Collaborative Research: Computational Modeling of How Living Cells Utilize Liquid-Liquid Phase Separation to Organize Chemical Compartments

真核细胞已经进化出多种隔离和维持局部化学或分子浓度的机制。最明显的是物理膜，如将细胞质与周围环境隔开的细胞膜，或将染色体DNA限制在细胞核内的核膜。分隔机制是必不可少的，因为它们凌驾于浓度梯度的扩散平滑之上，否则将使细胞内容物均一化，并无法允许对关键细胞过程进行空间调节。最近发现的和目前细胞生物学中强烈关注的是在没有物理膜的情况下形成的化学室。该项目侧重于一个具体的例子：通过液-液相分离(LLP)将细胞质蛋白质和RNA结合到形成富含蛋白质的液滴的复合体中。通过将数学、计算和生物科学家聚集在一起，研究人员的目标是开发一个通用的计算建模平台，以研究LLP产生的细胞质液滴及其空间分布。其目的是从力学上了解这些隔室如何在活细胞中建立和保持mRNA定位和表达的细胞质异质性，以及与此相关的分子种类、复合体和动力学时间尺度。通过将该平台应用于其他活细胞，有可能了解LLP的基本细胞特异性分子成分和化学动力学，从而有助于理解跨细胞生物学的细胞内区隔的多样性。在细胞生物学中，膜及其在建立细胞外和细胞内化学隔间中的作用的研究有着丰富的历史。然而，相对较少的人知道，在没有物理膜的情况下，分子蛋白质、细胞器和染色体DNA在细胞质或细胞核内如何化学相互作用和自组织来创建、维持和进化局部的化学和大分子隔间。借助已解析的初级分子物种和物种络合物的空间和时间实验数据，本项目的研究人员专注于三个特定的目标。1.探索初级分子(蛋白质、RNA、蛋白质-RNA复合体)和微观(核、膜)物种、化学物种亲和力和空间限制条件的输入空间的计算模拟平台。该平台将产生一个结果相图，模拟活细胞数据(络合物和分子物种的动态自组织，液-液相分离造成的液滴形成)，并揭示足够的成分和相互作用，用于无膜细胞内化学室及其稳健性。2.通过随机模型和连续模型相结合的方法，在体内外实验数据的条件下，发现足够的分子物种、复合体和隐藏的化学亲和力，重现活细胞的化学区隔。3.扩展多相建模的数值工具，以适应由化学动力学、液滴的粘弹性和液-液相分离引起的流动所驱动的强烈波动和失衡行为。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Eukaryotic cells have evolved multiple mechanisms for sequestering and maintaining localized chemical or molecular concentrations. The most obvious is a physical membrane, such as the cell membrane that separates the cytoplasm from its surrounding environment or the nuclear membrane that confines chromosomal DNA within the nucleus. Mechanisms for compartmentalization are essential as they override diffusive smoothing of concentration gradients that would otherwise homogenize cellular contents and fail to allow spatial regulation of critical cellular processes. A recently identified and current intense focus in cell biology is on chemical compartments that form in the absence of physical membranes. This project focuses on a specific example: the binding of cytoplasmic proteins and RNAs into complexes that form protein-rich droplets by way of liquid-liquid phase separation (LLPS). By bringing together mathematical, computational, and biological scientists, the investigators aim to develop a general computational modeling platform to study cytoplasmic droplets and their spatial distributions that arise from LLPS. The aim is to understand mechanistically how these compartments establish and preserve cytoplasmic heterogeneity in mRNA localization and expression in live cells, and the molecular species, complexes, and kinetic timescales that are responsible. By applications of this platform to other live cells, there is the potential to understand the essential cell-specific molecular ingredients and chemical kinetics for LLPS, thereby contributing to understanding of the diversity of intracellular compartmentalization across cell biology. There is a rich history in cell biology of the study of membranes and their role in establishing extracellular and intracellular chemical compartments. Yet, relatively little is known about how molecular proteins, organelles, and chromosomal DNA, within the cytoplasm or within the nucleus, chemically interact and self-organize to create, sustain, and evolve localized chemical and macromolecular compartments in the absence of physical membranes. Armed with resolved spatial and temporal experimental data of primary molecular species and species complexes, the investigators in this project focus on three specific aims. 1. A computational modeling platform to explore the input space of primary molecular (proteins, RNAs, protein-RNA complexes) and microscopic (nuclei, membranes) species, chemical species affinities, and spatial confinement conditions. This platform will produce a phase diagram of outcomes that mimics live cell data (dynamic self-organization of complexes and molecular species, droplet formation due to liquid-liquid phase separation), and that reveals sufficient ingredients and interactions for membrane-less, intracellular chemical compartments, and their robustness. 2. By way of coupled stochastic and continuum modeling, conditioning on ex vivo and in vivo experimental data, to discover sufficient molecular species, complexes, and hidden chemical affinities that reproduce the chemical compartmentalization of live cells. 3. To extend numerical tools for multiphase modeling to accommodate strong fluctuations and out-of-equilibrium behavior driven by chemical kinetics, viscoelasticity of droplets, and induced flow by liquid-liquid phase separation.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.