Computational Studies of Biomolecular Structure and Function

许多重要的生物学过程，如信号传导、识别和转运，以及感染和免疫反应，都涉及多肽或蛋白质与脂质之间的相互作用。由于在脂质双分子层提供的各向异性和化学复杂环境中难以阐明肽和蛋白质的结构，这些过程的详细表征目前受到阻碍。由于这些原因，膜蛋白的结构研究经常使用清洁剂分子，它模仿脂质双分子层的化学性质，来溶解蛋白质，使它们更容易结晶。然而，目前尚无通用的膜蛋白结构实验测定方法，因此需要可靠的预测算法。利用计算机模拟是了解生物分子结构和功能的一种有效方法，但目前这些研究的范围受到研究脂质中蛋白质和肽的折叠和自组装所需的长时间尺度的限制。

Many important biological processes, such as signalling, recognition, and transport, as well as infection and immune response, involve interactions between peptides or proteins and lipid. The detailed characterization of these processes is currently impeded by the difficulty in elucidating the structure of peptides and proteins in the anisotropic and chemically complex environment provided by lipid bilayers. For these reasons, structural studies of membrane proteins often employ detergent molecules, which mimic the chemical properties of lipid bilayers, to solubilize the proteins, making them more amenable to crystallization. However, at present there are no general methods for the experimental determination of membrane protein structure, underscoring the need for reliable predictive algorithms. A powerful approach to gain insight into the structure and function of biomolecules is to use computer simulations, but the scope of these studies is currently limited by the long time scales required to investigate the folding and self-assembly of proteins and peptides in lipid. In this proposal, we will address these limitations by combining novel sampling algorithms developed in our laboratory with massive computational resources awarded to us by Compute Canada. These extraordinary resources put us in a unique position to examine the interaction of peptides and proteins with lipid and detergent molecules in four different but complementary studies. In two of these projects, we will study how lipid and detergent self-assemblies solvate proteins and determine their fold. Reciprocally, in the other two studies, we will examine how the self-assembly of peptides affects the structure of lipid bilayers. This work will provide advanced computational tools for the much-needed prediction of membrane protein structure, and will help improve detergent-based approaches to elucidate the structure of membrane proteins experimentally. Just as important, the proposed studies will afford fundamental insight into the molecular basis for the action of antimicrobial peptides and the toxicity of peptides derived from prions implicated in "mad cow" and chronic wasting diseases, two serious threats to Canadian wildlife and agriculture.