Separation of high-purity plasma extracellular vesicles for investigating proteomic signatures in diabetic retinopathy

Extracellular vesicles (EVs) play a multifaceted role in intercellular communication and hold significant promise as bio-functional indicators for clinical diagnosis. Although plasma samples represent one of the most critical sources of circulating EVs, the existing technical challenges associated with plasma-EV isolation have restricted their application in disease diagnosis and biomarker discovery. In this study, we introduce a two-step purification method utilizing ultracentrifugation (UC) to isolate crude extracellular vesicle (EV) samples, followed by a phospholipid affinity-based technique for the selective isolation of small EVs, ensuring a high level of purity for downstream proteomic analysis. Our research demonstrates that the UC & TiO2-coated magnetic bead (TiMB) purification system significantly improves the purity of EVs when compared to conventional UC or TiMB along. We further revealed that proteomic alterations in plasma EVs effectively reflect key gene ontology components associated with diabetic retinopathy (DR) pathogenesis, including the VEGF-activated neuropilin pathway, positive regulation of angiogenesis, angiogenesis, cellular response to vascular endothelial growth factor stimulus, and immune response. By employing a comprehensive analytical approach, which incorporates both time-series analysis (cluster analysis) and differential analysis, we have identified three potential protein signatures including LGALS3, MYH10, and CPB2 that closely associated with the retinopathy process. These proteins exhibit promising diagnostic and severity-classification capabilities for DR disease. This adaptable EV isolation system can be regarded as an effective analytical tool for enhancing plasma-based liquid biopsies toward clinical applications.