Metal Binding of Alzheimer's Amyloid-β and Its Effect on Peptide Self-Assembly

Abstract

Metal ions have been identified as key factors modulating the aggregation of amyloid-β peptide (Aβ) implicated in Alzheimer's disease (AD). The presence of elevated levels of metal ions in the amyloid plaques in AD patients supports the notion that the dysfunction of metal homeostasis is connected to the development of AD pathology. Here, recent findings from high- and low-resolution biophysical techniques are put into perspective, providing detailed insights into the molecular structures and dynamics of metal-bound Aβ complexes and the effect of metal ions on the Aβ aggregation process. In particular, the development of theoretical kinetic models deducing different microscopic nucleation events from the macroscopic aggregation behavior has enabled deciphering of the effect of metal ions on specific nucleation processes. In addition to these macroscopic measurements of bulk aggregation to quantify microscopic rates, recent NMR studies have revealed details about the structures and dynamics of metal-Aβ complexes, thereby linking structural events to bulk aggregation. Interestingly, transition-metal ions, such as copper, zinc, and silver ions, form a compact complex with the N-terminal part of monomeric Aβ, respectively, where the metal-bound "folded" state is in dynamic equilibrium with an "unfolded" state. The rates and thermodynamic features of these exchange dynamics have been determined by using NMR relaxation dispersion experiments. Additionally, the application of specifically tailored paramagnetic NMR experiments on the Cu(II)-Aβ complex has been fruitful in obtaining structural constraints within the blind sphere of conventional NMR experiments. This enables the determination of molecular structures of the "folded" Cu(II)-coordinated N-terminal region of Aβ. Furthermore, the discussed transition-metal ions modulate Aβ self-assembly in a concentration-dependent manner, where low metal ion concentrations inhibit Aβ fibril formation, while at high metal ion concentrations other processes occur, resulting in amorphous aggregate formation. Remarkably, the metal-Aβ interaction predominately reduces one specific nucleation step, the fibril-end elongation, whereas primary and surface-catalyzed secondary nucleation mechanisms are less affected. Specific inhibition of fibril-end elongation theoretically predicts an enhanced generation of Aβ oligomers, which is an interesting contribution to understanding metal-Aβ-associated neurotoxic effects. Taken together, the metal binding process creates a metal-bound Aβ complex, which is seemingly inert to aggregation. This process hence efficiently reduces the aggregation-prone peptide pool, which on the macroscopic level is reflected as slower aggregation kinetics. Thus, the specific binding of metals to the Aβ monomer can be linked to the macroscopic inhibitory effect on Aβ bulk aggregation, providing a molecular understanding of the Aβ aggregation mechanism in the presence of metal ions, where the metal ion can be seen as a minimalist agent against Aβ self-assembly. These insights can help to target Aβ aggregation in vivo, where metal ions are key factors modulating the Aβ self-assembly and Aβ-associated neurotoxicity.

Figure 1

Metal ions specifically bind to the N-terminal part of Aβ and form a compact metal-Aβ complex. (a) ¹H–¹⁵N HSQC spectrum of 75 μM Aβ40 in 10 mM HEPES, pH 7.2, with (violet) and without (black) 100 μM Cu(II) at 281 K. (b–d) Relative ¹H–¹⁵N HSQC signal intensities of Aβ40 in the presence of different concentrations of Cu(II), Zn(II), and Ag(I). Metal ion concentrations were 40 (black-violet), 60 (dark violet), 75 (violet) and 100 μM (light violet) for Cu(II), 20 μM for Zn(II) (blue), and 20 (dark green), 40 (green) and 80 μM (light green) for Ag(I) for 75 to 80 μM Aβ40.⁻ (e) Relative translational diffusion coefficients of Cu(II), Zn(II), and Ag(I) with a global fit to a two-state model, where the diffusion coefficients for the “unfolded” and compact “folded” states are shared fitting parameters. Data were replotted from refs (−3).

Figure 2

Molecular structures of the Cu(II)-Aβ complex using paramagnetic NMR experiments. (a) Structural constraints from different paramagnetic NMR experiments where binding ligands (violet) are within the blind sphere of paramagnetic NMR experiments (orange). Nuclei detected with paramagnetic NMR but not with diamagnetic NMR pulse sequences are located within the decreased blind sphere of paramagnetic compared to diamagnetic experiments (green). PRE measurements provide additional structural constraints (cyan). (b) Molecular structures of two different binding modes with H13 or H14 as the fourth binding ligand (available as PDB structure 8B9Q or 8B9R, respectively). The assigned binding ligands are the nitrogen of the NH₂ terminus, the amide oxygen of D1, the N_ε of H6, and the N_δ of H13 or H14. Data were reproduced and the figure was adapted with permission from ref (1). Copyright 2022 the authors. American Chemical Society.

Figure 3

Exchange dynamics of Zn(II)- and Ag(I)-Aβ complexes characterized by NMR relaxation dispersion experiments. (a, b) Distinct residues exhibit significantly high amplitude ¹⁵N CPMG relaxation dispersion profiles for Zn(II)-Aβ40 (blue) and Ag(I)-Aβ40 (green) complexes, where the absolute chemical shift changes are plotted, obtained from a global fit analysis. The temperature dependence of the relaxation dispersion profile for two selected residues, E3 and V12, is displayed, where circles correspond to 278 (red), 281 (yellow), 284 (green), 287 (blue), and 290 K (violet) in the presence of metal ions. Black squares reflect the dynamics without any metal ions added. (c, d) The population of the NMR invisible state and the exchange rate between the two states exhibit very similar temperature dependences for Zn(II) and Ag(I). (e) The binding model is visualized where in the NMR invisible state the N-terminus encapsulates the metal ion, forming a more compact “folded” complex. Data were replotted from refs (2) and (3).

Figure 4

Aggregation kinetics analysis reveals a specific effect of metal ions on fibril-end elongation, leading to an enhanced oligomer generation rate. (a–c) Global fit analysis of aggregation kinetics, here shown for 3 μM Aβ42 in the presence of 0 to 10 μM Cu(II) (from violet to red colors), reveals the best fit for the elongation rate, k₊, as the sole free fitting parameter. (d, e) From the global fit results, the nucleation rate of new nucleation units can be calculated, exhibiting a shift of the maximum of the reaction and an increased area under the curve with increasing Cu(II) concentration, corresponding to an increased number of new nucleation units. (f) Aggregation model based on kinetic analysis, showing a specific inhibitory effect on fibril-end elongation by Cu(II) (violet) that results in an increased generation rate for new nucleation units (oligomers) in the presence of Cu(II) (violet) compared to the absence of Cu(II) (black). The figure was modified with permission from ref (1). Copyright 2022 the authors. American Chemical Society.

Figure 5

Model for mechanism of action of metal ion-modulated Aβ self-assembly. (a) Transition-metal ions, in particular referring to Cu(II), Zn(II), and Ag(I) ions, specifically prevent fibril-end elongation events by forming a seemingly aggregation-inert metal-bound Aβ complex. Inhibition of fibril elongation predicts an enhanced rate of oligomer generation. At high metal ion concentration, other aggregation processes dominate, and amorphous aggregates are formed. (b) An energy diagram shows the concentration-dependent formation of Aβ fibrils and amorphous Aβ aggregates, where an increased concentration of Aβ generally enhances aggregation and the energy barrier toward amorphous aggregate formation is determined by the metal ion concentration. The figure was modified with permission from ref (1). Copyright 2022 the authors. American Chemical Society.