Integrating WGCNA and machine learning to identify and validate key biomarkers in MASLD

Area of Science:

• Hepatology
• Systems Biology
• Molecular Biology

Background:

• Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) is the most common chronic liver disease globally.
• Its molecular heterogeneity hinders the development of effective pharmacotherapies.
• Identifying novel hub genes is crucial for understanding MASLD pathogenesis.

Purpose of the Study:

• To identify novel hub genes regulating MASLD pathogenesis using an integrative systems-biology approach.
• To elucidate the molecular mechanisms underlying lipogenesis and inflammation in MASLD.
• To validate potential therapeutic targets in vitro and in vivo.

Main Methods:

• Weighted Gene Co-expression Network Analysis (WGCNA) and machine learning algorithms (LASSO, Random Forest, SVM-RFE) were applied to public gene expression datasets (GSE89632, GSE63067).
• Hub genes were identified and validated in vitro using free fatty acid-treated HepG2 cells.
• In vivo validation was performed in a high-fat diet (HFD)-induced mouse model of MASLD.

Main Results:

• Two distinct pathological networks were identified: a lipogenesis cluster with upregulated FMO1 and C10orf140 and downregulated JUNB, linked to the SREBP-1c/FASN pathway.
• A co-expression module correlated with inflammation identified 11 inflammation-related hub genes (e.g., MAP3K8, PFKFB3).
• HFD-induced MASLD mice exhibited severe steatosis and paradoxical activation of both pro-lipogenic p-AKT and inhibitory p-AMPK pathways, suggesting "AMPK Resistance".

Conclusions:

• FMO1, C10orf140, and JUNB are identified as novel regulators of MASLD lipogenesis via the SREBP-1c pathway.
• Co-activated p-AKT and p-AMPK signaling in steatotic livers signifies "AMPK Resistance," a state of unabated pro-lipogenic signaling.
• These findings highlight potential therapeutic strategies targeting key signaling pathways in MASLD.