TMTCrunch: A Proteomic Atlas of Alternative Splicing for Predicting Splicing-Induced Implications in Aging and Alzheimer's Disease.

Abstract

Alzheimer's disease (AD) is the most prevalent form of dementia with incompletely understood pathogenesis. A major gap arises from the lack of proteomics tools capable of characterizing alternative splicing (AS)-derived proteoforms and their contributions to neurodegeneration. We developed a novel bioinformatics pipeline, TMTCrunch, tailored for rigorous quantitative meta-analysis of big proteomics data at the splice-proteoform level. TMTCrunch characterizes each proteoform by unique peptides, assessing similarity to canonical peptides and unique peptide coverage, employing SMD-based quantitation, and predicting proteoform-specific alterations of protein-protein interactions (PPIs) and novel post-translational modifications (PTMs) on spliced peptides. Applying TMTCrunch to 420 brain samples, we constructed the first atlas of splicing translatomes in AD, reproducibly identifying 870 noncanonical proteoforms. Differential analysis suggests splicing affecting proteoforms implicated in cytoskeletal regulation (e.g., MAPT, CLU, DPYSL3, ACTN2, SORBS1, FHL1), glutamatergic transmission (GRIA3), pre-mRNA splicing regulation (ARL6IP4), potassium channel modulation (DPP6), and cAMP signaling (PDE4D). Our analysis predicts disruption of PPIs within the Rho GTPase and EGFR signaling pathways and PTMs (deamidation, oxidation, phosphorylation) within AS regions, regardless of disease state. This approach implicates specific proteoforms in neurodegeneration: DPP6 (P42658-2), GRIA3 (P42263-2), the three-repeat isoforms of tau (3R-MAPT), and ASPH (Q12797-7). This study provides new insights into linking splicing to neurodegeneration.