Abstract
Alzheimer’s disease (AD) is a debilitating neurodegenerative disorder characterized by progressive memory loss, seizures, anxiety, personality changes and ultimately, death. Current animal models of AD have not yielded any effective treatment of AD, suggesting that these models may not faithfully replicate the entire suite of AD pathologies. The aim of this thesis was to develop a newer and more physiologically relevant mouse model of AD that encompassed the two main hallmarks related to AD: amyloid-beta accumulation and tau hyperphosphorylation. Current mouse models featuring amyloid and tau pathology do so by expressing non-physiological levels of mutant human amyloid and tau proteins to elicit an AD phenotype. Our new mouse model was intended to improve upon the existing method of tau pathology development by incorporating a pathophysiological observation in the brains of AD patients into an existing mouse model of AD. This observation was the reduction of protein phosphatase 2A activity (PP2A) and the upregulation of its inhibitor (I1PP2A). PP2A is responsible for the dephosphorylation of tau and a reduction of PP2A activity by I1PP2A was predicted to lead to tau hyperphosphorylation.
This project was designed as a preliminary investigation into the effects of I1PP2A overexpression in an APPswe/PS1dE9 mouse model of AD. This was carried out by the injection of a virus containing an I1PP2A or luciferase transgene into the hippocampus of 2-month-old C57BL/6 wild-type (WT) or transgenic (TG) APPswe/PS1dE9 mice and comparing the behavioural and neuropathological deficits between treatment groups at 3 months of age. The hippocampus was chosen as the site of experimental manipulation due to its dysfunction early in the course of AD. Mice were tested on a battery of behavioural tasks aimed at measuring anxiety, learning, exploratory behaviour, spatial memory acquisition, memory retrieval, locomotor function, and contextual fear memory. Post-mortem analyses were carried out to investigate I1PP2A expression levels, microglia activation, neurodegeneration and the level of tau phosphorylation in the hippocampi of these mice.
TG-I1PP2A mice exhibited an unexpected lethal phenotype within 3-7 weeks after injections which precluded complete behavioural characterisation of the new mouse model. There were no other casualties in the other treatment groups. During the period the TG-I1PP2A mice were alive, they demonstrated a significant reduction in exploratory behaviour compared to the other treatment groups. WT-I1PP2A mice displayed a significant impairment in spatial memory retrieval and a significant decrease in dentate hilar width compared to other treatment groups. In addition, activated microglia and neurodegeneration were qualitatively readily evident in the CA3, CA1, and hilus of TG-I1PP2A and WT-I1PP2A mice. Tau phosphorylation was not significantly increased in the WT-I1PP2A mice.
The lethality observed in TG-I1PP2A mice hints at a more intricate relationship between I1PP2A, PP2A, tau and amyloid than previously thought. While the unanticipated lethality of this model raises the question of the validity of this model for AD, it highlighted that we still do not fully understand the aetiology behind AD and more research on improved animal models must be carried out before we are able to devise a treatment for AD.