Backgrounds and objectives Mesenchymal stem cells (MSC) are multipotent cells with immunomodulatory functions that are of interest for therapeutic purposes in osteoarticular autoimmune diseases such as rheumatoid arthritis (RA). These functions are primarily mediated by soluble mediators that can be released within extracellular vesicles (EV). EVs consist of three main populations: exosomes, microvesicles and apoptotic bodies that mirror the effect of parental cells but little is known about the respective role of the various subsets of EVs secreted by a specific cell type. The aim of this study was to compare the immunomodulatory effects of exosomes and microvesicles released by MSCs and determine whether a pre-activation of MSCs before EV purification impacts their function.
Materials and methods Exosomes and microvesicles were isolated from culture supernatants of murine primary MSCs by differential ultracentrifugation at 100K and 18K, respectively. Size and structure of exosomes and microvesicles were evaluated by NTA and/or electron microscopy. Expression of membrane and endosomal markers was tested by flow cytometry and Western blot. Proliferation of freshly isolated murine splenocytes or subpopulations of immune cells was quantified after 72 hour of incubation with different quantities of EVs, using the Cell TiterGlo assay and cytometry. In vivo, EVs were injected systemically in the collagen-induced arthritis (CIA) model.
Results MSC-derived exosomes were around 120 nm in diameter and expressed CD9, CD81, TSG101 markers while microvesicles comprised a heterogeneous population of vesicles from 165 to 500 nm in diameter and expressed CD29, CD44 and Sca1 membrane markers. Addition of microvesicles or exosomes in proliferative assays significantly inhibited the proliferation of CD8+ T cells and increased Treg cell population with no effect on CD4+IFN+ T cells. In addition, proliferation and activation of B lymphocyte were decreased. These effects were independent of MSC pre-activation. The immunomodulatory function of both types of EVs was also observed in the CIA model of inflammatory arthritis with a significant reduction of clinical symptoms.
Conclusion Our data indicated that the immunosuppressive effect of MSCs is at least in part mediated by EVs with no major differential role of exosomes or microvesicles on inflammatory immune cells in vitro or in vivo.