Mechanism of membrane pore formation by antimicrobial peptides

抗菌肽提供了对微生物感染的天然防御，因此有望成为基础 新的抗生素。大量证据表明，它们以细菌膜为靶标，通过形成 毛孔或分解毛孔。然而，对这一过程的理解达到了预测行为的水平 缺乏特定的多肽。在过去的几年里，Pi的团队使用了新的隐式膜模型， 以及详细的全原子模拟，以获得对这些肽功能的重要见解。《少年派》 现在的目标是使用一种多管齐下的方法，包括全原子模拟，隐式溶剂模拟， 和分子热力学模型，以开发一种系统的方法来预测 多肽稳定的膜孔。与实验实验室的密切互动将有助于验证预测 模特们。这项工作有三个目标。第一个重点是β-发夹抗菌肽Protigin，目的是 以阐明它是否形成了完整的β桶、不完整的桶(弧形)或经典的环状孔隙。 将使用电生理学、染料渗漏和抗微生物活性测量来验证理论 结果。另外两个目标为多肽诱导膜的全面理论提供了关键要素 孔形成：多肽-多肽相互作用和膜变形的自由能。其影响范围 多肽在成孔过程中的相互作用是一个亟待解决的问题。派的群 将首先通过与实验和显式模拟势能的比较来验证隐式溶剂化模型 平均作用力，进而表征蜂毒素和麦芽菌素在膜上的聚集程度 表面和毛孔中。也将考虑将马尼宁与PGLa的混合物用于解决所观察到的 这两种多肽之间的协同作用。第三个目的是在理论中包含膜的自由能。 变形。多肽稳定的孔的自由能包含来自多肽-孔相互作用的贡献， 多肽-多肽相互作用和膜变形。孔结构具有四种特性： 混合膜的大小、形状、头部基团分布和电荷分布。全原子模拟将是 用于获取将这四个属性的分析函数参数化的数据。此函数一起使用 通过隐式溶剂模拟，将被用来确定自由能作为半径和 最低自由能多肽-孔结构。由此产生的结构将通过全原子模拟进行测试。这个 不同多肽在中性膜和带电性膜中的孔结构应该提供一个明确的答案 蜂毒素与马尼汀选择性差异的原因及设计依据 高效低毒的多肽或多肽类药物。

Antimicrobial peptides provide natural defense against microbial infection and thus hold promise as the basis of new antibiotics. Substantial evidence suggests that they target the bacterial membranes, by either forming pores or disintegrating them. However, an understanding of this process to the level of predicting the behavior of specific peptides is lacking. Over the last few years the PI’s group used novel implicit membrane modeling, as well as detailed all-atom simulations, to obtain significant insights into the function of these peptides. The PI now aims to use a multipronged approach that encompasses all-atom simulations, implicit-solvent simulations, and molecular thermodynamic modeling, to develop a systematic approach that predicts the structure of peptide-stabilized membrane pores. Close interaction with experimental labs will help validate the predicted models. This work has three aims. The first focuses on the β-hairpin antimicrobial peptide protegrin, and aims to elucidate whether it forms complete β-barrels, incomplete barrels (arcs), or classical toroidal pores. Electrophysiology, dye leakage, and antimicrobial activity measurements will be used to validate the theoretical results. The other two aims provide key elements in a comprehensive theory of peptide-induced membrane pore formation: peptide-peptide interactions and the free energy of membrane deformation. The extent of interactions between peptides in the process of pore formation is a crucial unsolved problem. The PI’s group will first validate the implicit solvation models by comparison to experiments and explicit simulation potentials of mean force and then proceed to characterize the extent of aggregation of melittin and magainin on membrane surfaces and in pores. Mixtures of magainin with PGLa will also be considered to address the observed synergy between these two peptides. The third aim is to include in the theory the free energy of membrane deformation. The free energy of a peptide-stabilized pore contains contributions from peptide-pore interactions, peptide-peptide interactions and membrane deformation. Pore structure is characterized by four properties: size, shape, headgroup distribution, and charge distribution in mixed membranes. All-atom simulations will be used to obtain data that will parameterize an analytical function of these four properties. This function, together with implicit-solvent simulations, will be used to determine profiles of free energy as a function of radius and lowest free energy peptide-pore structures. The resulting structures will be tested by all-atom simulations. The pore structure of different peptides in neutral and charged membranes should provide a definitive answer to the origin of the difference in selectivity between melittin and magainin and a firm basis for the design of peptides or peptidomimetics with high potency and low toxicity.