Integrative Multiomics Analysis Identifies HK2 as a Key Regulator of Metabolic Reprogramming in Hepatic Stellate Cells

TL;DR

This study identifies HK2 as a key gene involved in liver fibrosis, showing it promotes disease progression and could be a potential treatment target.

Contribution

The study reveals HK2 as a novel regulator of metabolic changes in hepatic stellate cells during liver fibrosis.

Findings

HK2 expression is significantly increased in hepatic stellate cells and correlates with liver fibrosis progression.

HK2 knockdown impairs HSC migration and wound healing, suggesting its role in fibrosis activation.

A machine learning model using glycolysis-related genes achieved high accuracy (AUC 0.889) in predicting liver fibrosis.

Abstract

Liver damage caused by chronic liver disease frequently leads to hepatic fibrosis. A pivotal step in the fibrotic process is the activation of hepatic stellate cells (HSCs). Previous studies have suggested that enhanced aerobic glycolysis is closely associated with HSC activation. However, a comprehensive analysis of the relationship between hepatic fibrosis and aerobic glycolysis remains lacking.

RNA sequencing of liver tissue from 30 patients with fibrosis or cirrhosis and 8 healthy controls was conducted as part of a comprehensive multiomics approach to discover differentially expressed genes (DEGs). Weighted gene coexpression network analysis (WGCNA) was conducted to detect gene modules associated with liver fibrosis. Functional analyses, including migration and wound healing, were subsequently performed. Furthermore, a machine learning model predicting fibrosis was constructed based on glycolysis-related gene expression and validated using an independent dataset. Its clinical significance was subsequently explored. Protein expression and localization were further validated via western blotting and immunohistochemistry techniques.

The expression of HK2 is notably increased in HSCs and is strongly linked to the advancement of liver fibrosis. Within the constructed machine learning model, the random forest algorithm demonstrated the highest predictive performance for liver fibrosis, achieving an area under the curve (AUC) of 0.889. HK2 expression levels also had a positive correlation with clinical signs of liver damage, such as ALT and AST levels. Knockdown of HK2 in HSCs markedly impaired their migratory capacity and wound healing ability.

HK2 is involved in activating HSCs, thus promoting the progression of liver fibrosis. These findings suggest that HK2 holds potential as a therapeutic target for liver fibrosis and as a biomarker for predicting its progression.

Figures