Analysis of gut microbiota-derived metabolites regulating pituitary neuroendocrine tumors through network pharmacology

Abstract

Pituitary neuroendocrine tumors (PitNETs) are a special class of tumors of the central nervous system that are closely related to metabolism, endocrine functions, and immunity. In this study, network pharmacology was used to explore the metabolites and pharmacological mechanisms of PitNET regulation by gut microbiota. The metabolites of the gut microbiota were obtained from the gutMGene database, and the targets related to the metabolites and PitNETs were determined using public databases. A total of 208 metabolites were mined from the gutMGene database; 1,192 metabolite targets were screened from the similarity ensemble approach database; and 2,303 PitNET-related targets were screened from the GeneCards database. From these, 392 overlapping targets were screened between the metabolite and PitNET-related targets, and the intersection between these overlapping and gutMGene database targets (223 targets) were obtained as the core targets (43 targets). Using the protein-protein interaction (PPI) network analysis, Kyoto encyclopedia of genes and genomes (KEGG) signaling pathway and metabolic pathway analysis, CXCL8 was obtained as a hub target, tryptophan metabolism was found to be a key metabolic pathway, and IL-17 signaling was screened as the key KEGG signaling pathway. In addition, molecular docking analysis of the active metabolites and target were performed, and the results showed that baicalin, baicalein, and compound K had good binding activities with CXCL8. We also describe the potential mechanisms for treating PitNETs using the information on the microbiota (Bifidobacterium adolescentis), signaling pathway (IL-17), target (CXCL8), and metabolites (baicalin, baicalein, and compound K); we expect that these will provide a scientific basis for further study.

Keywords: CXCL8; gut microbiota; metabolites; pituitary neuroendocrine tumors; tryptophan metabolism.

FIGURE 1

(A) Workflow of the steps in this study. (B) Overlapping targets (392) between metabolite targets predicted by the similarity ensemble approach (SEA) database and PitNET-related targets. (C) Identification of 43 core targets between the 392 overlapping targets and gutMGene database (223 targets).

FIGURE 2

Protein–protein interaction network in this study.

FIGURE 3

(A) KEGG enrichment analysis of the 43 core targets. (B) GOBP enrichment analysis of the 43 core targets. (C) Pathway interaction diagram. (D) Metabolic pathway analysis of the 31 metabolites identified from the 43 core targets. The colors of the dots represent their p-values from the pathway enrichment analyses; color variation from yellow to red indicates that the p-value is decreasing. The size of the dot represents the pathway impact value from the pathway topology analysis, where the larger the dot, the greater the pathway impact.

FIGURE 4

Molecular docking visualization of (A) baicalin–CXCL8, (B) baicalein–CXCL8, and (C) compound-K–CXCL8.

FIGURE 5

(A) Proposed microbiota-signaling pathways-targets-metabolites network. To show the relationships between the microbiota and pathways more clearly, only the top 10 microbiota are shown based on their interaction degrees. (B) Network showing the signaling pathways and targets.