Work

Integrative transcriptomic, genetic, single-cell, and experimental analyses prioritize HILPDA in preeclampsia

Year: 2026
Type: article
Abstract: Background Preeclampsia (PE) is a major cause of maternal and perinatal morbidity and mortality. Its molecular drivers remain incompletely understood, which limits the development of targeted therapies. This study aimed to prioritize PE-associated candidate genes and characterize their cell type-specific relevance in the placenta. Methods Eight placental transcriptomic datasets were integrated for differential expression analysis and weighted gene co-expression network analysis. Three gradient-boosting algorithms were used to construct PE prediction models. Five additional machine-learning methods were applied for biomarker selection. Mendelian randomization was performed using two independent PE genome-wide association studies. Single-cell RNA sequencing was used to examine cell type-specific expression, transcription factor activity, intercellular communication, trophoblast co-expression networks, virtual gene perturbation, and trophoblast subtypes. The function of the prioritized gene was evaluated in lipopolysaccharide-treated HTR-8/SVneo cells. Connectivity Map screening, molecular docking, and molecular dynamics simulations were used to identify candidate compounds. Results Twenty-two differentially expressed and disease-associated genes were identified. The prediction models showed good performance in the mixed validation dataset, with area under the curve values of 0.841, 0.789, and 0.825 for XGBoost, LightGBM, and GBM, respectively. HILPDA was selected by all five feature-selection methods. Mendelian randomization provided genetic evidence supporting an association between HILPDA expression and PE. Single-cell analysis showed that HILPDA was predominantly expressed in trophoblasts. HILPDA expression was associated with altered trophoblast communication, inflammatory signaling, mitochondrial and translational programs, cell adhesion, and subtype-specific regulatory networks. In LPS-stimulated HTR-8/SVneo cells, HILPDA knockdown partially restored cell migration and VEGF and PlGF expression and reduced inflammatory cytokine expression, NF-κB activation, and apoptosis-related changes. Four compounds, NVP-BEZ235, vanoxerine, NVP-TAE226, and cycloserine, showed favorable predicted interactions with HILPDA. Conclusions HILPDA was identified as a PE-associated candidate by transcriptomic, machine-learning, and genetic analyses and was mainly expressed in trophoblasts at the single-cell level. In LPS-stimulated HTR-8/SVneo cells, HILPDA knockdown attenuated inflammatory and pro-apoptotic responses and partially restored cell migration and angiogenic-factor expression. These findings support further investigation of HILPDA in PE-related trophoblast dysfunction, although its underlying mechanisms and therapeutic relevance require additional validation.
Language: English
Field: Medicine
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7235017563
FWCI:
Cites: 93
Cited by:
Related to: 10
Parts: 3