Affiliations: [a] Department of Neurology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, China
| [b] Department of Nuclear Medicine, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, China
Correspondence:
[*]
Correspondence to: Prof. Qiqiang Tang, Department of Neurology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230001, China. Tel.: +86 13625518339; E-mail: [email protected].
Abstract: Background:Early-onset Alzheimer’s disease (EOAD) exhibits a notable degree of heterogeneity as compared to late-onset Alzheimer’s disease (LOAD). The proteins and pathways contributing to the pathophysiology of EOAD still need to be completed and elucidated. Objective:Using correlation network analysis and machine learning to analyze cerebrospinal fluid (CSF) proteomics data to identify potential biomarkers and pathways associated with EOAD. Methods:We employed mass spectrometry to conduct CSF proteomic analysis using the data-independent acquisition method in a Chinese cohort of 139 CSF samples, including 40 individuals with normal cognition (CN), 61 patients with EOAD, and 38 patients with LOAD. Correlation network analysis of differentially expressed proteins was performed to identify EOAD-associated pathways. Machine learning assisted in identifying crucial proteins differentiating EOAD. We validated the results in an Western cohort and examined the proteins expression by enzyme-linked immunosorbent assay (ELISA) in additional 9 EOAD, 9 LOAD, and 9 CN samples from our cohort. Results:We quantified 2,168 CSF proteins. Following adjustment for age and sex, EOAD exhibited a significantly greater number of differentially expressed proteins than LOAD compared to CN. Additionally, our data indicates that EOAD may exhibit more pronounced synaptic dysfunction than LOAD. Three potential biomarkers for EOAD were identified: SH3BGRL3, LRP8, and LY6 H, of which SH3BGRL3 also accurately classified EOAD in the Western cohort. LY6 H reduction was confirmed via ELISA, which was consistent with our proteomic results Conclusions:This study provides a comprehensive profile of the CSF proteome in EOAD and identifies three potential EOAD biomarker proteins.
