Systematic Assessment of Small RNA Profiling in Human Extracellular Vesicles

MOTIVATION: Extracellular vesicles (EVs) are produced and released by most cells and are now recognized to play a role in intercellular communication through the delivery of molecular cargo, including proteins, lipids, and RNA. Small RNA sequencing (small RNA-seq) has been widely used to characterize the small RNA content in EVs. However, there is a lack of a systematic assessment of the quality, technical biases, RNA composition, and RNA biotypes enrichment for small RNA profiling of EVs across cell types, biofluids, and conditions. METHODS: We collected and reanalyzed small RNA-seq datasets for 2756 samples from 83 studies involving 55 with EVs only and 28 with both EVs and matched donor cells. We assessed their quality by the total number of reads after adapter trimming, the overall alignment rate to the host and non-host genomes, and the proportional abundance of total small RNA and specific biotypes, such as miRNA, tRNA, rRNA, and Y RNA. RESULTS: We found that EV extraction methods varied in their reproducibility in isolating small RNAs, with effects on small RNA composition. Comparing proportional abundances of RNA biotypes between EVs and matched donor cells, we discovered that rRNA and tRNA fragments were relatively enriched, but miRNAs and snoRNA were depleted in EVs. Except for the export of eight miRNAs being context-independent, the selective release of most miRNAs into EVs was study-specific. CONCLUSION: This work guides quality control and the selection of EV isolation methods and enhances the interpretation of small RNA contents and preferential loading in EVs.