Proteomics Profiling of Neuron-Derived Small Extracellular Vesicles from Human Plasma: Enabling Single-Subject Analysis

Abstract

Small extracellular vesicles have been intensively studied as a source of biomarkers in neurodegenerative disorders. The possibility to isolate neuron-derived small extracellular vesicles (NDsEV) from blood represents a potential window into brain pathological processes. To date, the absence of sensitive NDsEV isolation and full proteome characterization methods has meant their protein content has been underexplored, particularly for individual patients. Here, we report a rapid method based on an immunoplate covalently coated with mouse monoclonal anti-L1CAM antibody for the isolation and the proteome characterization of plasma-NDsEV from individual Parkinson's disease (PD) patients. We isolated round-shaped vesicles with morphological characteristics consistent with exosomes. On average, 349 ± 38 protein groups were identified by liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis, 20 of which are annotated in the Human Protein Atlas as being highly expressed in the brain, and 213 were shared with a reference NDsEV dataset obtained from cultured human neurons. Moreover, this approach enabled the identification of 23 proteins belonging to the Parkinson disease KEGG pathway, as well as proteins previously reported as PD circulating biomarkers.

Keywords: L1CAM; NES cells; Parkinson’s disease; central nervous system; exosomes; miR-155; neuron-derived vesicles; plasma; proteomics; small extracellular vesicles.

Figure 1

Cultured human neuron-derived small extracellular vesicles (NDsEV) isolation and characterization. (A) Size Exclusion Chromatography using in-flow Dynamic Light Scattering detection; the graph presents vesicle diameter (d.nm) in blue and the derived count rate in red. The orange box highlight indicates the fractions selected for NDsEV isolation. (B) Venn diagram of the proteins identified in the neuroepithelial stem cell neuron-derived sEV (NES-NDsEV) and their overlap with the ExoCarta Top 100 database. (C) Exosome markers identified in the NES-NDsEV. (D) Cellular component Gene Ontology enrichment analysis of the NES-NDsEV proteins using the DAVID database [33]. (E) Identified proteins in NES-NDsEV in common with the Human Brain Proteome (Human Proteome Atlas) database annotated as significantly enriched genes in the brain (488 entries).

Figure 2

Plasma neuron derived small extracellular vesicles (NDsEV) characterization. (A) Nanoparticle-tracking analysis of plasma-NDsEV isolated from healthy control plasma samples and (B) Parkinson disease patient plasma. (C) Representative TEM images of plasma-NDsEV from healthy controls and (D) PD patients. Scale bars: 100 nm. (E) miR155 expression in 4 PD patients’ NDsEV and 4 healthy controls’ NDsEV; the histograms report the mean of the densitometric values of the bands of controls (in black) and of PD (in grey) normalized with respect to the GAPDH ± SD * p < 0.05.

Figure 3

(A) NDsEV proteomics workflow: NDsEV were isolated from platelet-free plasma by L1CAM immune-precipitation, and S-traps used to digest the proteins and to purify the peptides that are then analyzed by LC-MS/MS for protein identification and quantification. (B) Number of protein groups identified in each plasma sample. (C) Venn diagram of the proteins identified in the replicates and the ExoCarta Top 100 database. (D) Cellular component Gene Ontology enrichment analysis of the proteins identified from plasma-NDsEV of PD patients using the DAVID database [33].

Figure 4

(A) Average number of identified proteins groups of PD-NDsEV and relative standard deviations for the technical replicates (PD1A, PD1B and PD1C) and biological replicates (PD2, PD3 and PD4). (B) Venn diagram showing the number of common and unique proteins among the technical replicates and among biological replicates of PD-NDsEV. (C) Protein intensity correlation among the technical replicates and among biological replicates; the figures report the Pearson’s correlation coefficient extracted with Perseus software [41].