*Corresponding Author:
Xue Bai
Department of Neurology,
The Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University,
Luzhou, Sichuan 646000,
E-mail: baixue@swmu.edu.cn
This article was originally published in a special issue,“New Advancements in Biomedical and Pharmaceutical Sciences”
Indian J Pharm Sci 2022:84(2) Spl Issue “180-190”

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms


Based on the theory of “Brain Xuan-fu Blood-Brain Barrier”, we use network pharmacology to study the potential mechanism of wind medicine to regulate the blood-brain barrier. The effective components of six wind medicines were screened using the traditional Chinese medicine systems pharmacology database and analysis platform and relevant targets were identified through this database; potential targets related to blood-brain barrier function were obtained through the GeneCards database; the “drug-component-target-blood-brain barrier” network map was constructed using Cytoscape software. The search tool for the retrieval of interacting genes/proteins database was used to construct a protein-protein interaction network map of target protein interactions and to screen the core targets using Cytoscape software; the database for annotation, visualization and integrated discovery database was used to perform gene ontology and Kyoto encyclopedia of genes and genomes enrichment analysis of the core targets at the drug-disease intersection; Vina was used for molecular docking. Six wind drugs with 43 effective components and 255 potential targets were screened, with a total of 217 targets where the drug intersected with the blood-brain barrier and 68 core targets. The top 20 biological processes obtained were related to pathways such as positive regulation of ribonucleic acid polymerase II promoter transcription, positive regulation of transcription, negative regulation of the apoptotic process, inflammatory response, lipopolysaccharide-mediated signaling pathway and cellular response to hypoxia. The top 20 pathways obtained involved cancer pathways, tumor necrosis factor signaling pathway, T cell receptor signaling pathway, nucleotide-binding oligomerization domain-like receptor signaling pathway, hypoxia-inducible factor 1 signaling pathway, toll-like receptor signaling pathway, etc. The molecular docking results showed that the main effective components bind well to the core targets. The mechanism of action of wind medicine is probably through the mechanism of anti-inflammation and anti-apoptosis to act on the blood-brain barrier, which is of great significance in the treatment of encephalopathy.


Wind medicine, blood-brain barrier, network pharmacology, encephalopathy

The Blood-Brain Barrier (BBB) maintains the homeostasis of the brain’s internal environment and protects the central nervous system from toxins, pathogens, inflammation, etc[1]. When brain diseases occur, its barrier effect is disrupted, on the one hand, the increased permeability will allow harmful substances to enter the central nervous system, etc., and on the other hand, its barrier function will create barriers to drug delivery to the central nervous system[2]. Studies have confirmed that BBB disruption is an important factor and pathological marker in the development of stroke, vascular dementia, Alzheimer’s disease and other central nervous system diseases[3,4]. Therefore,it is important to study the regulation of BBB function for the treatment of central nervous system diseases. The brain Xuan-fu is the manifestation of the Xuan-fu theory in the brain and has been widely used in recent years in the study of stroke, dementia and other brain diseases. Studies have shown that there is a significant correlation between the Xuan-fu of the brain and the BBB in terms of structure, physiological function and pathological characteristics[5]. The main pathogenesis of encephalopathy is the depression of the brain Xuan-fu and the treatment should be to open the Xuan-fu. Professor Mingjie Wang[6] believed that wind medicine is ascending and dispersing in nature, which is good at ascending and dispersing to the surface, can dispel and smooth the qi of the whole body and can draw medicine into the brain, so the use of wind medicine to open the Xuan-fu is a major rule in the treatment of encephalopathy. Studies have proved that the use of wind medicine has a significant regulatory effect on the function of BBB, which can reduce pathological BBB permeability and improve BBB function[7,8].

In this paper, we used the network pharmacology method to screen the effective components of selected wind medicine, predict their potential targets and reveal the network relationship between “wind drugcomponent- target-BBB” to explore the mechanism of action of wind drugs on the BBB.

Materials and Methods


The Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) (https://tcmspw.com/tcmsp.php); Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database (https://string-db.org/); Database for Annotation, Visualization and Integrated Discovery (DAVID) database (https://david. ncifcrf.gov/tools.jsp). GeneCards database (https:// www.genecards.org/); Uniprot database (www. uniprot.org/); Cytoscape software (Version3.7.0); Microbiology Letter online mapping website (http:// www. bioinformatics.com.cn).

