Quantitative and functional characterisation of extracellular vesicles after passive loading with hydrophobic or cholesterol-tagged small molecules

• DOI: 10.1016/j.jconrel.2023.08.010
• Publication Date: 2023-08-19T05:07:44Z
• Authors: Gwenola Tréton; Claudia Sayer; Melanie Schürz; Maria Jaritsch; Anna Müller; Cristian-Tudor Matea; Vesna Stanojlovic; Heloisa Melo-Beni...; Celine Be; Caroline Krembel; Stephane Rodde; Matthias Haffke; Samuel Hintermann; Andreas Marzinzik; Sébastien Ripoche; Constantin Blöchl; Julia Hollerweger; Daniela Auer; Chiara Cabrele; Christian G. Huber; Martin Hintersteiner; Trixie Wagner; Andreas Lingel; Nicole Meisner-Kober
• Keywords: Nanoparticle tracking analysis

Extracellular vesicles (EVs) are nanosized intercellular messengers that bear enormous application potential as biological drug delivery vehicles. Much progress has been made for loading or decorating EVs with proteins, peptides RNAs using genetically engineered donor cells, but post-isolation synthetic drugs and from natural sources remains challenging. In particular, quantitative unambiguous data assessing whether how small molecules associate versus other components in the samples still lacking. Here we describe systematic characterisation of passive EV based on hydrophobic interactions - either through direct adsorption compounds, by membrane anchoring hydrophilic ligands via cholesterol tags. As revealed single vesicle imaging, both ligand types bind to CD63 positive (exosomes), however also non-specifically vesicles, particles, serum proteins. The compounds Curcumin Terbinafine aggregate no apparent saturation up 106-107 per quantified liquid chromatography high resolution mass spectrometry (LC-HRMS). For density resulted formation a population large, electron-dense detected cryo-transmission electron microscopy (TEM), reduced cell uptake toxic gain function Curcumin-EVs. contrast, tagging mdm2-targeted cyclic peptide saturated at densities ca 104-105 vesicle, lipidomics showing addition to, rather than replacement endogenous cholesterol. Cholesterol anchored did not change EVs' size morphology, such retained their activity without inducing toxicity. However, cholesterol-anchored were rapidly shed presence serum. Based these data, conclude (1) methods allow prone unspecific compound binding redistribution if present sample, (2) needs substantial optimization formulation stability vivo applications, whereas (3) careful titration is warranted when relying mitigate changes physicochemical properties, loss