Abstract
The pathogenesis of preeclampsia (PE) is complex and placental internal homeostasis is regulated by cellular autophagy. However, there are fewer studies related to the role of placental autophagy in the pathogenesis of PE. The GSE75010 and GSE10588 datasets were downloaded from the gene expression omnibus (GEO) database. In the GSE75010 (test cohort), 103 differentially expressed genes (DEGs) were screened using "Limma" package, and 281 PE characteristic genes were screened by weighted gene coexpression network analysis (WGCNA). Combined with the autophagy gene set, a total of 5 autophagy-related hub genes were obtained. Three biomarkers (HK2, PLOD2, and TREM1) were then further screened by random forest(RF) model and least absolute shrinkage and selection operator(LASSO) algorithm as diagnostic of PE. In the unsupervised consensus clustering analysis, HK2, PLOD2, and TREM1 may be synergistically involved in hypoxia-induced autophagy and hypoxia-inducible factor 1(HIF-1) signaling pathway to induce PE. In addition, we constructed and evaluated a nomogram model for PE diagnosis using these three key diagnostic biomarkers, and the results showed that the model had significantly excellent predictive power (AUC values of GSE75010 and GSE10588 datasets were 0.869 and 0.876, respectively). In terms of immune infiltration, a higher proportion of T cells CD8, and a lower proportion of Macrophages M2 were found in PE placentas compared to normal tissue, and high expression of HK2, PLOD2, and TREM1 were accompanied by low levels of Macrophages M2 infiltration. HK2, PLOD2, and TREM1 may be associated with the development of pre-eclampsia, and their mechanisms of action in preeclampsia need to be further investigated.