Nonfibrillar neurotoxic amyloid β (Aβ) oligomer structures are typically rich in β-sheets, which could be promoted by metal ions like Zn2+. Here, using molecular dynamics (MD) simulations, we systematically examined combinations of Aβ40 peptide conformations and Zn2+ binding modes to probe the effects of secondary structure on Aβ dimerization energies and kinetics. We found that random conformations do not contribute to dimerization either thermodynamically or kinetically. Zn2+ couples with preformed secondary structures (α-helix and β-hairpin) to speed dimerization and stabilize the resulting dimer. Partial α-helices increase the dimerization speed, and dimers with α-helix rich conformations have the lowest energy. When Zn2+ coordinates with residues D1, H6, H13, and H14, Aβ40 β-hairpin monomers have the fastest dimerization speed. Dimers with experimentally observed zinc coordination (E11, H6, H13, and H14) form with slower rate but have lower energy. Zn2+ cannot stabilize fibril-like β-arch dimers. However, Zn2+-bound β-arch tetramers have the lowest energy. Collectively, zinc-stabilized β-hairpin oligomers could be important in the nucleation–polymerization of cross-β structures. Our results are consistent with experimental findings that α-helix to β-structural transition should accompany Aβ aggregation in the presence of zinc ions and that Zn2+ stabilizes nonfibrillar Aβ oligomers and, thus, inhibits formation of less toxic Aβ fibrils.

The identification of a secondary nucleation pathway in the early aggregation of amyloid peptides suggests that the generation of toxic oligomers involves both monomers and preformed fibril seeds. To elucidate the underlying molecular mechanism, a set of one-pot coarse-grained molecular dynamics simulations was performed to investigate the self-assembly of amyloid β peptides in the presence of fibril seeds. It was observed that fibril seeds alone randomly assemble into an elongated protofibril, whereas monomers alone form an elongated globular oligomer with various morphologies. In the mixture of monomers and fibril seeds, both the self-assembly of monomers into small oligomers and the association of monomers and oligomers on the surface of fibril seeds are primarily driven by hydrophobic interactions. The cooperativity of conformational selection and competition leads to different binding propensity of two hydrophobic surfaces of fibril seeds. The molecular architecture of the final aggregate shows that the fibril seeds establish the elongated framework, and oligomers cover them. Oligomers exposed to the solvent are less compact and unstable and can be disassociated from the fibril seeds, providing an origin for oligomers generated from the secondary nucleation pathway.

Extensive experimental and computational studies have suggested that multiple Zn2+ binding modes in amyloid β (Aβ) peptides could exist simultaneously. However, consistent results have not been obtained for the effects of Zn2+ binding on Aβ structure, dynamics, and kinetics in particular. Some key questions such as why it is so difficult to distinguish the polymorphic states of metal ions binding to Aβ and what the underlying rationale is, necessitate elucidation. In this work, two 3N1O Zn2+ binding modes were constructed with three histidines (His6, His13, and His14), and Asp1/Glu11 of Aβ40 coordinated to Zn2+. Results from molecular dynamics simulations reveal that the conformational ensembles of different Zn2+-Aβ40 complexes are nonoverlapping. The formation of turn structure and, especially, the salt bridge between Glu22/Asp23 and Lys28 is dependent on specific Zn2+ binding mode. Agreement with available NMR observations of secondary and tertiary structures could be better achieved if the two simulation results are considered together. The free energy landscape constructed by combining both conformations of Aβ40 indicates that transitions between distinct Aβ40 conformations thar are ready for Zn2+ binding could be possible in aqueous solution. Markov state model analyses reveal the complex network of conformational space of Aβ40 modeulated by Zn2+ binding, suggesting various misfolding pathways. The binding free energies evaluated using a combination of quantum mechanics calculations and the MM/3D-RISM method suggest that Glu11 is the preferred oxygen ligand of Zn2+. However, such preference is dependent on the relative populations of different conformations with specific Zn2+ binding modes, and therefore could be shifted when experimental or simulation conditions are altered.Keywords: Alzheimer’s disease; free energy calculation; Markov state model; metal ions; molecular dynamics simulations

