Characterization of Extracellular Vesicles Labelled with a Lipophilic Dye Using Fluorescence Nanoparticle Tracking Analysis

Abstract

Research on extracellular vesicles (EVs) has intensified over the past decade, including fluorescent membrane labeling of EVs. An optimal fluorescent method requires the size of EVs to be preserved after labeling. Lipophilic fluorescent dyes, such as CellMask™ Green (CMG), have been widely used for this purpose. Here, we investigated conditions affecting the optimum CMG labeling of EVs derived from human choriocarcinoma cells (JAr) and different biological fluids using fluorescence NTA (fl-NTA). The effect of CMG labeling on the size, concentration and zeta potential (ZP) on JAr EVs purified with different methods were measured along with biological fluid-derived EVs. With the increase of CMG dye concentration, a significant decrease in the mean size of fluorescent nanoparticles (fl-NPs) was observed. The ZP of fl-NPs originating from JAr cells with the lowest and highest dye concentrations showed a significant shift towards more and less negative ZP values, respectively. Differences in the concentration of fl-NPs were observed for JAr EVs purified using size-exclusion chromatography (SEC) alone and SEC in combination with tangential flow filtration. The proportion of CMG labeling of NPs varied across different biological sources. CMG labeling may be a reliable technique for the detection of EVs using fl-NTA.

Characterization of bovine uterine fluid extracellular vesicles proteomic profiles at follicular and luteal phases of the oestrous cycle

“…In the current study, we measured the proportion of particles labelled with CMG resulting to lower particle counts compared to measurements without the dye, which indicate that only a proportion of particles measured with the conventional NTA could be EV. This result is in line with previous studies, which have measured proportionately lower fluorescently labelled particles compared to total nanoparticles in cell culture media (Midekessa et al 2021 ) and human urine samples (Marcu et al 2020 ). Hence, alternative methods to conventional NTA and standardization of these measurements are necessary to accurately measure EV counts needed for biomedical function and for the understanding their biological properties (Hartjes et al 2019 ).…”

“…However, different methods have been developed to distinguish EV from other types of nanoparticles, one of them being membrane labelling of EV using lipophilic fluorescent dyes. These lipophilic dyes stain the lipid bilayer of the EV, which could be measured with fluorescence NTA (Midekessa et al 2021 ). In the current study, we measured the proportion of particles labelled with CMG resulting to lower particle counts compared to measurements without the dye, which indicate that only a proportion of particles measured with the conventional NTA could be EV.…”

Biomolecular interactions on densely coated nanoparticles: a single-molecule perspective

“…Additionally, the variation in surface charge on the particles could result in patchy particles, and the inter-particle variation in the degree of patchiness could result in such heterogeneity. Methods like nanoparticle tracking analysis 42 and tunable resistive pulse sensing 43 could in the future be used to quantify the heterogeneity in surface charge of bio-functionalized colloidal nanoparticles. However, correlating dispersion in zeta-potential with receptor density ( N ) in individual NPs, require advanced correlation methods.…”

Extracellular vesicles as carriers of mRNA: Opportunities and challenges in diagnosis and treatment

“…To capture an entire population of EVs, multiple surface markers can be used, but this increases overall expense and is relatively low-throughput 33 . Lipophilic and other complete membrane dyes may be used, however, these may alter EV properties, decrease stability, or increase particle aggregation 34 , 35 . Flow cytometry technologies have been expanded to the EV field, however, unbound dye may limit signaling from EVs resulting in low output readings after normalization 36 .…”