| This thesis aims to characterize some of the earliest events in the pathogenesis of Alzheimer's disease (AD), focusing on the biological importance of a defined population of amyloid beta-protein Abeta assemblies, Abeta protofibrils. Two distinct models of the disease are employed to complete this objective: the application of synthetic Abeta to cultured primary neurons, and biochemical analysis of the brains of transgenic mouse models of AD expressing the human amyloid beta-precursor protein (APP) and human presenilin-1 (PS1). Dissolution of Abeta in guanidine HCl and subsequent purification by size exclusion chromatography effectively dissociates all assemblies of synthetic Abeta resulting in a purified population of monomeric peptide. This can then be used for application to neuronal cultures, thereby eliminating uncertainty derived from Abeta peptide solutions containing heterogeneous mixtures of Abeta assemblies. A method was also developed far a quantitative immunohistochemical assay for axon and dendrite number in primary neuron cultures. This was intended as a more precise measure for neurodegeneration than those commonly used in neuron culture systems.; The purified solution of Abeta bearing the Arctic (E22G) mutation was used with quantitative neurite assay to analyze the earliest Abeta-mediated neurotoxic changes in neuritic architecture. Arctic Azbeta induced neurotoxicity reveals a progressive degeneration, beginning within 5--7 hours of first exposure to the peptide, specific to the axonal and dendritic network. The timing of this neurite loss corresponds to the formation of small protofibrillar assemblies in cell free Abeta aggregation assays, well before the formation of mature Abeta fibrils. After 15--17 hours of exposure, levels of lactate dehydrogenase are increased over baseline in the cultured media, suggesting a step-wise degeneration of these cultures, beginning with loss of neurite stability and culminating hours after in frank cell loss.; A centrifugal fractionation procedure was also developed by spiking purified populations of synthetic Abeta with defined biochemical sizes into wild-type brain homogenates, in order to isolate a fraction of brain homogenate free of monomeric and fibrillar Abeta---the protofibril fraction. Using this procedure to fractionate brains from APP/PS1 transgenic mice yields detectable quantities of Abeta in the protofibril fraction by ELISA. This thesis provides further evidence supporting a pathologic role for nonfibrillar Abeta assemblies in AD, and a framework from which to build new experimental investigation of the disease. |
