Quantitative and qualitative flow cytometric analysis of nanosized cell-derived membrane vesicles

Abstract

Nanosized cell-derived membrane vesicles are increasingly recognized as therapeutic vehicles and high-potential biomarkers for several diseases. Currently available methods allow bulk analysis of vesicles but are not suited for accurate quantification and fail to reveal phenotypic heterogeneity in membrane vesicle populations. For such analyses, single vesicle-based, multiparameter, high-throughput methods are needed. We developed a fluorescence-based, high-resolution flow cytometric method for quantitative and qualitative analysis of nanosized membrane vesicles. Proof of principle was obtained by single-particle analysis of virions and liposomes. Further validation was obtained by quantification of cell-derived nanosized membrane vesicles from cell cultures and body fluids. An important aspect was that the technology was extended to detect specific proteins on individual vesicles. This allowed identification of exosome subsets and phenotyping of individual exosomes produced by dendritic cells (DCs) undergoing different modes of activation. The described technology allows quantitative, multiparameter, and high-throughput analysis of a wide variety of nanosized particles and has broad applications.

From the clinical editor: The authors developed a fluorescence-based, high-resolution flow cytometric method for quantitative and qualitative analysis of nanosized cell-derived membrane vesicles that are increasingly recognized both as therapeutic vehicles and high-potential biomarkers for several diseases. A high throughput, easily available, and sensitive detection method such as the one discussed here is a critically important prerequisite for further refinements of this technology.

Graphical abstract

Figure 1

Quantification of nanosized particles by flow cytometric analysis. (A) Serial twofold dilutions of fluorescent 100-nm beads were mixed with a fixed number of fluorescent 200-nm beads. Samples were measured using fluorescence threshold triggering and the absolute numbers of 100-nm and 200-nm beads were analyzed. Expressed is the ratio of 100-nm beads versus 200-nm bead at each dilution (slope -0.9637 ± 0.064, R2 = 0.999, as determined by linear regression). One representative experiment out of three is shown. (B) Serial twofold dilutions of fluorescent 100-nm beads were prepared and measured using fluorescence threshold triggering. The absolute number of beads measured in a fixed time window (30 sec) was plotted against the dilution factor (slope -1.012 ± 0.012, R2 = 1.00, as determined by linear regression). The measured values (♦) are plotted together with the calculated amount of input beads (x) based on the specified concentration of beads in the stock solution. One representative experiment out of three is shown.

Figure 2

Analysis of nanosized virions and liposomes. (A-B, D-F) Flow cytometric analysis of CFSE-labeled mouse hepatitis virions (MHV) and calcein-labeled liposome preparations I and II using fluorescence threshold triggering. Wide-angle FSC levels were compared with those of 100- and 200-nm fluorescent beads as indicated. (C, G) Virions and liposomes were sized by NTA.

Figure 3

Flow cytometric analysis of exosomes from cell culture supernatant. (A) Schematic diagram of the method developed to quantify and characterize cellular membrane vesicles by flow cytometric analysis. Exosomes were isolated from culture supernatants of LPS-activated DC cultures by differential centrifugation. Pelleted (100,000g) exosomes were fluorescently labeled with PKH67 and loaded at the bottom of a sucrose gradient, after which the vesicles were floated to equilibrium by ultracentrifugation. Sucrose gradient fractions were analyzed using a threshold on PKH67 fluorescence. (B) Time-based quantification of fluorescent membrane vesicles detected in collected gradient fractions. Indicated are the numbers of events measured in 30 sec. One representative experiment out of five is shown. (C) Dot plot of wide-angle FSC versus PKH67 fluorescence of DC exosomes pooled from 1.11–1.18 g/mL fractions. (D) Histogram indicating the wide-angle FSC of DC exosomes relative to the scatter levels of 100- and 200-nm beads. (E) Exosome size was determined by NTA. The mean size of these DC exosomes was estimated to be 83 ± 47 nm.

Figure 4

Detection and quantification of human seminal fluid nanosized membrane vesicles by flow cytometric analysis. Membrane vesicles were isolated from seminal fluid, fluorescently labeled with PKH67, loaded onto a sucrose gradient and centrifuged to equilibrium density. One representative experiment out of two independent experiments is shown. (A) Gradient fractions were analyzed for the presence of CD9 by western blotting. (B) Sucrose gradient fractions were analyzed using a threshold on PKH67 fluorescence. Time-based (30 sec) quantification of fluorescent membrane vesicles (black bars) detected in different gradient fractions and dot plot of wide-angle FSC versus PKH67 fluorescence representing fluorescent vesicles pooled from fractions with densities of 1.09–1.17 g/mL. As a control for unbound dye aggregates, PKH67 dye-only without membrane vesicles was loaded onto a sucrose gradient (gray bars).

Figure 5

Flow cytometric characterization of exosome subsets. Exosomes were isolated from culture supernatants of murine DC, T cell or DC-T cell co-cultures. (A) Time-based quantification of fluorescent exosomes detected in different gradient fractions. Indicated are the numbers of events measured in 30 sec. One representative experiment out of four is shown. (B) Exosomes from LPS-activated DC cultures were additionally stained with R-PE-labeled anti-MHCII (black bars) or isotype control (gray bars) antibodies. Indicated are the geometric mean fluorescence intensities of events measured in each of the gradient fractions. One representative experiment out of three is shown. (C) Dot plots representing PKH67 and anti-MHCII-R-PE or isotype control antibody-labeled exosomes (pools of 1.11–1.18 g/ml sucrose fractions) derived from T cell cultures (left), LPS-activated DC cultures (middle) or DC/LPS-T cell co-cultures (right). One representative experiment out of three is shown. (D) Dot plots representing PKH67 and anti-MHCII (APC-labeled) / anti-MFG-E8 (B-PE-labeled) double-labeled exosomes or corresponding isotype control labeled exosomes derived from nonactivated or LPS-activated DC cultures.