Benchmarking a Microfluidic-Based Filtration for Isolating Biological Particles

Isolating particles from complex fluids is a crucial approach in multiple fields including biomedicine. In particular, biological matrices contain a myriad of distinct particles with different sizes and structures. Extracellular vesicles (EVs), for instance, are nanosized particles carrying vital information from donor to recipient cells, and they have garnered significant impact on disease diagnostics, drug delivery, and theranostics applications. Among all the EV types, exosome particles are one of the smallest entities, sizing from 30 to 100 nm. Separating such small substances from a complex media such as tissue culture and serum is still one of the most challenging steps in this field. Membrane filtration is one of the convenient approaches for these operations; yet clogging, low-recovery, and high fouling are still major obstacles. In this study, we design a two-filter-integrated microfluidic device focusing on dead-end and cross-flow processes at the same time, thereby minimizing any interfering factors on the recovery. The design of this platform is also numerically assessed to understand pressure-drop and flow rate effects over the procedure. As a model, we isolate exosome particles from human embryonic kidney cells cultured in different conditions, which also mimic complex fluids such as serum. Moreover, by altering the flow direction, we refresh the membranes for minimizing clogging issues and benchmark the platform performance for multitime use. By comprehensively analyzing the design and operation parameters of this platform, we address the aforementioned existing barriers in the recovery, clogging, and fouling factors, thereby achieving the use of a microfluidic device multiple times for bio-nanoparticle isolation without any notable issues.