Exploration of the effect of Celastrol on protein targets in nasopharyngeal carcinoma: Network pharmacology, molecular docking and experimental evaluations

Abstract

Background: Celastrol, an important extract of Tripterygium wilfordii, shows strong antitumor activity in a variety of tumors including nasopharyngeal carcinoma (NPC). However, little is known about its targets in NPC. We aimed to screen the key gene targets of Celastrol in the treatment of NPC by means of in silico analyses (including network pharmacology and molecular docking) and experimental evaluations. Methods: The main target genes of Celastrol and the genes related to NPC were obtained by retrieving the relevant biological databases, and the common targets were screened. Protein-protein interaction analysis was used to screen the hub genes. Then, a "compound-target-disease" network model was created and molecular docking was used to predict the binding of Celastrol to the candidate hub proteins. Afterward, the expression changes of the candidate genes under the administration of Celastrol were verified in vitro and in vivo. Results: Sixty genes common to Celastrol and NPC were screened out, which may be related to numerous biological processes such as cell proliferation, apoptosis, and tube development, and enriched in various pathways such as PI3K- Akt, EGFR tyrosine kinase inhibitor resistance, and Apoptosis. The tight binding ability of the candidate hub proteins (TNF, VEGFA, and IL6) to Celastrol was predicted by molecular docking [Docking energy: TNF, -6.08; VEGFA,-6.76; IL6,-6.91(kcal/mol)]. In vitro experiments showed that the expression of TNF and VEGFA decreased while the expression of IL6 increased in NPC cells (CNE2 and HONE1) treated with Celastrol. In vivo experiments suggested that Celastrol significantly reduced the weight and volume of the transplanted tumors in tumor-bearing mice in vivo. The expression of TNF, VEGFA, and IL6 in the transplanted tumor cells could be regulated by using Celastrol, and the expression trends were consistent with the in vitro model. Conclusion: Several gene targets have been filtered out as the core targets of Celastrol in the treatment of NPC, which might be involved in a variety of signaling pathways. Hence, Celastrol may exert its anti-NPC activity through multiple targets and multiple pathways, which will provide new clues for further research. Future experiments are warranted to validate the findings.

Keywords: Celastrol; drug targets; nasopharyngeal carcinoma; network pharmacolgy; traditional Chinese medicine.

FIGURE 1

The flowchart of this study. The target genes of Celastrol were firstly obtained by screening the TCMSP, BATMAN-TCM, and Swisstarget databases; then NPC-related genes were obtained by the Genecards database; the two gene sets were intersected to obtain 60 genes. These genes were then subjected to functional enrichment analysis and PPI network construction to filter the hub genes, and the candidate hub genes were then subjected to molecular docking and experimental validation.

FIGURE 2

(A) The 3D structure of Celastrol. (B) Venn analysis showed that the intersection involved the genes shared by NPC and Celastrol. There were 138 targets of Celastrol and 2150 NPC-related genes. The intersection included 60 genes common to both NPC and Celastrol. (C) The PPI network of the common genes constructed in the STRING. (D) The PPI network of the top 30 genes created by Cytoscape (sorted by the “degree” values). The node represents the genes/proteins and the edges stand for the relationships.

FIGURE 3

(A) GO analysis of the genes common to Celastrol and NPC. (B) KEGG pathway enrichment analysis. (C) The Compound-Target-Disease network. (D) Molecular docking between Celastrol and the top three hub genes (TNF, VEGFA, and IL6). The red arrows indicated the locations of Celastrol. The ribbons represented the 3D structures of the proteins.

FIGURE 4

(A) The cell viability of the cells in the Celastrol-treated group was significantly lower than that of the cells in the control group. (B,C) The mRNA and protein expression levels of the candidate hub genes (TNF, VEGFA, and IL6) assessed by qRT-PCR (C) and western blot (B), respectively, in CNE2 and HONE1 cells. (D) The excised transplanted tumors in two groups of mouse models. (E) The volumes of the transplanted tumors in the mice treated with or without Celastrol. (F) The weights of the tumors in the two groups. (G) The weight of the mice in the two groups. (H) The mRNA expression levels of the candidate hub genes (TNF, VEGFA, and IL6) in the transplanted tumors assessed by qRT-PCR.The t-test was used to compare between the two groups. (n = 3,*p < 0.05). The error bars represented the standard deviations.