多肽和磷脂复合体系相行为的多尺度模拟研究

Membrane interacting peptides (MIPs), such as antimicrobial peptides and fusion peptides are involved in numerous crucial biological processes, such as antimicrobial defense mechanisms, membrane fusion, transport of therapeutic compounds, and others. At present, the mechanism of interaction of MIPs with membrane still remains elusive. Recent experimental work suggests that the interactions of MIPs with lipids can be inferred from the phase behavior of peptide/lipid mixtures. Molecular dynamics (MD) simulation is a powerful tool to theoretically study peptide-lipid interactions, which can provide a detailed description of the phase behavior. However, considerable obstacles need to be faced to obtain accurate results, due to the time-consuming computation of long-range electrostatic interactions and to the relaxation times of these phenomena, which are typically in microsecond, longer than those presently accessible to MD. Thereby, we employ hybrid particle-field MD (MD-SCF) technique in this study. MD-SCF is a heterogeneous multi-scale method and is efficient to simulate some slowly evolving collective processes. By developing the theoretical framework and the GPU-accelerated algorithm for electrostatic interactions based on MD-SCF, we can efficiently simulate the phase behavior of peptide/lipid mixtures.The relationship between the phase behaviors and peptide-lipid interactions will be presented. The competitive or cooperative interactions between peptides and lipids will be revealed. Moreover, we will study the self-assembly of β-sheet peptides and the interactions of self-assembled nanostructures with membrane. The peptide-lipid interactions and peptide-peptide interactions need to be elucidated in these processes. Based on these studies, the molecular mechanism of MIPs in many biological processes can be inferred.

与生物膜有相互作用的一类多肽，如抗菌肽、融合肽等，在抗菌防御、促膜融合、跨膜输运等方面起着非常重要的作用，然而目前对于这类多肽的作用机理仍不是很清楚。由于该类体系稳定的相结构往往要在微秒的时间尺度上形成，且分子间存在着计算耗时的长程静电相互作用，因此常规的分子动力学模拟不适合研究这类典型的带电大分子复合体系。本项目拟在多尺度的粒子场-分子动力学方法（MD-SCF）的基础上发展合适的对静电相互作用的描述，并在已有GPU加速的MD-SCF程序的基础上开发高效率的静电计算模块，大大提高对这类体系的模拟效率，利用这种创新方法研究多肽和磷脂混合体系相行为的发生规律，认清多肽和磷脂分子间的多种相互作用的竞争和协同机制。同时拓展研究β-sheet多肽的自组装以及其自组装体在磷脂膜中的稳定性，进一步揭示多肽和多肽之间的相互作用，阐明多肽在膜介导的生物活动中的作用机理。