318 457 cases had been confirmed abroad, and the number of confirmed cases increased. The National Health Committee has published seven editions of the COVID-19 medical scheme, including Western medicine treatments with experimental drugs, such as α-interferon, lopinavir, ritonavir, and a series of respiratory and circulatory support therapies. Because the curative effect of this approach is poor and is associated with side effects, the third edition scheme included traditional Chinese medicines (TCMs). Many TCM prescriptions have been applied in COVID-19 patients, and treatments integrating traditional Chinese and Western medicine have been remarkably successful (Dong, Wang, & Li, 2020) .

According to the COVID-19 protocol (trial version 7) released by the National Health Commission, patients with different symptoms will be treated differently. The clinical manifestations were low fever, non-fluttery body heat or no heat, dry cough, less sputum, tiredness, fatigue, chest tightness, vomiting and loose stool, with a tongue coating of slightly light or slightly red, moss of thin white or white greasy, pulse of moisten moisten. The recommended prescription (hereinafter referred to as "Hanshi Zufei prescription") is: Atractylodis Rhizoma (15 g), Citri Reticulatae Pericarpium (10 g), Magnoliae Officinalis Cortex (10 g), Pogostemonis Herba (10 g), Tsaoko Fructus (6 g), Ephedrae Herba (6 g), Notopterygii Rhizoma et Radix (10 g), Zingiberis Rhizoma Recens (10 g), and Arecae Semen (10 g).

This prescription is a modification of the traditional Chinese medicine prescriptions Pingweisan and Dayuanyin, which have been used during disease outbreaks (Wen & Zhang, 2020) . Atractylodis Rhizoma, which is spicy, bitter, and warm, eliminates dampness to strengthen the spleen; Magnoliae Officinalis Cortex can aromatize turbidness, dehumidify and regulate qi; Citri Reticulatae Pericarpium can regulate qi and stomach dryness and dampening up the spleen; Pogostemonis Herba can prevent vomiting; Tsaoko Fructus have anti-retching and anti-evil effects; Ephedrae Herba can alleviate sweat and cold, relieve lung asthma, and subside swelling; Notopterygii Rhizoma et Radix has the effect of dispelling cold, dehumidification, and analgesia; Zingiberis Rhizoma Recens have the effects of dispelling cold and stopping vomiting, phlegm, and cough; Arecae Semen scatters damp evil, dissolves phlegm, and loosens knots. This prescription had been remarkably effective in treating COVID-19 (Peng, He, Sun, Wang, & Huang, 2019; Zhao, Li, & Xie, 2020; Shi et al., 2019) .

Network pharmacology is a systematic study to explore the interactions between active ingredients, targets, and diseases Li & Zhang, 2013; Zhang, Liang, Wang, & Yang, 2019; Zhang, Hong, Chen, Zhou, & Li, 2019) . Here we used network pharmacology and molecular docking techniques to explore the molecular basis for the Hanshi Zufei formula to treat COVID-19.

Using Atractylodis Rhizoma, Citri Reticulatae Pericarpium, Magnoliae Officinalis Cortex, Pogostemonis Herba, Tsaoko Fructus, Ephedrae Herba, Notopterygii Rhizoma et Radix, Zingiberis Rhizoma Recens, and Arecae Semen as the nine TCM keywords, using TCMSP (http://lsp.nwu.edu. cn/TCMSP. PHP) retrieval, the active ingredients and corresponding targets were screened to identify those with oral bioavailability (OB) > 30% and drug-likeness (DL) > 0.18. Based on the NCBI (https://www.ncbi.nlm.nih.gov/) and GeneCards (https://www.genecards.org/) databases, using "novel coronavirus" as keywords, genes associated with COVID-19 were searched (Ru et al., 2014; Zhou, Wang, Xiang, Tong, & Chen, 2019; Zeng, Li, Siu, Jin, & Wu, 2019) .

The corresponding targets of the selected active ingredients were annotated to obtain the gene symbols. The genes associated with COVID-19 in the GeneCards database were modified. The drug target genes were mapped to the disease genes, and common targets were obtained and visualized using a Venn diagram (Wei, Zhou, Niu, Zhang, & Zhao, 2019; Shawky, 2019; Bu, Su, Zou, Meng, Wang, 2019) . The shared drug and disease targets were used to construct the herb-active ingredient-target regulatory network using Cytoscape software (version 3.7.1).

Using the BisoGenet plug-in of Cytoscape (3.7.1), the protein type "Homo sapiens" was selected to construct a protein interaction network. Then, we used the CytoNCA plug-in for network topology analysis. Generally speaking, Degree centrality (DC) and Betweenness centrality (BC) are of great significance to the whole network when they are greater than the median value. To generate a more intuitive drug-disease core targets network and mine the information of biological value, this study selected DC > 61 and BC > 600 as screening criteria (Szklarczyk et al., 2015; Szklarczyk et al., 2017) .

