LAUSR

Amyloid-β (Aβ40/Aβ42) peptide with a length of 40 or 42 residues is naturally secreted as cleavage product of the amyloid precursor protein, and formation of Aβ aggregates in a patient's brain is a hallmark of Alzheimer's disease (AD). Therefore, disaggregation and disruption provide potential therapeutic approaches to reduce, inhibit, and even reverse Aβ aggregation. The disaggregation/inhibition effect of the inhibitors applies generally to both Aβ40 and Aβ42 aggregations. Here we capture the atomic-level details of the interaction between Aβ40/Aβ42 and either natural tanshinone compound TS1 or its derivative TS0, and observe novel results by using molecular dynamics simulations. We observe that the natural TS1 indeed inhibits the monomolecular Aβ42 (mAβ42) aggregation and disaggregates Aβ42 amyloid fibrils, being in good agreement with the experimental results. TS1 is favorable to stabilize mAβ40 and even Aβ40 fibril, playing an opposite role to that in the Aβ42 counterpart, however. TS0 can inhibit the misfolding of either mAβ40 or mAβ42 and disaggregate Aβ42 fibril but stabilize the Aβ40 fibril. Using a combination of secondary structural analysis, MM-PBSA binding energy calculations, and radial distribution functions computations, we find that both TS0 and TS1, especially the former, prefer to bind at the charged residues within disordered N-terminus with a scarce positive binding energy and disappear the characteristic C-terminal bend region of Aβ42 fibril, as well as twist the Aβ42 fibril seriously. It turns out to destabilize the Aβ42 fibril and enable the conversion of U-shaped Aβ42 fibril from the onefold to the twofold morphologies. The N-terminal binding preference helps us to identify N-terminal region as the specific epitope for specific inhibitors/drugs (such as TS0 and analogues), heralding unusual inhibition/disaggregation or stabilization mechanisms, and offering an alternative direction in engineering new inhibitors to treat AD.