A Metadynamics-Based Protocol for Sampling and Scoring Peptide Nanofibril Conformations

The morphology of a cross-β nanofibril, ranging from ribbon to rod conformations, is commonly determined by the amino acid sequence of its constituent peptides in addition to environmental factors such as pH and salt concentration. The use of all-atom molecular dynamics simulation of a given peptide nanofibril to explore its energetically favorable fibril conformations is a desirable but challenging goal, due to potential trapping in local energy minima. In this work, we explore a metadynamics-based approach to investigate morphologically distinct peptide nanofibrils, introducing collective variables based on supramolecular twist as a sampling coordinate. We artificially construct a search space consistent with integral fibril structures and perform metadynamics simulations to overcome energy barriers and traverse the rugged conformational landscape of the large peptide assemblies in water. Based on this, we compute a measure which we term the resistance score to identify the propensity of distinct fibril morphologies across the conformational space. This computational protocol is then used to investigate the impact of fibril length and peptide sequence on the supramolecular twist of peptide nanofibrils in aqueous solution, comparing these results with the fibril morphologies observed in electron microscopy. The proposed workflow demonstrates an ability to identify a preferred morphology for β-sheet assemblies, which could aid in the design of new peptide self-assembled architectures.