Detection and quantification of novel C-terminal TDP-43 fragments in ALS-TDP

Abstract

The pathological hallmark of amyotrophic lateral sclerosis (ALS) is the presence of cytoplasmic inclusions, containing C-terminal fragments of the protein TDP-43. Here, we tested the hypothesis that highly sensitive mass spectrometry with parallel reaction monitoring (MS-PRM) can generate a high-resolution map of pathological TDP-43 peptide ratios to form the basis for quantitation of abnormal C-terminal TDP-43 fragment enrichment. Human cortex and spinal cord, microscopically staged for the presence of p-TDP-43, p-tau, alpha-synuclein, and beta-amyloid pathology, were biochemically fractionated and analyzed by immunoblot and MS for the detection of full-length and truncated (disease-specific) TDP-43 peptides. This informed the synthesis of heavy isotope-labeled peptides for absolute quantification of TDP-43 by MS-PRM across 16 ALS, 8 Parkinson's, 8 Alzheimer's disease, and 8 aged control cases. We confirmed by immunoblot the previously described enrichment of pathological C-terminal fragments in ALS-TDP urea fractions. Subsequent MS analysis resolved specific TDP-43 N- and C-terminal peptides, including a novel N-terminal truncation site-specific peptide. Absolute quantification of peptides by MS-PRM showed an increased C:N-terminal TDP-43 peptide ratio in ALS-TDP brain compared to normal and disease controls. A C:N-terminal ratio >1.5 discriminated ALS from controls with a sensitivity of 100% (CI 79.6-100) and specificity of 100% (CI 68-100), and from Parkinson's and Alzheimer's disease with a sensitivity of 93% (CI 70-100) and specificity of 100% (CI 68-100). N-terminal truncation site-specific peptides were increased in ALS in line with C-terminal fragment enrichment, but were also found in a proportion of Alzheimer cases with normal C:N-terminal ratio but coexistent limbic TDP-43 neuropathological changes. In conclusion this is a novel, sensitive, and specific method to quantify the enrichment of pathological TDP-43 fragments in human brain, which could form the basis for an antibody-free assay. Our methodology has the potential to help clarify if specific pathological TDP-43 peptide signatures are associated with primary or secondary TDP-43 proteinopathies.

Keywords: Alzheimer’s disease; LATE; TDP-43; amyotrophic lateral sclerosis; proteomic biomarker.