Network Pharmacological Analysis of TCM Formulae Based on the Syndrome Ontology and Multidimensional Quantitative Association Computing Platform

Summary

We propose a "Formula-Ingredient-Target-Pathway-Syndrome-Disease" network construction method based on the SoFDA platform, network pharmacology, and Traditional Chinese Medicine (TCM). Compared to the currently popular network pharmacology methods, this network incorporates the critical element of syndrome, which embodies the fundamental principle of syndrome differentiation in TCM.

Abstract

Current network pharmacology methods for TCM formulae predict potential active ingredients and mechanisms of action for specific diseases by constructing a formula-ingredient-target-pathway-disease network.

Syndrome differentiation and treatment, a core principle of TCM, embodies its holistic and dynamic approach, emphasizing individual variability and disease progression. By integrating syndrome characteristics into this network, researchers can develop multi-target drugs tailored to each disease stage, enabling precise and personalized treatment. The TCM Syndrome Ontology and Multidimensional Quantitative Association Computing Platform (SoFDA) allow the incorporation of TCM syndromes into network pharmacology analysis.

In this study, we used Lingguizhugan Decoction (formula) as an example to demonstrate its application in treating obesity with Yang Deficiency Syndrome and Dampness-Phlegm Syndrome (syndromes). We first identified key active ingredients from the TCMSP database. We then constructed a network linking diseases, syndromes, target genes, and signaling pathways using SoFDA. Finally, we integrated these data to build a formula-ingredient-target-pathway-syndrome-disease network.

Introduction

By integrating the concept of dialectical treatment in Chinese medicine and using the SoFDA^1,2 platform, we have explored a novel network pharmacology approach, namely, integrative pharmacology, to construct a formula-drug-target-pathway-syndrome-disease network diagram of Chinese medicine formulations.

Compared with the popular network pharmacology methods in recent years, this method incorporates the element of syndrome in TCM. Evidence, or Syndrome, is one of the unique concepts of TCM, and is a generalization of the pathological properties of a disease at a certain stage in its development.

Syndrome is the basic mode of diagnosis and treatment in Chinese medicine. There are several syndromes for a disease, and as the disease occurs, develops, and heals, the syndrome will change accordingly.

TCM practitioners will use different formulae to treat different syndromes. Therefore, the incorporation of syndrome into network pharmacology is an important step in the original theoretical research and methodological improvement of TCM^3,4.

The TCM Syndrome Ontology and Multidimensional Quantitative Association Computing Platform (SoFDA) is an open-source platform developed by the Chinese Academy of Traditional Chinese Medicine^1,2. It is the first available collection of traditional Chinese medicine (TCM) syndrome ontology, the syndrome classification tool, and the related feature associations with disease and prescriptions for investigating the pathological links and therapeutic mechanisms.

It evaluates the association levels of syndrome-syndrome, syndrome-disease, syndrome-formula, and syndrome-disease-formula, via calculating the symptom, target gene, the enriched gene ontology item, pathway, and network module-based similarities and the network density, and further utilizes a variety of visualization types, such as heatmap, multi-level network, and upset-view, to illustrate the association calculation results.

The current basic pathway for network pharmacological analysis is roughly as follows: the first step is retrieving TCM/herb-chemical databases, disease-targets databases, and disease-targets databases to collect TCM formulae/herbs ingredients/chemicals, targets of chemicals, and diseases. The protein interaction database is then used for protein-protein interaction (PPI) expansion and analysis. The results are finally validated in vitro, in vivo, or in silico.

The results of cyberpharmacology are presented as herbal medicine/formulation-gene target-pathway-component-modern disease, and its ultimate goal is to predict the basis and mechanism of action of TCM formulae. In order to further reveal the material basis, mechanism of action, and the theory of medicinal properties of TCM, we chose to introduce the SoFDA platform into cyberpharmacology. We call this computer pharmacological analysis method, which results in prescription-drug-target-pathway-syndrome-disease, integrated pharmacology^1,2.

Obesity as a common chronic liver disease worldwide, imposes a huge economic burden on the global society.

In the concept of traditional Chinese medicine, obesity is a complex disease with syndromes of Yang deficiency, phlegm-dampness syndrome, spleen and stomach
deficiency syndrome, and blood stasis syndrome etc^5,6. So it is appropriate to adopt the treatment of strengthening the Yang Qi, resolving phlegm and dampness, and regulating the cardinal organs⁷.

Lingguizhugan decoction (Fuling, Guizhi, Baizhu, Gancao, mass ratio 4:3:3:2) is derived from the Essentials of the Golden Chamber, and is effective in the treatment of obesity through clinical practice. It is also be called FU LING GUI ZHI BAI ZHU GAN CAO TANG in Chinese.

