HYDROCARBON PERMEATION THROUGH LIPID BILAYERS

这个子项目是许多利用 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， 因此可以在其他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。 动机：了解分子穿透细胞膜的机制对于深入了解许多生物过程（如药物输送和毒性）非常重要。虽然通过脂质双层（模拟细胞膜）的渗透已被广泛研究，但对渗透的基本热力学贡献尚未得到很好的理解1，2。本计画的目标是利用先进的分子模拟方法和高效能计算，来获得渗透物和脂质双层之间的生物物理学的详细知识。建议理由：多环芳烃（PAH）广泛存在于环境中，众所周知具有毒性和致癌性，但其毒性机制尚不清楚3，4。已知PAH与生物膜相互作用，并且怀疑由此引起的膜结构和功能的变化损害细胞生长和活性3。在这个项目中，我们将使用分子动力学（MD）方法研究一系列含有6到20个碳原子的PAH在脂质双层中的转运。该研究的主要目的是深入了解脂质双层的大小和形状对渗透的影响，以系统的方式，尚未充分显示在文献中。自由能方法：已经开发了许多用于从MD模拟计算自由能的方法并在文献中使用。这些方法中最常用的一些包括自由能微扰（FEP）、热力学积分（TI）、带加权直方图分析方法的伞形采样（US-WHAM）和约束力（CF）5。由于膜的不均匀性，采样将沿平行于双层法线6的反应坐标沿着进行。CF和US-WHAM方法都已成功地用于文献中确定通过双层的平均力（PMF）的潜力。这些方法之间的主要区别之一是US-WHAM计算特定相空间窗口中的态密度，而CMF平均来自周围系统的粒子上的力。力平均通常提供更好的收敛状态密度，因此我将使用CF方法来找到PMF 6。CF也更容易促进并行模拟。与CF方法一起，TI方法将用于计算双层中几个离散点之间以及在不同温度下的自由能差。然后，TI方法的结果将用于验证CF模拟并确定自由能7的组成部分。高性能计算：该项目是利用高性能计算资源的理想选择。这些模拟中的每一个都需要在单个处理器上花费几个月的时间，但是当并行执行时，挂钟的时间可以缩短到几周。除了模拟所需的计算时间外，还需要大量数据来计算有意义的PMF。对于由己烷渗透二油酰磷脂酰胆碱模型膜组成的系统，总共需要500 ns的模拟时间来产生PMF 8。在单个处理器上，模拟可能需要一年多的时间才能完成，并且在两个飞秒的时间步长下，将需要存储多达2500万个时间步长的数据。这些要求仅用于产生一个分子的PMF。时间轴CF模拟：第1-4个月计算PMF：第4-6个月TI模拟：第6-8个月各种温度下TI：第9-12个月参考文献：1）Ke P.C.，乔河J. Phys. Condensed Matter. 19：（2007）2）Ginzberg V.V.，巴利耶帕里湾纳米字母。0（0）：（2007）; 3）Plant A.L.，克纳普R. D.，史密斯公司J.Biol.Chem.262（6）：2514（1986）; 4）Bemporad D.，Luttmann C.，埃塞克斯J.W. Biophys. J. 87：1（2004）; 5）Ghoufi A.，马尔弗雷·P·莫莱克。104（22-24）：3787（2006）; 6）Trzesniak D.，Kunz A.E.，货车冈斯特伦ChemPhysChem. 8：162（2007）; 7）Peter C.，奥斯滕布林克角，货车Dorp A.，货车冈斯特伦120（6）：2653（2004）; 8）MacCallum J.L.，蒂勒曼D. P. J. Am. 128：125（2005）

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. Motivation: Understanding the mechanisms through which molecules penetrate cell membranes is important for gaining insight into many biological processes such as drug delivery and toxicity. While permeation through lipid bilayers, which model cell membranes, has been studied extensively, the underlying thermodynamic contributions to penetration are not well understood1,2. The goal of this project is to gain detailed knowledge of the biophysics between a permeant and lipid bilayer using advanced molecular modeling methods and high performance computing. Proposed Reasearch: Polycyclic Aromatic Hydrocarbons (PAH) are widespread in the environment and are well known to be toxic and carcinogenic but mechanisms for their toxicity are not clear3,4. It is known that PAH interact with biological membranes, and it is suspected that resulting changes to the structure and functioning of the membrane impair cell growth and activity3. In this project we will study the transport of a series of PAH containing from six to twenty carbon atoms, in lipid bilayers to gain using molecular dynamic (MD) approaches. The main objective of the study is to gain insights into effects of size and shape on permeation of a lipid bilayer in a systematic way that has not been adequately shown in literature. Free Energy Methods: A number of methods for calculating free energies from MD simulations have been developed and used in literature. Some of the most used of these methods include free energy perturbation (FEP), thermodynamic integration (TI), umbrella sampling with the weighted histogram analysis method (US-WHAM), and constraint force (CF)5. Due to the inhomogeneity of the membrane, the sampling will be carried out along a reaction coordinate parallel to the bilayer normal6. Both CF and US-WHAM methods have been successfully used in literature for determining a potential of mean force (PMF) through a bilayer. One of the major differences between these methods is that US-WHAM counts the density of states in a particular phase space window whereas CMF averages forces on a particle from the surrounding system. Force averaging generally provides better convergence density of states therefore I will use the CF method to find the PMF6. CF also more easily facilitates parallel simulations. Together with the CF method, the TI approach will be used to calculate the free energy differences between a few discrete points in the bilayer as well as at various temperatures. The results from the TI method will be then used to validate the CF simulations and determine components of the free energy7. High-performance Computing: This project is ideal for utilizing high-performance computational resources. Each of these simulations would take months on a single processor but, when performed in parallel, the wall clock time could be shortened to weeks. In addition to the computational time required for the simulations, large amounts of data are needed to calculate a meaningful PMF. For a system composed of hexane permeating a dioleoyl phosphatidylcholine model membrane, a total of 500ns of simulation time was required for generating the PMF8. On a single processor that simulation could take over a year to complete and with a two femtosecond time step, would require storage of data for up to twenty-five million time steps. These requirement are for generating the PMF of one molecule only. Timeline CF simulations: Months 1-4 calculating PMF: Months 4-6 TI simulations: Months 6-8 TI at various temperatures: Months 9-12 References: 1) Ke P.C., Qiao R. J. Phys. Condensed Matter. 19: (2007) 2) Ginzberg V.V., Balijepalli S. Nano Letters. 0 (0): (2007); 3) Plant A.L., Knapp R.D., Smith L.C. J. Biol. Chem. 262 (6): 2514 (1986); 4) Bemporad D., Luttmann C., Essex J.W. Biophys. J. 87: 1 (2004); 5) Ghoufi A., Malfreyt P. Molec. Phys. 104 (22-24): 3787 (2006); 6) Trzesniak D., Kunz A.E., van Gunsteren W.F. ChemPhysChem. 8: 162 (2007); 7) Peter C., Oostenbrink C., van Dorp A., van Gunsteren W.F. J. Chem. Phys. 120 (6): 2653 (2004); 8) MacCallum J.L., Tieleman D.P. J. Am. Chem. Soc. 128: 125 (2005)