TDP-43 based biomarker development in amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis is a devastating neurodegenerative disease which is clinically and aetiologically heterogenous. However, TDP-43 neuropathology is a unifying hallmark found in 97% of ALS cases. Understanding the common pathological pathways downstream of TDP-43 dysregulation is essential for the development of novel drug targets, and the development of a TDP-43 based biomarker for clinical translation would have profound implications for early diagnosis and monitoring of disease progression. The work presented in the first part of this thesis addresses the need to develop TDP-43 as a unifying disease-specific biomarker. Mass spectrometry was used to detect TDP-43 from insoluble protein fractions extracted from ALS post mortem tissue with TDP-43 pathology which led to the discovery of ALS disease-specific TDP-43 peptides. These peptides were then validated in an independent targeted proteomic experiment which, for the first time, allowed the quantification of disease-relevant peptides. This enabled the measurement of an ALS-specific TDP-43 peptide signal that reflects sites of pathological truncation and an increased C-terminal TDP-43 protein ratio. This now provides a platform with which to investigate ALS biofluids with the aim to ultimately develop a disease-specific TDP-43 in vivo assay for patient stratification. In the second part of the thesis I describe the use of affinity-enrichment mass spectrometry to investigate the human wild-type and M337V mutant TDP-43 interactome in a low expression (BAC)-transgenic mouse motor neuron model. I demonstrate that the C-terminal mutation of TDP-43 disrupts specific protein-protein interactions associated with human wild-type TDP-43. In response to oxidative stress, wild-type TDP-43 interactions shift towards proteins that are relevant to cellular stress pathways, while mutant TDP-43 is bound to proteins that are independently linked to pathways known to drive ALS pathogenesis. The key interactions that were disrupted by the M337V mutation involve proteins that are important for stress granule formation and extracellular vesicle secretion. Furthermore, I was able to demonstrate correlative phenotypic abnormalities in stress granule assembly and exosome secretion in motor neurons from the human TDP-43 (BAC)-transgenic mouse model and also in human iPSC-derived motor neurons with the same TDP-43 M337V mutation. In conclusion, work in this thesis provides significant novel data to support the further development of TDP-43 as an in vivo biomarker in ALS. Analysis of the TDP-43 interactome in a physiologically relevant mouse model has defined core cellular pathways matched to relevant phenotypes which will be important for the identification of pathways relevant for drug target development