Date of Award

12-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Physics and Astronomy

Committee Chair/Advisor

Feng Ding

Committee Member

Emil Alexov

Committee Member

Hugo Sanabria

Committee Member

Xian Lu

Abstract

Amyloid fibril formation is a hallmark of numerous degenerative diseases, including Alzheimer’s disease (AD) and type-2 diabetes (T2D). Despite extensive research, the molecular mechanisms that drive amyloid fibril polymorphism and determine its pathological consequences remain poorly understood. This dissertation employs a comprehensive set of computational approaches based on replica-exchange discrete molecular dynamics (rexDMD) and replica-permutation DMD (rpDMD) simulations to elucidate the thermodynamic and kinetic principles governing fibril formation and to determine how polymorphism and environmental regulators regulate this process, using T2D-associated human islet amyloid polypeptide (hIAPP) and AD-associated amyloid-β (Aβ) peptides as representative amyloid systems.

In the first part, rexDMD simulations of human islet amyloid polypeptide (hIAPP) revealed that distinct fibril morphologies, i.e., the S-shaped and extended forms, exhibit different elongation mechanisms and stability profiles. The S-shaped fibril demonstrated faster seeded growth and greater thermodynamic stability, whereas the extended fibril required cooperative stabilization of multiple monomers, which may account for its slower propagation and lower population observed experimentally. In addition, we also used DMD simulations to study asymmetric amyloid fibril growth, a common experimental observation. Our analysis suggested that the asymmetry between the two fibril ends in terms of their available hydrogen-bond donors/acceptors and the surface local curvatures resulted in asymmetric binding and conformational conversion dynamics of the incoming peptide, i.e., fibril growth.

In the second part, studies of Aβ42 fibrils associated with familial and sporadic Alzheimer’s subtypes uncovered morphology-dependent differences in thermodynamic stability and kinetic behavior. Familial AD (fAD) fibrils displayed lower energy barriers and faster growth rate but reduced stability compared with sporadic AD (sAD) fibrils, which may contribute to the distinct temporal progression of these two subtypes. Furthermore, simulations revealed that the molecular chaperone Bri2 BRICHOS preferentially binds and inhibits fAD fibril formation, thereby modulating strain distributions.

Finally, comparative simulations of Aβ40 and Aβ42 fibrils demonstrated that Aβ42 fibrils are more stable and grow more rapidly than Aβ40, consistent with its predominance in parenchymal plaques. Cross-seeding simulations confirmed the capability of Abeta42-Abeta40 co-aggregation observed experimentally, while underscoring the need to incorporate additional factors to explain the absence of parenchymal Abeta40 aggregates. Additionally, analyses of chaperone modulation using Bri2 BRICHOS as an example further highlighted how intrinsic peptide properties and extrinsic factors cooperate to shape amyloid heterogeneity.

Altogether, these findings establish a unified molecular framework that links amyloid fibril morphology, stability, aggregation kinetics, and environmental modulators to experimentally and clinically observed population differences and potentially to disease-specific pathology. The DMD-based methodologies developed here provide atomistic insight into amyloid assembly processes and open new avenues for the rational design of therapeutic agents targeting the propagation of specific fibril strains.

Author ORCID Identifier

0000-0001-8631-7014

Available for download on Thursday, December 31, 2026

Share

COinS