The Bioconductor software package was used to perform enrichment analyses of gene ontology (GO), molecular function (MF), biological process (BP), and cell composition (CC). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were used to identify common targets. Hits with P-values ≤ 0.05 were retained and used to construct the GO histogram and KEGG signal pathway bubble map. Finally, Cytoscape software was used to build the common target-signal pathway network diagram (Singh, Vennila, Snijesh, George, & Sunny, 2016; Li, Guo, & Sun, 2019; Ning et al., 2017) .

We used ZINC (https://http://zinc.docking.org/) to confirm the database name of the compound and its molecular weight while also retrieving the compound's three-dimensional structural formula (mol2 format), then used AutoDock to convert to *PDBQT format. The 3D structure files of SARS-CoV-2 3CL hydrolase and Angiodensin Converting Enzyme II (ACE2) were downloaded from the RSCB PDB database (https://www.rcsb.org/), and then AutoDock 4.2.6 (http: // autodock. Scripps. edu/news/autodock42.6) was used to remove water, and to perform the hydrotreating and other operations. The result was saved as a *PDBQT format file for later use. Ultimately, for the protein analysis, we used the AutoDock docking operation, then used the Pymol compounds dock Zong, Yao, Shan, & Wang, 2020; Zheng, Mo, Wu, Guo, & Bao, 2019) .

The TCMSP database was used to retrieve the active ingredients contained in nine traditional Chinese medicines, and the parameters of OB > 30% and DL > 0.18 were used as the screening conditions. We found that 86 active ingredients were obtained: Arecae Semen (n = 8), Atractylodis Rhizoma (n = 9), Tsaoko Fructus (n = 8), Citri Reticulatae Pericarpium (n = 5), Pogostemonis Herba (n = 11), Magnoliae Officinalis Cortex (n = 2), Ephedrae Herba (n = 23), Notopterygii Rhizoma et Radix (n = 15), and Zingiberis Rhizoma Recens (n = 5) ( Table 1) . A further 228 related targets were obtained by predicting "related targets" and removing duplications. 

A total of 394 potential targets of novel coronavirus were collected through NCBI and GeneCards databases, and 46 duplicates were eliminated. A total of 348 novel coronavirus-related targets were obtained.

The 228 active ingredient targets predicted from the TCMSP database and 348 novel coronavirus disease targets were plotted as a Venn diagram, and 49 overlapping targets were obtained (Fig. 1) , which are the targets related to the active ingredients of traditional Chinese medicine acting on diseases. Cytoscape software was used to construct a herbs-active ingredients-targets network. The network has 91 nodes and 212 edges, among which the light blue rectangular nodes represent overlapping targets (Gene), the triangles represent compounds, and different colours represent different TCMs. Each edge represents the interaction between compounds and targets (Fig. 2) . According to degree ranking, the core active ingredients were quercetin (MOL000098, value = 36), luteolin (MOL000006, value = 18), kaempferol (MOL000422, value = 14), wogonin (MOL000173, value = 14), naringenin (MOL004328, value = 12), nobiletin (MOL005828, value = 11), and irisolidone (MOL005916, value = 10).

The common target information was imported into Cytoscape (3.7.1) software. The network relationship between target proteins was constructed using BisoGenet to obtain a PPI network (Fig. 3) , including 2 148 proteins and 47 330 interrelation channels. The network topology analysis was performed on 2148 proteins, including DC and BC (Fig. 4A ). When the DC value was greater than 61 (Fig. 4B ) and BC value was greater than 600, 65 target proteins and 1006 interaction relationships were obtained (Fig. 4C) . The top six target proteins were NTRK1, CUL3, TP53, FN1, MCM2, and CDK2. 

GO and KEGG enrichment analyses were carried out on 49 common targets. According to the retaining results (P＜0.05), the first 20 items were visualized and analyzed to generate a histogram and bubble graph (Ma et al., 2020) . The vertical coordinates of the bubble diagram indicate the KEGG name, and the horizontal coordinates indicate the gene proportion. The co-acting targets in the biological process (BP) are mainly response to toxic substances, response to the metal ion, and response to oxidative stress (Fig. 5A ). The common targets in the molecular function (MF) are mainly cytokine receptor binding, phosphatase binding, cytokine activity, and other processes (Fig. 5B) . The common targets in cell composition (CC) mainly involve membrane raft, membrane microdomain, and membrane region (Fig. 5C ). Through KEGG pathway enrichment analysis (P < 0.05), the common targets were mainly concentrated in the MAPK signalling pathway, endocrine resistance, Ras signalling pathway, and other relevant signalling pathways (Fig. 5D) . The screened targets and pathways were imported into Cytoscape software to construct the targets-KEGG network diagram (Fig. 6) . Red represents the targets, and yellow represents the KEGG pathway. The results showed that the pathways with many adjacent nodes included the MAPK signalling pathway, endocrine resistance, and Ras signalling pathway, and the targets include MAPK1, MAPK3, BAD, and EGFR Fig. 6 . Target-pathway network.

