Abstract
The aggregation of misfolded proteins into β-sheet-rich fibrils constitutes a characteristic feature of neurodegenerative disorders and represents a therapeutic target. While cryo-electron microscopy has elucidated ordered binding patterns of small molecules on fibril surfaces, the mechanisms of ordered aggregate formation generally remain unclear. This study employs molecular dynamics (MD) simulations of the model ligand GTP-1 to examine fibril-templated ligand aggregation and elucidate the molecular determinants governing the aggregation process. Our results showed that in aqueous solution, GTP-1 molecules form dynamic clusters without preferential configurations, whereas tau fibril surfaces induce organized aggregation through protein-ligand hydrogen bonding and ligand-ligand π-π stacking interactions. 1000 independent 100 ns simulations were initiated from diverse ligand conformations to comprehensively sample the conformational landscape. Analysis of the MD trajectories revealed two distinct aggregation pathways. Starting from random initial configurations, on-pathway trajectories spontaneously sampled crystal-structure-like conformations during the simulation, and these conformations exhibited high kinetic stability after formation. In contrast, off-pathway trajectories were characterized by ligands adopting non-native binding geometries, with continuous interconversions between multiple disordered states. The conformational stability of on-pathway states was attributed to optimal surface complementarity and enhanced intermolecular interactions, while off-pathway configurations exhibited reduced structural order and increased conformational flexibility. Quantitative analysis demonstrated differential hydrogen-bonding patterns, with on-pathway aggregates forming 2.01 bonds per structure compared to 0.74 in off-pathway configurations. Energy decomposition identified protein-ligand interactions as the primary determinant of binding energetics, highlighting the direct influence of fibril surface properties on ligand aggregation. These findings provide a mechanistic basis for fibril-templated aggregation and offer a rational foundation for designing diagnostic agents targeting pathological protein fibrils in neurodegenerative diseases.