exosomes which are formed in endosomal compartments and microvesicles formed by budding of the cell membrane. [1] While ascertaining the exact biogenesis of these vesicles is extraordinarily difficult, the umbrella term of EVs is established within the research community and will be continued to be used here when mammalian vesicles are addressed. [2] Gram-negative bacteria are also generating a type of vesicle, called outer membrane vesicles (OMVs) whose structural elements and content are comparable to that of their bacterial origin. [3,4] Roughly summarizing, both types of vesicles consist of a phospholipid bilayer and contain, depending on their origin, nucleic acids, proteins, and enzymes. The main biological function of mammalian or bacterial vesicles is to deliver content, enabling the cargo to be transported for longer distances, protected from the environment. [5] This allows EVs to be involved in tissue repair or immune modulation by transporting miRNA or antigens for example. [6,7] In pathogens, OMVs play a crucial role in the transfer of resistances, transport of degrading enzymes, or genetic information. [8,9] OMVs may also influence biofilm formation and architecture, providing an alternative treatment avenue for such difficult-to-treat infections. [10] With this in mind, it was only a matter of time that EVs and OMVs were considered as potential therapeutics, such as drug delivery systems.Recently, extracellular vesicles (EVs) sparked substantial therapeutic interest, particularly due to their ability to mediate targeted transport between tissues and cells. Yet, EVs' technological translation as therapeutics strongly depends on better biocompatibility assessments in more complex models and elementary in vitro-in vivo correlation, and comparison of mammalian versus bacterial vesicles. With this in mind, two new types of EVs derived from human B-lymphoid cells with low immunogenicity and from non-pathogenic myxobacteria SBSr073 are introduced here. A large-scale isolation protocol to reduce plastic waste and cultivation space toward sustainable EV research is established. The biocompatibility of mammalian and bacterial EVs is comprehensively evaluated using cytokine release and endotoxin assays in vitro, and an in vivo zebrafish larvae model is applied. A complex three-dimensional human cell culture model is used to understand the spatial distribution of vesicles in epithelial and immune cells and again used zebrafish larvae to study the biodistribution in vivo. Finally, vesicles are successfully loaded with the fluoroquinolone ciprofloxacin (CPX) and showed lower toxicity in zebrafish larvae than free CPX. The loaded vesicles are then tested effectively on enteropathogenic Shigella, whose infections are currently showing increasing resistance against available antibiotics.