Diabetic retinopathy (DR) is a major microvascular complication of diabetes mellitus, representing a significant cause of blindness in the working-age population worldwide (Cheung et al., 2010). Proliferative diabetic retinopathy (PDR), the more severe form of DR, is characterized by the formation of new, fragile blood vessels on the retina, leading to bleeding, retinal detachment, and vision loss (Chaudhary et al., 2021). The pathogenesis of DR is multifactorial, involving hyperglycemia-induced oxidative stress, inflammation, and vascular dysfunction (H. Li et al., 2023; Yue et al., 2022). Despite advancements in understanding the molecular mechanisms underlying DR, effective treatments remain limited, necessitating further research into novel therapeutic targets.

Emerging evidence has highlighted the role of gut microbiota and their metabolites in various metabolic diseases, including diabetes and DR (Cai and Kang, 2023; Iatcu et al., 2021). Trimethylamine-N-oxide (TMAO), a gut microbiota-derived metabolite, has garnered considerable attention due to its association with cardiovascular diseases and metabolic disorders (Chen et al., 2019; Thomas and Fernandez, 2021). Recent studies suggest that elevated levels of TMAO may exacerbate inflammatory responses and oxidative stress, potentially linking it to the progression of DR (Benson et al., 2023; Luo et al., 2024). However, the precise mechanisms by which TMAO influences DR, particularly PDR, remain unclear.

Previous research has primarily focused on the role of hyperglycemia and associated metabolic disturbances in DR. Studies have demonstrated that chronic hyperglycemia leads to the accumulation of advanced glycation end-products, increased production of reactive oxygen species (ROS), and activation of inflammatory pathways, all contributing to retinal damage (Kaur and Harris, 2023; Sahajpal et al., 2019). However, the involvement of gut microbiota and their metabolites, such as TMAO, in DR pathogenesis is a relatively new area of investigation. Recent findings suggest that TMAO may influence vascular inflammation and endothelial dysfunction, both critical factors in DR development. For instance, TMAO has been shown to promote atherosclerosis by modulating cholesterol and bile acid metabolism, as well as inflammatory signaling pathways (Florea et al., 2022; Wang et al., 2021). In addition, increased concentrations of TMAO were linked to a greater likelihood and more severe manifestations of DR in individuals with type 2 diabetes mellitus (W. Liu et al., 2021). Despite these insights, there is a paucity of studies directly linking TMAO to DR, particularly using comprehensive network-based approaches to uncover potential molecular targets and pathways.

Given the complexity of DR pathogenesis and the emerging role of gut microbiota metabolites, there is a compelling need to explore the potential impact of TMAO on DR progression. Understanding the molecular interplay between TMAO and DR could unveil new therapeutic targets and strategies for managing this debilitating condition. This study aims to fill the gap by systematically investigating the potential role of TMAO in PDR using a network toxicology approach, integrating multiple databases and analytical techniques.

The primary objective of this study was to systematically investigate the role of TMAO in the progression of DR using a network toxicology approach. To achieve the objective, we utilized a network toxicology approach to systematically identify TMAO-related targets by integrating data from CTD, SuperPred, and GeneCards databases. Differential expression analysis was performed on two publicly available GEO datasets (GSE60436 and GSE102485) to identify differentially expressed genes (DEGs) between normal and PDR groups. Intersection analysis was conducted to pinpoint key genes potentially implicated in PDR through TMAO-mediated mechanisms. Functional enrichment and pathway analyses were carried out to elucidate the biological roles of these targets. Immune cell infiltration levels were assessed using the ssGSEA algorithm. Machine learning techniques, including LASSO and random forest (RF), were applied to identify key marker genes. A predictive nomogram was constructed for DR risk estimation. Finally, molecular docking analysis was performed to investigate the interactions between TMAO and key toxicity targets. The flow chart of this study is shown in Fig. 1.

This study unveils TMAO as a novel mediator of DR progression, linking gut microbiota metabolism to retinal inflammation and apoptosis. The identification of CASP3, CXCR4, and MAPK1 as central biomarkers provides actionable targets for therapeutic intervention. The predictive nomogram offers clinical utility for early PDR detection, while molecular docking insights pave the way for designing TMAO-targeted inhibitors. These findings advance our understanding of the gut-retina axis and highlight TMAO as a potential therapeutic frontier in diabetic complications.