Search

Despite the tremendous advances of the last century, the cause of Alzheimer disease (AD) remains unclear. Genetic analysis of families with Alzheimer disease has revealed a disease-associated variant of the APOE gene, which encodes apolipoprotein E, a transporter of lipids in the blood and central nervous system. The effect of the AD-associated isotype, termed ApoE-E4, on disease risk has been validated, though it is unclear by what mechanism apoE-E4 confers AD risk. Mitochondria have long been implicated in AD pathogenesis, as the canonical histopathological findings of amyloid plaques and tau tangles occur in the setting of mitochondrial dysfunction. The disrupted processes include calcium homeostasis, cholesterol metabolism, phospholipid synthesis, and mitochondrial dynamics, and are all regulated by a subcompartment of the ER that is in physical contact with mitochondria. This compartment, called the mitochondria-associated ER membrane, or MAM, has been found to be overactive in AD patient cell lines and cell models of AD. Given that MAM is dysfunctional in AD and that ApoE-ε4 is the most important risk factor for AD, this dissertation examines if ApoE4 contributes to the MAM dysfunction seen in AD. The MAM dysfunction seen in AD patients and in cell models of AD has been best characterized in the context of familial AD, and it is the purpose of this study to extend those findings to the more common, sporadic, form of the disease. Familial AD is the result of autosomal dominant mutations in one of three genes, amyloid precursor protein (APP), presenilin 1 (PSEN1), or presenilin 2 (PSEN2). APP is the protein from which amyloid-beta, the component of amyloid plaques, is cleaved. The presenilins constitute the enzymatic core of the γ-secretase complex, which cleaves amyloid-beta from a precursor APP molecule. Both PSEN1 (PS1) and PSEN2 (PS2) localize at the MAM, and their action is speculated to influence MAM activity. Fibroblasts from familial AD patients, which contained mutations in APP, PSEN1 or PSEN2, showed a marked increase in MAM activity when compared to that of age-matched controls. In mouse embryonic fibroblasts, one can recapitulate this increased MAM activity by knocking out presenilins 1 and 2. In these Psen1/2 double knockout (DKO) cells, the typical measures of MAM function, i.e. increased cholesteryl ester and phosphatidylethanolamine synthesis, calcium transport from ER to mitochondria, and co-localization of ER and mitochondria by confocal and electron microscopy, mimicked the same phenotype found in fibroblasts obtained from familial AD patients, which suggests that the presenilins are negative regulators of ER-mitochondrial apposition. The precise mechanism by which they regulate the ER-mitochondria interface, whether directly as part of a tethering complex, or indirectly though the metabolism of APP-derived substrates, is unclear. While the effect of familial AD mutations on MAM has been characterized, the mechanism of mitochondrial dysfunction seen in the more common sporadic form of the disease remains obscure. Sporadic AD patients harbor no mutations in APP, PSEN1, or PSEN2, but rather inherit mutations in other genes which do not guarantee the development of the disease, but are instead considered risk factors. The most important of these risk factors, in terms of both amount of AD risk conferred and prevalence in the population, is ApoE. Embedded in the phospholipid monolayer of lipoproteins, ApoE is involved in the transport of phospholipids, cholesterol, and cholesteryl esters in plasma and the central nervous system (CNS). In the CNS, it is the most abundant apolipoprotein, and is secreted primarily by astrocytes and taken up by neurons. Once endocytosed, ApoE can follow three different pathways: degradation by the lysosome, intracellular retention in early endosomes, or re-lipidation and re-secretion out of the cell. Our approach takes advantage of the physiological role of ApoE as part of a high densitylike lipoprotein particle (HDL). Using astrocytes from ApoE targeted gene replacement mice in which murine APOE has been replaced by either human APOE-E3 or human APOE-E4, cultured media containing ApoE3 and ApoE4-lipoproteins can be produced and applied to target cells that do not express ApoE, such as neurons or fibroblasts. These target cells can then be analyzed for MAM activity. To examine the contribution of ApoE towards MAM dysfunction, target cells, either neurons or fibroblasts, were grown in the presence of astrocyte conditioned media (ACM) from ApoE targeted gene replacement mice. Several measures of phospholipid and cholesteryl ester synthesis were performed to analyzed MAM function. To confirm that the alterations in phospholipid synthesis were the result of altered MAM activity, the same assay was performed in cells in which a protein tethers that bind mitochondria and ER were genetically ablated. Finally, to confirm that the effects seen were the result of the HDL particles and not the result of other components of the ACM, lipoproteins were extracted from ACM by density ultracentrifugation and applied to fibroblasts. In all of the assays performed, ApoE4 conditioned media or ApoE4 isolated lipoproteins were able to induce a significant increase in MAM activity, whereas ApoE4 from recombinant sources did not. These data suggest a contribution of ApoE4 towards MAM dysfunction seen in AD. The mechanism of these ApoE4 induced MAM alterations remains to be deduced. One may speculate that given the role of ApoE in cholesterol transport outside of the cell, its intracellular retention may impact the distribution of cholesterol within the cell. MAM is a cholesterol rich subdomain with lipid raft-like properties, and any change in the cholesterol content or lipid nature of this membrane may alter its activity. To test this hypothesis, MAM was biochemically extracted from ApoE3 and ApoE4 treated cells and analyzed for cholesterol and lipidomic content. The results described in this thesis demonstrate an AD-like effect in wildtype cells when treated with ApoE-E4, and that the mechanism for these alterations may be due to disturbances in cholesterol distribution in the MAM.

