摘要

The aggregation of amyloid-beta (A beta) peptide induced by Cu2+ is a key factor in development of Alzheimer's disease (AD), and metal ion chelation therapy enables treatment of AD. Three CQ(i) (i = 1, 2, and 3 with R = H, Cl, and NO2, respectively) drugs had been verified experimentally to be much stronger inhibitors than the pioneer clioquinol (CQ) in both disaggregation of A beta(40) aggregate and reduction of toxicity induced by Cu2+ binding at low pH. Due to the multiple morphologies of Cu2+-A beta(40) complexes produced at different pH states, we performed a series of molecular dynamics simulations to explain the structural changes and morphology characteristics as well as intrinsic disaggregation mechanisms of three Cu2+-A beta(40) models in the presence of any of the three CQ(i) drugs at both low and high pH states. Three inhibition mechanisms for CQ(i) were proposed as "insertion", "semi-insertion", and "surface" mechanisms, based on the morphologies of CQ(i)-model x (CQ(i)-x, x = 1, 2, and 3) and the strengths of binding between CQ(i) and the corresponding model x. The insertion mechanism was characterized by the morphology with binding strength of more than 100 kJ/mol and by CQ(i) being inserted or embedded into the hydrophobic cavity of model x. In those CQ(i)-x morphologies with lower binding strength, CQ(i) only attaches on the surface or inserts partly into A beta peptide. Given the evidence that the binding strength is correlated positively with the effectiveness of drug to inhibit A beta aggregation and thus to reduce toxicity, the data of binding strength presented here can provide a reference for one to screen drugs. From the point of view of binding strength, CQ(2) is the best drug. Because of the special role of Asp23 in both A beta aggregation and stabilizing the A beta fibril, the generation of a H-bond between CQ(3) and Asp23 of the A beta(40) peptide is believed to be responsible for CQ(3) having the strongest disaggregation capacity. Therefore, besides strong binding, stronger propensity to H-bond with Asp23 would be another key factor to be taken seriously into account in drug screens. Meanwhile, the structural characteristics of drug CQ(i) itself are also worthy of attention. First, the increasing polarity from CQ(1) and CQ(2) to CQ(3), in turn results in increasing probability and strength of the interaction between the drug and the N-terminal (NT) region of A beta(40), which obviously inhibits A beta peptide aggregation induced by Cu2+ binding. Second, both the benzothiazole ring and phenol ring of CQ(i) can overcome the activation energy barrier (similar to 16 kJ/mol) to rotate flexibly around the intramolecular C7-N14 bond to achieve the maximum match and interaction with the ambient A beta(40) residues. Such a structural feature of CQ(i), paves the new way for ones in selection and modification of a drug.