Open Access

ARTICLE

Identification of EML4 as a key hub gene for endometriosis and its molecular mechanism and potential drug prediction based on the GEO database

by XIANBAO FANG^1,#, MINGYAN TANG^1,#, ZIYANG YU^1,#, JIAQI DING¹, CHONG CUI², HONG ZHANG^1,*

1 Reproductive Medicine Center, The Second Affiliated Hospital of Soochow University, Suzhou, 215002, China
2 Imaging Diagnosis Department, The Second Affiliated Hospital of Soochow University, Suzhou, 215002, China

* Corresponding Author: Hong Zhang,
# Equal contribution

(This article belongs to the Special Issue: Bioinformatics Study of Diseases)

BIOCELL 2023, 47(9), 2059-2068. https://doi.org/10.32604/biocell.2023.030565

Received 13 April 2023; Accepted 29 May 2023; Issue published 28 September 2023

Abstract

Objective: Key genes were screened to analyze molecular mechanisms and their drug targets of endometriosis by applying a bioinformatics approach. Methods: Gene expression profiles of endometriosis and healthy controls were obtained from the Gene Expression Omnibus database. Significant differentially expressed genes were screened using the limma package. Correlation pathways were screened by Spearman correlation analysis on the echinoderm microtubule-associated protein-like 4 (EML4) and enrichment in endometriosis pathways and estimated by the GSVA package. Immune characteristics were assessed by the “ESTIMATE” R package. Potential regulatory pathways were determined by enrichment analysis. The SWISS-MODE website was used in homology modeling with EML4 and EML4 protein activity was predicted. VarElect was employed in molecular docking for screening potential compound inhibitors targeting endometriosis. Results: Ten endometriosis and 10 normal samples were included. EML4 was significantly upregulated in endometriosis (p < 0.05). Thirty significantly correlated pathways involving 18 positive and 12 negative correlations, including GLYCOSAMINOGLYCAN_BIOSYNTHESIS_HEPARAN_SULFATE and GLYCOSPHINGOLIPID_BIOSYNTHESIS_GANGLIO_SERIES were screened between EML4 and endometriosis. Immunocorrelation analysis showed a significant difference in immune-related pathways in endometriosis and normal samples (p < 0.05). In endometriosis, EML4 was associated with T-cell CD4 resting memory, activated mast cells, plasma cells, activated NK cells, M2 macrophages, and follicular helper T cells (p < 0.05). Molecular docking identified five potential inhibitors of EML4, and compound DB05104 (asimadoline) bound well to EML4 protein to exert its physiological effects. Conclusion: Differential gene expression and immune correlation analyses revealed that EML4 may affect endometriosis through multiple targets and pathways, the mechanism of which involved immune cell activation and infiltration. Molecular docking and dynamics simulation verified DB05104 as a potential inhibitor of EML4 and a powerful target for endometriosis treatment.

---

Graphic Abstract

---

Keywords

---