The aggregation of amyloid β-protein (Aβ) has been associated with the pathogenesis of Alzheimer’s disease. A number of single point mutations at residues A21, E22, D23, and M35 have been identified to show increased or decreased aggregation tendency. Although the effects of point mutations on the structural properties of Aβ peptides have been intensively studied, how single point mutation affects the kinetics of Aβ remains unknown. In this work, dihedral dynamics analyses, which combine dihedral principle component analysis (dPCA), potential of mean force (PMF) calculations, and Markov state models (MSMs), were proposed to elucidate the different global free energy landscapes (FELs), the PMF of individual dihedral angle, and microstates/macrostates for a number of Aβ42 mutants (Flemish A21G, Arctic E22G, Italian E22K, Dutch E22Q, Iowa D23N, Japanese E22Δ, and M35 oxidation Met35OX). Our simulation results show that one point mutation is sufficient to change the rugged FEL of Aβ42 by altering the energy barriers around basins. This alteration was also observed in the potential of each dihedral angle to varying degrees, although most minima of PMF do not shift. MSMs further reveal that E22 mutants (E22Δ, E22G, E22K, and E22Q) and D23N generate more hub-like microstates than wild type Aβ42, thus creating diverse alternative pathways for conformational transitions and increasing subsequent aggregation. In contrast, transitions are more preferred within the same microstate of A21G and Met35OX. Mapping MSM to FEL suggests that transitions between different sets of microstates are kinetically feasible but thermodynamically difficult.

Amyloid β (Aβ) peptides and metal ions have been associated with the pathogenesis of Alzheimer’s disease. The conformational space of Aβ fragments of different length with and without binding of metal ions has been extensively investigated by replica-exchange molecular dynamics (REMD) simulation. However, only trajectories extracted at relatively low temperatures have been used for this analysis. The capability of REMD simulations to characterize the internal dynamics of such intrinsically disordered proteins (IDPs) as Aβ has been overlooked. In this work, we use an approach recently developed by Xue and Skrynnikov (J Am Chem Soc 133:14614–14628, 2011) to calculate NMR observables, including 15N relaxation rates and 15N–1H nuclear Overhauser enhancement (NOE), from the high-temperature trajectory of REMD simulations for zinc-bound Aβ peptides. The time axis of the trajectory was rescaled to correct for the effect of the high temperature (408 K) compared with the experimental temperature (278 K). Near-quantitative agreement between simulated values and experimental results was obtained. When the structural properties and free-energy surfaces of zinc-bound Aβ(1–40) and Aβ(1–42) were compared at the physiological temperature 310 K it was found that zinc-bound Aβ(1–42) was more rigid than Aβ(1–40) at the C terminus, and its conformational transitions were also more preferred. The self-consistent results derived from trajectories at high and low temperatures demonstrate the capability of REMD simulations to capture the internal dynamics of IDPs.

The pathogenesis of Alzheimer’s disease (AD) has been suggested to be related with the aggregation of amyloid β (Aβ) peptides. Metal ions (e.g. Cu, Fe, and Zn) are supposed to induce the aggregation of Aβ. Recent development of bifunctional molecules that are capable of interacting with Aβ and chelating biometal ions provides promising therapeutics to AD. However, the molecular mechanism for how Aβ, metal ions, and bifunctional molecules interact with each other is still elusive. In this study, the binding mode of Zn2+-bound Aβ with bifunctional molecules was investigated by the combination of conformational sampling of full-length Aβ peptides using replica exchange molecular dynamics simulations (REMD) and conformational selection using molecular docking and classical MD simulations. We demonstrate that Zn2+-bound Aβ(1–40) and Aβ(1–42) exhibit different conformational ensemble. Both Aβ peptides can adopt various conformations to recognize typical bifunctional molecules with different binding affinities. The bifunctional molecules exhibit their dual functions by first preferentially interfering with hydrophobic residues 17–21 and/or 30–35 of Zn2+-bound Aβ. Additional interactions with residues surrounding Zn2+ could possibly disrupt interactions between Zn2+ and Aβ, which then facilitate these small molecules to chelate Zn2+. The binding free energy calculations further demonstrate that the association of Aβ with bifunctional molecules is driven by enthalpy. Our results provide a feasible approach to understand the recognition mechanism of disordered proteins with small molecules, which could be helpful to the design of novel AD drugs.