Alzheimer’s disease (AD) is characterised by a series of biomolecular aggregation events, which include the formation of neurotoxic protein structures composed of the β-amyloid (Aβ) peptide. In this thesis, fluorescence self-quenching (FSQ) between fluorescently-labelled peptides is introduced as a strategy for detecting and characterizing Aβ aggregates in solution, and for overcoming limitations associated with conventional methods. Using a combination of steady-state, picosecond time-resolved fluorescence and transmission electron microscopy, the fluorescence response of HiLyte Fluor 555-labelled Aβ peptides is characterised to demonstrate that Aβ self-assembly organizes the covalently attached probes in close proximity to trigger the self-quenching sensing process over a broad range of conditions. Importantly, N-terminal tagging of β-amyloid peptides is shown to not alter the self-assembly kinetics or the resulting aggregated structures. When performed in Förster resonance energy transfer (FRET) format, this method becomes a ratiometric platform to gain insights into amyloid structure and for standardizing in vitro studies of amyloid self-assembly. The ability of FSQ-based methods to monitor the inhibition of Aβ aggregation by model test compounds including the small heat shock protein (Hsp), the amyloid-binding alcohol dehydrogenase protein (ABAD) and bovine serum albumin (BSA) is also demonstrated. Given that Aβ is formed within the cell membrane and is known to induce its disruption, sophisticated single-molecule fluorescence spectroscopy methods were developed to quantify membrane dynamics induced by the presence of disrupting agents, such as Aβ and detergents. The solubilisation dynamics of single liposomes induced by the non-ionic surfactant Triton-X 100 (TX-100) were studied in real-time. Using this approach, the swelling and permeabilization steps of the solubilisation process were unambiguously separated within single FRET trajectories, and their kinetic details as a function of Triton-X 100 and presence of cholesterol within the membrane structure were examined. Finally, single-molecule stepwise-photobleaching techniques were employed to study the effect of Aβ oligomers interacting with supported-lipid bilayers, establishing a platform from which to investigate how the presence of a membrane layer affects Aβ oligomerization at the level of individual molecules. Overall, the fluorescence-based strategies for amyloid- and liposome-sensing presented in this work bridges the gap between current morphology-specific techniques and highly-specialized single-molecule methods to provide a biophysical toolbox to investigate the changes in structure, size and molecular interactions accompanying the amyloid aggregation pathway and for the screening of novel therapeutic and diagnostic agents.

Electronic version excludes material for which permission has not been granted by the rights holder Dementia causes an increasing social and economic burden worldwide, demanding action regarding its diagnosis, treatment and everyday management. Recent years have seen many advances in neurodegeneration research, but the search for new truly disease modifying therapies for Alzheimer's disease (AD) and frontotemporal dementia (FTD) has so far not been successful. This is mainly due to a lack of understanding of the precise intracellular events that lead up to neuronal dysfunction in early and in late stages of the disease. This thesis describes the approaches taken to extend the current knowledge about the intracellular effects of neuronal amyloid-beta and the signalling pathways causing neuronal death or disturbed synaptic function in dementia. Endophilin-1(Ep-1), amyloid-binding alcohol dehydrogenase (ABAD), peroxiredoxin-2 (Prx-2) and the EF-hand domain family, member D2 (EFHD2) have been found to be elevated in the human brain with dementia and in mouse models for frontotemporal lobar degeneration (FTLD) or AD. The expression of these proteins as well as the expression of c-Jun N-terminal kinase (JNK), c-Jun and APP were analysed by western blotting and real-time PCR in human brains affected by AD or FTLD as well as in mouse models for AD. This provided a new insight into the regulation of these proteins in relation to each other in the ageing brain and uncovered a new potential link between elevated levels of EFHD2, Prx-2 and APP in FTLD. By studying the effects of the overexpression of Ep-1 in neurons, this research has led to a better understanding of its role in JNK-activation. It furthermore verified a protective role for Prx-2 against neurotoxicity and pointed towards a new function for Prx-2 in the regulation of JNK-signalling. The analysis of the effect of increased levels of EFHD2 uncovered for the first time its involvement in the PI3K-signalling cascade in neuronal cells. The current work has therefore contributed to the knowledge about the cellular processes that are affected by Ep-1, Prx-2 and EFHD2 in different types of dementia and will greatly benefit future research into their actions in the neuronal network.