Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 9 (1), 2042

High Sensitivity Detection of Extracellular Vesicles Immune-Captured From Urine by Conventional Flow Cytometry

Affiliations

High Sensitivity Detection of Extracellular Vesicles Immune-Captured From Urine by Conventional Flow Cytometry

Carmen Campos-Silva et al. Sci Rep.

Abstract

Extracellular vesicles (EVs) provide an invaluable tool to analyse physiological processes because they transport, in biological fluids, biomolecules secreted from diverse tissues of an individual. EV biomarker detection requires highly sensitive techniques able to identify individual molecules. However, the lack of widespread, affordable methodologies for high-throughput EV analyses means that studies on biomarkers have not been done in large patient cohorts. To develop tools for EV analysis in biological samples, we evaluated here the critical parameters to optimise an assay based on immunocapture of EVs followed by flow cytometry. We describe a straightforward method for EV detection using general EV markers like the tetraspanins CD9, CD63 and CD81, that allowed highly sensitive detection of urinary EVs without prior enrichment. In proof-of-concept experiments, an epithelial marker enriched in carcinoma cells, EpCAM, was identified in EVs from cell lines and directly in urine samples. However, whereas EVs isolated from 5-10 ml of urine were required for western blot detection of EpCAM, only 500 μl of urine were sufficient to visualise EpCAM expression by flow cytometry. This method has the potential to allow any laboratory with access to conventional flow cytometry to identify surface markers on EVs, even non-abundant proteins, using minimally processed biological samples.

Conflict of interest statement

CSIC and UAM participate in an R&D contract with Immunostep, S.L. RJA is CEO in Immunostep, S.L.

