Mapping protein variants associated with Alzheimer's disease by mass-spectrometry based de novo sequencing assisted strategy

Background Alzheimer's disease (AD) is an incurable neurodegenerative disease with poorly understood pathogenesis. Understanding changes in protein sequences due to amino acid substitutions (AASs) may be important for uncovering molecular mechanisms of this disease. Objective The study aimed at developing a bioinformatic pipeline for searching AASs in proteomic data and revealing the AD-specific ones, highlighting potential biomarkers and/or therapeutic targets. Methods The developed pipeline integrates peptide de novo sequencing approach, database searches, and retention time prediction. It was applied to a large collection of AD proteomic data from global consortium studies obtained for post-mortem brain tissue samples. Results Proteins with identified AASs were clustered by functionality. Proteins heavily enriched with AASs were the ones associated with ion transport activity, ATP binding, and G-protein signaling, aligning with known AD mechanisms. Further we classified the identified AASs by their origin (tRNA misacylation, post-translational modifications, single nucleotide polymorphisms) and by Braak stage and sex. Pathogenicity analysis, cross-referenced with clinical information, identified pathogenic mutations in HBD (W38S), GFAP (N77D), and NEFL (N98D). Conclusions The developed pipeline successfully maps disease-relevant protein variants by uncovering novel molecular features of a disease with biomarker and therapeutic potential. AASs identified in the work for AD samples reveal specific pathogenic mutations and implicate important biological processes.