|
Abstract INTRODUCTION:
Clinical application of currently available ALS disease pathology markers is limited due to an overall lack of specificity, creating a pressing need for improved markers. Extracellular vesicles (EVs) have been implicated in neurodegenerative diseases as a potential propagation mechanism for protein aggregates and have shown promise as peripheral markers of disease burden. As such, EVs represent a valuable reservoir of potential biomarkers of ALS disease pathology that could be leveraged to improve ALS diagnosis and disease monitoring. EVs are small membranous vesicles released from most cell types under both physiological and pathological conditions, comprising three main subtypes: exosomes (small EVs [SEVs]), microvesicles (large EVs [LEVs]), and apoptotic bodies, differentiated by their biogenesis, size, and mechanism of release. In this study, we utilized a combination of untargeted and targeted proteomics to assess the potential of cerebrospinal fluid (CSF) derived EVs as a source of ALS-specific markers.
METHODS:
CSF samples from 28 ALS cases and 21 disease controls (14 normal pressure hydrocephalus [NPH] and 7 non-NPH controls) were obtained from the neuroimmunology laboratory at Mayo Clinic in Rochester MN, with IRB consent. EVs were isolated from 0.5 mL of CSF using Izon qEV1 size exclusion chromatography (SEC) columns after removing cellular debris with centrifugation at 2000 x g for 5 min. EV containing fractions (fractions 5-7.5) were collected from the SEC column, transferred to a 10 mL conical bottom ultracentrifugation tube and pelleted with centrifugation at 20,000 x g for 20 min to collect the LEV fraction and 100,000 x g for 1 hr to collect the SEV fraction. SEV pellets were proteolytically digested with trypsin/LysC mix overnight. After digestion, peptides were separated on a reversed-phase C-18 (75 µm x 25 cm, IonOpticks) column on a Vanquish Neo UHPLC using a 37 min method. Eluting peptides were analyzed on an Orbitrap Astral mass spectrometer operated in data independent acquisition (DIA) mode. MS1 scans were acquired in the Orbitrap analyzer at 240K with 500% AGC between 380-980 m/z and MS2 scans were acquired in the Astral analyzer with 3 m/z - 7ms and 500% AGC. Resulting raw files were processed in Spectronaut software and protein group intensities were extracted from the resulting text file output. Differential expression analysis was performed by fitting the log2 transformed normalized MS2 protein intensities from each group to a Gaussian-linked generalized linear model and an ANOVA was implemented to detect differentially expressed proteins between the experimental groups. Differentially expressed proteins were defined using a criterion of FDR ≤ 0.05 and a log2 fold change ≥1.0. All computations were performed in R version 4.3.2.
Targeted mass spectrometry was performed on an Orbitrap Exploris 480 mass spectrometer coupled to an Ultimate 3000 RSLC-nano HPLC with a trap and elute configuration. Briefly, peptides were separated on a C-18 column (15cm x 180um, IonOptiks) using a 30min gradient and analyzed using scheduled PRMs with an isolation window of ±3min of the target retention time. The top-3 most abundant product ions were selected for monitoring and all data was processed in Skyline software.
RESULTS:
In total, 5772 proteins were identified across all samples in this study. Among the identified proteins were several EV markers, including CD9, CD81, CD63, TSG101, PDCD6IP and FLOT1, indicating the presence of bona fide EVs. Using a criterion of FDR ≤ 0.05 and a log2 fold change ≥1.0, 434 proteins were identified as significantly differentially expressed (325 upregulated, 109 downregulated). Clustering analysis of significantly regulated proteins resulted in two distinct clusters, largely separating ALS patients and controls. A targeted mass spectrometry assay was developed to confirm the top 11 most promising protein candidates utilizing 2-stable isotope-labeled (SIL) peptides per protein. Preliminary analysis utilizing residual digest from a subset of the discovery cohort samples (cases: n = 5, controls: n = 5) found that 4 of the 11 proteins displayed good separation between groups resulting in AUCs ranging from 0.76 – 0.9, while the remaining 6 proteins showed expected trends however lacked enough signal for robust quantification.
CONCLUSION:
This study demonstrates that SEC-based EV isolation followed by nano-LC-MS/MS enables comprehensive and quantitative comparisons of the CSF-derived EV proteome. Using this workflow, we identified over 400 proteins that were differentially expressed between ALS cases and disease controls, providing a rich source of potential novel ALS markers. Targeted mass spectrometry analysis of residual discovery cohort digest confirmed the trends observed in the discovery experiment, while also underscoring the need for a more sensitive platform for target confirmation. Further studies in an independent cohort are ongoing and will be required to validate these candidates and evaluate their performance in ALS diagnosis and prognostication.
|
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
