Prerequisites for the analysis and sorting of extracellular vesicle subpopulations by high-resolution flow cytometry.

Abstract

Submicron-sized vesicles released by cells are increasingly recognized for their role in intercellular communication and as biomarkers of disease. Methods for high-throughput, multi-parameter analysis of such extracellular vesicles (EVs) are crucial to further investigate their diversity and function. We recently developed a high-resolution flow cytometry-based method (using a modified BD Influx) for quantitative and qualitative analysis of EVs. The fact that the majority of EVs is <200 nm in size requires special attention with relation to specific conditions of the flow cytometer, as well as sample concentration and event rate. In this study, we investigated how (too) high particle concentrations affect high-resolution flow cytometry-based particle quantification and characterization. Increasing concentrations of submicron-sized particles (beads, liposomes, and EVs) were measured to identify coincidence and swarm effects, caused by the concurrent presence of multiple particles in the measuring spot. As a result, we demonstrate that analysis of highly concentrated samples resulted in an underestimation of the number of particles and an interdependent overestimation of light scattering and fluorescence signals. On the basis of this knowledge, and by varying nozzle size and sheath pressure, we developed a strategy for high-resolution flow cytometric sorting of submicron-sized particles. Using the adapted sort settings, subsets of EVs differentially labeled with two fluorescent antibodies could be sorted to high purity. Moreover, sufficient numbers of EVs could be sorted for subsequent analysis by western blotting. In conclusion, swarm effects that occur when measuring high particle concentrations severely hamper EV quantification and characterization. These effects can be easily overlooked without including proper controls (e.g., sample dilution series) or tools (e.g., oscilloscope). Providing that the event rate is well controlled, the sorting strategy we propose here indicates that high-resolution flow cytometric sorting of different EV subsets is feasible.