Jerzy Haber Institute of Catalysis and Surface Chemistry, Polish Academy of Sciences, Niezapominajek 8, 30-239 Krakow, Poland.
Department of Biopharmacy, Medical University of Lublin, Chodzki 4a, 20-093 Lublin, Poland.
J Chem Theory Comput. 2022 Aug 9;18(8):5089-5107. doi: 10.1021/acs.jctc.2c00553. Epub 2022 Jul 29.
Carbohydrates play an essential role in a large number of chemical and biochemical processes. High structural diversity and conformational heterogeneity make it problematic to link their measurable properties to molecular features. Molecular dynamics simulations carried out at the level of classical force fields are routinely applied to study the complex processes occurring in carbohydrate-containing systems, while the usefulness of such simulations relies on the accuracy of the underlying theoretical model. In this article, we present the coarse-grained force field dedicated to glucopyranose-based carbohydrates and compatible with the recent version of the Martini force field (v. 3.0). The parameterization was based on optimizing bonded and nonbonded parameters with a reference to the all-atom simulation results and the experimental data. Application of the newly developed coarse-grained carbohydrate model to oligosaccharides curdlan and cellulose displays spontaneous formation of aggregates of experimentally identified features. In contact with other biomolecules, the model is capable of recovering the protective effect of glucose monosaccharides on a lipid bilayer and correctly identifying the binding pockets in carbohydrate-binding proteins. The features of the newly proposed model make it an excellent candidate for further extensions, aimed at modeling more complex, functionalized, and biologically relevant carbohydrates.
碳水化合物在许多化学和生化过程中起着至关重要的作用。由于其结构高度多样化和构象异质性,将其可测量的性质与分子特征联系起来具有一定的挑战性。在经典力场水平上进行的分子动力学模拟通常用于研究含碳水化合物系统中发生的复杂过程,而这些模拟的有用性依赖于基础理论模型的准确性。本文介绍了一种专用于基于吡喃葡萄糖的碳水化合物的粗粒化力场,并与最近版本的 Martini 力场(v.3.0)兼容。参数化是基于参考全原子模拟结果和实验数据对键合和非键合参数进行优化。将新开发的粗粒化碳水化合物模型应用于寡糖昆布多糖和纤维素,可自发形成实验确定特征的聚集。与其他生物分子接触时,该模型能够恢复葡萄糖单糖对脂质双层的保护作用,并正确识别碳水化合物结合蛋白中的结合口袋。新提出的模型的特点使其成为进一步扩展的理想候选,旨在模拟更复杂、功能化和更具生物学相关性的碳水化合物。