Abstract

Objective: The traditional Chinese medicine Caulis Sargentodoxae is widely used in the treatment of ulcerative colitis (UC), but the mechanism remains unknown. The present study aims to reveal its effective components, targets and pathways through network pharmacology and bioinformatics approaches.

Materials and methods: Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) was used to identify effective components. The ligand-based targets prediction was achieved through SwissTargetPrediction and TargetNet. UC-related targets were identified using Gene Expression Omnibus (GEO) data and DisGeNET. The common targets of disease and components were constructed and analyzed by PPI network. Lastly, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses are used to explain the functions of these common targets. Components-Targets-Pathways network was visualized and analyzed to further reveal the connection between the components and targets.

Results: Eight active components and 102 key targets were identified to play an important role in UC. These targets were related to regulation of protein serine/threonine kinase activity, positive regulation of cell motility, response to molecule of bacterial origin, response to toxic substance, ERK1 and ERK2 cascade, peptidyl-tyrosine modification, inositol lipid-mediated signaling, cellular response to drug, regulation of inflammatory response and leukocyte migration. Moreover, HIF-1 signaling pathway and PI3K-Akt signaling pathway were the key targets involved in UC-related signaling pathways.

Conclusion: The eight active components of Caulis Sargentodoxae mainly play a therapeutic role for UC through synergistic regulation of HIF-1 signaling pathway and PI3K-Akt signaling pathway.

Introduction

Ulcerative colitis (UC) is a disease caused by a genetically predisposed host’s immune disorder to antigens in the intestine, and its incidence and prevalence are increasing globally [1,2]. The disease has the characteristics of remission and flare, and in most cases drugs are used to induce and maintain remission. However, up to 15% of cases require surgery due to medical treatment failure or inflammation that develops into dysplasia [3]. The primary treatment goal of UC is to achieve disease remission, prevent disease-related complications, surgery and tumor, and maintain the quality of life of patients [4]. The currently available data indicated that UC is an intestinal barrier disease, which is caused by epithelial cells or structural intestinal epithelial dysfunction, or strong inflammatory mediators and cells in the lamina propria to produce an inflammatory cascade. Therefore, treatment strategies can target epithelial cells or inflammatory cells to restore the intestinal barrier function and achieve clinical remission [5]. Currently, drugs used for UC include 5-aminosalicylic acid, steroids and immunosuppressants. However, long-term use of these drugs may cause a variety of side effects, such as liver and kidney toxicity, drug resistance and allergic reactions [6].

The use of herbs to treat various diseases has a long history. In addition, traditional Chinese medicine shows obvious advantages in the treatment of UC, such as low recurrence rate and fewer side effects [6]. Some natural active compounds have good therapeutic effects on experimental UC models. SM934 is a water-soluble artemisinin analog with anti-inflammatory and immunomodulatory effects. It was found to improve DSS-induced UC in mice by inhibiting neutrophils and macrophages [7]. Indigo Naturalis has been proven effective in a randomized, placebo-controlled trial [8]. However, there were potential adverse effects, including pulmonary hypertension [8]. This also reflects the complexity of Chinese herbal medicine.

Network pharmacology methods are rapidly evolving and applied to find new treatment opportunities and reposition of approved drugs. They can also effectively map the unexplored targets of natural products, thereby providing drug space for proteins related to various complex diseases [9]. The purpose of constructing the network is to realize the interaction between bioactive compounds and target proteins and the interaction between various target proteins, and then to identify and verify the key nodes through network analysis and verification [10]. Network pharmacology has been successfully used to reveal the active ingredients and mechanisms of indigo naturalis in the treatment of UC [11]. Caulis Sargentodoxae is also a high-frequency herb used in the treatment of UC. As a heat-clearing and detoxifying Chinese medicine, it has been found to have significant anti-inflammatory, anti-thrombotic, anti-tumor and anti-bacterial effects [12]. However, its active ingredients and targets are still unclear. The present study is based on the network pharmacology approaches and bioinformatics technologies to clarify the mechanism of Caulis Sargentodoxae in the treatment of UC, and to provide new candidate compounds for the treatment of UC. The entire workflow is shown in Figure 1.

Work flow of the present study

Figure 1
Work flow of the present study
Figure 1
Work flow of the present study

Materials and methods

Components acquisition and screening

Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) is a manually curated database containing 12144 herbal compounds and related pharmacokinetic properties (https://tcmspw.com/tcmsp.php) [13]. Oral bioavailability (OB), Drug-likeness (DL), Caco-2 permeability and other parameters information can be obtained from it. The components and characteristics of Caulis Sargentodoxae were retrieved from TCMSP. According to its suggested drug screening criteria, the components were screened. Finally, the compounds with molecular weight (MW) from 180 to 500 Dalton, OB ≥ 20%, DL ≥ 0.1 were retained.

