Cells release subpopulations of exosomes with distinct molecular and biological properties

Abstract

Cells release nano-sized membrane vesicles that are involved in intercellular communication by transferring biological information between cells. It is generally accepted that cells release at least three types of extracellular vesicles (EVs): apoptotic bodies, microvesicles and exosomes. While a wide range of putative biological functions have been attributed to exosomes, they are assumed to represent a homogenous population of EVs. We hypothesized the existence of subpopulations of exosomes with defined molecular compositions and biological properties. Density gradient centrifugation of isolated exosomes revealed the presence of two distinct subpopulations, differing in biophysical properties and their proteomic and RNA repertoires. Interestingly, the subpopulations mediated differential effects on the gene expression programmes in recipient cells. In conclusion, we demonstrate that cells release distinct exosome subpopulations with unique compositions that elicit differential effects on recipient cells. Further dissection of exosome heterogeneity will advance our understanding of exosomal biology in health and disease and accelerate the development of exosome-based diagnostics and therapeutics.

Figure 1. B16F10 melanoma cells release distinct subpopulations of exosomes.

(A) Schematic overview of the isolation protocol used to obtain different exosome populations. (B,C) P110 was loaded on top (B) or at the bottom (C) of a sucrose density gradient and subjected to ultracentrifugation for 16 h. The resulting fractions (1–10) with increasing density were analyzed for particle number by NTA (upper panels) and the presence of exosome marker proteins ALIX and TSG101 by Western blotting (lower panels). For Western blots, an equal volume of each sample was analyzed. Fractions 3–5 (LD-Exo) and 8–9 (HD-Exo) were pooled for further analysis. (D) Exosomes or MVs were negatively stained with uranyl acetate and visualized by transmission electron microscopy. Scale bars represent 100 nm. (E) Size distribution profiles as determined by NTA. Data shown are representative of three independent experiments. (F,G) Exosomes from B16F10 melanoma cells cultured for 48h in the absence or presence of 7.5 μM GW4869 were isolated as in Fig. 1A. P110 was loaded at the bottom of a sucrose gradient and subjected to ultracentrifugation for 16 h. The resulting fractions (1–10) with increasing density were analyzed for particle number by NTA (F) and the presence of exosome marker proteins ALIX and TSG101 by Western blotting (G). A control sample (Ctrl) was included to show consistent analysis across the membranes. For Western blots, an equal volume of each sample was analyzed. Data shown are representative of two independent experiments.

Figure 2. Exosome subpopulations are released by different cell types and are detected in plasma.

Exosomes derived from (A) N2a cells, (B) A431 cells, (C) H5V cells, (D) MSC cells, or (E) plasma were isolated as in Fig. 1A. P110 was loaded at the bottom of a sucrose density gradient and ultracentrifugated for 16 h. The resulting fractions (1–10) with increasing density were analyzed for particle number by NTA (upper panels) and the presence of exosome marker proteins ALIX and TSG101 by Western blotting (lower panels). For Western blots, an equal volume of each sample was analyzed. Data shown are representative of two independent experiments.

Figure 3. Proteome characterization of EV subpopulations by nanoLC-MS/MS reveals differences in protein composition.

(A) Overlap of identifications between MV, LD-Exo and HD-Exo. Proteins were considered identified if they had quantitative protein area in at least 2 of the 3 replicates. (B) Heatmap showing protein intensity (area) for each replicate. Subpopulations were grouped based on overlap in identifications as in (A), and subsequently ranked from highest to lowest average protein area. (C) Validation of proteomics data for selected proteins (left panel) by Western blotting (right panel). Equal amounts of protein were analyzed. (D) GO enrichments of all identified proteins and proteins identified in MV, LD-Exo and HD-Exo. (E) GO enrichments of proteins uniquely identified in MV, LD-Exo and HD-Exo.

Figure 4. EV subpopulations have different RNA profiles.

RNA from MV, LD-Exo and HD-Exo was extracted using Trizol and analyzed using capillary electrophoresis with the Agilent RNA 6000 Pico chip (left panel) and Agilent small RNA chip (right panel) on an Agilent 2100 Bioanalyzer®. The y-axis of the electropherograms represents fluorescence units (FU) and the x-axis represents the nucleotide length of the RNA (nt). Peaks at 25 nt (left panels) or at 4 nt (right panels) represent internal standards. Data shown are representative of two independent experiments.

Figure 5. LD-Exo and HD-Exo have different effects on gene expression in recipient cells.

LD-Exo and HD-Exo were isolated as in Fig. 1A and incubated with H5V endothelial cells for 24 h. (A) Heat map displaying genes that were upregulated (red) or downregulated (green) in response to LD-Exo or HD-Exo exposure (>1.5 fold change, q-value < 0.15). (B) RT-qPCR validation of array results. Transcript levels were measured relative to GAPDH and plotted relative to levels in PBS-treated cells. Dot plots represent mean ± SD. (n = 4). (C) Representative GO enrichments differentially regulated (upregulated,  + , or downregulated, -) by LD-Exo versus HD-Exo. #, number of genes.