Traditional Chinese medicine (TCM) has a longstanding history and has gained widespread clinical applications. San Cao Decoction (SCD) is an experience prescription first formulated by Prof. Duzhou Liu. We previously demonstrated its antihypertensive effects; however, to systematically explain the underlying mechanisms of action, we employed a systems pharmacology approach for pharmacokinetic screening and target prediction by constructing protein-protein interaction networks of hypertension-related and putative SCD-related targets, and Database for Annotation, Visualization, and Integrated Discovery enrichment analysis. We identified 123 active compounds in SCD and 116 hypertension-related targets. Furthermore, the enrichment analysis of the drug-target network showed that SCD acts in a multidimensional manner to regulate PI3K-Akt-endothelial nitric oxide synthase signaling to maintain blood pressure. Our results highlighted the molecular mechanisms of antihypertensive actions of medicinal herbs at a systematic level.
Hypertension is a chronic condition that triggers various fatal cardiovascular disorders, including heart failure, coronary artery disease, stroke, peripheral artery disease, and renal failure [
The WHO global atlas of traditional, complementary, and alternative medicine reports a global upsurge in herbal and traditional medicinal practices [
Because herbal formulations contain multiple components exhibiting multidimensional pharmacological effects, it is a challenge to identify the effects of individual components using traditional methods of analyses. Moreover, the pharmacological effects of herbal formulations depend on the complex and dynamic interactions among its components [
In the present study, a systems pharmacology approach was employed to screen the oral bioavailability (OB) and drug-likeness (DL) of the individual compounds of SCD. The potential biological targets of these active ingredients and interaction networks were obtained from public databases, which has been previously used to study the biological mechanisms of Niao Du Qing granules [
Data on the individual herbs in SCD were mined from the TCM systems pharmacology (TCMSP) database [
Although formulations in TCM are composed of multiple compounds, all of them may not be pharmacologically active. Therefore, the identification of the pharmacokinetic properties of each compound in TCM is essential. We screened various compounds present in SCD based on their pharmacokinetic absorption, distribution, metabolism, and excretion (ADME) parameters, such as OB (systemic bioavailability after oral absorption and distribution) [
For the current study, we chose a public database interrogation strategy performed as previously described to predict the pharmacological targets of the individual compound in SCD [
Known hypertension-related targets were identified from the following six existing resources: (1) the DrugBank database; we identified interactions between FDA-approved drugs for hypertension treatment and human gene/protein targets [
A PPI network was constructed using Bisogenet, a Cytoscape plugin, for the analysis of five existing PPI databases, including the Biological General Repository for Interaction Datasets, the Biomolecular Interaction Network Database, the Molecular INTeraction Database, the Human Protein Reference Database, and the Database of Interacting Proteins [
We constructed an interaction network for the known hypertension-related targets and putative pharmacological targets of SCD based on data obtained using the Cytoscape plugin, Bisogenet. Further, the interaction network was visualized using the Cytoscape software (Version 3.2.1), and the topological properties of each node in the interaction network were assessed using another Cytoscape plugin (CytoNCA) on the basis of betweenness centrality (BC), degree centrality (DC), closeness centrality (CC), eigenvector centrality (EC), network centrality (NC), and local average connectivity (LAC). The definitions and computational formulas of these parameters have been previously described [
We analyzed 172 putative targets of SCD using GO enrichment with DAVID to identify their involvement based on three different terms including biological process (BP), cell component (CC), and molecular function (MF) terms. With p <0.05, we applied a hypergeometric test to identify enriched GO terms. An overview of the GO analysis with up to 10 significantly enriched terms in each of these three categories is shown. Further, we performed a DAVID-based enrichment analysis of 116 candidate targets of SCD with the KEGG signaling pathway; we only selected terms with p <0.05.
Based on previous reports, 60, 77, 51, 63, and 280 compounds for each of the five medicinal herbs,
Analysis of the active compounds of SCD and preliminary GO analysis of putative SCD targets. (a) Active compounds in SCD were preliminarily screened for two ADME parameters. (b) ADME parameter distribution for different herbs. (c) The compound-target network plotting.
We identified 172 putative therapeutic targets for 112 of the 123 candidate compounds of SCD by integrating available chemical, genomic, and pharmacological information [
GO analysis for targets of SCD. Biological processes, cell component, and molecular function terms were performed on putative SCD targets; the top 10 terms with P < 0.05 are shown. Terms in the same category are ordered by p values (95% confidence level) starting with the most significant values on top. The percentage of genes/proteins involved in a term is presented at the bottom of the figure.