Targets collection of components in Caulis Sargentodoxae

Drug-targets prediction technology, which can predict component targets with high throughput and quickly, has been widely used in the preliminary research of drug development. The current prediction technologies mainly include ligands-based prediction methods (such as chemical similarity search, pharmacophore models etc.), targets-based prediction methods (such as reverse molecular docking) and machine learning methods. SwissTargetPrediction is a ligand-based targets prediction tool for bioactive small molecules (http://www.swisstargetprediction.ch/). The principle is to train the prediction model by multivariate logistic regression fitting of various size related subsets of known active molecules to weighted 2D and 3D similarity parameters [14]. TargetNet builds a large number of QSAR models based on current chemical and genomics data for molecular target prediction (http://targetnet.scbdd.com/calcnet/index/). Its server will predict the activity of user-submitted molecules in 623 human proteins by building a high-quality QSAR model for each human protein [15]. The structures and SMILES of each component were obtained from PubChem (https://pubchem.ncbi.nlm.nih.gov/). Subsequently, the SMILES of each component were uploaded to two databases for targets prediction.

Acquisition of UC disease targets

Gene Expression Omnibus (GEO) is an ever-increasing platform for storing high-throughput sequencing data. Integrating sequencing information from different centers and performing gene differential expression analysis will facilitate the identification of disease treatment targets. We identified three series from GEO that all came from GPL6244, among which GSE73661 contains 23 active UC tissues and 12 normal tissues, GSE75214 contains 74 active UC tissues and 11 normal tissues, GSE48958 contains 7 active UC tissues and 8 normal tissues. Genes with adjusted P<0.05 and log2 (fold change) > 2 or log2 (fold change) < −2 were considered as differential genes, they have played an important role in the development of UC.

DisGeNET integrated data from public databases, GWASs, animal models and scientific literature, and collected a large number of variants and genes associated with human diseases (https://www.disgenet.org/). It currently covers more than 24000 diseases and traits, 17000 genes and 117000 genomic variants [16]. UC-related genes were also obtained via using ‘ulcerative colitis’ as the keyword to retrieve relevant genes in DisGeNET.

Construction and analysis of target protein–protein interaction network

STRING (https://string-db.org/) can carry out direct (physical) and indirect (functional) interactions of proteins to achieve a comprehensive and objective network [17]. The targets were uploaded in the form of gene symbol, the species was selected as ‘Homo sapiens’, and the minimum required interaction score threshold ‘highest confidence (≥0.9)’ was set to obtain protein interaction information. Cytoscape v3.7.2 was used for visualization and in-depth analysis of the network. The hub nodes were identified in each network by calculating the size of the degree.

Gene ontology and Kyoto Encyclopedia of Genes and Genomes signal pathway enrichment analysis

Gene ontology (GO) is a gene function classification system that comprehensively describes the attributes of genes and gene products of various species through dynamically updated GO terms. It is used to describe the biological process (BP), the cellular component (CC) and molecular function (MF) of gene sets of interest. Kyoto Encyclopedia of Genes and Genomes (KEGG) is a knowledge base for systematic analysis of gene function, which can describe pathway data involved in genes of interest, including connection of metabolism, membrane transport, signal transduction and cell cycle, etc. We used WebGestalt (http://www.webgestalt.org/#), a network enrichment analysis tool, for GO and KEGG enrichment analyses. The reference set was selected as human genome protein encoding. And the enrichment method was selected as over-representaion analysis.

Construction and analysis of components-targets-pathways network

Components-Targets-Pathways network was constructed and visualized via Cytoscape v3.7.2 to identify key molecules and targets. The degrees of each component and target were calculated using CytoHubba.

Toxicity prediction of components in Caulis Sargentodoxae

AdmetSAR 2.0 was a free tool and continually updated database for the prediction of ADMET properties, and has been widely used in chemical and pharmaceutical fields [18,19]. Each molecule’s SMILES was input to obtain organ toxicity and other toxic properties.

Results

Components of Caulis Sargentodoxae

Nine eligible valid results were screened through TCMSP database, with OB ≥ 20%, DL ≥ 0.1 and 500 ≥ MW ≥ 180 Dalton. These molecules have good OB and DL properties (Table 1 and Supplementary Figure S1). We have also made a toxicity prediction for compounds with SMILES. Analysis found that they have no carcinogenicity, but emodin and physcion may have hepatotoxicity. And physcion and (-)-catechin were positive in Ames mutagenesis (Supplementary Table S1).

Table 1
Parameter information of molecules in Caulis Sargentodoxae
TCMSP IDMolecule namePubchem CidMWOB (%)DL
MOL000358 β-sitosterol 222284 414.79 36.91 0.75 
MOL000359 Sitosterol 12303645 414.79 36.91 0.75 
MOL000472 Emodin 3220 270.25 24.4 0.24 
MOL000476 Physcion 10639 284.28 22.29 0.27 
MOL007920 Meso-dihydroguaiaretic acid 476856 330.46 31.32 0.26 
MOL007923 3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester 16119626 284.33 93.36 0.21 
MOL007925 Glucosyringic acid 10383888 360.35 24.29 0.3 
MOL000096 (-)-Catechin 73160 290.29 49.68 0.24 
MOL007922 Syringaresnol diglucoside_qt Not available 418.48 29.84 0.72 
TCMSP IDMolecule namePubchem CidMWOB (%)DL
MOL000358 β-sitosterol 222284 414.79 36.91 0.75 
MOL000359 Sitosterol 12303645 414.79 36.91 0.75 
MOL000472 Emodin 3220 270.25 24.4 0.24 
MOL000476 Physcion 10639 284.28 22.29 0.27 
MOL007920 Meso-dihydroguaiaretic acid 476856 330.46 31.32 0.26 
MOL007923 3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester 16119626 284.33 93.36 0.21 
MOL007925 Glucosyringic acid 10383888 360.35 24.29 0.3 
MOL000096 (-)-Catechin 73160 290.29 49.68 0.24 
MOL007922 Syringaresnol diglucoside_qt Not available 418.48 29.84 0.72 

