RESUMEN
BACKGROUND: Preeclampsia is a life-threatening disease of pregnancy that lacks effective pharmaceuticals which can target its pathogenesis. Since preeclampsia involves complex pathological processes, including autophagy, this study aims to explore autophagy-related mechanisms of preeclampsia and to screen potential drugs. METHODS: Firstly, the datasets GSE75010, GSE24129, GSE66273, and autophagic genes lists were downloaded from public databases. Then, a weighted gene co-expression network analysis (WGCNA) was applied to filter autophagic-related hub genes of preeclampsia. The differential expression levels of the hub genes were validated with datasets GSE24129 and GSE66273. Next, the GO and KEGG enrichment, protein-protein interacting (PPI) network, as well as the downstream pathways was analyzed via the starBase, STRING and Cytoscape to determine the functions and regulatory network of the hub genes. Additionally, the immune microenvironment of preeclampsia was investigated by the CIBERSORTX database. Finally, three herb ingredients, berberine, baicalein, and luteolin were screened by molecular docking in comparison to pravastatin, metformin, and aspirin, to predict potential drugs for treating preeclampsia. RESULTS: A total of 54 autophagy-related genes were filtered by WGCNA. After filtering with |GS| > 0.5 and |MM| > 0.8, three hub genes, namely PKM, LEP, and HK2, were identified and validated. Among these genes, PKM and LEP were overexpressed in women older than 35 years old ( p<0.05; p<0.05); the expression of PKM, LEP, and HK2 differed remarkably in women with different BMI (all p<0.05); PKM overexpressed in women with hypertension (p<0.05). The regulatory network of hub genes demonstrated that they were mainly enriched in metabolic pathways, including the AMPK signaling pathway, glucagon signaling pathway, adipocytokine signaling pathway, and central carbon metabolism. Then, immune microenvironment analysis turned out that M2 macrophages were reduced in preeclampsia women (p<0.0001) and were negatively correlated with the expression of PKM (r=-0.2, p<0.05), LEP (r=-0.4, p<0.0001), and HK2 (r=-0.3, p<0.001). Lastly, molecular docking showed baicalein and luteolin could bind intimately to hub genes. CONCLUSION: PKM, LEP, and HK2 could be promising biomarkers for preeclampsia, which might regulate the pathogenesis of preeclampsia via metabolism pathways and immune microenvironment. Baicalein and luteolin could be potential therapeutics for preeclampsia.