Stability differences in the NMR ensembles of amyloid β fibrils

Abstract

Alzheimer’s Disease is characterized by the formation of amyloid beta (A[Formula: see text] fibril plaques in the brain. These fibrils can be probed by solid state NMR (ssNMR), which leads to an ensemble of configurations that are compatible with the NMR signals. Typically, only the lowest energy conformer is considered in computer simulations that probe the stability of fibrils and their binding with drug candidates. This restriction could produce data that are not physiologically relevant if the NMR entries differ significantly in stability or binding affinities. In order to study this effect, we have investigated the variance in stability between members of NMR ensembles. Our test cases are a patient-derived A[Formula: see text]-fibril model and two in vitro A[Formula: see text]-fibril models from a previous study we performed on comparative stability. The latter two models allow us also to compare different staggering patterns. We observe significant variations in molecular flexibility, compactness and secondary structure, suggesting that the full NMR ensemble must be considered for a physiologically relevant description of A[Formula: see text] fibrils.