Putative targets of components

In both SwissTargetPrediction and TargetNet, we selected targets with a probability greater than 0 for subsequent analysis. Due to the lack of structural information and SMILES, we failed to obtain the targets of syringaresnol diglucoside_qt. The other 8 components get 396 targets through SwissTargetPrediction, and 719 targets through TargetNet. A total of 387 unique targets of 8 components were obtained after removal of the repeats. The relationship among the 8 components and the 387 targets was shown in Figure 2.

Components-targets-network of Caulis Sargentodoxae

Figure 2
Components-targets-network of Caulis Sargentodoxae

The yellow ellipses represent the components, and the red ellipses represent the targets of components in Caulis Sargentodoxae.

Figure 2
Components-targets-network of Caulis Sargentodoxae

The yellow ellipses represent the components, and the red ellipses represent the targets of components in Caulis Sargentodoxae.

UC-related targets

GSE73661, GSE75214 and GSE48958 were respectively analyzed for differential genes. Through the Venn diagram, we obtained 87 common up-regulated genes and 42 common down-regulated genes from three series (Figure 3A,B). In addition, we obtained 1458 UC-related targets from DisGeNET. A total of 1528 unique targets of UC were obtained after removal of the repeats (Figure 3C). By comparing the components targets and the disease targets through the Venn diagram, 125 targets for the treatment of UC were identified (Figure 3D). These targets may be the key proteins of Caulis Sargentodoxae in the treatment of UC.

Work flow of targets screening

Figure 3
Work flow of targets screening

(A) Venn diagram of up-regulated genes in UC from three GEO series. (B) Venn diagram of down-regulated genes in UC from three GEO series. (C) Venn diagram of targets of UC from differential genes (GEO) and DisGeNET. (D) Venn diagram of targets from UC and Caulis Sargentodoxae.

Figure 3
Work flow of targets screening

(A) Venn diagram of up-regulated genes in UC from three GEO series. (B) Venn diagram of down-regulated genes in UC from three GEO series. (C) Venn diagram of targets of UC from differential genes (GEO) and DisGeNET. (D) Venn diagram of targets from UC and Caulis Sargentodoxae.

PPI network of 125 common targets

We constructed PPI networks with threshold ‘highest confidence (≥0.9)’ in STRING. A network with 102 nodes and 618 edges was obtained (Figure 4). Based on the degree, we identified the key targets among them, including SRC, PIK3CA, MAPK1, HSP90AA1, MAPK8, ESR1, EGFR, LYN, RAC1, RELA, TNF, MAPK14, S1PR1, MMP2, MMP9, PRKCB (Table 2).

PPI network of common targets in UC and Caulis Sargentodoxae (interaction score ≥ 0.9)

Figure 4
PPI network of common targets in UC and Caulis Sargentodoxae (interaction score ≥ 0.9)

The darker nodes represent higher degrees.

Figure 4
PPI network of common targets in UC and Caulis Sargentodoxae (interaction score ≥ 0.9)

The darker nodes represent higher degrees.

Table 2
Hub genes in PPI network
RankNameScore
SRC 56 
PIK3CA 54 
MAPK1 48 
HSP90AA1 36 
MAPK8 34 
ESR1 34 
EGFR 30 
LYN 30 
RAC1 30 
10 RELA 28 
11 TNF 24 
11 MAPK14 24 
13 S1PR1 22 
14 MMP2 20 
14 MMP9 20 
14 PRKCB 20 
RankNameScore
SRC 56 
PIK3CA 54 
MAPK1 48 
HSP90AA1 36 
MAPK8 34 
ESR1 34 
EGFR 30 
LYN 30 
RAC1 30 
10 RELA 28 
11 TNF 24 
11 MAPK14 24 
13 S1PR1 22 
14 MMP2 20 
14 MMP9 20 
14 PRKCB 20 

GO and KEGG enrichment analyses

Using WebGestalt to conduct bioinformatics analysis of the above key targets. We identified the top ten GO terms and pathways that were significantly enriched, respectively (Figures 5 and 6). In UC, the BPs mainly regulated by Caulis Sargentodoxae were regulation of protein serine/threonine kinase activity, positive regulation of cell motility, response to molecule of bacterial origin, response to toxic substance, ERK1 and ERK2 cascade, peptidyl-tyrosine modification, inositol lipid-mediated signaling, cellular response to drug, regulation of inflammatory response and leukocyte migration (Table 3).

