ABSTRACT
OBJECTIVE: Soluble oligomers of amyloid beta (Abeta), rather than amyloid fibrils, have been proposed to initiate synaptic and cognitive dysfunction in Alzheimer's disease (AD). However, there is no direct evidence in humans that this mechanism can cause AD. Here, we report a novel amyloid precursor protein (APP) mutation that may provide evidence to address this question. METHODS: A Japanese pedigree showing Alzheimer's-type dementia was examined for mutations in APP, PSEN1, and PSEN2. In addition, 5,310 Japanese people, including 2,121 patients with AD, were screened for the novel APP mutation. The pathogenic effects of this mutation on Abeta production, degradation, aggregation, and synaptotoxicity were also investigated. RESULTS: We identified a novel APP mutation (E693Delta) producing variant Abeta lacking gulutamate-22 (E22Delta) in Japanese pedigrees showing Alzheimer's-type dementia and AD. Although the secretion of total Abeta was markedly reduced by this mutation, the variant Abeta was more resistant to proteolytic degradation. The mutant peptides showed the unique aggregation property of enhanced oligomerization but no fibrillization, and inhibited hippocampal long-term potentiation more potently than wild-type peptide in rats in vivo. Consistent with the nonfibrillogenic property of the variant Abeta, a very low amyloid signal was observed in the patient's brain on positron emission tomography using Pittsburgh compound-B. INTERPRETATION: The E693Delta mutation has been suggested as a cause of dementia because of enhanced formation of synaptotoxic Abeta oligomers. Our findings may provide genetic validation in humans for the emerging hypothesis that the synaptic and cognitive impairment in AD is primarily caused by soluble Abeta oligomers.