In the following protocol, we will demonstrate the use of an innovative SoFDA-based networked pharmacological analysis of Chinese herbal formulae (integrated pharmacology) on Lingguizhugan decoction for the treatment of Yang Deficiency Syndrome and Dampness-phlegm Syndrome in obesity. We will demonstrate the methodology by conducting a network analysis of components, targets, and pathways of Lingguizhugan decoction for treating nonalcoholic fatty Yang Deficiency Syndrome and Dampness-Phlegm Syndrome.

Protocol

Taking Lingguizhugan decoction to treat "Yang deficiency Syndrome" and "Dampness-Phlegm Syndrome" of abdominal obesity as an example. Lingguizhugan decoction comprises 4 herbs: Poria cocos (Fuling), Cinnamomi ramulus (Guizhi), Rhizoma Atractylodis macrocephalae (Baizhu), Glycyrrhizae radix et Rhizoma (Gancao). The website of SoFDA platform is http://www.tcmip.cn/Syndrome/front/#/. The website of the traditional Chinese medicine systems pharmacology (TCMSP) database is https://old.tcmsp-e.com/tcmsp.php

1. Screening out the active ingredients of the formula

1. Access the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP)⁸ database.
2. Enter Fuling, Guizhi, Baizhu, and Gancao, respectively, under the Herb name column, and then click search.
3. Click the button next to OB, select is greater than or equal to under Show items with value that:, and enter 30.00 in the number column below.
4. Click the filter button of DL, select is greater than or equal to under Show items with value that:, and enter 0.18 in the number column below.
   NOTE: Screen oral bioavailability (OB) ≥ 30% and drug-like properties (DL) ≥ 0.18 chemical ingredients of the 4 herbs^2,8according to absorption, distribution, metabolism, and excretion (ADME)^9,10.
5. Copy the molecular names of the screened chemical components of the four herbs Fuling, Guizhi, Baizhu, and Gancao, in sequence, and paste them in a spreadsheet to summarize them into a table.

2. Screening out targets of the formula, syndromes, and diseases

1. Enter the SoFDA platform, select Disease-Syndrome-Formula Association in the Association Evaluation column.
   NOTE: The term Association level encompasses five categories: Completely Different, Low, Moderate, High, and Identical. Selecting Identical, High and Moderate indicates a high and moderate degree of correlation between the Association item and the disease or herbal formula, respectively. The greater the correlation of the Association item with the disease or herbal formula, the more accurate the analysis results will be. Therefore, Identical, High and Moderate should be chosen.
2. In the Disease-Syndrome-Formula Association with the Same Disease subgroup, select obesity as the Disease. Select Yang Deficiency Syndrome and Dampness-Phlegm Syndrome for Syndrome. Select Primary & Secondary Symptom for Symptom.
3. Select Symptom-related Genes for Association item. Select Identical, High and Moderate for Association level. Select FU LING GUI ZHI BAI ZHU GAN CAO TANG/formula 134 for the Formula.
   NOTE: By entering abdominal obesity into the disease search bar on the SoFDA platform, two results are retrieved: Abdominal Obesity-Metabolic Syndrome 1/ABD014 and Abdominal Obesity-Metabolic Syndrome 3/ABD013. Both types of obesity are characterized by abdominal obesity, but they differ in their respective symptom complexes. Syndrome 3 is associated with the symptoms: Myocardial Infarction, Truncal Obesity, Hypercholesterolemia, Coronary Artery Stenosis, Abdominal Obesity, Autosomal Dominant Inheritance, Hypertension, Stroke, while Syndrome 1 is associated with: Autosomal Dominant Inheritance, Hypertension, Abdominal Obesity. Therefore, Syndrome 1 was selected for subsequent analysis.
4. Click submit, then get the set of genes related to Yang deficiency Syndrome, Dampness-Phlegm Syndrome, abdominal obesity, and formula.
5. Click Yang Deficiency Syndrome and Dampness-Phlegm Syndrome, respectively, copy the Primary & Secondary Symptom-related Genes of the two syndromes and save them to the files.
6. Access the STRING online website, click the multiple protein section on the left side, and enter the genes obtained from the two syndromes in step 2.2 into the search box. Select homo sapiens for the organisms, then click on search-continue to obtain the Protein-Protein Interaction (PPI) network diagram. Under the settings section, choose the high confidence score core (0.700), and finally export the TSV files separately^11,12.
7. Launch the Cytoscape software, click on File > Import > Network from File to open the TSV file obtained in the previous step within Cytoscape. Then, under the Style tab of the software, customize the color and layout settings for the PPI network. Finally, click File > Export > Export to Image to obtain the final PPI network diagram.