Volcano plot of the top ten significantly enriched GO terms of the common targets in UC and Caulis Sargentodoxae

Figure 5
Volcano plot of the top ten significantly enriched GO terms of the common targets in UC and Caulis Sargentodoxae
Figure 5
Volcano plot of the top ten significantly enriched GO terms of the common targets in UC and Caulis Sargentodoxae

Volcano plot of the top ten significantly enriched pathways of the common targets in UC and Caulis Sargentodoxae

Figure 6
Volcano plot of the top ten significantly enriched pathways of the common targets in UC and Caulis Sargentodoxae
Figure 6
Volcano plot of the top ten significantly enriched pathways of the common targets in UC and Caulis Sargentodoxae
Table 3
Top ten significantly enriched GO terms of key targets
Gene setDescriptionSizeExpectRatioP-valueFDR
GO:0071900 Regulation of protein serine/threonine kinase activity 497 3.323 8.7272 
GO:2000147 Positive regulation of cell motility 493 3.2962 9.7081 
GO:0002237 Response to molecule of bacterial origin 330 2.2064 10.878 
GO:0009636 Response to toxic substance 499 3.3363 7.793 2.22E-16 3.77E-14 
GO:0070371 ERK1 and ERK2 cascade 326 2.1796 10.093 2.22E-16 3.77E-14 
GO:0018212 Peptidyl-tyrosine modification 389 2.6009 8.8432 6.66E-16 8.09E-14 
GO:0048017 Inositol lipid-mediated signaling 165 1.1032 15.41 6.66E-16 8.09E-14 
GO:0035690 Cellular response to drug 349 2.3334 9.4282 8.88E-16 9.44E-14 
GO:0050727 Regulation of inflammatory response 361 2.4137 9.1148 1.78E-15 1.68E-13 
GO:0050900 Leukocyte migration 419 2.8014 8.21 3.55E-15 3.02E-13 
Gene setDescriptionSizeExpectRatioP-valueFDR
GO:0071900 Regulation of protein serine/threonine kinase activity 497 3.323 8.7272 
GO:2000147 Positive regulation of cell motility 493 3.2962 9.7081 
GO:0002237 Response to molecule of bacterial origin 330 2.2064 10.878 
GO:0009636 Response to toxic substance 499 3.3363 7.793 2.22E-16 3.77E-14 
GO:0070371 ERK1 and ERK2 cascade 326 2.1796 10.093 2.22E-16 3.77E-14 
GO:0018212 Peptidyl-tyrosine modification 389 2.6009 8.8432 6.66E-16 8.09E-14 
GO:0048017 Inositol lipid-mediated signaling 165 1.1032 15.41 6.66E-16 8.09E-14 
GO:0035690 Cellular response to drug 349 2.3334 9.4282 8.88E-16 9.44E-14 
GO:0050727 Regulation of inflammatory response 361 2.4137 9.1148 1.78E-15 1.68E-13 
GO:0050900 Leukocyte migration 419 2.8014 8.21 3.55E-15 3.02E-13 

In UC, the pathways mainly regulated by Caulis Sargentodoxae were Pathways in cancer, Proteoglycans in cancer, Endocrine resistance, Prostate cancer, AGE-RAGE signaling pathway in diabetic complications, HIF-1 signaling pathway, MicroRNAs in cancer, Fluid shear stress and atherosclerosis, PI3K-Akt signaling pathway and EGFR tyrosine kinase inhibitor resistance (Table 4). Among them, the changes of HIF-1 signaling pathway and PI3K-Akt signaling pathway are most closely related to UC.

Table 4
Top ten significantly enriched pathways of key targets
Gene setDescriptionSizeExpectRatioP-valueFDR
hsa05200 Pathways in cancer 524 7.1346 5.0458 
hsa05205 Proteoglycans in cancer 198 2.6959 8.9024 
hsa01522 Endocrine resistance 98 1.3343 14.239 
hsa05215 Prostate cancer 97 1.3207 13.629 4.44E-16 3.62E-14 
hsa04933 AGE-RAGE signaling pathway in diabetic complications 99 1.348 12.612 1.02E-14 6.66E-13 
hsa04066 HIF-1 signaling pathway 100 1.3616 11.751 2.13E-13 1.16E-11 
hsa05206 MicroRNAs in cancer 150 2.0424 8.8133 9.90E-13 4.61E-11 
hsa05418 Fluid shear stress and atherosclerosis 138 1.879 9.0475 2.93E-12 1.20E-10 
hsa04151 PI3K-Akt signaling pathway 354 4.82 4.9793 3.10E-11 1.02E-09 
hsa01521 EGFR tyrosine kinase inhibitor resistance 79 1.0756 12.086 3.11E-11 1.02E-09 
Gene setDescriptionSizeExpectRatioP-valueFDR
hsa05200 Pathways in cancer 524 7.1346 5.0458 
hsa05205 Proteoglycans in cancer 198 2.6959 8.9024 
hsa01522 Endocrine resistance 98 1.3343 14.239 
hsa05215 Prostate cancer 97 1.3207 13.629 4.44E-16 3.62E-14 
hsa04933 AGE-RAGE signaling pathway in diabetic complications 99 1.348 12.612 1.02E-14 6.66E-13 
hsa04066 HIF-1 signaling pathway 100 1.3616 11.751 2.13E-13 1.16E-11 
hsa05206 MicroRNAs in cancer 150 2.0424 8.8133 9.90E-13 4.61E-11 
hsa05418 Fluid shear stress and atherosclerosis 138 1.879 9.0475 2.93E-12 1.20E-10 
hsa04151 PI3K-Akt signaling pathway 354 4.82 4.9793 3.10E-11 1.02E-09 
hsa01521 EGFR tyrosine kinase inhibitor resistance 79 1.0756 12.086 3.11E-11 1.02E-09 

