.png){kind=icon}
.png){kind=icon}
Return
Multilayer Validation Reveals a Glia-Associated Secretome Signature in Temporal Lobe Epilepsy
.png){kind=icon}
PRE
AI
Y
Z
Q
X
C
Y
Y
G
DOI：10.1007/s12031-026-02564-3.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Although multi-omics studies have increasingly revealed molecular alterations associated with epilepsy, clinically accessible cerebrospinal fluid (CSF) biomarkers that are stably detectable and biologically relevant remain scarce. This study aimed to identify measurable biomarkers associated with epilepsy-related molecular alterations. Kainic acid (KA)-induced mouse model, single-cell RNA sequencing, and public transcriptomic datasets identified secreted protein-coding genes associated with inflammation, glial activation, and metabolic dysregulation. Candidate biomarkers were validated across mouse and human datasets and quantified in CSF of patients with temporal lobe epilepsy (TLE) and non-epileptic controls using enzyme-linked immunosorbent assays. Machine learning models were built to evaluate diagnostic performance and interpretability using SHAP analysis. Six proteins (C-X-C motif chemokine ligand 10 [CXCL10], transthyretin [TTR], apolipoprotein E [APOE], apolipoprotein D [APOD], galectin-3-binding protein [LGALS3BP], and lysozyme [LYZ]) were dysregulated in human CSF. CSF levels of CXCL10, APOE, APOD, LGALS3BP, and LYZ were significantly elevated in TLE, whereas TTR was markedly reduced (all p < 0.001). Machine learning models demonstrated promising classification performance, with TTR and CXCL10 emerging as major contributing features. CSF CXCL10 levels positively correlated with preoperative seizure frequency (Spearman’s ρ = 0.698, p < 0.0001). Multilevel evidence from animal models and clinical CSF samples indicates that microglial activation and glia-related molecular alterations are associated with TLE progression, and related secreted proteins may provide candidate biomarkers for future translational investigation.
Keywords:
Epilepsy
Biomarker
Single-cell RNA sequencing
Machine learning diagnosis
Journal
IF:
2.7
Papers:
315
Citations:
6.4K
Organization
Cited Papers