Wind medicine screening:

Six wind medicines were selected based on “New knowledge of wind medicine” by Professor Mingjie Wang[9], namely Chuanxiong rhizoma, Ephedrae herba, Bupleuri radix, Asari Radix et Rhizoma, Puerariae lobatae radix, Cinnamomi ramulus.

Screening of effective components and their potential targets:

All effective components of the six wind medicines were retrieved using the TCMSP platform and the effective components were screened by Oral Bioavailability (OB)≥30 %; Drug-Like properties (DL)≥0.18; crossing BBB coefficient≥-3 and Half- Life (HL)≥4. The potential targets of each effective components were then derived from this database and the obtained target protein names were normalized to human gene names in the Uniprot database.

Acquisition of BBB related targets and screening of drug intersection targets with the BBB:

The targets related to the BBB were obtained from the GeneCards database using the keyword "Bloodbrain barrier". The targets related to the effective components that have been normalized to human gene names in the Uniprot database were intersected with the targets related to the BBB to find the targets of the effective components of wind medicine acting on the BBB.

Drug-component-target-BBB network:

The data from screening of effective components and their potential targets and acquisition of BBB related targets and screening of drug intersection targets with the BBB were compiled and uploaded to Cytoscape 3.7.0 software to construct a “drug-componenttarget- BBB” network map for visualization and reconciliation.

Drug-BBB, Protein-Protein Interaction (PPI) network construction and screening of core targets:

The intersection targets of wind medicine effective components and the BBB were entered into the STRING database, and the species was defined as “Homo sapiens” to construct the PPI network of wind medicine action on the BBB. The obtained STRING files in Tab-Separated Values (TSV) format were uploaded to Cytoscape 3.7.0 software for network topology analysis and the core targets were filtered out with the condition that they were larger than the average value of degree.

Functional enrichment and biological pathway analysis:

Enter the core targets into the DAVID database, select "OFFICE_GENE_SYMBOL", "gene list" and select "Homo sapiens" for the species. After data submission, Gene Ontology (GO) bioprocess analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was performed. The top 20 biological processes and KEGG pathways with the smallest p-value were selected with p0.05 as the screening criterion.

Molecular docking:

The top 10 core targets were selected by degree value for molecular docking with the key effective components and the Protein Data Bank (PDB) format files of the targets were obtained in the PDB database (https://www.rcsb.org/), and then the water molecules and small molecule ligands of the target proteins were removed by Pymol software and saved as PDB format files. Download the Two Dimensional (2D) structure file of the effective components from the PubChem database (https://pubchem.ncbi.nlm. nih.gov/), save it in mol2 format and convert it to a Three-Dimensional (3D) structure mol2 format file using Chem3D software. The PDB format receptor files and mol2 format ligand files were converted into PDBQT format using AutoDock Tools 1.5.6 software, and then molecular docking was performed using Vina.

Results and Discussion

After removing duplicate data, a total of 43 effective components were screened for 6 wind medicines. The effective components were numbered according to which drug they belonged to, for example, (-)-taxifolin belonged to Cinnamomi ramulus and was numbered GZ1, Peroxyergosterol also belonged to Cinnamomi ramulus and was numbered GZ2. The components belonging to Chuanxiong rhizoma started with CX; those belonging to Ephedrae herba started with MH, those belonging to Bupleuri radix started with CH; those belonging to Asari Radix et Rhizoma started with XX and those belonging to Puerariae lobatae radix started with GG. The effective components are common to several drugs and are numbered with capital letters plus Arabic numerals, for example, Mandenol is a common component of Chuanxiong rhizoma and Ephedrae herba and is numbered A1 (Table 1).

