AUTHOR=Gao Fan , Li Jinzi 

TITLE=Identification of ferroptosis-related gene signatures in temporal lobe epilepsy with hippocampal sclerosis

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 19 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2025.1530182

DOI=10.3389/fnins.2025.1530182

ISSN=1662-453X

ABSTRACT=BackgroundFerroptosis is a form of regulated cell death that damages neurons in the central nervous system. In this study, we aimed to construct ferroptosis-related gene signatures in temporal lobe epilepsy with hippocampal sclerosis (TLE-HS) and explore their diagnostic role in TLE-HS.MethodsThe GSE205661 dataset was acquired for training purposes, while the GSE71058 was obtained to serve as the validation dataset. Subsequently, ferroptosis-related differentially expressed genes (FR-DEGs) in TLE-HS were further analyzed. We used weighed gene co-expression network analysis (WGCNA) algorithm, single-factor logistic regression analysis, and LASSO algorithm to screen characteristic FR-DEGs. Then, the receiver operating characteristic (ROC) was used to evaluate the value of these characteristic genes in disease diagnosis. Finally, a long non-coding RNA (lncRNA)–microRNA (miRNA)–messenger RNA (mRNA) network was constructed.ResultsWe identified 141 FR-DEGs in TLE-HS, and these genes were enriched in T-cell activation involved in immune response and signaling pathways related to lipids and atherosclerosis. Further WGCNA was performed to select 47 overlapping FR-DEGs, which were significantly enriched in 13 biological processes and 14 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, including the negative regulation of apoptotic process and ferroptosis. Four genes, namely PDK4, SMPD1, GPT2, and METTL14, were identified as signature genes in TLE-HS. Moreover, the ROC derived from the four genes in GSE205661 and GSE71058 for predicting TLE-HS had an area under the curve (AUC) of 0.988 and 0.929, respectively. Furthermore, the lncRNA–miRNA–mRNA network constructed from the 4 FR-DEGs consisted of 5 lncRNAs and 14 miRNAs. The signatures based on four FR-DEGs were found to be a strong predictor of TLE-HS, and they may represent valuable therapeutic targets for TLE-HS.