Components-targets-pathways network

The key targets of Caulis Sargentodoxae enriched to HIF-1 signaling pathway were BCL2, EGFR, ERBB2, FLT1, IGF1R, INSR, MAPK1, MKNK2, NOS2, NOS3, PIK3CA, PIK3CD, PRKCB, RELA, RPS6KB1 and SERPINE1 (Supplementary Figure S2). The key targets of Caulis Sargentodoxae enriched to PI3K-Akt signaling pathway were BCL2, CASP, CCND1, CCNE1, EGFR, ERBB2, FGFR1, FLT1, HSP90AA1, IGF1R, INSR, KDR, MAPK1, MCL1, MDM2, MET, NOS3, NTRK1, PIK3CA, PIK3CD, PIK3CG, RAC1, RELA, RPS6KB1 (Supplementary Figure S3). In the PPI network, EGFR, PRKCB, MAPK1, HSP90AA1, RELA, RAC1, PIK3CA were also identified as hub genes.

Moreover, we constructed components-targets-pathways network with 8 components, 102 hub targets and HIF-1 signaling pathway and PI3K-Akt signaling pathway (Figure 7). The network showed that these eight molecules were involved in regulating HIF-1 signaling pathway and PI3K-Akt signaling pathway. In this network, we further identified key molecules and targets using degrees (Table 5). Physcion, emodin, 3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester and meso-dihydroguaiaretic acid were the most important components that regulates the above two pathways.

Components-targets-pathways network of Caulis Sargentodoxae

Figure 7
Components-targets-pathways network of Caulis Sargentodoxae

The triangles represent the components, the quadrangles represent the pathways, and the squarenesses represent the targets that can be regulated by the components and are in the pathways. The darker nodes represent higher degrees.

Figure 7
Components-targets-pathways network of Caulis Sargentodoxae

The triangles represent the components, the quadrangles represent the pathways, and the squarenesses represent the targets that can be regulated by the components and are in the pathways. The darker nodes represent higher degrees.

Table 5
Hub genes/molecules in components-targets-pathways network
RankNameScore
Physcion 18 
Emodin 17 
3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester 12 
Meso-dihydroguaiaretic acid 
CASP9 
RELA 
HSP90AA1 
(-)-Catechin 
RAC1 
10 MDM2 
10 NOS3 
10 NOS2 
10 EGFR 
14 MET 
14 BCL2 
14 PIK3CA 
14 Glucosyringic acid 
14 RPS6KB1 
14 ERBB2 
14 KDR 
14 MCL1 
14 INSR 
14 IGF1R 
24 β-sitosterol 
24 PIK3CG 
24 PIK3CD 
24 SERPINE1 
24 FLT1 
RankNameScore
Physcion 18 
Emodin 17 
3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester 12 
Meso-dihydroguaiaretic acid 
CASP9 
RELA 
HSP90AA1 
(-)-Catechin 
RAC1 
10 MDM2 
10 NOS3 
10 NOS2 
10 EGFR 
14 MET 
14 BCL2 
14 PIK3CA 
14 Glucosyringic acid 
14 RPS6KB1 
14 ERBB2 
14 KDR 
14 MCL1 
14 INSR 
14 IGF1R 
24 β-sitosterol 
24 PIK3CG 
24 PIK3CD 
24 SERPINE1 
24 FLT1 

Discussion

Although drugs such as aminosalicylate, glucocorticoids, antibiotics and anti-TNFα play a good role in the treatment of UC, they still fail to achieve healing results and often cause side effects [20]. In China, traditional Chinese medicine prescriptions are widely used to treat recurrent UC. The mechanism of these drug combination based on the basic theory of traditional Chinese medicine is often difficult to elucidate due to the complexity of the components. The holistic view of Chinese medicine has a lot in common with the key ideas of network pharmacology and systems biology. This network-based approach can explain the combination rules and network regulation effects of herbal formulas [21]. Network pharmacology strategies were successfully used to reveal mechanism of Lang Chuang Wan in treating systemic lupus erythematosus [22] and Yizhiqingxin formula on Alzheimer’s disease [23].

