蛋白质专一性极化力场在分子动力学模拟中的应用

Electrostatic interaction is the most important interaction in biomolecules,such as protein folding,protein-ligand binding, protein-protein interaction,electron transfer,proton binding and release, enzyme reaction, et al. A well-known deficiency of the widely used force fields is the lack of electronic polarization of proteins. In standard nonpolarizable force fields, the partial atomic charges of the protein are pre-fixed regardless of the change of the electrostatic environment. Our recent work that includes polarization effect in molecular dynamics simulation has demonstrated some important effect of electronic polarization on structure and dynamics of proteins. In this APPC (adaptive polarized protein-specific charge) approach, the partial atomic charges of protein are determined from quantum chemistry calculation of protein in solvent using a fragment approach in which the atomic charges of relevant residues are fitted at regular intervals in the MD simulation. Comparison of simulation results with the available experimental data demonstrated that intra-protein polarization plays a significant role in stabilizing protein structure and dynamics. A major effect of polarization results from the electronic polarization of intra-protein hydrogen bonds. Recent studies demonstrated that electronic polarization energetically stabilizes hydrogen bonds in proteins. Since hydrogen bonding is a prevalent feature in protein structure, the stabilizing effect of hydrogen bond has a significant impact on protein structure,dynamics and folding. In the project,the application of APPC will focuse on the study about the metalloprotein: Correctly describe the change about the ligancy and charge of the metallic ion in molecular dynamics. However, on-the-fly quantum calculation is computationally expensive,in which quantum mechanical calculation must be periodically carried out to obtain new atomic charges. It is describe to employ a simpler and effective method to include the effect of electrostatic polarization and thus we refitted the result of quantum chemistry calculation to obtain a simpler analytical polarizable hydrogen bond (PHB) model. Then the PHB is used to study the following two aspects. 1)protein-protein interaction: Accurately describe the intramolecular and intermolecular interactions and study the electrostatic contribution to binding free energy. 2)protein folding: Accuartely build the free energy landscape of protein folding and deeply understand the kinetic and thermodynamical behavior about protein folding. The results of current project will provide important theoretical basis to make quantitatively accurate description of protein's structural and functional properties.

静电相互作用是生物分子中最重要的相互作用，但目前分子力场对静电相互作用的描述都不包含关键的静电极化效应。新发展的基于全量子力学计算得到具有静电极化效应的原子电荷，对不同体系大量的分子动力学模拟计算证实了静电极化效应的重要性。特别是对于蛋白质中氢键的影响尤其关键，这对于其结构的稳定性，动力学性质以及折叠都起到了不可忽略的作用。本项目在前期工作基础上，将其进一步应用到金属蛋白的研究中，正确描述金属蛋白离子配位数和电荷在动力学过程中的变化。为了应用于更复杂的体系，提高计算效率，对量化计算的结果进行分析拟合，将新的极化电荷模型应用到蛋白质和蛋白质的相互作用：精确的描述蛋白质分子内和分子间的相互作用，研究静电极化效应对结合自由能的影响和蛋白质折叠中：精确的构建蛋白质折叠的自由能面，深入理解蛋白质折叠的热力学和动力学行为。本项目的研究结果将对定量可靠的描述蛋白质分子的结构功能和动力学行为提供重要的理论

由于传统分子力场缺乏显式的静电极化效应，严重制约着动力学模拟的可靠性。基于精确量子力学计算的新型的动态的蛋白质极化力场克服了目前广泛使用的分子力场的不足，提高了蛋白质等生物大分子理论计算的可靠性和精度。本项目主要致力于以下几方面的研究。（1）用动态极化力场研究了villin headpiece蛋白质折叠的机理；研究了一些重要蛋白的稳定性；研究了人类凝血酶蛋白（thrombin）与配体的相互作用机理；研究了桥梁水分子W301在介导HIV-protease与配体结合的物理机制。（2）由于动态极化力场在模拟过程中需要实时量化计算蛋白质的原子电荷，计算量非常昂贵，很难适用于大体系的折叠。所以我们采用了加速MD方法对一些大体系进行折叠研究，研究发现这些体系在100ns内成功到达了折叠态，而传统分子力场对这些体系的研究是失败的。（3）研究了DNA探针和不同的microRNA相互作用机制；DNA与石墨烯分子的相互作用机制，研究结论与实验观察是一致的。（4）由于传统分子力场对于蛋白质折叠研究有一定的二级结构倾向性，所以一个具有均衡二级结构分布的分子力场对成功的蛋白质折叠是必要的。我们利用高精度的量化计算，发展了一种基于二维傅里叶展开来拟合二面角项参数的力场。研究表明该力场对于计算的J耦合值、化学位移、二级结构分布等与实验吻合的很好，较传统力场有较大的改进。（5）利用最新发展的精确可靠高效率的半浮动电荷的可极化力场（EPB）研究了7个重要体系（2I9M (PDB: 2IM9), Trpcage (PDB: 1L2Y), 1WN8 (PDB: 1WN8), C34 (PDB:1AIK:C-terminal), N36(PDB:1AIK:N-terminal), 2KES (PDB:2KES), 2KHK (PDB:2KHK)）的折叠机理，而传统的非极化的AMEBR电荷对这些体系的折叠研究是失败的。.现已发表相关的SCI论文9篇，三篇论文在审稿中，完成了预期的科研目标。