The molecular docking results showed that the minimum binding energy between the core active ingredient (herbacetin, genkwanin and wogonin) and SARS-CoV-2 3CL hydrolase were below −5 kcal/mol, indicating that they had good binding activity and can form stable binding conformation. Molecular docking of the core compound with ACE2 (PDB ID: 1R42) was performed, and the results were shown in Fig. 7 and Table 2 . 

Since ancient times, TCM has played an important role in the prevention and treatment of epidemic infectious diseases, such as smallpox, plague, meningitis, haemorrhagic fever, and swine flu, and many effective methods have been widely used in clinical practice (Huang, Wang, Xu, & Liu, 2020) . In Dayuanyin, Arecae Semen, Tsaoko Fructus, and Magnoliae Officinalis Cortex play a key role in treating plague. Pingwei powder is widely used to dry dampness, promote qi, and regulate the stomach. In the Hanshi Zufei formula, Dayuanyin and Pingwei powder have been combined to dialectically add and reduce the drug flavour, and the drug treatment has a good effect from the comprehensive and holistic concept of traditional Chinese medicine.

This study screened 228 potential drug targets, 49 overlapping targets and 15 core compounds (quercetin, luteolin, kaempferol, wogonin, naringenin, nobiletin, irisolidone, β-sitosterol, pectolinarigenin, diosmetin, genkwanin, taxifolin, (+)-catechin, herbacetin, and (-)-catechin) . Twelve of these are flavonoids, (+)-catechin and (-)-catechin are phenols, and β-sitosterol is a phytosterol . Flavonoids have anti-inflammatory, antibacterial, antiviral, and immunological effects, and these activities might be related to their anti-free radical or anti-oxidation activities (Ma, & Wu, 2013) . According to PPI network topological analysis, the drug-disease core targets network was screened, such as NTRK1, CUL3, TP53, and FN1. Among them, NTRK1 is the medium nerve growth factor for promoting survival signal transduction, and it is also a known oncogene that has been generally changed in human cancer, which can be used as a drug target (Kim et al., 2015) . CUL3 is a ubiquitinated ligase E3 protein, which belongs to the Cullin protein family. Deletion or mutation in animal models has a lethal embryonic phenotype. The TP53 gene encodes a tumour suppressor protein containing transcriptional activation, DNA binding, and oligonucleotide domains, which regulates the expression of target genes in response to various cellular stress responses, thereby inducing cell cycle arrest, apoptosis, senescence, DNA repair, and metabolic changes. The FN1 gene encodes fibronectin, which is involved in cell adhesion and migration, including embryogenesis, wound healing, coagulation, host defence, and metastasis.

According to the GO enrichment analysis results, the main enrichment items in biological processes are a response to toxic substances, response to metal ions, and response to oxidative stress. According to our KEGG enrichment analysis results, the main pathways involved in the treatment process include MAPK signalling, endocrine resistance, and Ras signalling pathways. The targets were MAPK1, MAPK3, BAD, and EGFR. Among them, the BAD gene is an important regulatory molecule of cell apoptosis, and the EGFR gene is a receptor for epithelial growth factor cell proliferation and signal transduction. Mutations or overexpression of these genes will generally cause tumours (He & Xue, 2007; Wang, 2019) . The MAPK signalling pathway regulates a variety of important cellular physiological/pathological processes, including cell growth, differentiation, stress adaptation to the environment, and inflammatory response, and plays a key role in gene expression regulation and cytoplasmic functional activities (Chen, Wang, Wang, & Cao, 2011) . The above targets and pathways play a key role in treating common COVID-19, but further validation studies are needed.

