Extracellular Vesicle (EV) Array: microarray capturing of exosomes and other extracellular vesicles for multiplexed phenotyping

Abstract

Background: Exosomes are one of the several types of cell-derived vesicles with a diameter of 30-100 nm. These extracellular vesicles are recognized as potential markers of human diseases such as cancer. However, their use in diagnostic tests requires an objective and high-throughput method to define their phenotype and determine their concentration in biological fluids. To identify circulating as well as cell culture-derived vesicles, the current standard is immunoblotting or a flow cytometrical analysis for specific proteins, both of which requires large amounts of purified vesicles.

Methods: Based on the technology of protein microarray, we hereby present a highly sensitive Extracellular Vesicle (EV) Array capable of detecting and phenotyping exosomes and other extracellular vesicles from unpurified starting material in a high-throughput manner. To only detect the exosomes captured on the EV Array, a cocktail of antibodies against the tetraspanins CD9, CD63 and CD81 was used. These antibodies were selected to ensure that all exosomes captured are detected, and concomitantly excluding the detection of other types of microvesicles.

Results: The limit of detection (LOD) was determined on exosomes derived from the colon cancer cell line LS180. It clarified that supernatant from only approximately 10(4) cells was needed to obtain signals or that only 2.5×10(4) exosomes were required for each microarray spot (~1 nL). Phenotyping was performed on plasma (1-10 µL) from 7 healthy donors, which were applied to the EV Array with a panel of antibodies against 21 different cellular surface antigens and cancer antigens. For each donor, there was considerable heterogeneity in the expression levels of individual markers. The protein profiles of the exosomes (defined as positive for CD9, CD63 and CD81) revealed that only the expression level of CD9 and CD81 was approximately equal in the 7 donors. This implies questioning the use of CD63 as a standard exosomal marker since the expression level of this tetraspanin was considerably lower.

Keywords: EV Array; antigenic capturing; exosomes; extracellular vesicles; nanoparticle tracking analysis; phenotyping; plasma; protein microarray.

Fig. 1

(A) Extracellular vesicle detection using a customized protein microarray “EV Array”. A microarray print with spots of 21 different antibodies is used to capture exosomes from, for example, plasma or cell culture supernatants. The captured vesicles (exosomes) are detected with a cocktail of biotinylated antibodies against the tetraspanins CD9, CD63 and CD81 followed by fluorescence-labelled streptavidin. (B) The technology of using protein microarray for exosome detection was validated using NTA and a series of experiments with different exosome sources and microarray print setups as illustrated in the flowchart. Microarray print with a cocktail of the capturing antibodies CD9, CD63 and CD81 was used to detect the signals from all exosomes present in the samples in a semi-quantitative manner in order to obtain the limit of detection (LOD, Fig. 3). This setup was also used to capture exosomes prior to elution (supplemental material) and validation by NTA. Microarray prints with a panel of 21 different capturing antibodies was used to phenotype the exosomes (positive for CD9, CD63 and/or CD81) present in plasma from 7 healthy individuals (Fig. 4).

Fig. 2

(A) Nanoparticle Tracking Analysis of the starting material from the EV Array analysis compared to the eluted particles after capturing on a microarray printed with a cocktail of antibodies against CD9, CD63 and CD81. The analyses were performed on a plasma sample (Donor N) and on cell media containing exosomes from the cancer cell lines SW948 (colon) and OAW42 (ovarian). No signals were obtained above 500 nm. The mode and mean values for each sample are given. (B) EV Array detection of exosomes captured on slides printed with a cocktail of antibodies against CD9, CD63 and CD81. The fluorescence intensities (mean±SD) are shown in relation to the number of LS180 (colon cancer) cells producing the exosomes within 48 h. A linear correlation (R ²>99%) is seen and the red line indicates 2 SD of the negative control demonstrating a limit of detection (LOD).

Fig. 3

Microvesicle analysis of plasma from 7 healthy donors. (A) A series of dilutions were tested in triplicates on EV Array slides printed with a cocktail of capturing antibodies (against CD9, CD63 and CD81). The relative fluorescence intensities of the spots (mean±SD) are plotted against the volume of plasma added. (B) Nanoparticle Tracking Analysis of the plasma revealed a great variance of the microvesicle distribution and particle concentration. Donors 4 and 5 are observed to have a higher number of vesicles <100 nm (exosomes) as indicated by the grey line.

Fig. 4

Summary of the phenotyping of the exosomal (positive for CD9, CD63 and/or CD81) population in plasma from 7 healthy donors. The exosomes were profiled using an EV Array printed with 21 different capturing antibodies. The relative fluorescence intensity was log2 transformed and a hierarchical clustering was performed to illustrate the phenotypes of the plasma-derived exosomes.