Investigating TDP-43 in the Cerebellum in Health and Models of Amyotrophic Lateral Sclerosis

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

TAR DNA-binding protein 43 (TDP-43) was first identified as ubiquitinated deposits within the spinal cord of patients with amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) in 2006 (Neumann et al., 2006). Since then, the cytoplasmic aggregation of phosphorylated and p62-labelled TDP-43 within the motor cortex and spinal cord has become a hallmark of these neurodegenerative diseases with ~97% of ALS and 45% of FTD cases displaying aggregated TDP-43 (Ling, Polymenidou and Cleveland, 2013). In health, TDP-43 is a predominantly nuclear RNA binding protein that is involved in many facets of RNA-processing, and shuttles to the cytoplasm to participate in processes such as stress granule formation (Bhardwaj et al., 2013; Walker et al., 2013). A region of the brain that seems relatively spared from TDP-43 proteinopathy is the cerebellum, despite sharing vast connections with affected areas (Al-Sarraj et al., 2011). The cerebellum functions in fine motor tuning through the concise regulation of motor regions such as the motor cortex, spinal cord and brainstem (Daskalakis et al., 2004).

The main aim of this project was to investigate TDP-43 expression, interaction, and function within the cerebellum and motor cortex of nontransgenic mice, transgenic mice over-expressing human wild-type TDP-43 (hTDP-43WT) and an ALS-relevant transgenic mouse model (hTDP-43Q331K). Western blot and immunohistochemistry (IHC) were used to analyse TDP-43 expression in the three mouse models throughout ageing, whilst mass spectrometry analysis was conducted on TDP-43 immunoprecipitations to identify potential differences in the interactome of TDP-43 between brain regions. Proximity ligation assays (PLA) were used in nontransgenic mouse tissue and transfected primary neurons to validate key interactors and explore the impact of the TDP-43 Q331K mutation on these interactions. Since cellular stress is implicated in the pathogenesis of ALS, potential differences in G3BP-1 and TIA-1 stress granule (SG) formation/dissolution were explored in cortical and cerebellar neurons. Total and cytoplasmic TDP-43 expression was highest within the cerebellum of the nontransgenic mice with ageing having little impact on TDP-43 levels. Over-expression of human wild-type and Q331K TDP-43 caused expression of TDP-43 to increase in both brain regions however in different cellular distributions. Total and cytoplasmic TDP-43 remained higher in the cerebellum whereas nuclear TDP-43 was highest in the cortex of the hTDP-43Q331K mice (P=0.0187). Qualitative analysis of TDP-43, phosphorylated TDP-43 (pTDP-43) and p62 IHC saw cytoplasmic puncta of TDP-43 and large cytoplasmic aggregates of colocalised pTDP-43/p62 within the motor cortex of hTDP-43Q331K mice that were absent from the cerebellum and aged-matched controls. The TDP-43 interactome study identified several groups of proteins that were distinct interactors of TDP-43 within the cortex and cerebellum. A short-list of seven interactors was made for further analysis that included FUS, Matrin-3, PSPC-1, SFPQ, NONO, FMRP and Synaptotagmin-1. There were clear brain region differences in the cellular localisation of TDP-43 interaction with PSPC-1, SFPQ and NONO (P=0.0047, P=0.0449, P=0.0456), with similar non-significant trends with FUS, Matrin-3 and FMRP, specifically showing increased cytoplasmic interactions within the motor cortex. PLA spots were consistently larger in the motor cortex, suggesting large cytoplasmic complexes that were absent in the cerebellum. The presence of the TDP-43 Q331K mutation significantly altered its capacity to interact with some of these interactors, in some cases in a regionally specific manner. Interactions with FUS and Matrin-3 were specifically reduced within the nucleus of cortical neurons, whilst PSPC1, SFPQ and NONO interaction was reduced in both cortical and cerebellar cells. SFPQ, NONO and Matrin-3 were also identified in cytoplasmic TDP-43 puncta in the motor cortex of the Q331K TDP-43 mouse model. Cellular stress evoked a larger G3BP-1 and TIA-1 SG response in cortical cells (P=0.0103, P=0.0368), however, indexed analysis of TDP-43 recruitment to these granules from control levels were higher within the cerebellum.

The results of this study therefore suggested that despite the cerebellum possessing higher total and cytoplasmic levels of TDP-43, it was not susceptible to TDP-43 proteinopathy. Key differences in the cellular distribution and nature of these interactions between brain regions were identified, suggesting functional differences. A number of these interactions, notably with the paraspeckle proteins, were reduced in the presence of the Q331K mutation particularly within cortical cells, signalling to the potential cessation of important homeostatic processes. In conclusion the results of this study have highlighted several important differences within TDP-43 expression and interaction between the two brain regions that may indicate key areas of vulnerability within the motor cortex. Together with future research, the full understanding of how these differences may lead to motor cortical vulnerability to TDP-43 aggregation could lead to new avenues of therapeutics for ALS.
Date of Award1 Jun 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorJackie Mitchell (Supervisor) & Caroline Vance (Supervisor)

Cite this

'