Identification of Glycolysis-Related Genes in MAFLD and Their Immune Infiltration Implications: A Multi-Omics Analysis with Experimental Validation

作者：Jiawei Chen, Siqi Yang, Diwen Shou, Bo Liu, Shaohan Li, Tongtong Luo, Huiting Chen, Chen Huang, Yongjian Zhou · 发表于：Biomedicines · 年份：2025 · DOI：10.3390/biomedicines13071636 · 被引用次数：4 · 研究领域：Liver Disease Diagnosis and Treatment、Lipid metabolism and disorders、RNA modifications and cancer

Background: Metabolic-associated fatty liver disease (MAFLD) is characterized by metabolic syndrome and immune infiltration, with glycolysis pathway activation emerging as a pivotal contributor. This study aims to identify glycolysis-associated key genes driving MAFLD progression and elucidate their crosstalk with immune infiltration through bioinformatics analysis and experimental validation. Methods: Integrative multi-omics analysis was performed on bulk RNA-seq, single-cell RNA-seq, and spatial transcriptomic datasets from MAFLD patients and controls. Differential expression analysis and WGCNA were employed to pinpoint glycolysis-correlated key genes. The relationship with immune infiltration was analyzed using single-cell and spatial transcriptomics technologies. Machine learning was applied to identify feature genes for matching shared TFs and miRNAs. External cohort validation and in vivo experiments (methionine choline-deficient diet murine models) were conducted for biological confirmation. Results: Five glycolysis-associated key genes (ALDH3A1, CDK1, DEPDC1, HKDC1, SOX9) were identified and validated as MAFLD discriminators. Single-cell analysis revealed that the hepatocyte–fibroblast–macrophage axis constitutes the predominant glycolysis-active niche. Spatial transcriptomics showed that CDK1, SOX9, and HKDC1 were colocalized with the monocyte-derived macrophage marker CCR2. Using four machine learning models, four feature genes were identified, along with their comm...