Integrated transcriptomic analysis identifies liver aging-driven fibrosis signatures and reveals therapeutic strategies based on medicine-food homology.

Abstract

Despite its strong regenerative capacity, liver aging paradoxically increases susceptibility to fibrosis and metabolic dysfunction-associated steatotic liver disease through dysregulated inflammation, senescence-associated secretory phenotypes, and immune-metabolic crosstalk. To systematically characterize these processes, we integrated longitudinal transcriptomics, single-cell RNA sequencing, and machine-learning approaches. We identified 252 aging-associated genes and developed an Aging Gene Score (AGS) to quantify senescence burden across hepatic cell populations. Single-cell analysis revealed macrophages as key drivers of fibrosis progression, with high-AGS myeloid subsets markedly expanded in cirrhotic livers. Using combined Boruta and LASSO algorithms, we established a five-gene biomarker panel (EFEMP1, LUM, DKK3, GPRC5B, NCAM2) that accurately predicts advanced fibrosis (AUC > 0.77). Furthermore, a network pharmacology framework was applied to screen medicine-food homology (MFH) herbs, identifying Fagopyrum dibotrys (Jinqiaomai) and Astragalus membranaceus (Huangqi) as top candidates. Molecular docking demonstrated strong binding between the bioactive compound MOL000098 and the fibrosis-related target COL3A1. Functional assays showed that Jinqiaomai-containing serum alleviates oxidative stress, improves HepG2 cell viability, reduces ALT and AST levels, and suppresses macrophage lipid accumulation, accompanied by reduced expression of inflammatory and fibrosis-related markers. Collectively, our findings highlight macrophage-centered mechanisms linking liver aging and fibrosis and suggest MFH-derived compounds as promising anti-aging and anti-fibrotic dietary interventions.