All-Atom Modeling of Methacrylate-Based Multi-Modal Chromatography Resins for Langmuir Constant Prediction of Peptides
Tim Ballweg,Modan Liu,Julian Grimm,Elaheh Sedghamiz,Wolfgang Wenzel,Matthias Franzreb
DOI: https://doi.org/10.1016/j.chroma.2024.465089
IF: 4.601
2024-06-16
Journal of Chromatography A
Abstract:In downstream processing, the intricate nature of the interactions between biomolecules and adsorbent materials presents a significant challenge in the prediction of their binding and elution behaviors. This complexity is further heightened in multi-modal chromatography (MMC), which employs two distinct binding mechanisms. To gain a deeper understanding of the involved interactions, simulating the adsorption of biomolecules on resin surfaces is a focal point of ongoing research. However, previous studies often simplified the adsorbent surface, modeling it as a flat or slightly curved plane without including a realistic backbone structure. Here, we introduce and validate two novel workflows aimed at predicting peptide binding behaviors in MMC, specifically targeting methacrylate-based resins. Our first achievement was the development of an all-atom model of a commercial MMC resin surface, incorporating its polymethacrylic backbone. Furthermore, we established and tested a workflow for rapid calculations of binding free energies (Δ G ) with 10 linear peptides as target molecules. These Δ G calculations were effectively used to predict Langmuir constants, achieving a high coefficient of determination (R2) of 0.96. In subsequent benchmarking tests, our model outperformed established, simpler resin surface models in terms of predictive capabilities.
chemistry, analytical,biochemical research methods