Polarizable Force Field for Proteins and Lipids

描述(申请人提供)：基于原子模型的计算在理解生物分子系统方面发挥着越来越重要的作用。到目前为止，这些计算通常是使用势函数来执行的，这些势函数平均使用原子部分电荷的有效参数来解释多体极化效应。为了克服这一限制，在第一个资助期内，我们开发了蛋白质和脂类的势能函数，其中包括通过经典的Drude振子模型显式处理诱导电子极化。本赠款申请中建议的研究重点是完成针对代表蛋白质和脂类的模型化合物的参数优化，然后测试已开发的大分子系统中的力场，其中存在大量的实验数据。目标1中基于小分子的优化将以代表氨基酸电离状态的化合物为目标，这些化合物是使用双肽和多肽量子力学和实验数据计算和优化phi、psi主链和chi侧链参数所需的。在目标1中开发的参数将在一系列具有不同螺旋、β折叠和β转角倾向的模型多肽上进行测试，通过在显式溶剂中的哈密顿回火副本交换以及在溶液和晶体环境中的高分辨率蛋白质的模拟中来验证力场可以重现实验中可获得的结构和动力学性质。目的3将侧重于定量评估力场复制能级的能力，包括选定蛋白质的PKA位移、Rubredosin的氧化还原电位和电子传递速率、Calbindin D9k中的Ca~(2+)与EF-Hand的协同结合以及脂单分子层和双层的界面电位。在拟议的研究完成后，一个最先进的蛋白质和脂类的可极化经验力场将提供给计算化学界。此外，还将获得有关电子极化对许多生物现象的贡献的新见解。 公共卫生相关性：这项研究项目的目标是完成一个力场的发展，明确地说明蛋白质和膜的激发极化。这样的力场将具有更高的精确度，将允许对具有生物医学重要性的广泛分子系统进行真实的计算机模拟。这些类型的计算机模拟在药物发现和线索优化中也发挥着关键作用。

DESCRIPTION (provided by applicant): Computations based on atomistic models are playing an increasingly important role in understanding biomolecular systems. To date, these computations have typically been performed using potential functions that account for many-body polarization effects in an average way using an effective parameterization of the atomic partial charges. To overcome this limitation during the first funding period we have undertaken the development of a potential energy function for proteins and lipids that includes the explicit treatment of induced electronic polarization via the classical Drude oscillator model. The studies proposed in the present grant submission focus on completion of the optimization of parameters targeting model compounds representative of proteins and lipids followed by testing of the developed force field in macromolecular systems for which extensive experimental data exist. Small molecule based optimization in Aim 1 will target compounds representing ionization states of amino acids required for pKa calculations and optimization of the phi, psi backbone and chi sidechain parameters using di- and polypeptide quantum mechanical and experimental data. Parameters developed in Aim 1 will be tested on a series of model polypeptides with different helical, beta sheet and beta turn propensities via Hamiltonian tempering replica-exchange in explicit solvent and in simulations of high-resolution proteins both in solution and crystal environments to validate that the force field can reproduce experimentally accessible structural and dynamic properties. Aim 3 will focus on quantitative evaluation of the ability of the force field to reproduce energetic observables including pKa shifts in selected proteins, redox potentials and electron transfer rates in rubredoxin, cooperative binding of Ca2+ to the EF-hands in calbindin D9k, and interfacial potentials of lipid monolayers and bilayers. Upon completion of the proposed study a state-of-the-art polarizable empirical force field for proteins and lipids will be available to the computational chemistry community. In addition, novel insights on the contribution of electronic polarization to a number of biological phenomena will be obtained. PUBLIC HEALTH RELEVANCE: The goal of this research project is to complete the development of a force field accounting explicitly for induced polarization for proteins and membranes. Such a force field will have an improved accuracy that will permit realistic computer simulations of a wide range of molecular systems that have biomedical importance. These types of computer simulations also play a critical role in the drug discovery and lead optimization.