The pharmacological target of a drug determines its indication. We identified hypertension-related targets using various databases, including DrugBank, OMIM, GAD, KEGG Pathway, TTD, and T-HOD database. After compensating for data redundancy, 913 hypertension-related targets were identified (Table
Identification of candidate targets for SCD against hypertension. (a) SCD shared 75 putative targets with known antihypertension drugs. The compound-putative target network was constructed by linking the overlapped targets (between SCD putative and known hypertension-related) and the homologous candidate compounds of SCD. The nodes representing candidate compounds are shown as polychrome rhombus and the targets are presented as grey circles. (b) Identification of candidate SCD targets for hypertension treatment through PPI network. 116 candidate targets are finally predicted.
Genes and proteins do not function independently; instead they work on multiple levels via interconnected molecular networks and pathways [
First, a putative target PPI network of SCD-related genes was constructed with 6,634 nodes and 1,50,547 edges on a systems pharmacology platform. After further extraction of hypertension-related targets, a disease-specific network was constructed with 7,509 nodes and 1,65,579 edges. We then merged these two networks to obtain a core PPI network with 4,563 nodes and 1,21,159 edges. Subsequently, candidate hypertension-related proteins modified by SCD were screened based on the topological features of the core PPI. A node is identified as a significant target if its degree was more than twice the median degree of all nodes in the network [
Based on DAVID enrichment, we correlated BP and MF terms with proteins involved in cellular processes, such as the regulation of NOS expression, cell adhesion mediated by cadherin binding, and the regulation of apoptosis by p53 signaling. Various signaling molecules related hypertension, such as PI3K-Akt, MAPK, ErbB, FoxO, TGF-beta, Wnt, NOD-like receptor, Rap1, Toll-like receptor, and Ras signaling pathways, were identified (ordered in P-value, Figure
Enrichment analysis of candidate targets for SCD against hypertension.
Modulating PI3K-Akt signaling pathway of SCD. Targets of SCD were colored in pink, targets of hypertension were colored in yellow, and proteins in the pathway were colored in green.
Long-established Chinese herbal formulations not only stabilize the blood pressure, but also improve the quality of life, minimize hypertension-related risk factors, and prevent organ damage to improve patient survival [
To understand potential biological mechanism of SCD, a PPI network was constructed, and 116 potential targets were recognized. Through the KEGG pathway analysis, we recognized ten hypertension-related signaling pathways, PI3K-Akt, MAPK, ErbB, FoxO, TGF-beta, Wnt, NOD-like receptor, Rap1, Toll-like receptor, and Ras signaling pathways. Actually, these pathways may be involved in the progress of hypertension. Based on P-Value, we choose PI3K-Akt signal pathway as most candidate signal for further study. We constructed a concept map containing SCD targets and hypertension targets in PI3K/Akt signaling pathway and found a synergistic effect of SCD targets in this pathway to treat hypertension. We highlighted PI3K-Akt-eNOS as an important signal pathway based on its role in regulating blood pressure [
We previously demonstrated that SCD increases eNOS activation to increase NO levels
In this study, we predicted the mechanisms underlying the antihypertensive effects of SCD to verify its potential to treat hypertension. We recognized that SCD acts by modulating the PI3K-Akt-eNOS pathway to produce antihypertensive effects. This study demonstrates the usefulness of a systems pharmacology-based approach to elucidate relationships between complex diseases, such as hypertension, and Chinese herbal medicines. A limitation of our study is that the target-prediction tools used in our systems pharmacology analysis only reveal indeterminate connections between compounds and their corresponding target genes. Therefore, further experimental studies are required to accurately determine and validate the predicted mechanisms of action.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare no conflicts of interest.
Chongyang Ma, Changming Zhai, and Tian Xu contributed equally to this work.
This research was supported by the National Natural Science Foundation of China [Grants nos. 81430102, 81774122, 81774030, and 81874448].
Supplementary materials contain five tables. Table S1: detailed information of active compounds in SCD; Table S2: relationship between active compounds and putative targets; Table S3: detailed information on these known therapeutic targets; Table S4: topological features of antihypertension targets related active compounds in SCD; Table S5: topological features of 116 candidate targets.