Figures

Figure 1
Figure 1
Characterization of PC3-derived EVs. (A) Nanoparticle tracking analysis (NTA). Average size and concentration were obtained in a Nanosight equipment capturing 3 videos of 60 s per measurement, with a focus −15 to +15 and camera level 12. ɸ: diameter. (B) Western Blot. EVs were loaded on SDS-PAGE and immunoblotted for β-actin (Sigma) and antibodies against tetraspanins [anti-CD9 (MEM62), -CD63 (MEM259) and -CD81 (MEM-38)]. Three gels were loaded: one gel, under non-reducing conditions with 2.2·109 particles, for CD9 and CD81 detection, exposed for 2 min; a second non-reducing gel with 6.8 · 109 particles, for CD63 detection, exposed for 1 h; and the third gel under reducing conditions, for actin detection, exposed for 40 s. The experiment shown is representative of 4.
Figure 2
Figure 2
(A) Theoretical calculation of the number of EVs immobilised per bead. The graph represents at real scale a 6 μm-diameter bead and an EV of 150 nm of diameter (ɸ). The external surface of a bead (Sbead) could capture a maximum of N times the circle area (Cexo) of 150 nm EVs. N is calculated by dividing Sbead/Cexo. (B) Schematic representation, not to scale, of the bead binding assay. 6 μm-diameter fluorescent magnetic beads were coated with capture antibody and used for immune-capture of the EVs. A second antibody directed against the same or a different molecule, either biotinylated or directly conjugated with PE, was used for detection. In the case of biotinylated antibodies, PE-conjugated streptavidin was subsequently used. Samples were analysed by conventional flow cytometry.
Figure 3
Figure 3
Specificity of EV immunocapture on antibody-coated microbeads. (A) Gating strategy. EVs immobilised on 6 μm APC-beads were stained using biotinylated antibody followed by PE-conjugated streptavidin and analysed by flow cytometry. A gate containing only single beads was created in the Forward Scatter (FSC)/Side Scatter (SSC) plot. A second gate, within single beads, confirmed the APC fluorescence of microbeads. 1500–2000 events from this combined gating were acquired and analysed for PE labelling. (B) Negative control, IgG1. 109 particles of PC3-derived EVs were captured onto either anti-CD63 (Clone TEA3/18) or IgG1-coated beads followed by detection with biotinylated antibody directed against CD9. A sample with no EVs is also shown for comparison. (C) Antibody blocking. 109 particles of PC3-derived EVs were pre-incubated with increasing amounts of the indicated soluble blocking antibody [anti-CD63 (Clone TEA3/18), anti-CD9 (Clone VJ1/20)] before being incubated for capture on CD63- (left) or CD9- (right) coated beads. Experiments are representative of 3 independent repetitions.
Figure 4
Figure 4
Optimization of tetraspanin antibody combination. 4∙109 particles of PC3-derived EVs were captured onto either anti-CD9 (Clone VJ1/20) or anti-CD63 (Clone TEA3/18) coated beads followed by detection with PE-conjugated antibodies directed against CD9, CD81 or CD63. The sensitivity of each antibody combination is compared using the Stain Index (SI) SI=MFIpositiveMFIbackground2σbackground, where σ is the standard deviation and MFI Mean Fluorescence Intensity; and the Relative fluorescence Index (RFI), RFI=MFIpositiveMFIbackground, indicated on the upper right corner of each panel. A representative experiment out of 3 is shown.
Figure 5
Figure 5
Dynamic range and limit of detection. EVs were captured on anti-CD63-coated beads followed by detection with biotinylated anti-CD9 antibody (Clone VJ1/20). (A) Flow cytometry analysis profiles. Increasing amounts (30 ng to 10 μg; 1.37∙107–4.59∙109 particles) of PC3-derived EVs were captured on 6 μm beads and detected by flow cytometry. The total volume of the assay was 100 μl. The graph represents the overlay of the curves obtained. The legend indicates the amount of EVs and the RFI for each curve. (B) Regression analysis. RFI (right) and SI values (left) were plotted as a scatter graph and fitted to a polynomic curve (upper panel). Linearity can be observed between 0.2–8 µg of PC3-derived EVs with a r2 > 0.98 (lower panels). The minimal amount of EVs detected (RFI > 1), corresponds to 30 ng yielding a RFI of 1.08. A representative experiment out of 3 is shown.
Figure 6
Figure 6
Detection of EVs from healthy donors urine. (A) Purified EVs from urine. 68 µg (6.8∙1010 particles) of EVs from healthy donors (Hansa Biomed) were analysed by Western Blot for detection of CD63 (1 h exposure), CD81 (10 min exposure), CD9 (10 s exposure) and β-actin (10 min exposure) (left). 10 µg (109 particles) were captured onto anti-CD63-coated beads followed by detection with biotinylated anti-CD9 antibody by flow cytometry (right). (B) EVs contained in 500 µl of urine analysed by flow cytometry. 500 µl of healthy donor urine was pre-treated by mild reduction (see methods) and incubated with either anti-CD63- or IgG1-coated beads followed by detection with biotinylated anti-CD9 antibody by flow cytometry. Stain Index (SI) and the Relative fluoresce Index (RFI) are indicated inside each panel. A representative experiment out of 3 is shown.
Figure 7
Figure 7
Direct detection of EpCAM in urinary EVs. (A) EpCAM detection on PC3-derived EVs. 10–20 µg (4.59–9.18∙109 particles) of PC3-derived EVs were captured onto anti-CD63-coated beads followed by detection with biotinylated anti-EpCAM antibody by flow cytometry. (B) 5–10 µg (2.29–4.59∙109 particles) of PC3-derived EVs were captured onto anti-EpCAM-coated beads followed by detection with biotinylated anti-CD9 antibody by flow cytometry. (C) Specificity of anti-EpCAM-coated beads: antibody blocking. To confirm the specificity of anti-EpCAM-coated beads, PC3-derived EVs were pre-incubated with anti-EpCAM antibody, previously to their incubation with microbeads. (D) Specificity of anti-EpCAM-coated beads: negative control. 1.78∙109 particles of SK-Mel-28-derived EVs (EpCAM-negative cell line) were captured onto either anti-CD63-coated beads, anti-EpCAM- or IgG1-coated beads followed by detection with biotinylated anti-CD81 antibody by flow cytometry. (E) Detection of EpCAM by WB in EVs from 8 ml of urine. 8 ml of pre-treated urine from 3 healthy donors (HD) and 6 patients (P1- P6) were used to purify EVs by ultra-centrifugation and they were analysed by WB. The number under the EpCAM panel corresponds to the Relative fluoresce Index (RFI) in the flow cytometry experiment. (F) Detection of EpCAM by flow cytometry in EVs from 500 µl of urine. 500 µl of pre-treated urine were incubated with either IgG1 or anti- EpCAM-coated beads followed by detection with biotinylated anti-CD9 antibody and flow cytometry analysis. Three flow cytometry experiments were performed using the same patient samples and the results from a representative experiment are shown.

Similar articles

See all similar articles

Cited by 2 articles

References

    1. Yanez-Mo M, et al. Biological properties of extracellular vesicles and their physiological functions. Journal of extracellular vesicles. 2015;4:27066. doi: 10.3402/jev.v4.27066. - DOI - PMC - PubMed
    1. Colombo M, et al. Analysis of ESCRT functions in exosome biogenesis, composition and secretion highlights the heterogeneity of extracellular vesicles. J Cell Sci. 2013;126:5553–5565. doi: 10.1242/jcs.128868. - DOI - PubMed
    1. Leblanc P, et al. Isolation of Exosomes and Microvesicles from Cell Culture Systems to Study Prion Transmission. Methods Mol Biol. 2017;1545:153–176. doi: 10.1007/978-1-4939-6728-5_11. - DOI - PubMed
    1. Simpson, R. J., Kalra, H. & Mathivanan, S. ExoCarta as a resource for exosomal research. Journal of extracellular vesicles1, 10.3402/jev.v1i0.18374 (2012). - PMC - PubMed
    1. Kim DK, et al. EVpedia: a community web portal for extracellular vesicles research. Bioinformatics. 2015;31:933–939. doi: 10.1093/bioinformatics/btu741. - DOI - PMC - PubMed

Publication types

Feedback