Integrative analysis identifies lactylation-related biomarkers in sepsis via bioinformatics and machine learning

Protein lactylation is a novel post-translational modification driven by lactate and has emerged as a critical link between cellular metabolism and epigenetic regulation in inflammatory diseases, including sepsis. However, a systematic profiling of lactylation-related genes (LRGs) and their diagnostic potential in sepsis remains unexplored.

We conducted an integrative study leveraging transcriptomic data from public repositories (GSE236713 as training set; GSE28750 as validation set). Differential expression analysis identified sepsis-associated genes, which were intersected with LRGs from GeneCards to obtain candidate genes. Functional enrichment was performed. Three machine learning algorithms-LASSO regression, Support Vector Machine (SVM), and Random Forest (RF)-were employed to screen for core biomarkers. Their diagnostic efficacy was validated using ROC curve analysis. Immune infiltration profiles were assessed via CIBERSORT. Key findings were further corroborated in a murine sepsis model induced by cecal ligation and puncture (CLP) using qRT-PCR.

We identified 10 candidate LRGs from the initial 3252 differentially expressed genes. Machine learning algorithms narrowed this down to three core biomarkers: SLC4A4, SLC16A4, and CRYAB. These biomarkers demonstrated consistent differential expression patterns (down-regulation of SLC4A4 and up-regulation of SLC16A4 and CRYAB) in both the training and independent validation cohorts. ROC analysis confirmed their robust diagnostic performance, with AUCs all exceeding 0.7. Furthermore, these biomarkers showed significant correlations with specific immune cell subsets, particularly CD8⁺ T cells and activated mast cells, suggesting a role in immune dysregulation. Validation in the CLP mouse model confirmed the significant dysregulation of these three genes in septic lung tissue.

This multi-disciplinary study identifies SLC4A4, SLC16A4, and CRYAB as promising lactylation-related diagnostic biomarkers for sepsis. Our findings indicate the intricate interplay between lactate metabolism, epigenetic modification, and immune dysfunction in sepsis, providing new insights for early diagnosis and mechanistic investigation.