Skip to main navigation Skip to search Skip to main content

Characterization of extracellular vesicles from biofluids using a nanofluidic trap

  • Pavlina Senikoglou

Research output: Book/ReportPh.D. thesis

Abstract

Exosomes, nanometer-sized spherical vesicles once considered waste disposing vesicles, are now recognized as important biomarkers of disease. The vesicles are present in all biofluids - including blood, saliva, and urine - which serve as minimally invasive liquid biopsies. The exosome concentration is very low compared to other highly abundant components, and many challenges present in the way of isolation of the vesicles of interest. Standard downstream applications for exosome purification and characterization typically require multiple costly, time-consuming, and quality-compromising steps.

In this thesis, an integrated approach that couples nanofluidics and fluorescence microscopy is presented. Using an inverted fluorescence microscope and an injection-molded nanofluidic device, we capture, trap, and characterize biological nanoparticles. A salinity gradient is established along a nanochannel, where the balance of diffusioosmosis and diffusiophoresis spatially traps the particles. The detected fluorescence signal of the concentration profile in the nanofluidic trap enables our model to characterize the particles over size and zeta potential simultaneously.

Given the low exosome concentrations in liquid biopsies we investigated how the strength of the salinity gradient, the passivation of the channel walls, and the design of the trap can influence trapping position and can enhance the detection signal. We found that capture efficiency increases as a function of gradient strength, and that by tuning our system, liposomes - used as model particles - can be trapped at exosome-related concentrations within short times. Furthermore, we showed that reducing the length of the funnel-shaped nanochannel increases the strength of the salinity gradient. This in turn moves the trapping position to a smaller volume, thus enhancing the detected signal. Moreover, we developed a passivation protocol that successfully inhibits surface interactions with complex vesicles in protein solutions in the micromolar range.

In this work, we were specifically interested in purifying exosomes from human blood plasma. The challenge of the protein background in the biofluid was addressed by employing our protocol that passivates the nanochannel walls with Pluronic. Here, we use a model system with liposomes and albumin, where the introduction of a protein solution has a reversible effect. The Pluronic coated nanochannel successfully enables liposome purification from micromolar protein solutions, providing the first step towards liquid biopsy purification in the nanofluidic trap.

Immuno-staining is often used for exosome detection, and here a model system that mimics the antigen-antibody complex was portrayed with biotinylated liposomes and streptavidin. We showed that the Pluronic coated nanochannel can trap functionalized, thus more complex, liposomes without observing interactions between surface and vesicles. Furthermore, we investigated the kinetics of tetrameric streptavidin binding on the biotin sites of the liposomes, and the effect of binding on the trapping position. Such effects can be investigated in the trap where the liposomes are freely diffusing compared to functionalized surfaces that immobilize them.

A highly concentrated component of plasma is the lipoproteins. These vesicles are considered a contaminant in almost all purification methods, hindering recovery and specificity. Here, we showed that lipoproteins can trap in the nanofluidic device, and moreover that the interactions of lipoproteins with the channel walls are massively influenced by the plasma proteins in solution. Most importantly, we were able to show that antibodies targeting common biomarkers on the exosomal surface do not bind unspecifically on the lipoprotein surface. Antibody aggregation is eliminated by introducing the antibody solution in a high salinity solution, thus we avoid detection of false-positive signal. This serves as another step towards exosome purification from blood.

Cancer-derived extracellular vesicles were identified and investigated in a collagen layer using spinning disk microscopy. Investigating extracellular vesicles in collagen allows us to study vesicles prior to any processing in a medium resembling the extracellular matrix. Using indirect immunofluorescence staining we can significantly amplify the detected signal. With this method, we can quantify biomarkers on extracellular vesicles shed from cells embedded in collagen with high sensitivity.

Finally, the growing need to investigate liquid biopsies was addressed here. We developed an application that can handle exosomes from sources with varied biological complexity. These were commercial exosomes, exosomes derived from cell culture media, and exosomes derived from a liquid biopsy with minimal processing. Blood-derived exosomes from a liquid biopsy were purified from plasma proteins in the micromolar range successfully using our protocol. We used immuno-staining to target biomarkers on the exosomal surface and detect the signal from the fluorophore-conjugated antibodies. Using our model, we were able to characterize the varied exosome populations over their size and zeta potential. Overall, this work shows that diffusiophoretic trapping can serve as a highly specific and minimally invasive approach for purification and characterization of exosomes from a liquid biopsy. The method is label-compatible, sensitive to surface charge and size, and adaptable to various sample sources.
Original languageEnglish
PublisherDTU Health Technology
Number of pages220
Publication statusPublished - 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 3 - Good Health and Well-being
    SDG 3 Good Health and Well-being

Fingerprint

Dive into the research topics of 'Characterization of extracellular vesicles from biofluids using a nanofluidic trap'. Together they form a unique fingerprint.

Cite this