Multi-scale, model-driven exploration of sub-generational gene expression in bacteria: individual consequences, population benefits

研究摘要/摘要 我们的目标是破译分子级别的事件或属性如何在 种群，以及这种异质性如何为整个种群带来不可用的优势 发送给个别成员。我们建议确定亚世代的基因表达--不仅仅是个体 基因，但也包括包含多个具有协调功能的基因的完整操纵子--创造混合 更适合对各种环境线索做出反应的人群。这一提议，深度融入了 计算模型和实验测量是我们在E。 这是今年早些时候在《科学》杂志上报道的。大肠杆菌模型预测了一些令人惊讶的 行为；最相关的是发现在大肠杆菌中明显的大多数基因以一种 每个细胞周期少于一次--这种现象我们称之为“亚代基因表达”。这样的表述 可能会对个别细菌产生负面影响，但对整个细菌种群有利。 因为细菌无法可靠地预测未来的情况，所以种群必须时刻做好准备。 但没有一种细菌能够表达所有需要做出反应的基因 在足够的水平上对任何环境产生影响。相反，我们的工作假设是种群是异质性的， 由单独的成员组成，每个成员都为少数可能的环境做好了准备。因此， 虽然没有单个细胞为所有环境做好准备，但作为一个整体，整个群体已经为大多数可能发生的情况做好了准备。 因此，群体由个体主导，通过亚世代的表现随机出现 表达的基因，他们最适合在任何给定的时刻生存。我们的团队结合了这两个领域的专业知识 全细胞和基于代理的模型，并一直致力于全细胞群体模拟，在 其中成百上千的细胞每个都运行大肠杆菌模型的实例化。我们的目标是：(1) 确认模型预测的基因是亚代表达的；(2)计算预测和 实验确定操纵子结构对功能相关基因亚代表达的影响 基因对；以及(3)通过计算预测和实验确定产生的表型异质性 通过在细胞群体中分离操纵子。我们的建议最具影响力和开创性的方面是 我们将揭示操纵子结构在原核基因调控中的基本新作用；我们将 产生前所未有的复杂性的扩展全细胞模型，以及高度创新的新 建模技术；最后，这项工作将是第一次利用新的多尺度仿真平台 它结合了全细胞模型和基于代理的模型，包括最令人兴奋的实验模型 展示全细胞和全菌落模型的主要潜力：预测大规模涌现 属性，以深入了解复杂的细胞行为。

Research Summary/Abstract Our goal is to decipher how a molecular-level event or property can create heterogeneous behavior within a population, and how this heterogeneity leads to advantages for the population as a whole that are not available to individual members. We propose to determine how sub-generational gene expression - not only of individual genes, but also of entire operons containing multiple genes with coordinated functions - creates mixed populations that are more fit to respond to various environmental cues. This proposal, which deeply integrates computational modeling and experimental measurement, arose out of our efforts in “whole-cell” modeling of E. coli, which were reported in Science earlier this year. The E. coli model has predicted a number of surprising behaviors; most relevant is the finding that a clear majority of the genes in E. coli are transcribed at a rate of less than once per cell cycle - a phenomenon we call “sub-generational gene expression”. Such expression can have negative consequences for individual bacteria, but benefits the bacterial population as a whole. Because bacteria are unable to reliably anticipate future conditions, the population must always be prepared for any environmental change - but no single bacterium is able to express all of the genes required to respond to any environment at sufficient levels. Instead, our working hypothesis is that the population is heterogeneous, comprised of individual members who are each prepared for a small number of possible environments. Thus, while no single cell is ready for all environments, as a whole the population is prepared for most eventualities. The colony is thus dominated by individuals, emerging stochastically via expression of sub-generationally expressed genes, who are the most fit to survive at any given moment. Our groups combine expertise in both whole-cell and agent-based models, and have been working towards whole-cell population simulations, in which hundreds or thousands of cells each run an instantiation of the E. coli model. Our Aims are to: (1) confirm that model-predicted genes are expressed sub-generationally; (2) computationally predict and experimentally determine the effect of operon structure on sub-generational expression of functionally related gene pairs; and (3) computationally predict and experimentally determine the phenotypic heterogeneity created by operon separation in cell populations. The most impactful and pioneering aspects of our proposal are that we will uncover a fundamental new role for operon structure in prokaryotic gene regulation; that we will produce an expanded whole-cell model of previously unseen complexity, as well as highly innovative new modeling technology; and finally, that this work will be the first to utilize a novel multi-scale simulation platform that combines whole-cell models with agent-based models, including the most exciting experimental demonstration of whole-cell and whole-colony modeling’s major potential: predicting large-scale emergent properties to generate insights into complex cellular behaviors.