Corsi, J.; Peroni, D.; Belli, R.; Lassandro, M.; Sidarovich, V.; Adami, V.; Grosskreutz, J.; Fabbiano, F.; Grossmann, D.; Hermann, A.; Tell, G.; Basso, M.; D'Agostino, V. G.

Extracellular vesicles (EVs) are cell-secreted membranous particles contributing to intercellular communication. Coding and non-coding RNAs are widely detected EV cargo, and RNA-binding proteins (RBPs), such as hnRNPA2B1, have been circumstantially implicated in sorting vesicular RNAs. However, the contribution of competitive RBP-RNA interactions responsible for RNA-sorting outcomes still needs to be deciphered, especially for EV-RNA interference and predictability. We conducted a reverse proteomic analysis that prioritized heterogeneous nuclear ribonucleoproteins recognizing purine-rich RNA sequences representing a subset of previously identified EXO motifs. A screening campaign using a full-length human hnRNPA2B1 protein and artificial purine-rich RNA brought to small molecule inhibitors orthogonally validated through biochemical and cell-based approaches. Selected drugs effectively interfered with a post-transcriptional layer impacting secreted EV-RNAs, reducing the vesicular pro-inflammatory miR-221 while counteracting the hnRNPA2B1- or TDP43Q331K-dependent paracrine activation of NF-kB in EV-recipient cells. This study demonstrates the possibility of predicting the EV-RNA quality for developing innovative strategies targeting discrete paracrine functions.