Mol ID Molecule name Number OB (%) BBB DL HL
MOL001736 (-)-taxifolin GZ1 60.51 -1.02 0.27 14.37
MOL011169 Peroxyergosterol GZ2 44.39 0.43 0.82 4.06
MOL000392 Formononetin GG1 69.67 0.02 0.21 17.04
MOL002959 3'-Methoxydaidzein GG2 48.57 -0.32 0.24 17.04
MOL002135 Myricanone CX1 40.6 -0.08 0.51 4.39
MOL002140 Perlolyrine CX2 65.95 0.15 0.27 12.62
MOL002157 Wallichilide CX3 42.31 0.73 0.71 6.85
MOL000433 Folsaeure CX4 68.96 -2.59 0.71 213.28
MOL001645 Linoleyl acetate CH1 42.1 1.08 0.2 7.48
MOL002776 Baicalin CH2 40.12 -1.74 0.75 17.36
MOL000354 Isorhamnetin CH3 49.6 -0.54 0.31 14.34
MOL004598 3,5,6,7-tetramethoxy-2-(3,4,5-trimethoxyphenyl)chromone CH4 31.97 0.08 0.59 15.54
MOL004609 Areapillin CH5 48.96 -0.29 0.41 16.52
MOL013187 Cubebin CH6 57.13 -0.41 0.64 12.4
MOL004718 α-spinasterol CH7 42.98 0.79 0.76 6.46
MOL004628 Octalupine CH8 47.82 0.3 0.28 4.17
MOL004644 Sainfuran CH9 79.91 0.23 0.23 8.58
MOL004648 Troxerutin CH10 31.6 -0.38 0.28 4.36
MOL004702 Saikosaponin c_qt CH11 30.5 -0.85 0.63 6.12
MOL002823 Herbacetin MH1 36.07 -0.65 0.27 14.8
MOL000006 Luteolin MH2 36.16 -0.84 0.25 15.94
MOL001755 24-Ethylcholest-4-en-3-one MH3 36.08 1.22 0.76 5.49
MOL001771 Poriferast-5-en-3beta-ol MH4 36.91 1.14 0.75 5.07
MOL002881 Diosmetin MH5 31.14 -0.66 0.27 16.34
MOL004328 Naringenin MH6 59.29 -0.37 0.21 16.98
MOL005043 Campest-5-en-3beta-ol MH7 37.58 0.94 0.71 4.43
MOL005190 Eriodictyol MH8 71.79 -0.54 0.24 15.81
MOL005573 Genkwanin MH9 37.13 -0.24 0.24 16.1
MOL005842 Pectolinarigenin MH10 41.17 -0.09 0.3 16.56
MOL011319 Truflex OBP MH11 43.74 0.6 0.24 4.9
MOL012140 4,9-dimethoxy-1-vinyl-b-carboline XX1 65.3 0.72 0.19 9.85
MOL012141 Caribine XX2 37.06 -0.15 0.83 10.44
MOL001460 Cryptopin XX3 78.74 0.41 0.72 21.25
MOL001558 Sesamin XX4 56.55 -0.08 0.83 13.44
MOL002962 (3S)-7-hydroxy-3-(2,3,4-trimethoxyphenyl)chroman-4-one XX5 48.23 0.01 0.33 16.64
MOL009849 ZINC05223929 XX6 31.75 0.12 0.83 14.12
MOL001494 Mandenol A1 42 1.14 0.19 5.39
MOL000359 Sitosterol A2 36.91 0.87 0.75 5.37
MOL000358 b-sitosterol B1 36.91 0.99 0.75 5.36
MOL004576 Taxifolin C1 57.84 -0.8 0.27 14.41
MOL000449 Stigmasterol D1 43.83 1 0.76 5.57
MOL000422 Kaempferol D2 41.88 -0.55 0.24 14.74
MOL000098 Quercetin D3 46.43 -0.77 0.28 14.4

Table 1: The Active Components of Chuanxiong, Ephedra, Pueraria Lobata, Guizhi, Bupleurum, Asarum

The relevant targets of 43 effective components were obtained from the TCMSP database, duplicate data were removed and entered into the Uniprot database normalized to human gene names, and targets for which the exact human gene names were not queried were removed. A total of 255 relevant targets were obtained. The 5303 BBB targets were obtained from the Genecards database and 217 intersecting targets were obtained by intersecting the wind drug-related targets with the BBB-related targets (fig. 1), including Interleukin-4 (IL-4), Tumour Necrosis Factor (TNF), Caspase-3 (CASP3), Mitogen-Activated Protein Kinase (MAPK) 8, etc.


