A Lactylation-Based Diagnostic Model Reveals Molecular Subtypes and Therapeutic Targets in Dilated Cardiomyopathy

Dilated cardiomyopathy (DCM) is a progressive myocardial disorder lacking reliable molecular biomarkers for early diagnosis. Given the emerging role of protein lactylation in cardiovascular disease, we investigated lactylation-related genes (LRGs) in DCM using transcriptomic data from the GEO database. Differential gene expression and lactylation databases were integrated to identify 18 dysregulated LRGs. Using ensemble machine learning algorithms, six core LRGs (G6PD, PPP1CC, MBP, LSP1, HMGN1, HMGN2) were selected to construct a diagnostic model, which showed robust performance across five external cohorts. Unsupervised clustering revealed two molecular DCM subtypes. Immune infiltration and ssGSEA analyses suggested that core LRGs modulate immunometabolic remodeling. Single-cell RNA-seq analysis confirmed their cell-type-specific distribution, with fibroblasts identified as dominant signaling sources via CellChat analysis. Ferrostatin-1 and Resveratrol were predicted as potential therapeutic compounds targeting core genes. Validation in human and mouse myocardial tissues confirmed differential gene and protein expression. This study uncovers lactylation-driven molecular signatures in DCM and establishes a robust diagnostic model with promising translational potential. By linking lactate metabolism to immune regulation and cardiac remodeling, our findings highlight novel diagnostic markers and potential therapeutic targets for precision intervention in DCM.