The results of this study were verified by network pharmacology and the guidance of TCM. Our network pharmacology analysis indicates that the 15 prescription compounds with the highest score are all from herbs with warm nature and spicy flavours, such as Pogostemonis Herba, Tsaoko Fructus, and Ephedrae Herba. These core compounds were used for molecular docking with SARS-CoV-2 3CL hydrolase. All of them could bind to SARS-CoV-2 3CL hydrolase, and the binding energies were lower for herbacetin, genkwanin, and wogonin. It has been proposed that both SARS-CoV-2 and SARS-CoV infect humans by binding S-Protein to ACE2 (Tang, Li, Xu, & Shen, 2020) . Therefore, molecular docking of the core compound with ACE2 was conducted in this study, and the binding energies of irisolidone, (-)-catechin, and (+)-catechin were lower than other core compounds. These results suggest that the core compounds [herbacetin, genkwanin, wogonin, irisolidone, (-)-catechin, and (+)-catechin] are potentially useful in the treatment of COVID-19 patients .

Our network pharmacology and molecular docking approaches show that several core compounds (herbacetin, genkwanin, and wogonin) can regulate the MAPK signalling pathway, endocrine resistance, and the Ras signalling pathway by binding with the SARS-CoV-2 3CL hydrolase and ACE2 on multiple targets (MAPK1, MAPK3, BAD, and EGFR), which might play a therapeutic role in COVID-19 patients. In summary, using a theory and network prediction approach, this study describes the roles and mechanisms of the Hanshi Zufei formula on COVID-19 in multiple components, multiple targets, and multiple pathways to provide guidance and ideas for further research.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Study of the mechanism underlying therapeutic effect of Compound Longmaining on myocardial infarction using a network pharmacology-based approach

Progress in MAPK signal pathway research

Thinking of integrated Chinese and western medicine for severe patients with COVID-19

Regulatory effect BAD protein to cell apoptosis

Study on mechanism of Huanglian Jiedu Decoction in treating novel coronavirus pneumonia based on network pharmacology

NTRK1 fusion in glioblastoma multiforme

Traditional Chinese medicine network pharmacology: Theory, methodology and application

Integration of multi-omics data to mine cancer-related gene modules

Virtual screening of small molecular inhibitors of SARS-CoV-2 3CL hydrolase based on high-throughput molecular docking and prediction of Chinese materia medica and its compound against COVID-19

Research progress about pharmacological effect and mechanism of flavonoids in traditional Chinese medicine

Computational molecular networks and network pharmacology

Research progress on network pharmacology in Chinese material medical

Research progress on COVID-19

Tcmsp: A database of systems pharmacology for drug discovery from herbal medicines

Prediction of potential cancer-related molecular targets of north African plants constituents using network pharmacology-based analysis

A network pharmacology approach to investigate the blood enriching mechanism of Danggui Buxue decoction

Implying analytic measures for unravelling rheumatoid arthritis significant proteins through drug-target interaction

STRING v10: protein-protein interaction networks, integrated over the tree of life

The STRING database in 2017: Quality-controlled protein-protein association networks, made broadly accessible

Searching therapeutic strategies of COVID-19 from Angiotensin-converting Enzyme 2: The receptor of 2019-nCoV and SARS-CoV

Study on the network pharmacology and preliminary evidence of Lianhua Qingwen in the treatment of novel coronavirus (2019-nCoV) pneumonia

Research progress of EGFR gene mutation and targeted therapy in non-small cell lung cancer

Exploring the active compounds of Huanglian Jiedu Decoction in the treatment of coronavirus disease 2019(COVID-19) based on network pharmacology and molecular docking method

Network pharmacology exploration reveals the bioactive compounds and molecular mechanisms of li-ru-kang against hyperplasia of mammary gland

A brief discussion on the prevention and treatment of COVID-19 from spleen and stomach based on treatise on pestilence

Mechanism of Qingfeipaidu decoction on COVID-19 based on network pharmacology

Potential material basis of Kangbingdu Granules for treatment of coronavirus disease 2019 (COVID-19) based on network pharmacology and molecular docking technology

A combined molecular biology and network pharmacology approach to investigate the multi-target mechanisms of Chaihu Shugan San on Alzheimer's disease

Effects and mechanisms of Danshen-Shanzha herb-pair for atherosclerosis treatment using network pharmacology and experimental pharmacology

Target identification of active constituents of shen qi wan to treat kidney yang deficiency using computational target fishing and network pharmacology

Application of network pharmacology in research on mechanism of action of TCM compounds

Study on the anti-hyperlipidemia mechanisms and signaling pathways of traditional Chinese medicine Polygonum cuspidatum based on network pharmacology

Network pharmacology analysis of traditional Chinese medicine formula Xiao ke yin shui treating type 2 diabetes mellitus. Evidence Based Complementary & Alternative Medicine

Discussion on the active compounds of Qingxuan Zhike granule in treatment of pediatric coronavirus disease 2019 based on network pharmacology and molecular docking method

This work was supported by the Research Foundation of Administration of Traditional Chinese Medicine of Hebei Province (No. 2018105).