Fig. 1: Common targets of drugs and the BBB

The data of the six wind medicines, the effective components (excluding those without relevant targets), the BBB and the intersection targets were input into Cytoscape 3.7.0 software for visualization, and the "drug-component-target-BBB" network diagram was obtained (fig. 2). The triangle represents the BBB, the hexagon represents the effective components, the circle represents the wind medicine and the diamond represents the intersection target (the color is to distinguish each part, no practical significance). CH represents Bupleuri radix, XX represents Asari Radix et Rhizoma, GG represents Puerariae lobatae radix, GZ represents Cinnamomi ramulus, CX represents Chuanxiong rhizoma and MH represents Ephedrae herba (Table 1). The node size and transparency are arranged according to degree values, the larger the degree value, the darker the color and the larger the node, the larger the degree value indicate the more important the node. From fig. 1, it can be seen that there are multiple compounds and multiple targets in six wind medicines, reflecting the existence of multiple components and multiple targets of wind medicines acting on the BBB, and it can be found that the first seven major components of the six wind medicines acting on the BBB are quercetin,kaempferol, beta (β)-sitosterol, isorhamnetin, luteolin and formononetin (Table 2).


Fig. 2: Network diagram of "drug-component-target-BBB"

Serial number Number Components Degree
1 D3 Quercetin 138
2 D2 Kaempferol 54
3 MH2 Luteolin 53
4 MH6 Naringenin 34
5 B1 Beta-sitosterol 33
6 GG1 Formononetin 30
7 CH3 Isorhamnetin 30

Table 2: Main Components of Wind Medicine Acting on the BBB

PPI network diagram of drug-target and core targets was described below. The 217 intersecting targets were imported into the STRING database (among which Immunoglobulin Heavy Constant Gamma 1 (IGHG1) was not recognized by the database), the species was limited to "Homo sapiens", the medium confidence was set >0.9 and the free targets were hidden to obtain the PPI network graph (fig. 3). There are 216 nodes (including the hidden free targets) and 859 edges, the nodes represent the target proteins and the edges represent the proteinprotein interactions, the average node degree is 7.95 and the average local clustering coefficient is 0.47. The obtained STRING file in TSV format was uploaded to Cytoscape 3.7.0 software for network topology analysis. The average value of degree was 9.187 and 68 core targets were screened with degree greater than 9.187 (Table 3).


Fig. 3: PPI network of active components-BBB

Number Target Degree Number Target Degree
1 AKT1 42 35 MDM2 14
2 TP53 40 36 AR 14
3 JUN 38 37 CYP1A1 14
4 MAPK1 38 38 SIRT1 14
5 MAPK3 38 39 CCL2 14
6 TNF 35 40 TGFB1 14
7 RELA 34 41 CXCL2 14
8 MAPK8 28 42 CHRM2 14
9 IL-6 28 43 BCL2L1 13
10 MAPK14 27 44 CCNA2 13
11 APP 26 45 NCOA2 13
12 EGFR 25 46 BCL2 13
13 FOS 23 47 CXCL10 13
14 MYC 23 48 IFNG 13
15 CXCL8 23 49 PPARA 13
16 ESR1 23 50 CCNB1 12
17 RB1 22 51 PCNA 12
18 VEGFA 21 52 PPARG 12
19 RXRA 21 53 IL-10 12
20 CCND1 20 54 APOB 12
21 NR3C1 20 55 CYP3A4 12
22 EGF 19 56 CDK4 11
23 PRKCA 18 57 XIAP 11
24 CASP8 17 58 E2F1 11
25 CASP3 17 59 GSK3B 11
26 CDKN1A 17 60 HIF1A 11
27 IL-1b 17 61 ERBB2 11
28 PRKCB 17 62 CHUK 10
29 CDK1 16 63 IGF2 10
30 IL-4 16 64 MMP9 10
31 STAT1 16 65 CXCL11 10
32 CDK2 15 66 PTGS2 10
33 NFKBIA 15 67 NOS2 10
34 IL-2 15 68 RASA1 10

