SIMULATION OF ION TRANSPORT THROUGH FERRITINS

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得了主要资金， 因此可在其他CRISP条目中表示。所列机构为 研究中心，而研究中心不一定是研究者所在的机构。 该研究项目的目的是利用哺乳动物铁调节蛋白铁蛋白作为分子机器开发新型纳米组装体。这个想法是利用铁蛋白的能力，在从外部通过蛋白质结构运输后，将离子和复合物螯合在其核心。与杨百翰大学的瓦特小组合作，我们目前正在使用合成和分子模拟相结合的方法来了解新离子是如何被纳入铁蛋白的。瓦特小组发现阴离子和金属络合离子可以共沉积在铁蛋白中，这为在铁蛋白中制备纳米材料开辟了新的合成可能性。这种合成方法已经证实，可以在铁蛋白核心合成全新的纳米结构，我们的主要目标是使用分子动力学模拟和理论来确定与这些过程相关的基本机制问题，并使用这些信息来指导新的合成策略和方向。为了回答一些基本的科学问题，需要分子尺度的模拟来补充合成和结构研究。特别是，理论工作的重点将是进行原子分子动力学模拟铁蛋白，多亚基21 kDa的蛋白质。铁蛋白的24个亚基形成具有通道的中空球体。推测有机小分子和过渡金属络离子会穿过通道并参与球体内的氧化还原反应。要做到这一点，通道或孔隙必须大大加宽或打开。蛋白质中孔打开的时间尺度被认为在几十纳秒到微秒的范围内。为了使用原子分子动力学轨迹模拟这些事件，我们的研究小组将被要求运行分子动力学模拟到至少40纳秒（ns），并进行后续的主成分分析（PCA）和引导基本动力学运行沿着主成分。从2ns模拟获得的初步结果已经显示具有相当大的方差的主成分，这意味着大尺度运动。为了获得更准确的大尺度运动表示，我们需要运行更长的模拟以及进行频繁的采样。所需的模拟只能在具有足够磁盘空间的非常快速的多处理器机器上有效地进行，以确保轨迹可以运行足够长的时间，并有效地存储和处理数据。此外，PCA分析涉及求解大型本征系统，以在计算中包含蛋白质中足够数量的原子。对于像铁蛋白这样的大蛋白质，内存需求是相当大的，并且超过了我们目前能够访问的资源。我们的初步结果都是使用分子动力学模拟包GROMACS模拟产生的。我们的计划是继续使用这个软件包，因为它已经在BIGBEN和Rachel上可用。我们可能会为每个轨迹使用<10个处理器，因为这通常为分子动力学模拟提供了最佳缩放。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. This research project is aimed at developing novel nanoassemblies using the mammalian iron-regulatory protein, ferritin, as a molecular machine. The idea is to use the ability of ferritin to sequester ions and complexes at its core following transport from the outside through the protein structure. In collaboration with the Watt group at BYU, we are currently using a combined synthetic and molecular simulation approach to understand how novel ions are incorporated into ferritins. The discovery that anions and metal complex ions can be co-deposited in ferritin by the Watt group has opened new synthetic possibilities for preparing nanomaterials in ferritin. This synthetic approach has confirmed that entirely novel nanostructures can be synthesized at the ferritin core, and our major goal is to use molecular dynamics simulations and theory to determine the basic mechanistic questions associated with these processes, and to use the information to guide new synthetic strategies and directions. In order to answer a number of the basic scientific questions, molecular scale simulations will be needed to complement the synthetic and structural studies. In particular, the focus of the theoretical effort will be to perform atomistic molecular dynamics simulations on Ferritin, a multi-subunit 21 kDa protein. The 24 subunits of ferritin form a hollow sphere with channels. Small organic molecules and transition metal complex ions are presumed to pass through the channels and participate in redox reactions within the sphere. For this to happen, the channels or pores would have to widen or open considerably. The timescale of pore opening in proteins is thought to range from tens of nanoseconds to microseconds. In order to simulate these events using atomistic molecular dynamics trajectories, our research group will be required to run molecular dynamics simulations to at least 40 nanoseconds (ns) and carry out subsequent Principal Component Analysis (PCA) and Guided Essential Dynamics runs along the principal components. Preliminary results obtained from a 2ns simulation already show Principal Components with fairly large variances, implying large scale motions. In order to get more accurate representations of the large scale motion, we need to run longer simulations as well as carry out frequent sampling. The required simulations can be effectively carried out only on very fast, multi-processor machines with adequate disk space to ensure that the trajectories can be run to long enough times and the data stored and processed efficiently. In addition, PCA analysis involves solving large eigensystems to include a sufficient number of atoms in the protein in the calculation. For a large protein like ferritin, the memory requirements are considerable, and exceed the resources which we are currently able to access. Our preliminary results have all been generated using the molecular dynamics simulations package GROMACS simulation. Our plan is to continue to use this package since it is already available on both BIGBEN and RACHEL. We will likely use <10 processors for each trajectory as that provides the best scaling in general for molecular dynamics simulations.