Combined with the bioinformatics research on diseases, we can more clearly define the targets and specific mechanisms of Chinese medicine on the diseases, and identify key active molecules. This is of great significance to the development of natural medicines and the treatment of diseases. In our study, we successfully identified eight active molecules and two important pathways related to UC through network pharmacology and bioinformatics. They played a synergistic role in regulating HIF-1 signaling pathway and PI3K-Akt signaling pathway. Current evidences showed that Treg/Th17 imbalance is closely related to UC. The mTOR-HIF-1α-Th17 and STAT3-HIF-1α-Th17 pathways play an important role in the development of Th17 and the activation of IL-17. In addition, HIF-1α can bind to Foxp3, thereby affecting the development and function of Treg [24]. PI3K can respond to LPS and TNF-α, thereby activating Akt and affecting NF-κB to produce anti-apoptotic and pro-inflammatory effects [25]. Therefore, Caulis Sargentodoxae can improve UC by regulating them.

According to Components-Targets-Pathways network, the degrees of physcion, emodin, 3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester and meso-dihydroguaiaretic acid were the highest. Physcion is an anthraquinone compound commonly found in various plants, and shows anti-cancer properties in a variety of tumors. It can significantly inhibit the production of LPS/IFN-γ-induced NO in various cells via attenuating iNOS [26]. It can also reduce the gene expression of TNF-α, IL-6 and IL-1β in hepatocellular carcinoma cells induced by LPS in a dose-dependent manner by regulating p38 [27]. This indicated that it has good anti-inflammatory ability. And in our study, it was also identified as the most critical component for the treatment of UC. Emodin is a natural anthraquinone derivative with a wide range of pharmacological properties, including anticancer, liver protection, anti-inflammatory, antioxidant and antibacterial activities [28]. Emodin was found to induce autophagy in L02 human hepatocytes by inhibiting PI3K/AKT/mTOR signaling pathway [29]. Emodin was found to protect mice from DSS-induced colitis via the regulation of TLR5/NF-κB signaling pathway [30], and significantly inhibit LPS-mediated NF-κB activation and DNA binding activity in RAW264.7 cells [31]. This is consistent with our computer-based research, and indicates that emodin can regulate PI3K-Akt signaling pathway to exert anti-inflammatory effects. Meso-dihydroguaiaretic acid was found can attenuate airway inflammation and mucus hypersecretion in an ovalbumin-induced murine model of asthma via inhibiting the activation of NF-κB, ERK1/2 and p38 [32]. However, the pharmacological activity of 3-(4-Hydroxyphenyl)acrylic acid 4-hydroxyphenethyl ester has not been reported yet, and further experiments are needed.

Overall, our results suggested that Caulis Sargentodoxae can treat UC through regulating HIF-1 signaling pathway and PI3K-Akt signaling pathway, but this effect is a synergistic effect of eight active components. It is necessary to further compare the therapeutic effects of Caulis Sargentodoxae and its different components in animal models, and further evaluate these molecules in terms of safety and higher efficacy. Analyzing and verifying the effective dose and toxicity of different molecules by using pharmacology and toxicology experiments will also have important significance for clinical applications.

Data Availability

The data used to support the findings of the present study are included within the article and the supplementary files.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was supported by the Sichuan Medical Youth Innovative Scientific Research Project [grant number Q19045]; and the Chengdu Medical College Fund Natural Science General Project [grant numbers CYZ19-19, CYZ19-36].

Author Contribution

G.Z. and B.L. designed the present study. B.L., X.Z., J.Y. and W.L. collected and organized the data. B.L., X.Z., J.L., X.L. and R.W. completed the data analysis and chart processing. B.L. and X.Z. wrote the manuscript. G.Z. reviewed the manuscript.