3. Functional prediction

1. Go to the Functional Enrichment of Common Genes/Network Module among Associated Syndromes-Diseases-formulae section on the SoFDA platform.
2. Select biological process, cellular component, molecular function, and reactome pathway under Enrichment type.
3. Click submit to obtain the results.
4. Click the ↓ button at the bottom right corner of the image to download the spreadsheet files related to biological process, cellular component, molecular function, and reactome pathway.
5. After finishing the selection, click submit to see the Functional Enrichment of Common Genes/ Network Module among Associated Syndromes-Diseases-formulae module, the data of the graph of biological process (BP), cellular component (CC), molecular function (MF) description, and reactome pathway are all downloaded.
6. Open the Wei Sheng Xin online website (https://www.bioinformatics.com.cn/), choose the bar with color gradient column, then import the BP, CC, MF description, and reactome pathway data obtained into this website to obtain the bar chart with color gradient.

4. Formula-Target-Pathway-Syndrome-Disease network construction

1. Following the 2.1 step, after finishing the selection, click Go to Network at the bottom of the first page and enter the second page.
2. Choose Yang Deficiency Syndrome and Dampness-Phlegm Syndrome for Syndrome. Choose Primary & Secondary Symptom for Symptom and choose Abdominal Obesity for Disease. Choose FU LING GUI ZHI BAI ZHU GAN CAO TANG/formula 134 for the Formula. Then, under Show, choose all the Disease, Syndrome, Formula, Symptom, and Reactome pathway.
3. Then click Submit and get the network diagram of Formula-Target-Pathway-Syndrome-Disease.

5. Formula-Ingredient-Target-Pathway-Syndrome-Disease network construction

NOTE: On the SoFDA platform, searching for the formula Poria (Fu Ling), Cinnamon Twig (Gui Zhi), Atractylodes Macrocephala (Bai Zhu), and Licorice (Gan Cao) Decoction does not yield the primary active ingredients. Therefore, it is recommended only to retrieve the main active ingredients of Poria, Cinnamon Twig, Atractylodes Macrocephala, and Licorice from the TCMSP database. Follow the steps 1.1.3-1.1.4 to obtain the main active ingredients of the Poria (Fu Ling), Cinnamon Twig (Gui Zhi), Atractylodes Macrocephala (Bai Zhu), and Licorice (Gan Cao) Decoction. Then, import them into the Formula-Target-Pathway-Syndrome-Disease network to generate the Formula-Ingredient-Target-Pathway-Syndrome-Disease network diagram. Consequently, there is no data heterogeneity between gene IDs and pathways in SoFDA and TCMSP.

1. Download the network data of Formula-Target-Pathway-Syndrome-Disease from SoFDA platform generated by section 4 and put it into a spreadsheet.
2. After that, add the active ingredients of Lingguizhugan decoction screened in section 2 into this spreadsheet, and rename the spreadsheet as Network. Create another new spreadsheet named type to save the types of each node produced in section 4.
3. Open the Cytoscape (type 3.9.1) software, click Import Network from file to import the network file, and click import table from file to import the type file. Then, click the style button to adjust and modify the Formula-Ingredient-Target-Pathway-Syndrome-Disease network¹³.
4. Click the Export with image file button to get the Formula-Ingredient-Target-Pathway-Syndrome-disease network.

Results

We selected 103 active ingredients of four Chinese herbs in the Lingguizhugan decoction by the criteria of bioavailability ≥30% and pharmacophore ≥0.18%. The main active ingredients were quercetin, kaempferol, naringenin, β-sitosterol, isorhamnetin. It is consistent with the laboratory extraction and screening components of Jiang et al.¹⁴.

For Dampness-Phlegm Syndrome and Yang Deficiency Syndrome of Lingguizhugan decoction (Figure 1), 328 t...

Discussion

SoFDA is the first open-source ontology database for TCM syndromes, with its data and analysis results fully accessible to the public. This open-source nature provides researchers with greater flexibility. The platform employs a similarity-based association evaluation system, utilizing Jaccard or cosine similarity measures to quantitatively assess the relationships among disease-syndrome, and formula-syndrome associations. This dynamic modeling capability allows SoFDA to better simulate the dynamic changes in diseases an...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Cytoscape(type 3.9.1)	National Human Genome Rearch Institute (NHGRI)	https://cytoscape.org/
Gene Ontology (GO) knowledgebase	Global Core Biodata Resources	https://www.geneontology.org/
KEGG: Kyoto Encyclopedia of Genes and Genomes	Kanehisa Laboratories	https://www.genome.jp/kegg/
SoFDA platform	China Academy of Chinese Medical Science	http://www.tcmip.cn/Syndrome/front/#/
STRING12.0 website	Global Biodata Coalition?ELIXIR	https://cn.string-db.org/
TCMSP platform	Zhejiang Jiuwei Health Co., LTD	https://old.tcmsp-e.com/tcmsp.php
Wei Sheng Xin online website	NEWCORE	https://www.bioinformatics.com.cn/