Search

Alzheimer's disease (AD) is an insidious age-related neurodegenerative disease and the most common type of dementia. AD neuropathology typically involves the accumulation of extracellular plaques as well as intracellular neurofibrillary tangles that mainly consist of amyloid-beta (Abeta) and hyperphosphorylated tau aggregates, respectively. Previous studies in animal models lacking the expression of Mapt, which encodes for microtubule-associated protein tau that is implicated in AD neuropathology, demonstrated that Abeta-driven toxicity is tau-dependent. However, inconsistent reports of Abeta-driven toxicity in Mapt homozygous knockout (Mapt-/-) animal models have emerged while a human MAPT-/- model remains unavailable to corroborate whether tau is essential to mediate the toxic phenotypes caused by Abeta in human genetic backgrounds. In this thesis, I describe the generation of the first human MAPT-/- induced pluripotent stem cell (iPSC) lines in collaboration with the Cowley research group in Oxford. Two isogenic panels of MAPT-/-, MAPT+/- and MAPT+/+ iPSC lines, which were derived from two parental iPSC lines of healthy individuals, were identified. These iPSC lines were differentiated into cortical neurons in parallel and these neurons exhibited similar levels of cortical marker and synapse expression regardless of their MAPT genotypes. However, tau ablation in the iPSC-derived cortical neurons was sufficient to cause neurite outgrowth impairments, alterations in mitochondrial transport along axons, and reduced electrophysiological activities. Upon the exposure to extrinsic Abeta insults in the forms of synthetic oligomers or AD brain-derived Abeta, the MAPT+/+ neurons experienced cytotoxicity, reduced number of anterograde motile mitochondria along axons, synapse loss, and hyperactivation in vitro. These adverse phenotypes could be reversed in the MAPT-/- neurons whereas Abeta-induced neurite outgrowth impairments were not rescued in the MAPT-/- neurons. Notably, having a reduced dosage of tau expression in the MAPT+/- neurons was enough to be protective against cell death caused by Abeta. The study on Abeta-driven toxicity was extended to iPSC lines derived from the sporadic AD patients who participated in the Deep and Frequent Phenotyping pilot cohort. Synapse losses caused by extrinsic Abeta insults were quantified in thirteen lines of patient iPSC-derived cortical neurons and these neurons displayed varying levels of synaptic vulnerability to Abeta across patient lines. Their individual vulnerability to Abeta in vitro was demonstrated to reflect clinical vulnerability to Abeta burden measured by cognitive readouts from the same patients in vivo, suggesting that AD modelling with iPSCs is a tractable method to study individual levels of disease vulnerability. Here I present the experimental evidence on the effects of tau deletion in human iPSC-derived cortical neurons and show that tau is required in certain aspects of Abeta-driven toxicity. The clinical relevance of studying Abeta-driven toxicity using sporadic AD patient iPSC-derived cortical neurons was also investigated by demonstrating that individual clinical vulnerability to Abeta can be modelled in vitro using these neurons.