Memory in Bacterial Responses to Fluctuating Stress

项目总结： 细菌细胞有一系列的反应，可以用来在不同类型的环境中生存 环境压力的影响。碳源的变化导致细胞启动特定的代谢 基因，当这些来源耗尽时，这些基因后来被抑制。接触抗生素可能会 触发从细胞中移除抗生素的分子泵的表达，或者 产生专门降解它的酶。在不断变化的环境中， 开启和关闭基因的过程可能代价高昂，特别是在接触抗生素的情况下 如果反应基因关闭，细胞就会迅速死亡。我们的研究表明，细菌结合了它们的 具有分子记忆机制的反应，使细胞避免频繁的 波动环境中的基因调控。 该项目将确定分子记忆是一种有益策略的条件， 通过使用合成生物学、微流体、显微镜和建模的组合，我们 将在实验中扰乱和衡量记忆的成本和好处。我们将建造 具有不同记忆水平的细菌菌株，并进行竞争实验 确定蜂窝内存配置文件如何根据外部环境进行调整。 拟议的实验利用了一种定制的微流控‘化学通量’系统，我们 已开发，其中细菌种群在单层中生长，以单细胞分辨率跟踪 在显微镜下，生长介质可以在时间上任意波动。使用 化学通量和我们的图像分析算法，我们能够量化数万个细胞 数百代人，从而测量波动中的种群动态 环境以一种以前无法达到的分辨率。 我们将使用两个不同级别的建模，包括粗粒度方法 时间标度和速率常数是主要参数，目标是预测最优的 针对给定响应和波动环境的内存量；以及详细的单个单元 随机模型，其中细胞的伸长和分裂过程被精确地量化 并在不断变化的条件下进行建模。这两种表述将针对不同的 记忆的方方面面，并允许我们从详细的实验室测量到 细菌生存的一般生物学原理。

Project Summary: Bacterial cells have a repertoire of responses that can be used to survive under different types of environmental stress. Changes in carbon sources cause cells to turn on specific metabolic genes, which are later repressed when those sources are depleted. Antibiotic exposure can trigger the expression of molecular pumps that remove the antibiotic from the cell, or the production of enzymes that specifically degrade it. In a continually fluctuating environment, the process of turning genes on and off can be costly, especially under antibiotic exposures when cells are rapidly killed if the response gene is off. Our work shows that bacteria combine their responses with molecular memory mechanisms that allow cells to avoid the costs of frequent gene regulation in a fluctuating environment. The project will determine the conditions under which molecular memory is a beneficial strategy, and by using a combination of synthetic biology, microfluidics, microscopy, and modeling, we will experimentally perturb and measure the costs and benefits of memory. We will construct bacterial strains with a range of memory levels, and perform competition experiments to determine how cellular memory profiles are tuned to the external environment. The proposed experiments make use of a custom-built microfluidic ‘chemoflux’ system that we developed, in which bacterial populations grow in monolayers, tracked at single cell resolution under the microscope, while the growth media can be arbitrarily fluctuated in time. Using the chemoflux and our image analysis algorithms, we are able to quantify tens of thousands of cells over hundreds of generations, and thereby measure population dynamics in fluctuating environments at a resolution that was previously unattainable. We will use two different levels of modeling, including a coarse-grained approach in which timescales and rate constants are the main parameters, and the goal is to predict the optimal amount of memory for a given response and fluctuating environment; and a detailed, single cell stochastic model, in which the process of cellular elongation and division is precisely quantified and modeled under changing conditions. These two representations will address different aspects of memory, and allow us to bridge from detailed laboratory measurements to the general biological principles that underlie bacterial survival.