Measurement of α-synuclein as protein cargo in plasma extracellular vesicles

Abstract

Extracellular vesicles (EVs) are released by all cells and hold great promise as a class of biomarkers. This promise has led to increased interest in measuring EV proteins from both total EVs as well as brain-derived EVs in plasma. However, measuring cargo proteins in EVs has been challenging because EVs are present at low levels, and EV isolation methods are imperfect at separating EVs from free proteins. Thus, knowing whether a protein measured after EV isolation is truly inside EVs is difficult. In this study, we developed methods to measure whether a protein is inside EVs and quantify the ratio of a protein in EVs relative to total plasma. To achieve this, we combined a high-yield size-exclusion chromatography protocol with an optimized protease protection assay and Single Molecule Array (Simoa) digital enzyme-linked immunoassays (ELISAs) for ultrasensitive measurement of proteins inside EVs. We applied these methods to analyze α-synuclein and confirmed that a small fraction of the total plasma α-synuclein is inside EVs. Additionally, we developed a highly sensitive Simoa assay for phosphorylated α-synuclein (phosphorylated at the Ser129 residue). We found enrichment in the phosphorylated α-synuclein to total α-synuclein ratio inside EVs relative to outside EVs. Finally, we applied the methods we developed to measure total and phosphorylated α-synuclein inside EVs from Parkinson's disease and Lewy body dementia patient samples. This work provides a framework for determining the levels of proteins in EVs and represents an important step in the development of EV diagnostics for diseases of the brain, as well as other organs.

Keywords: Parkinson’s Disease; alpha synuclein; biomarker; exosomes; extracellular vesicles.

Fig. 1.

The distribution of α-synuclein in SEC fractions isolated from plasma. (A) Schematic of the overall goal: comparing the level of α-synuclein in EVs relative to α-synuclein in total plasma. (B) Pooled plasma was fractionated by SEC and Simoa was used to measure levels of the EV surface marker CD9 (Top), soluble protein marker Albumin (soluble protein marker, purple), and α-synuclein (blue) in each fraction. All the data are displayed as mean ± SD. (C) Levels of tetraspanins CD9, CD63, and CD81 were measured by Simoa in EVs isolated from pooled plasma by SEC (fractions 7 to 10) as well as in unfractionated plasma. EV recovery percentage was calculated as the average of the ratios of each tetraspanin in EVs relative to unfractionated plasma. All the data are displayed as mean ± SD of duplicate measurements. (D) Levels of α-synuclein were measured by Simoa in EVs isolated from six individual plasma samples by SEC (fractions 7 to 10) as well as in unfractionated plasma. The percentage of α-synuclein in the EV fractions relative to the total plasma α-synuclein was calculated for each sample.

Fig. 2.

Measurement of plasma α-synuclein, Tau, Aβ40, and Aβ42 levels in healthy and disease samples and corresponding EVs isolated by SEC. Levels of (A) Aβ40, (B) Aβ42, and (C) Tau were measured by Simoa in EVs isolated by SEC (fractions 7 to 10) from three individual plasma samples. Levels of each protein were also measured in corresponding unfractionated plasma samples, and a percentage of each protein in EVs was calculated. All the data are displayed as mean ± SD. The dotted line indicates limit of detection for each Simoa assay. (D) Levels of Aβ40, (E) Aβ42, (F) Tau, and (G) α-synuclein were measured by Simoa in EVs isolated by SEC (fractions 7 to 10) from pooled plasma samples of the following: healthy controls (HC), PD, or AD. Levels of each protein were also measured in corresponding unfractionated plasma samples, and a percentage of each protein in EVs was calculated. All the data are displayed as mean ± SD of duplicate measurements. The dotted line indicates the limit of detection for each Simoa assay.

Fig. 3.

Protease protection assay for estimating the internal and external α-synuclein in plasma EVs isolated by SEC. (A) Schematic illustration of the Protease protection assay. Proteins that are not inside of EVs and protected by the lipid membrane are digested by PK. PK is then inhibited and the contents of the EV are released by lysing the EVs with detergent. (B) Protease protection assay was performed on individual SEC fractions isolated from plasma. The levels of α-synuclein in each fraction were measured by Simoa after protease protection (blue bars), or with PK treatment after the addition of detergent (red bars). (C) Protease protection assay was performed on pooled EV fractions (fractions 7 to 10; Right) isolated from plasma; levels of α-synuclein were measured by Simoa in EVs isolated by SEC from pooled plasma and three conditions were compared: NT, Protease protection assay, or PK treatment after EV lysis with TritonX-100 (Triton-PK). All the data are displayed as mean ± SD. The dotted line indicates limit of detection for α-synuclein Simoa assay. (D) The percentage of protease-protected α-synuclein (internal α-synuclein) relative to α-synuclein in untreated EVs isolated from six individual plasma samples by SEC (fractions 7 to 10) using the same samples as those used in Fig. 1D. (E) Schematic illustrating that cargo protein (α-synuclein) in an immuno-isolated subset of EVs (such as cell-type-specific EVs), must be less than or equal to the level of that cargo protein in total EVs. (F) Schematic illustrating Protease protection assay after EV immuno-isolation to ensure cargo protein is inside EVs. Three conditions are compared: NT, Protease protection assay, and PK treatment after EV lysis with TritonX-100 (Triton-PK). (G) α-synuclein was measured by Simoa in EVs purified from plasma by SEC and then immuno-isolated using antibodies against CD9, CD63, L1CAM, GFP, or an additional isotype control. For each immuno-isolation, the following were compared: NT, Protease protection assay, or PK treatment after EV lysis with Triton X-100 (Triton-PK). Each experimental condition was performed in two different pools of plasma and each Simoa measurement was performed in duplicate. Results from two pools are averaged and displayed as mean ± SD.

Fig. 4.

Measurement of α-synuclein and pser129 in clinical samples. α-synuclein and pSer129 were measured in samples from a cohort of 45 patients, comprising 15 PD (purple), 15 DLB (blue), and 15 controls (green). Each sample underwent fractionation using our optimized SEC protocol, and the EV fractions (SEC fractions 7 to 10) as well as free alpha-synuclein fractions (SEC fractions 17 to 20) were collected. Protease protection assay was performed on the EVs for measuring the internal protein levels. (A) Ratio between pSer129 and total α-synuclein in the EVs and the free protein fractions is indicated on the Left axis. Fold increase of the percentage of phosphorylated α-synuclein between EVs and free protein fractions, showing that phosphorylated α-synuclein is enriched in EVs is indicated on the Right axis. Levels of (B) α-synuclein in EVs, (C) α-synuclein in the free protein fractions, (D) ratio between α-synuclein in EVs and free α-synuclein, (E) pSer129 in EVs, (F) pSer129 in the free protein fractions, (G) ratio between pSer129 in EVs and free pSer129 were measured by Simoa. Each data point represents the average of duplicate measurements. Sample groups were compared by the nonparametric t test. *P < 0.05. Mean value for each group is indicated.