Disentangling liquid-liquid phase separation and aggregation of intrinsically disordered proteins

蛋白质和核酸可以在没有膜的帮助下进行物理分裂。这些组件的行为就像液滴，为需要特定蛋白质浓度的生物分子反应提供了一个动态环境。已知其中一些蛋白质聚集并引起神经退行性疾病。这就提出了一个关键问题，即功能性液滴如何转变为病理固体聚集体。在本研究中，将使用先进的计算方法来观察这些蛋白质在液滴和聚集体中的结构差异。从这个项目中学到的知识对于设计能够破坏病理聚集但不破坏功能性液滴的新策略至关重要。计算建模的概念将被引入到亚利桑那州立大学（ASU）的本科研究和高中生物物理学徒计划中，这所大学的学生人数明显不足，而这所大学的学生人数又来自STEM中代表性不足的群体的大量个人。PI将特别关注通过亚利桑那州立大学专家数据库推广各种传播方法，以介绍生物物理学的概念，并促进向广泛受众传播材料。本质上无序的蛋白质，缺乏明确的折叠结构的蛋白质，已被证明参与液-液相分离（LLPS）和聚集。前者是可逆的，提供重要的细胞内功能，而后者通常是不可逆的，并且与疾病有关。人们对显示液滴形成如何诱导聚集越来越感兴趣，但对于能够区分这两种组装状态的结构特性知之甚少。一个主要的挑战是缺乏高分辨率的建模方法来整合多种实验技术的观测结果。在这项工作中，pi将开发一种新的计算建模框架来集成多种技术。具体来说，全原子显式溶剂模拟将用于研究无序蛋白质在分散和凝聚相中的结构性质，以及这些性质如何受到它们的相互作用伙伴的影响。粗粒度模型将进一步用于研究液滴和聚集体的形成。液滴形成对其聚集偏好的影响将进行定量评估。该项目将利用计算工具建立控制LLPS和聚合之间结构（非）相似性的基本物理原理。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Proteins and nucleic acids can physically partition without the help of membranes. These assemblies behave like liquid droplets and provide a dynamic environment for biomolecular reactions requiring specific protein concentrations. Some of these proteins are also known to aggregate and cause neurodegenerative diseases. This raises the key question how the functional liquid droplets transform into the pathological solid aggregates. In this research, advanced computational methods will be used to view how the structures of these proteins differ in the liquid droplets and aggregates. The knowledge learned from this project will be essential in designing novel strategies that can disrupt pathological aggregates but do not disrupt functional droplets. The concepts of computational modeling will be introduced into both the undergraduate research and high school biophysics apprenticeship programs at Arizona State University (ASU), a university with a significant underrepresented student population in a location that is home to a large number of individuals from underrepresented groups in STEM. The PI will pay specific attention to promoting a variety of dissemination approaches through the ASU expert database to introduce the concepts of biophysics and to facilitate the dissemination of materials to a broad audience.Intrinsically disordered proteins, proteins that lack a well-defined folded structure, have been shown to participate in both liquid-liquid phase separation (LLPS) and aggregation. The former is reversible, providing important intracellular functions, whereas the latter is often irreversible and disease-related. There has been increasing interest in showing how droplet formation could induce aggregation, but little is known regarding the structure properties capable of distinguishing these two assembly states. One major challenge is the lack of high-resolution modeling methods for integrating observations from multiple experimental techniques. In this work, the PIs will develop a novel computational modeling framework to integrate multiple technologies. Specifically, all-atom explicit-solvent simulations will be used to investigate the structure properties of disordered proteins in both dispersed and condensed phases and how these properties are influenced by their interaction partners. Coarse-grained models will further be used to investigate both the formation of liquid droplets and aggregates. The impact of droplet formation on its aggregation preference will be quantitatively assessed. The project will leverage computational tools to establish fundamental physics principles that govern the structural (dis)similarities between LLPS and aggregation.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.