GCLM as a novel biomarker for preeclampsia: Integrating bioinformatics and mechanistic validation.

INTRODUCTION

Preeclampsia (PE) is a pregnancy-specific disorder associated with hypertension and multi-organ dysfunction, posing serious risks to maternal and fetal health. Early detection remains challenging, highlighting the urgent need to identify reliable molecular biomarkers for improved diagnosis and therapeutic intervention.

METHODS

We integrated Gene Expression Omnibus (GEO) datasets (GSE10588, GSE25906, and GSE48424) and identified 671 differentially expressed genes (312 upregulated and 359 downregulated). Weighted Gene Co-expression Network Analysis (WGCNA) identified 165 genes highly correlated with PE, of which 74 overlapped with the DEGs. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted on these 74 genes. Additionally, 4 machine learning models were applied to prioritize diagnostic biomarkers.

RESULTS

GO analysis revealed enrichment in Wnt signaling, ER-to-Golgi vesicle transport, COPII-coated vesicle components, and ubiquitin-protein transferase activity. KEGG analysis indicated significant involvement in cysteine and methionine metabolism and protein processing in the endoplasmic reticulum. Among the top 20 genes from each machine learning model, Glutamate-cysteine ligase modifier subunit (GCLM) was the only overlapping gene. Its downregulation was validated in clinical samples and PE models. Functional experiments showed that lentiviral GCLM overexpression restored GSH/GPX4 levels, enhanced HUVEC viability, and reduced sFLT-1 secretion.

CONCLUSION

Our study identifies GCLM as a potential biomarker and therapeutic target for PE, offering new insight into its molecular pathogenesis and suggesting clinical relevance for diagnosis and intervention.