Filamentous amyloid aggregates are crucial for the pathology of Alzheimer's disease. Despite the tremendous biomedical importance of amyloid fibrils, the molecular mechanism and the dynamic pathways involved in their formation remain elusive and challenging to investigate in experiments and simulations.
We use a combination of detailed MD simulations and energy-landscape theory to overcome the challenge due to the different timescales involved in fibril growth.
In the first step, we calculate the free energy profile for fibril elongation by a single monomer at the two struc-turally unequal fibril tips and for the association of larger fibril fragments. The forces driving fibril formation are investigated in a detailed enthalpy/entropy decomposition providing insight into the role of solvent entropy as the main driving force for assembly.
In the second step, we calculate the local diffusion profile which gives insight into the degrees of freedom perpendicular to the distance between fibril and peptide. The combination of the free energy and diffusion profile uniquely determines the growth kinetics. The dissociation rates calculated from this two-step approach agree well with experimental results. The predicted kinetics is also consistent with the dock/lock mechanism characteristic for the growth of amyloid fibrils. Our results show that the breakage of long-lived non-native hydrogen bonds causing formation of kinetically trapped configurations is rate limiting in the assembly process and fundamental for the much slower locking.
This microscopic insight into different growth mechanism is indispensable for understanding and controlling fibril formation in nature and experiments and can help in the efficient design of tailored drugs.