Multiscale Exploration of Concentration-Dependent Amyloid-β(16-21) Amyloid Nucleation

Abstract

Atomic descriptions of peptide aggregation nucleation remain lacking due to the difficulty of exploring complex configurational spaces on long time scales. To elucidate this process, we develop a multiscale approach combining a metadynamics-based method with cluster statistical mechanics to derive concentration-dependent free energy surfaces of nucleation at near-atomic resolution. A kinetic transition network of nucleation is then constructed and employed to systematically explore nucleation pathways and kinetics through stochastic simulations. This approach is applied to describe Aβ16–21 amyloid nucleation, revealing a two-step mechanism involving disordered aggregates at millimolar concentration, and an unexpected mechanism at submillimolar concentrations that exhibits kinetics reminiscent of classical nucleation but atypical pathways involving growing clusters with structured cores wrapped by disordered surface. When this atypical mechanism is operative, critical nucleus size can be reflected by the nucleation reaction order. Collectively, our approach paves the way for a more quantitative and detailed understanding of peptide aggregation nucleation.

Self-Assembly of Amyloid-Beta (Aβ) Peptides from Solution to Near In Vivo Conditions

“…Perfect 1SN or 2SN pathway has Ω = 0 or 1, respectively (D). 23 filament, which each layer composed of in-register antiparallel (Aβ16-22) and parallel (PHF6) β-strands, consistent with the experimental fibrils. 20 It is unclear why the PACE-ASMgenerated free energy landscape of both peptides is funneled toward fibrillar states, as MD is sufficient to explore very rapidly fibrillar states, and peptides should spend most of their times in misregistered β-sheets.…”

“…It is unclear why the PACE-ASM-generated free energy landscape of both peptides is funneled toward fibrillar states, as MD is sufficient to explore very rapidly fibrillar states, and peptides should spend most of their times in misregistered β-sheets. , Using the PACE force field, BE-metadynamics with multiple CVs, and cluster statistical mechanics, Han et al showed that Aβ16-21 self-assembly follows a 2SN mechanism at 10 mM or above. At the critical oligomer concentration of 360 μM, aggregation follows a mechanism reminiscent of 1SN mechanism, but with atypical pathways characterized by growing oligomers with structured cores wrapped by disordered surfaces (Figure D) …”

“…At the critical oligomer concentration of 360 μM, aggregation follows a mechanism reminiscent of 1SN mechanism, but with atypical pathways characterized by growing oligomers with structured cores wrapped by disordered surfaces (Figure 1D). 23 Much less is known on the self-assembly of full-length Aβ peptides at atomic resolution and understanding transient disorder along pathways to amyloid is a real challenge experimentally, 24,25 and computationally. It was demonstrated that the difference between Aβ42 and Aβ40 aggregation kinetics is due to a shift of more than one order of magnitude in the relative importance of primary nucleation versus fibril-catalyzed secondary nucleation.…”

“…Ω probability distributions of Aβ16-21 pathways at various concentrations. Perfect 1SN or 2SN pathway has Ω = 0 or 1, respectively (D) 23.…”

Exploring the Barriers in the Aggregation of a Hexadecameric Human Prion Peptide through the Markov State Model

“…Such a reorganizational event is also prominent in the postnucleation stage. 52 Therefore, Lag2 can be considered as the rate-determining stage for the present aggregation process. At 300 K and 10 mM peptide concentration, the aggregation process is found to be trapped in the Lag2 phase in the case of the GROMOS-54a7 force field (Figure 3B).…”