Chinese herbal medicine, Tongxieyaofang, alleviates diarrhea via gut microbiota remodeling: evidence from network pharmacology and full-length 16S rRNA gene sequencing

Abstract

Background: Tongxieyaofang (TXYF) was a traditional Chinese medicine (TCM) formula for the treatment of diarrhea with liver stagnation and spleen deficiency syndrome, but the potential targets and mechanisms have not been fully clarified. This study aims to explore the potential mechanisms of TXYF in alleviating diarrhea using network pharmacology and full-length 16S rRNA gene sequencing.

Methods: Network pharmacology was applied to identify bioactive compounds and potential targets involved in the role of TXYF in alleviating diarrhea. Meanwhile, a model of diarrhea with liver stagnation and spleen deficiency syndrome was constructed by intragastric administration of Folium senna extract combined with restraint and tail pinch stress. The effect of TXYF on intestinal mucosal microbiota of diarrhea mice was analyzed by full-length 16S rRNA gene sequencing.

Results: Network pharmacology analysis showed that kaempferol, wogonin, naringenin, and nobiletin were compounds associated with the efficacy of TXYF. TXYF may alleviate diarrhea via multiple BPs and pathways, including TNF signaling pathway, IL-17 signaling pathway, and Toll-like receptor signaling pathway, which are involved in TCM-gut microbiota-host interactions. Then, we found that TXYF administration reshaped the diversity and composition of the intestinal mucosal microbial community of diarrhea mice. Lactobacillus, primarily Lactobacillus johnsonii, was enriched by the administration of TXYF. After TXYF administration, the abundance of Lactobacillus, particularly Lactobacillus johnsonii, was enriched.

Conclusion: Oral administration of TXYF may alleviate diarrhea through remodeling intestinal mucosal microbiota. Promoting the colonization of beneficial commensal bacteria in the intestinal mucosa through gut microbiota-host interactions may be a potential mechanism of TXYF in the treatment of diarrhea.

Keywords: Chinese medicine; Tongxieyaofang (TXYF); diarrhea; full-length 16S rRNA gene sequencing; intestinal mucosal microbiota.

Figure 1

The herb-compound-target (CT) network. The orange nodes stand for herbs in TXYF. The red nodes stand for bioactive compounds in TXYF. The blue nodes stand for targets responding to the efficacy of TXYF. The grey lines stand for corresponding connections between different category nodes. The size of each node indicates the number of connections. The larger the node, the more the connections.

Figure 2

Enrichment analysis. (A) Bar plot for top 10 enriched GO terms of each sub-class. The length of each bar stands for enriched counts, and the color of each bar stands for adjusted p-value. The abscissa represents the enriched gene counts. (B) Dot plot for top 20 enriched KEGG pathway terms. The size of each node stands for enriched counts, and the color of each node stands for adjusted p-value. The abscissa represents the enriched gene ratio.

Figure 3

The PPI network analysis (A), the Venn diagram of core hub targets (B), and the Sankey diagram of core hub targets and corresponding compounds (C). In (A), (1) to (4) networks represent the analysis results based on four algorithms, Betweenness, Degree, MCC, and Closeness, respectively. The highlight red-to-yellow nodes are hub targets (top 10). The blue nodes are also protein targets with the highest confidence (0.9) from the PPI network analysis. In (B), hub targets (top 10) identified from four algorithms were intersected to obtain the core hub targets for further analysis. In (C), the relationships between core hub targets obtained from the Venn diagram and corresponding bioactive compounds were shown directly. The up blocks stand for the core hub targets, and the underneath blocks stand for compounds, as follows: kaempferol, naringenin, wogonin, and nobiletin.

Figure 4

(A) Venn diagram of OTUs (at distance 0.03). (B) Rarefaction curve based on observed species. (C) Box plots of α-diversity index of OTUs number (1), Chao1 (2), and Simpson (3). (D) NMDS of OTU-based unweighted (1) and weighted (2) UniFrac distances. gtcm, the control group; gtmm, the placebo group; gttm, the TXYF group.

Figure 5

(A) Histogram of the number of microbial species of each sample at different taxonomic levels. (B) Radar graph of the relative abundance of dominant phylum in the intestinal mucosa. (C) Box plots of relative abundance of Bacteroidetes (1) and F/B ratio (2). (D, E) Stack diagram of relative abundance of genus (D) and species (E) in the intestinal mucosa. *p < 0.05, **p < 0.01, Kruskal-Wallis H test. gtcm, the control group; gtmm, the placebo group; gttm, the TXYF group.

Figure 6

The potential biomarkers associated with the administration of TXYF. (A) Identified taxonomic features of significant differences across three groups by LEfSe (LDA score ≥ 4 and p-value ≤ 0.05). (B) Identified discriminative species between the TXYF (gttm) and placebo (gtmm) groups by Metastats analysis.

Figure 7

Function prediction based on the KEGG database by PICRUSt. (A) Heatmap for the relative abundance of discriminative KOs between the TXYF (gttm) and placebo (gtmm) groups. (B) The main KEGG pathway of the KOs with significantly increased expression in the TXYF group. (C) The main KEGG pathway of the KOs with significantly decreased expression in the TXYF group.