Mitochondrial dysfunction is a prevalent feature in many neurodegenerative diseases including Amyotrophic Lateral Sclerosis (ALS) and Frontotemporal Dementia (FTD). ALS is a debilitating and incurable disease characterised by the loss of upper and lower motor neurons leading to symptoms such as muscle weakness and paralysis. Most patients die from respiratory failure after 2–5 years however, only one globally licensed treatment (Riluzole) is available which only prolongs survival by a modest 2–3 months.

A hexanucleotide repeat expansion consisting of GGGGCC (G4C2) in the first intron of C9orf72 is the most common pathogenic mutation in ALS/FTD. Three disease mechanisms have been proposed including haploinsufficiency and the sequestration of RNA binding proteins at accumulations (foci) of the transcribed RNA. Although intronic, the repeats are translated to produce 5 dipeptide repeat proteins (DPRs) through a mechanism known as repeat associated non-AUG (RAN) translation. Various pathogenic mechanisms have been proposed and it is generally accepted that mitochondrial dysfunction is an early alteration in ALS. Mitochondria are vital organelles important for cellular processes regulating energy metabolism and cell survival. The role of mitochondria specifically in C9orf72 ALS/FTD has been relatively understudied, especially in an in vivo system. Excess production of reactive oxygen species (ROS) and defective mitochondrial dynamics are common features of ALS, but it is not clear whether these phenomena are causative or a consequence of the pathogenic process.

In this thesis, I have utilised 3 different Drosophila models of C9orf72, including a (i) 36 repeat GGGGCC (G4C2x36), (ii) poly GR36 DPR-only model and (iii) GR1000-eGFP DPR-only model. Firstly, I recapitulated established phenotypic characterisations that have been previously published. Briefly, pan-neuronal expression of the various transgenes exhibit locomotor deficits which I used as a readout for testing different genetic manipulations to modulate and ultimately rescue these behavioural phenotypes. Next, I performed a thorough characterisation of mitochondrial dysfunction in all the models, analysing impacts on ROS, morphology and mitochondrial turnover (mitophagy). I found alterations in mitochondrial morphology, specifically hyperfusion, a reduction in mitophagy, increased ROS production and impaired respiration in these models. Unexpectedly, genetic manipulation to restore mitochondrial fission/fusion dynamics or boosting mitophagy were unable to rescue the locomotor deficits in larvae. However, genetic upregulation of antioxidants such as mitochondrial superoxide dismutase 2 (SOD2) and catalase were able to rescue impaired larval locomotion. Surprisingly, overexpression of cytosolic superoxide dismutase 1 (SOD1) exacerbated larval crawling phenotypes. Together, these data suggest a causative link between mitochondrial dysfunction, ROS and behavioural phenotypes.

To elaborate on this connection, I investigated whether the nuclear factor erythroid 2–related factor 2 (NRF2)/Keap1 signalling pathway might play a role. I found that NRF2 was translocated to the nucleus suggesting an activation of the pathway. However, there were minimal changes to NRF2 targeting transcript genes although changes were observed using a glutathione S-transferase D1 (gstD1-GFP) reporter for NRF2 activity. Despite these variable effects, both genetic reduction in Keap1 and pharmacological treatment with an NRF2 activator, dimethyl fumarate (DMF), showed a behavioural rescue in climbing activity of G4C2x36 and GR36 flies. While more research is needed, these results provide compelling evidence that mitochondrial oxidative stress is a major upstream pathogenic mechanism leading to downstream mitochondrial dysfunction such as alterations in mitochondrial function and turnover. Consequently, targeting one of the main intracellular defence mechanisms to counteract oxidative stress – the NRF2/Keap1 signalling pathway – could be a viable therapeutic strategy for ALS/FTD.