A Microfluidic Approach for Analyzing Extracellular Vesicles from Single Cells

Eukaryotic organisms consist of a heterogeneous entirety of cells organized into functional entities of tissues and organs. Each cell embodies an independent, but complex network of biochemical reactions and information flow. Eukaryotes effectively combine complex processes such as energy production, waste recycling, and immune response, and are also able to react to external stimuli, due to the fact that each task is allocated to separate, membrane-contained spaces. A distinct, but only partially understood role of membrane encapsulation in cells is the transport and exchange of information. The controlled release or reception of cellular content depends on the nature and function of the signal or biochemical molecule. Cells secrete a myriad of extracellular vesicles (EVs), which share some biochemical properties, but differ significantly in their formation. We know that cells are armed with protein complexes which can bend membranes and which can recognize and sort content. These complexes also create a distinct subpopulation of vesicles released by cells. How do cells form these vesicles and why? How do cells decide what is sorted into those vesicles and how does the sorting work? And how do those vesicles differ from other vesicles shed directly from the plasma membrane? Scientists around the globe are trying to answer these questions in order to unveil the details of EV formation and its diversity. While enough details to answer How? and Why? are still missing, we know that EVs can have enourmous impacts on health and disease. EVs secreted by cells carry valuable information about the cell and tissue of origin and are therefore not only potential biomarkers in diagnostics, but are also highly promising carriers for a more effective delivery of biomedical compounds. Substantial progress was achieved in the fields of EVs by observing intracellular protein interactions and analyzing the entirety of enriched EVs. However, with these advances, scientists exhausted the potential of current analysis methods and cannot compensate for their limitations. Besides the advantages of smaller volumetric dimensions, the rich toolbox of microfluidics easily allows for creating experimental environments to further push the boundaries for EV analysis. Thus, microfluidics offers opportunities to step beyond known analytical and experimental limitations in the quest of understanding cellular membrane trafficking. In this thesis, we combine microfluidics with a substantially sensitive microscopy method to challenge the current status quo in the field of EVs. For this, we introduce two optimized microfluidic devices standardized to not only separate and isolate single cells, but more importantly, to access cell-dependent EV heterogeneity from individual cells by preventing sample cross-contaminations. In a multiplexed approach, we reveal the heterogeneity of single cell-derived EVs produced by different pathways. We then use a significantly optimized version of our method to focus on the diagnostic potential of EVs: We identify the phenotypic fingerprints of different human tissues to demonstrate the power of single cell-derived EVs as diagnostic biomarkers for a faster and more sensitive detection and characterization of cancer with liquid-biopsies. Then, we analyze the membrane-specific composition of EVs for a subset of target epitopes which are involved in EV formation and are informative of their origin. Finally, we create stable cell lines to discover particularly weak and rare signals derived from native EVs—an aspect critical in fundamental science