Identification and validation of key biomarkers of the glycolysis-ketone body metabolism in heart failure based on multi-omics and machine learning

作者：Na Xiao, Jing Liu, Zhe Chen, Xiaoyong Geng · 发表于：Frontiers in Cardiovascular Medicine · 年份：2025 · DOI：10.3389/fcvm.2025.1672513 · 被引用次数：1 · 研究领域：Diet and metabolism studies、Cardiovascular Function and Risk Factors、Congenital heart defects research

Background: Metabolic remodeling, particularly involving glycolysis and ketone body metabolism, is a hallmark of heart failure (HF) pathophysiology. However, the regulatory network linking energy metabolism with immune dysregulation remains poorly understood. Objectives: This study aimed to identify and validate key biomarkers within the glycolysis-ketone body metabolism axis that contribute to the progression of HF, and to explore their association with immune microenvironment alterations. Methods: Transcriptomic data from HF patients were integrated with glycolysis and ketone metabolism gene sets. Differentially expressed genes (DEGs) were identified and analyzed through Weighted Gene Co-expression Network Analysis (WGCNA). Candidate genes were refined using machine learning algorithms (LASSO regression and Boruta), with functional enrichment assessed via Gene Set Enrichment Analysis (GSEA). Immune infiltration was profiled using ssGSEA, and regulatory networks were constructed by integrating miRNA and transcription factor predictions. Experimental validation was conducted in a murine myocardial infarction model using qPCR and cardiac ultrasound imaging. Results: Five candidate genes related to glycolysis and ketone metabolism were identified, among which TIMP1 emerged as the key hub gene. TIMP1 expression was significantly elevated in HF and correlated with enriched pathways including inflammatory signaling and mitochondrial dysfunction. Immune profiling revealed that TIMP...