Search for a command to run...
ABSTRACT Salivary extracellular vesicles (EVs) represent a powerful, non-invasive source of biomarkers for disease diagnosis and monitoring. Their molecular cargo reflects systemic and local physiological states, offering a window into neurological and inflammatory disorders. However, the diversity of EV isolation protocols and the possibility that each enriches distinct EV subpopulations remains a major barrier to reproducibility and data comparability. We conducted a comprehensive comparison of three EV isolation methods: ultracentrifugation (UC), PEG-based precipitation (Q), and immunoaffinity capture (M) to evaluate their impact on EV yield, purity, and molecular composition. Salivary EVs from healthy volunteers were analysed using proteomic and small-RNA sequencing approaches. Principal component analysis revealed clear isolation method-dependent clustering, where M-derived EVs displayed the most distinct profile. UC and Q produced broader proteomic repertoires with higher total protein content, whereas M-isolated EVs exhibited greater purity and enrichment of trafficking-and lysosome-associated proteins. Over 731 miRNAs selected, 28 were consistently altered across methods and 65 uniquely enriched in M isolates. RT-qPCR confirmed key directional trends. These 93 method-dependent miRNAs have predicted targets associated with synaptic structure and neurodegenerative pathways. These findings show that isolation methodology deeply shapes salivary EV cargo and suggest that immunoaffinity capture can isolate specific EV populations meeting diagnostic requirements.