Wang Xiaoyu, Liu Linlin, Meng Peiran, Zhao Jian, Wang Lei, Liu Cui, Gong Lidong, Yang Zhongzhi
School of Chemistry and Chemical Engineering, Liaoning Normal University, Dalian 116029, People's Republic of China.
J Chem Theory Comput. 2025 Jul 22;21(14):6933-6949. doi: 10.1021/acs.jctc.4c01704. Epub 2025 Jul 3.
In additive force fields, the charge is a tunable parameter designed to represent average polarization effects through a mean-field average, which could not accurately respond to different environments. The polarizable force field (PFF) offers enhanced accuracy in representing intermolecular interactions by dynamically capturing electronic polarization effects. The ABEEM PFF for lipids, built on the fluctuating charge model, offers reasonable charge distributions and specific characterization of the hydrogen bonding interaction to improve the electrostatic interactions. The hierarchical parameterization strategy requires optimizing parameters for small molecules of lipid functional groups and subsequently applying these parameters to seven PC lipids, including 1,2-dipalmitoyl--glycero-3-phosphocholine (DPPC), dimyristoylphosphatidylcholine (DMPC), dilauroylphosphatidylcholine (DLPC), 1,2-dioleoyl--glycero-3-phosphocholine (DOPC), 1-palmitoyl-2-oleoyl--glycero-3-phosphocholine (POPC), 1,2-distearoyl--glycero-3-phosphocholine (DSPC), and 1-stearoyl-2-oleoyl-phosphatidylcholine (SOPC). The ABEEM-DBSS method was refined for the specific characteristics of lipid bilayers by charge partitioning calculations for solutes, thereby enhancing the computational efficiency in molecular dynamics (MD) simulations. Results show that the ABEEM PFF can reproduce the quantum mechanical (QM) data of model compounds representing phospholipids as well as the experimental condensed-phase properties of lipid bilayers. In particular, the optimization of dihedral parameters for hydrocarbons has improved the accuracy of the NMR deuterium order parameters of carbon atoms. Based on the advantages of ABEEM PFF, unsupervised machine learning methods were employed to reveal the correlation, which showed the regular impact between the fluctuating charge of phosphorus (P) atoms in lipids and the tilt of the vector between the P and nitrogen (N) atoms in the PC group with respect to the interface plane, as well as the dynamic behavior of water molecules. This work may lay the foundation for further investigations on the structures and properties of membrane proteins in terms of the ABEEM PFF.
在加和力场中，电荷是一个可调参数，旨在通过平均场平均来表示平均极化效应，而这种效应无法准确响应不同的环境。可极化力场（PFF）通过动态捕捉电子极化效应，在表示分子间相互作用方面提供了更高的准确性。基于波动电荷模型构建的用于脂质的ABEEM PFF，提供了合理的电荷分布以及对氢键相互作用的特定表征，以改善静电相互作用。分层参数化策略要求先优化脂质官能团小分子的参数，随后将这些参数应用于七种磷脂酰胆碱脂质，包括1,2 - 二棕榈酰 - sn - 甘油 - 3 - 磷酸胆碱（DPPC）、二肉豆蔻酰磷脂酰胆碱（DMPC）、二月桂酰磷脂酰胆碱（DLPC）、1,2 - 二油酰 - sn - 甘油 - 3 - 磷酸胆碱（DOPC）、1 - 棕榈酰 - 2 - 油酰 - sn - 甘油 - 3 - 磷酸胆碱（POPC）、1,2 - 二硬脂酰 - sn - 甘油 - 3 - 磷酸胆碱（DSPC）和1 - 硬脂酰 - 2 - 油酰磷脂酰胆碱（SOPC）。针对脂质双层的特定特性，通过对溶质进行电荷划分计算对ABEEM - DBSS方法进行了优化，从而提高了分子动力学（MD）模拟中的计算效率。结果表明，ABEEM PFF能够重现代表磷脂的模型化合物的量子力学（QM）数据以及脂质双层的实验凝聚相性质。特别是，对烃类二面角参数的优化提高了碳原子的核磁共振氘序参数的准确性。基于ABEEM PFF的优势，采用无监督机器学习方法来揭示相关性，该相关性显示了脂质中磷（P）原子的波动电荷与磷脂酰胆碱基团中P和氮（N）原子之间的向量相对于界面平面的倾斜度以及水分子的动态行为之间的规律影响。这项工作可能为进一步研究基于ABEEM PFF的膜蛋白的结构和性质奠定基础。
