Investigating the mechanism of artemisinin in treating osteoarthritis based on bioinformatics, network pharmacology, and molecular docking

Abstract

This study aims to explore the mechanism of artemisinin in treating osteoarthritis (OA) through bioinformatics and network pharmacology. The targets of artemisinin were obtained from databases such as TCMSP, and the disease targets of OA were screened from OMIM, TTD, DisGeNET, and GEO databases. The predicted targets of artemisinin were intersected with OA disease targets to obtain drug-disease common targets, which were visualized using a Venn diagram. Gene ontology (GO) analysis and KEGG functional analysis was performed on the 68 common target genes, and protein interaction network analysis was conducted to analyze their interaction relationships. The key genes were identified using the Cytohubba algorithm, followed by molecular docking with AutoDockTools 1.5.7 software and PyMOL software. Through database screening, 464 targets of artemisinin were identified, and 1654 OA target genes were screened from databases and GEO chip databases. The intersection of drug targets and disease targets yielded 68 drug-disease common targets. GO and KEGG analysis showed that these common target genes are mainly involved in oxidative stress response, bone formation, response to bacterial molecules, response to lipopolysaccharide, response to hypoxia, response to xenobiotic stimuli. Their molecular functions include regulation of transcription factor binding, ubiquitin-protein ligase activity, cytokine receptor binding. These common targets are enriched in 36 signaling pathways, including MAPK signaling pathway, PI3K-Akt signaling pathway, TNF signaling pathway, IL-17 signaling pathway, NF-Kappa B signaling pathway, which are key regulatory pathways in the development of OA. Through protein interaction analysis and Cytohubba algorithm, 10 key genes were obtained. Furthermore, the top 5 key genes (BCL-2, IL-6, CASP3, HIF1A, TNF) were molecular-docked with artemisinin, and the results showed that these molecules could form stable binding through hydrogen bonding and hydrophobic interaction. Artemisinin may exert drug efficacy through multi-target and multi-pathway synergism in the treatment of OA. This study provides an effective theoretical basis for the treatment of OA with artemisinin.

Keywords: artemisinin; bioinformatics; molecular docking; network pharmacology; osteoarthritis.

Figure 1.

The differentially expressed genes in the GSE55457 dataset was illustrated by volcano plot. Upregulated genes were indicated in red and downregulated genes in blue, and genes with no significant difference in expression were shown in gray.

Figure 2.

The artemisinin-osteoarthritis network was depicted and a total of 68 common targets were identified.

Figure 3.

Bubble chart of GO analysis for disease-drug common target (top 10). GO = gene ontology.

Figure 4.

Bubble chart of KEGG analysis of disease-drug common target (top 20).

Figure 5.

Network diagrams of signal pathways enriched by common targets and their corresponding targets determined by KEGG enrichment analysis. The signal pathways were represented by diamond nodes, and the targets are represented by rectangular nodes. The darker the color of the rectangular node indicated the target in node involved in more signaling pathways.

Figure 6.

Drug-disease common target genes interaction network diagrams, the darker the node color represents the greater the number of interacting proteins.

Figure 7.

Key genes in drug-disease common targets were obtained using Cytohubba plugin MCC algorithm. The darker the node color, the higher the gene ranking. MCC = maximum modularity centrality.

Figure 8.

Molecular docking results. The binding enegeres (kcal/mol) of key targets and artemisinin were showed in bar chart (A). The binding sites (B–F).