Poster Y12 - Testing and Modification of a Next-Generation Nano-Pocket Membrane for Enhanced Isolation and Purification of Extracellular Vesicles

Introduction: Extracellular vesicles (EVs) are heterogeneous nanoscale membranous vesicles secreted from most cells and biological fluids and have been recognized as novel intercellular communications tools and biomarkers for several diseases including cancer and diabetes. EVs of various subtypes have particle diameters that typically range from 30 to 1000 nm; however, a major obstacle for research and clinical translations has been isolating EVs from contaminants, with additional difficulties being presented when separating EV subpopulations by size.
Various methods, such as ultracentrifugation, size exclusion chromatography, and ultrafiltration, have been employed for EV isolation based on size, but each approach presents drawbacks. Porous membranes, in particular, have been used for size-based EV isolation, though multiple membranes are required to capture specific size ranges of EVs, thus increasing EV losses as a result of particle aggregation and pore clogging.
To combat the aforementioned concerns, we previously developed methods to create a membrane featuring conical “nano-pocket” pores, rather than the typical cylindrical pores, which taper from a wide opening to a narrower opening on the bottom side of the membrane (Figure 1b). This format aims to capture EVs within a specific size range while preventing non-target EVs from blocking pores during tangential flow filtration (TFF) (Figure 1a). We sought to validate the efficacy of the fabricated membrane and present data demonstrating the capture and release behavior of assembled nano-pocket devices. Furthermore, we evaluate preliminary results surrounding surface modification of the membranes through polymer grafting that aims to enhance release patterns by reducing non-specific binding.

Materials and

Methods: The fabrication process involved various microfabrication steps, which were optimized in prior studies, ultimately resulting in a patterned Parylene C membrane. In order to be relevant for EV separations, the final target nano-pocket membrane shape has a wide opening (0.5 -1 µm) at the top, narrows to a smaller opening (0.05-0.15 µm) at the bottom, and has a (1-2.5 µm) total thickness.
Coupling pressure sensors to the inlet and outlets of the system, the permeability of the membrane was assessed, given known pore geometry and distribution. Further flow experiments were conducted using two syringe pumps and negatively charged fluorescent polystyrene beads of sizes 0.05, 0.5, and 10 µm as a model for protein aggregates, medium-sized EVs, and cells respectively. The beads were supplied to the system both individually, as well as in a 1:1:1 concentration mixture to evaluate capture and release efficacy. These steps are outlined in Figure 1a. Importantly, the membrane was inverted such that larger particles sediment rather than clogging pores.
To enhance EV release by decreasing non-specific binding, as well as further manipulating pore size, the membranes were further modified through polymer grafting. The parylene surfaces were activated through hydrogen peroxide priming which negatively charged the surface, prior to PLL-PEG (PLL(20)-g(3.5)-PEG(2)) grafting. Characterization efforts including flow analyses were replicated according to the aforementioned methods used for the original membrane.
Scanning electron microscopy (SEM), atomic force microscopy (AFM), and fluorescence microscopy were used to validate membrane geometry, as well as the capture and release of fluorescent beads.

Results, Conclusions, and Discussions: We successfully produced nano-pocket membranes featuring a wide opening at the top, measuring 600 nm, which narrows down to a smaller opening on the bottom side, ranging from 0.1 to 0.15 µm (Figure 1b).
The selective nature of the membrane was proven through the flow experiments involving polystyrene beads. Fluorescent analysis of the beads introduced to the membranous system demonstrated that while 0.05 µm beads passed through the membrane, 0.5 um beads were captured in the membrane as initially posited. This capture was further validated by SEM analysis, though it was also determined that some beads remained adhered to the membrane despite backflow during release (Figure 1c). Conversely, 10 um beads continued to flow across the membrane rather than being captured in the nano-pockets. This capture and release behavior was found consistent when a 1:1:1 mixture of the beads was introduced.
Hydrogen peroxide was subsequently found to decrease parylene contact angle which is known to increase PLL-PEG binding on substrates, though this was confirmed using SEM and AFM. Furthermore, the presence of PLL-PEG on the membranes was verified during bead capture and release analysis, where prior methods were replicated. Upon release of 0.5 um beads with a PEGylated membrane, less strong adherence of the beads to the parylene surface was observed, leading to larger quantities of beads released and thus higher filtrate yields.
The nano-pocket membranes developed through this work aim to provide an alternate and enhanced EV isolation method. Through previous fabrication optimization and recent flow analysis efforts, targeted pore geometries and capture and release of polystyrene beads as EV models were successful. Future work aims to validate the proposed membrane geometry for TFF of liposomes and subsequently EVs, as well as further characterize the impact of PEGylated membranes on pore geometry and capture/release mechanics, in addition to further vary nano-pocket sizes to allow further size-based purification of EV subpopulations.

Acknowledgements (Optional):