Computational studies of conformational switching in proteins: function and evolution

蛋白质是细胞分子机器的“运动部件”。它们执行一系列高度专业化的任务，如催化生化反应和传递信号。然而，无论是单个蛋白质的结构还是蛋白质复合体的结构，今天都不能可靠地预测--我们对蛋白质折叠和相互作用的物理知识仍然不完整。阻碍进展的一个关键因素是传统分子动力学模拟的巨大计算成本，使得许多问题无法计算。使这个问题更加紧迫的是，目前的生物物理实验越来越重视大型和复杂的构象转变，包括有序和无序状态之间的构象转变，以此作为了解蛋白质如何执行其功能的一种方式。在接下来的五年里，我的研究计划的一个组成部分是开发一种系统的程序，在分子模拟中形成粗粒度的势能函数。我将应用它来进一步增强一个独特的蛋白质折叠和相互作用的中等分辨率模型，该模型是我的团队最近开发的，它击中了细节和计算速度之间的权衡。利用这个粗粒度的模型，我将在两个互补的区域中研究蛋白质构象转换的生物物理基础。(A)我们将调查哪些因素允许“变质”蛋白质在完全不同的三维形状之间突然和可逆地切换，由结合事件或--如在进化过程中--由突变触发。我将探索的一个假设是，这些形状变化的蛋白质是否代表了通往新蛋白质形状的进化“桥梁”。在第二个领域，(B)我们将研究无序蛋白质中缺乏稳定三维形状的构象转变如何形成这类柔性蛋白质功能行为的分子基础，包括分子识别。这一研究计划将为分子模拟中的系统粗粒化提供亟需的新技术。同样重要的是，它将提供对蛋白质构象转换物理学的更深层次的理解，从而增强我们对细胞机制的理解。该计划还将通过为基于构象转换的生物传感器或靶向药物输送系统的开发奠定基础，从而使生物技术行业受益。

Proteins are the ‘moving parts’ of the molecular machinery of cells. They carry out a range of highly specialized tasks, such as catalyzing biochemical reactions and transducing signals. Yet neither structures of individual proteins nor protein complexes can be reliably predicted today — our knowledge of the physics of protein folding and interaction remains incomplete. A key factor hampering progress is the large computational cost of traditional molecular dynamics simulations, leaving many problems out of reach to computation. Adding urgency to this problem is the increased emphasis of current biophysical experiments on large and complex conformational transitions, including between ordered and disordered states, as a way to understand how proteins carry out their functions. One component of my research program over the next five years is the development of a systematic procedure for formulating coarse-grained potential energy functions in molecular simulations. I will apply it to further enhance a unique intermediate-resolution model for protein folding and interaction that was recently developed in my group and that hits a ‘sweet-spot’ in the tradeoff between detail and computational speed. With this coarse-grained model, I will investigate the biophysical basis of protein conformational switching in two complementary areas. (A) We will investigate the factors that allow ‘metamorphic’ proteins to abruptly and reversibly switch between entirely different three-dimensional shapes, triggered by a binding event or — as in evolutionary processes — by mutations. A hypothesis that I will explore is whether these shape-shifting proteins represent evolutionary ‘bridges’ to new protein shapes. In the second area, (B) we will study how conformational transitions in disordered proteins, which lack a stable three-dimensional shape, form the molecular basis for the functional behavior of this type of flexible proteins, including molecular recognition. This research program will provide much needed new techniques for systematic coarse-graining in molecular simulations. Equally important, it will provide a deeper understanding of the physics of conformational switching in proteins and thereby enhance our understanding of the cellular machinery. The program will also benefit the biotechnology industry by laying the foundation for the development of biosensors or targeted drug-delivery systems based on conformational switching.