Abbreviations

     
  • BP

    biological process

  •  
  • DL

    drug-likeness

  •  
  • DSS

    dextran sulphate sodium

  •  
  • EGFR

    epidermal growth factor receptor

  •  
  • GEO

    gene expression omnibus

  •  
  • GO

    gene ontology

  •  
  • GWAS

    genome-wide association studies

  •  
  • IFN

    interferon

  •  
  • IL

    interleukin

  •  
  • iNOS

    inducible nitric oxide synthase

  •  
  • KEGG

    Kyoto Encyclopedia of Genes and Genomes

  •  
  • LPS

    lipopolysaccharide

  •  
  • MW

    molecular weight

  •  
  • OB

    oral bioavailability

  •  
  • PPI

    protein-protein interaction

  •  
  • TCMSP

    Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform

  •  
  • Th17

    T helper cell 17

  •  
  • TLR5

    toll like receptor 5

  •  
  • Treg

    regulatory T cell

  •  
  • UC

    ulcerative colitis

References

1.
Ungaro
R.
,
Mehandru
S.
,
Allen
P.B.
,
Peyrin-Biroulet
L.
and
Colombel
J.F.
(
2017
)
Ulcerative colitis
.
Lancet
389
,
1756
1770
[PubMed]
2.
Hindryckx
P.
,
Jairath
V.
and
D’Haens
G.
(
2016
)
Acute severe ulcerative colitis: from pathophysiology to clinical management
.
Nat. Rev. Gastroenterol. Hepatol.
13
,
654
664
[PubMed]
3.
Feuerstein
J.D.
,
Moss
A.C.
and
Farraye
F.A.
(
2019
)
Ulcerative colitis
.
Mayo Clin. Proc.
94
,
1357
1373
[PubMed]
4.
Kayal
M.
and
Shah
S.
(
2019
)
Ulcerative colitis: current and emerging treatment strategies
.
J. Clin. Med.
9
,
[PubMed]
5.
Kobayashi
T.
,
Siegmund
B.
,
Le Berre
C.
,
Wei
S.C.
,
Ferrante
M.
,
Shen
B.
et al.
(
2020
)
Ulcerative colitis
.
Nat. Rev. Dis. Primers
6
,
74
[PubMed]
6.
Cao
S.Y.
,
Ye
S.J.
,
Wang
W.W.
,
Wang
B.
,
Zhang
T.
and
Pu
Y.Q.
(
2019
)
Progress in active compounds effective on ulcerative colitis from Chinese medicines
.
Chin. J. Nat. Med.
17
,
81
102
[PubMed]
7.
Yan
Y.X.
,
Shao
M.J.
,
Qi
Q.
,
Xu
Y.S.
,
Yang
X.Q.
,
Zhu
F.H.
et al.
(
2018
)
Artemisinin analogue SM934 ameliorates DSS-induced mouse ulcerative colitis via suppressing neutrophils and macrophages
.
Acta Pharmacol. Sin.
39
,
1633
1644
[PubMed]
8.
Naganuma
M.
,
Sugimoto
S.
,
Mitsuyama
K.
,
Kobayashi
T.
,
Yoshimura
N.
,
Ohi
H.
et al.
(
2018
)
Efficacy of indigo naturalis in a multicenter randomized controlled trial of patients with ulcerative colitis
.
Gastroenterology
154
,
935
947
[PubMed]
9.
Kibble
M.
,
Saarinen
N.
,
Tang
J.
,
Wennerberg
K.
,
Mäkelä
S.
and
Aittokallio
T.
(
2015
)
Network pharmacology applications to map the unexplored target space and therapeutic potential of natural products
.
Nat. Prod. Rep.
32
,
1249
1266
[PubMed]
10.
Luo
T.T.
,
Lu
Y.
,
Yan
S.K.
,
Xiao
X.
,
Rong
X.L.
and
Guo
J.
(
2020
)
Network pharmacology in research of chinese medicine formula: methodology, application and prospective
.
Chinese J. Integr. Med.
26
,
72
80
[PubMed]
11.
Gu
S.
,
Xue
Y.
,
Gao
Y.
,
Shen
S.
,
Zhang
Y.
,
Chen
K.
et al.
(
2020
)
Mechanisms of indigo naturalis on treating ulcerative colitis explored by GEO gene chips combined with network pharmacology and molecular docking
.
Sci. Rep.
10
,
15204
[PubMed]
12.
Yang
L.
,
Yin
P.
,
Cao
X.
and
Liu
Y.
(
2019
)
Screen for potential candidate alternatives of sargentodoxa cuneata from its six adulterants based on their phenolic compositions and antioxidant activities
.
Int. J. Mol. Sci.
20
,
13.
Ru
J.
,
Li
P.
,
Wang
J.
,
Zhou
W.
,
Li
B.
,
Huang
C.
et al.
(
2014
)
TCMSP: a database of systems pharmacology for drug discovery from herbal medicines
.
J. Cheminformatics
6
,
13
[PubMed]
14.
Daina
A.
,
Michielin
O.
and
Zoete
V.
(
2019
)
SwissTargetPrediction: updated data and new features for efficient prediction of protein targets of small molecules
.
Nucleic Acids Res.
47
,
W357
W364
[PubMed]
15.
Yao
Z.J.
,
Dong
J.
,
Che
Y.J.
,
Zhu
M.F.
,
Wen
M.
,
Wang
N.N.
et al.
(
2016
)
TargetNet: a web service for predicting potential drug-target interaction profiling via multi-target SAR models
.
J. Comput. Aided Mol. Des.
30
,
413
424
