Identification and Validation of a Lactylation-Related Gene Expression Signature for Diagnosis of Acute Ischemic Stroke.

Acute ischemic stroke (AIS) is a major cerebrovascular disorder with significant health impact. Current diagnostic methods lack sufficient sensitivity and specificity, highlighting the need for novel biomarkers. Recent studies suggest lactylation modification may be involved in AIS pathophysiology. We analyzed AIS-related gene expression data from the GEO database. Differential gene expression analysis and WGCNA identified key genes and co-expression modules. Unsupervised clustering stratified AIS subtypes, and machine learning models were developed for diagnosis. Immune infiltration was evaluated using CIBERSORT, xCell, and MCP-counter. qPCR and ELISA validation were performed on clinical samples. Four lactylation-related genes (HNRNPL, HSDL2, ARID3A, and SSB) were identified as significantly dysregulated in AIS and OGD/R cell model. The SVM model achieved high diagnostic accuracy (AUC = 0.969). Clustering analysis revealed two AIS subtypes with distinct immune infiltration and functional pathways. Validation by qPCR and ELISA confirmed significant upregulation of HNRNPL, HSDL2, and ARID3A, as well as downregulation of SSB, in AIS patients. These findings highlight the diagnostic potential of a four-gene expression signature associated with lactylation biology in AIS. These results provide hypothesis-generating insights into stroke pathophysiology involving lactate metabolism and monocyte-mediated inflammation, while acknowledging that direct evidence of site-specific lactylation modification on these proteins remains to be established. This candidate blood-based biomarker signature warrants validation in larger, multicenter cohorts inclusive of clinically relevant differential diagnoses (e.g., TIA, hemorrhagic stroke, stroke mimics) before clinical utility can be inferred.