Table 3: Core targets of Drug-BBB

The information of 68 core targets was entered into the DAVID database and the GO biological process analysis and KEGG pathway enrichment analysis were performed on the 68 core targets, and 512 biological processes and 116 pathways were obtained with p≤0.05 as the screening criteria. The top 20 biological processes and pathways were selected according to the order of p-value from smallest to largest and displayed in enrichment bubble plots (fig. 4 and fig. 5), which were drawn by http://www. bioinformatics.com.cn (an online platform for data analysis and visualization). The biological processes involved are mainly positive regulation of Ribonucleic Acid (RNA) polymerase II promoter transcription, positive regulation of transcription, negative regulation of the apoptotic process, inflammatory response, lipopolysaccharidemediated signaling pathway and cellular response to hypoxia. The pathways involved include cancer pathways, hepatitis B, T cell receptor signaling pathway, Nucleotide- Binding Oligomerization Domain (NOD)-like receptor signaling pathway, Hypoxia-Inducible Factor 1 (HIF-1) signaling pathway, Toll-like receptor signaling pathway, etc.


Fig. 4: GO biological process analysis


Fig. 5: KEGG enrichment analysis

Molecular docking of the 7 main effective components (quercetin, kaempferol, luteolin, naringenin, β-sitosterol, formononetin, isorhamnetin) with the top 10 core targets (AKT Serine/Threonine Kinase 1 (AKT1), Tumor Protein p53 (TP53), Jun Proto-Oncogene, AP-1 Transcription Factor Subunit (JUN), MAPK1, MAPK3, TNF, RELA Proto-Oncogene, NF-κB Subunit (RELA), MAPK8, IL6, MAPK14). It is generally believed that the lower the binding energy, the better the ligand and receptor binding effect. It was found that the lowest binding energy of the binding complexes of the seven effective components and the 10 target proteins were all less than 0 kcal/mol (Table 4) and the molecular docking diagrams of the targets with the best binding effect to the effective components were presented (fig. 6).

Target Ingredient Binding energy (kcal/mol)
AKT1 Quercetin -9.8
AKT1 Kaempferol -9.5
AKT1 Luteolin -9.8
AKT1 Naringenin -9.6
TP53 Quercetin -8.8
TP53 Luteolin -8.9
JUN Quercetin -8.0
JUN Kaempferol -8.2
JUN Luteolin -8.1
JUN Beta-sitosterol -7.0
JUN Formononetin -7.4
MAPK1 Quercetin -7.5
MAPK1 Luteolin -7.8
MAPK1 Naringenin -7.4
MAPK3 Naringenin -7.7
TNF Quercetin -8.8
TNF Kaempferol -8.5
TNF Luteolin -8.6
RELA Quercetin -9.2
RELA Kaempferol -9.2
RELA Luteolin -9.1
RELA Naringenin -8.9
RELA Isorhamnetin -8.9
MAPK8 Kaempferol -9.2
IL-6 Quercetin -7.8
IL-6 Luteolin -7.4
MAPK14 Isorhamnetin -9.3
MAPK14 Formononetin -10.2

Table 4: The Binding Energy of the Core targets and their Associated Effective Components


Fig. 6: Molecular docking simulation of the effective components and its lowest binding energy target

In this study, we analyzed the effective components of six wind medicines, namely, Chuanxiong rhizoma, Ephedrae herba, Bupleuri radix, Asari Radix et Rhizoma, Puerariae lobatae radix, Cinnamomi ramulus and concluded that the key components of the six wind medicines acting on the BBB were quercetin, kaempferol, isorhamnetin, luteolin, formononetin, β-sitosterol and naringenin. These components were found to have a significant protective effect on the BBB. Jin et al.[10] demonstrated that quercetin could improve BBB dysfunction by upregulating the expression of Zonula Occludens-1 (ZO-1), Claudin-5, etc., and decreasing the expression of Matrix Metalloproteinase-9 (MMP-9). Li et al.[11] found that kaempferol could inhibit pro-inflammatory proteins and improve BBB dysfunction in rats with cerebral ischemia-reperfusion. Zhao et al.[12] found that isorhamnetin could improve BBB function and upregulate the expression of occludin, ZO-1 and claudin-5. Zhang et al.[13] found that luteolin could protect the BBB by inhibiting the inflammatory response after BBB injury. Bai et al.[14] found that naringenin could downregulate the expression of Nuclear Factor kappa Light Chain Enhancer of Activated B cells (NF-κB) and MMP-9 and upregulate the expression of claudin-5 to protect the BBB function in rats with cerebral ischemia.