[PubMed]
16.
Piñero
J.
,
Ramírez-Anguita
J.M.
,
Saüch-Pitarch
J.
,
Ronzano
F.
,
Centeno
E.
,
Sanz
F.
et al.
(
2020
)
The DisGeNET knowledge platform for disease genomics: 2019 update
.
Nucleic Acids Res.
48
,
D845
D855
[PubMed]
17.
Szklarczyk
D.
,
Gable
A.L.
,
Lyon
D.
,
Junge
A.
,
Wyder
S.
,
Huerta-Cepas
J.
et al.
(
2019
)
STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets
.
Nucleic Acids Res.
47
,
D607
D613
[PubMed]
18.
Cheng
F.
,
Li
W.
,
Zhou
Y.
,
Shen
J.
,
Wu
Z.
,
Liu
G.
et al.
(
2012
)
admetSAR: a comprehensive source and free tool for assessment of chemical ADMET properties
.
J. Chem. Inf. Model.
52
,
3099
3105
[PubMed]
19.
Yang
H.
,
Lou
C.
,
Sun
L.
,
Li
J.
,
Cai
Y.
,
Wang
Z.
et al.
(
2019
)
admetSAR 2.0: web-service for prediction and optimization of chemical ADMET properties
.
Bioinformatics
35
,
1067
1069
[PubMed]
20.
Cao
F.
,
Liu
J.
,
Sha
B.X.
and
Pan
H.F.
(
2019
)
Natural products: experimental efficient agents for inflammatory bowel disease therapy
.
Curr. Pharm. Des.
25
,
4893
4913
[PubMed]
21.
Li
S.
and
Zhang
B.
(
2013
)
Traditional Chinese medicine network pharmacology: theory, methodology and application
.
Chinese J. Nat. Med.
11
,
110
120
[PubMed]
22.
Gao
Y.
,
Wang
K.X.
,
Wang
P.
,
Li
X.
,
Chen
J.J.
,
Zhou
B.Y.
et al.
(
2020
)
A novel network pharmacology strategy to decode mechanism of Lang Chuang Wan in treating systemic lupus erythematosus
.
Front. Pharmacol.
11
,
512877
[PubMed]
23.
Zhang
T.
,
Pan
L.
,
Cao
Y.
,
Liu
N.
,
Wei
W.
and
Li
H.
(
2020
)
Identifying the mechanisms and molecular targets of Yizhiqingxin formula on Alzheimer’s disease: coupling network pharmacology with GEO database
.
Pharmacogenom. Personal. Med.
13
,
487
502
[PubMed]
24.
Yao
J.
,
Wei
C.
,
Wang
J.Y.
,
Zhang
R.
,
Li
Y.X.
and
Wang
L.S.
(
2015
)
Effect of resveratrol on Treg/Th17 signaling and ulcerative colitis treatment in mice
.
World J. Gastroenterol.
21
,
6572
6581
[PubMed]
25.
Wei
J.
and
Feng
J.
(
2010
)
Signaling pathways associated with inflammatory bowel disease
.
Recent Patents on Inflammation & Allergy Drug Discovery
4
,
105
117
[PubMed]
26.
Li
X.
,
Liu
Y.
,
Chu
S.
,
Yang
S.
,
Peng
Y.
,
Ren
S.
et al.
(
2019
)
Physcion and physcion 8-O-β-glucopyranoside: A review of their pharmacology, toxicities and pharmacokinetics
.
Chem. Biol. Interact.
310
,
108722
[PubMed]
27.
Selim
N.M.
,
Elgazar
A.A.
,
Abdel-Hamid
N.M.
,
El-Magd
M.R.A.
,
Yasri
A.
,
Hefnawy
H.M.E.
et al.
(
2019
)
Chrysophanol, Physcion, Hesperidin and Curcumin modulate the gene expression of pro-inflammatory mediators induced by LPS in HepG2: in silico and molecular studies
.
Antioxidants
8
,
28.
Dong
X.
,
Fu
J.
,
Yin
X.
,
Cao
S.
,
Li
X.
,
Lin
L.
et al.
(
2016
)
Emodin: a review of its pharmacology, toxicity and pharmacokinetics
.
Phytother. Res.
30
,
1207
1218
[PubMed]
29.
Zheng
X.Y.
,
Yang
S.M.
,
Zhang
R.
,
Wang
S.M.
,
Li
G.B.
and
Zhou
S.W.
(
2019
)
Emodin-induced autophagy against cell apoptosis through the PI3K/AKT/mTOR pathway in human hepatocytes
.
Drug Des. Dev. Ther.
13
,
3171
3180
[PubMed]
30.
Luo
S.
,
Deng
X.
,
Liu
Q.
,
Pan
Z.
,
Zhao
Z.
,
Zhou
L.
et al.
(
2018
)
Emodin ameliorates ulcerative colitis by the flagellin-TLR5 dependent pathway in mice
.
Int. Immunopharmacol.
59
,
269
275
[PubMed]
31.
Zhu
T.
,
Zhang
W.
,
Feng
S.J.
and
Yu
H.P.
(
2016
)
Emodin suppresses LPS-induced inflammation in RAW264.7 cells through a PPARγ-dependent pathway
.
Int. Immunopharmacol.
34
,
16
24
[PubMed]
32.
Song
J.W.
,
Seo
C.S.
,
Cho
E.S.
,
Kim
T.I.
,
Won
Y.S.
,
Kwon
H.J.
et al.
(
2016
)
meso-Dihydroguaiaretic acid attenuates airway inflammation and mucus hypersecretion in an ovalbumin-induced murine model of asthma
.
Int. Immunopharmacol.
31
,
239
247
[PubMed]

Author notes

*

These authors contributed equally to this work.

This is an open access article published by Portland Press Limited on behalf of the Biochemical Society and distributed under the Creative Commons Attribution License 4.0 (CC BY).