Collaborative Research: Multiscale molecular simulations of protein-mediated bilayer fusion

智力可能是我们所知的细胞生命最重要的结构，它就是脂双层。脂类分子由一个可溶于水的“头”和一个不溶于水的“尾”组成，自发地组装成三明治状的双层膜，包围着所有活细胞，并进一步划分所有真核生物的细胞内部，这是植物、真菌、动物和人类所属的生命领域。典型的真核细胞的膜网络是非常复杂和高度动态的：小隔间像气泡一样从某些膜上萌发出来，将货物从细胞的一个部分带到另一个部分，在那里它们可以与其他膜融合，包括细胞的外膜。因此，双层融合是一个无处不在的生物过程，与物质和信息的运输紧密相连，因此它受到几类膜相关蛋白的精确控制。这些蛋白质显然在融合膜上发挥作用，但它们在分子水平上诱导的复杂的几何和拓扑形状转换序列不可能在实验中直接观察到。相比之下，分子模拟提供了一个了解这些细节的窗口，但到目前为止，相关的长度和时间尺度被证明太大，对于现实的系统规模来说，甚至无法观察到一个单一的聚变事件。这个项目建立了两个研究人员之间的合作，目的是通过将多尺度粗粒度建模的最新进展与增强采样分子模拟相结合来应对这一挑战。由于这一策略允许在同时表示大规模膜变形的同时纳入重要的化学细节，研究人员将能够阐明分子水平的机制如何在相关的生理长度和时间尺度上驱动融合事件。该项目分三个阶段进行，即：(1)通过加强采样对原始双层的融合进行建模，(2)开发模型融合蛋白的粗粒度模型，即SNARE系统，以及(3)将这两个步骤合并为单一方法。该项目将寻求许多与能量、形态和机械相关的主题，特别是围绕所谓的半融合中间状态的问题，在这种状态下，两个外部双层小叶已经融合，但由两个内部小叶形成的膜仍将两个间隔分开。由于双层融合在各种生物过程中的核心重要性，包括细胞内运输、病毒进入、神经递质释放、受精等，该项目将影响生物科学中的许多主题。除了研究中的具体问题，这里设想的计算方法朝着更复杂的多蛋白质/多膜现象的有效模拟迈出了早期步骤，因此将有助于未来对更广泛类别的分子生物学主题的研究。为了扩大研究成果的适用性，本项目开发的模拟框架将免费提供教程，这些教程将支持有效的学习，并促进现有技术和模块向新的应用转变。该项目建立了工程学和(生物)物理之间的跨学科交流，为该项目指导的学生的学术成长培养了一个激励的跨学科环境。它将进一步将理论和计算方法从工程学和物理学转移到生命科学及其日益量化的问题集。双层融合的普遍存在及其与生物物理学中一大类引人入胜的主题的联系，本身就是一个耐人寻味的跨学科学科，也为该项目中开发的专门知识提供了极好的机会，以支持专门针对STEM领域中代表性不足的群体的外联活动，例如通过课堂材料、讲座演示和公开演讲，两位研究人员将利用他们的经验和各自机构现有的成功项目来实施此类活动。

INTELLECTUAL MERITPerhaps the most important structure for cellular life as we know it is the lipid bilayer. Lipid molecules, consisting of a water-soluble "head" and water-insoluble "tails", spontaneously assemble into sandwich-like bilayer membranes, which surround all living cells and further compartmentalize the cellular interiors of all eukaryotic organisms the domain of life to which plants, fungi, animals, and humans belong. The network of membranes in a typical eukaryotic cell is very complex and highly dynamic: small compartments bud off from certain membranes like bubbles, carrying cargo from one part of the cell to another, where they can fuse with yet other membranes, including the outer membrane of the cell. Bilayer fusion is therefore a ubiquitous biological process, tightly linked to the transport of material and information, and therefore it is exquisitely controlled by several classes of membrane-associated proteins. These proteins clearly perform work on the fusing membranes, but the intricate sequence of geometric and topological shape transformations they induce on the molecular scale are impossible to observe directly in experiment. In contrast, molecular simulation offers a window onto these details, but until now the relevant length- and time-scales have proven too big to observe even a single fusion event for a realistic system size. This project establishes a collaboration between two investigators with the aim to meet this challenge by combining recent advances in multiscale coarse-grained modeling with enhanced-sampling molecular simulation. Since this strategy allows incorporating important chemical detail while simultaneously representing large-scale membrane deformations, the investigators will be able to elucidate how molecular-level mechanisms drive fusion events across the relevant physiological length- and time-scales. The project proceeds through three phases, namely: (i) modeling the fusion of pristine bilayers with enhanced sampling, (ii) development of coarse-grained models of model fusogenic proteins, the SNARE system, and (iii) combining these two steps into a single methodology. The project will pursue many topics of energetic, morphological, and mechanistic relevance, in particular questions revolving around the so-called hemifusion intermediate state, for which the two outer bilayer leaflets have already fused but a membrane formed by the two inner leaflets still separates the two compartments.BROADER IMPACTSThis project will impact many topics in the biological sciences due to the central importance of bilayer fusion in a variety of biological processes, including intracellular trafficking, viral entry, neurotransmitter release, fertilization, and more. Beyond the specific questions under study, the computational approach envisioned here takes early steps towards efficient simulation of more complicated multiple-protein/multiple-membrane phenomena and will therefore benefit future studies of a wider class of molecular biological topics. To broaden applicability of the research outcomes, the simulation framework developed in this project will be made freely available with tutorials that will support efficient learning and facilitate the transformation of existing techniques and modules towards novel applications. This project establishes cross-disciplinary exchange between engineering and (bio)physics, fostering a stimulating interdisciplinary environment for the academic growth of students mentored in this project. It will further the transfer of theoretical and computational methodologies from engineering and physics into the life sciences and their increasingly quantitative set of problems. The ubiquity of bilayer fusion and its connection to a wide class of fascinating themes in biological physics, which is in itself an intriguing cross-disciplinary subject, also present excellent opportunities for the expertise developed in this project to feed outreach specifically tailored towards groups underrepresented in STEM fields for instance through classroom material, lecture demonstrations, and public talks and both investigators will implement such activities, building on both their experience and existing successful programs at their respective institutions.