The core targets of wind drugs acting on the BBB screened in this study are all important targets for regulating the BBB and most of them are related to the immune-inflammatory response and apoptosis, such as apoptosis-related proteins (pro- or antiapoptotic) CASP8, CASP3, B-Cell Lymphoma 2 (BCL2), Bcl-2-like protein 1 (BCL2L1), X-linked Inhibitor of Apoptosis Protein (XIAP), etc.; immune inflammation-related proteins TNF, IL-6, IL-1B, IL- 4, IL-10, C-X-C Motif Chemokine Ligand (CXCL) 8, C-C Motif Chemokine Ligand 2 (CCL2), CXCL2, CXCL10, Interferon Gamma (IFNG), CXCL11, etc.; MAPK family, MAPK1, MAPK3, MAPK8, MAPK14; NF-κB family, RELA; MMP9; Vascular Endothelial Growth Factor A (VEGFA), etc. In addition, transcription factor, Signal Transducer and Activator of Transcription 1 (STAT1), HIF1A, sirtuin protein family member, Sirtuin 1 (SIRT1) and other proteins are closely related to the BBB. Current studies have confirmed that inflammation is an important cause of BBB disruption, inflammatory factors can directly damage endothelial cells, IL-6 can make endothelial cell damage by activating NF-κB and TNF-alpha (α) can also induce RELA overexpression and increase BBB permeability[15]. The sirtuin protein family is closely associated with apoptosis and inflammation, and studies have demonstrated that reduced expression of SIRT1 leads to increased permeability of the BBB[16,17]. Production of MMP, especially MMP-9, degrades tight junction and basement membrane proteins and exacerbates BBB disruption, while inhibition of MMP- 9 significantly enhances the protective effect on the BBB[18,19]. MAPK is closely associated with cellular processes such as inflammation and stress responses, and MAPK is an important signaling pathway for BBB disruption and activation of inflammatory responses, and inhibition of the MAPK pathway can significantly improve BBB function[20,21]. Chemokines associated with the immune-inflammatory response are also involved in BBB disruption and CCL2 and CXCL8 are associated with BBB dysfunction[22,23]. Endothelial growth factor induces increased vascular permeability and astrocytes can produce VEGFA to enhance the permeability of the BBB, while inhibition of VEGFA improves BBB function[24]. Studies have confirmed that pro-inflammatory cytokines and chemokines upregulate the expression of VEGFA, thereby exacerbating the disruption of the BBB[25]. When cerebrovascular disease occurs, apoptotic signals including p53, CASP3 and CASP9 are activated, thus promoting apoptosis and disrupting the BBB integrity[26]. GO biological processes include positive regulation of RNA polymerase II promoter transcription, negative regulation of the apoptotic process, lipopolysaccharidemediated signaling pathway, positive regulation of gene expression, inflammatory response, cellular response to hypoxia, positive regulation of nitric oxide biosynthesis process, etc. It mainly involves a series of biological processes such as inflammatory response, apoptosis, signal transduction, lipopolysaccharideinduced response and gene expression. The top 20 signaling pathways involved include cancer pathway, hepatitis B, TNF signaling pathway, toxoplasmosis, Toll-like receptor signaling pathway, NOD-like receptor signaling pathway, HIF-1 signaling pathway, T cell receptor signaling pathway, etc., which are mainly involved in viral or parasitic infection, immuneinflammatory response, apoptosis, etc.

Therefore, the present study confirmed that the wind drug has multi-component, multi-target and multi-pathway characteristics through network pharmacological analysis. It may act on the BBB through anti-inflammatory and anti-apoptotic mechanisms, which is important for the treatment of encephalopathy. However, this study is relatively one-sided and needs to be verified by follow-up experimental studies and so on.


This study is supported by Sichuan Science and Technology Project: 2020YJ0437; the Project of Sichuan Provincial Administration of Traditional Chinese Medicine: 2020JC0139; Luzhou Science and Technology Project: 2020-SYF-30; the Project of National Traditional Chinese Medicine Clinical Research Base: XNYDZYY [2020] No.33; the Southwest Medical University and Affiliated Traditional Chinese Medicine Hospital Joint Program: 2018XYLH-006.

Conflict of interests:

All authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.