UPDF AI

Anti-diabetic retinopathy molecular mechanism of Dihuang Yinzi: insights from network pharmacology, metabolomics, and microbiome analysis

Jinmiao Chai,Wanwei Gui,2 Authors,Qinqing Li

2026 · DOI: 10.3389/fmed.2026.1793936
Frontiers in Medicine · 0 Citations

TLDR

This study elucidates five key target genes, 30 key metabolites, and 18 key microbes through which DHYZ may exert its therapeutic effects in DR, a major microvascular complication arising from diabetes mellitus.

Abstract

Background Diabetic retinopathy (DR) represents a major microvascular complication arising from diabetes mellitus, characterized by multifactorial pathogenesis encompassing genetic, metabolic, and microbial components. While Dihuang Yinzi (DHYZ), a traditional Chinese herbal formulation, exhibits therapeutic promise for DR management, the precise mechanistic underpinnings warrant further investigation. This research sought to elucidate critical target genes, metabolic compounds, and microbial species implicated in DHYZ's therapeutic action against DR. Methods We constructed a murine DR model and procured biological specimens from 18 animals distributed across three experimental cohorts (control, disease model, and DHYZ-treated groups, n = 6 per group) for metabolomic profiling and microbiome characterization. Candidate genes emerged from overlapping DHYZ-associated targets with DR-linked genes. Through six computational algorithms within the cytoHubba plugin, we pinpointed pivotal target genes. Molecular docking studies examined binding affinity between essential target proteins and bioactive constituents. Metabolomic and microbiome datasets underwent differential expression analysis to enumerate candidate metabolites and microbial taxa, respectively. Finally, Spearman correlation-based integrative omics analysis distinguished critical metabolites and key microbial species. Results We identified 110 candidate genes and five key target genes (STAT3, IL6, TNF, ESR1, and IL1B). Molecular docking analysis revealed strong binding interactions between ESR1 and six corresponding active compounds, with the highest binding affinity observed for naringenin. Additionally, metabolomic analysis identified 50 candidate metabolites, and microbiome analysis revealed 24 candidate microbes. Spearman correlation analysis further pinpointed 30 key metabolites and 18 key microbes. Conclusion This study elucidates five key target genes, 30 key metabolites, and 18 key microbes through which DHYZ may exert its therapeutic effects in DR. These findings provide valuable insights and a foundational reference for understanding the multi-omics mechanism of DHYZ in the treatment of diabetic retinopathy.