Deciphering dynamic signals in control of cell fate decisions

项目摘要 从长远来看，我们的目标是了解单细胞如何整合和处理信息， 决定是否扩散，分化或死亡。炎症因子参与许多正常的 并且患病细胞命运决定通过动态重组细胞内的蛋白质来启动信号。为 例如，配体结合的TNF受体在质膜附近瞬时组织大的蛋白质复合物， 这些蛋白质在细胞内以离散的点状结构可见，而其他蛋白质则在 细胞间室，如细胞质和细胞核。这是一个新兴的原则， 信号转导电路内的分子的变化提供时间代码（包括变化率，幅度， 持续时间或频率等），这对每个细胞对刺激的反应至关重要。鉴于有 即使在克隆细胞系中，细胞间的异质性也很大，固定时间点的静态测量不能 揭示动态信息加工的机制。我们假设相同的成分 信号通路在单个细胞中确定性地相互联系，即使存在大量的 细胞间的异质性。在这里，我们提出了活细胞荧光报告基因的多重表达， 和同一细胞中同一信号通路的下游成分，并将时变信号 活细胞显微镜数据。使用定量成像，微流体和计算 技术，我们将在每个实验条件下从100 - 1000个单细胞中提取时变数据， 在几种不同的细胞系中进行比较。我们还将比较不同的细胞反应， 共享信号传导模块并汇聚于NF-κB转录系统的炎性因子，如 TNF、LPS或IL-1等。使用丰富的单细胞数据集，我们将使用传递熵来测量相互作用。 在同一通路中时变信号特征之间的信息，并推断信号的机制 除了与细胞命运的相关性之外，还包括转导。来自活细胞实验的数据将被纳入 机械模型，以形式化我们对信息如何通过信令网络中继的理解 转录，并建议扰动测试预测的机制。我们预计， 精确的模型可能导致非直观的策略来操纵单个单元中的决策。通过详细 了解动态分子信号如何编码，处理和解码信息，我们有 潜在的理解深深植根于疾病的生物学问题，并利用这些知识， 设计影响细胞命运决定的疗法。

PROJECT SUMMARY In the long-term, our goal is to understand how single cells integrate and process information to make irreversible decisions such as whether to proliferate, differentiate or die. Inflammatory factors that participate in many normal and diseased cell fate decisions initiate signals by dynamically re-organizing proteins within the cell. For example, ligand-bound TNF receptors transiently organize large protein complexes near the plasma membrane, and these are visible within the cell as discrete punctate structures, whereas other proteins translocate between cellular compartments such as the cytoplasm and the nucleus. It is an emerging principle that dynamic properties of molecules within signal transduction circuits provide temporal codes (including rate of change, amplitude, duration or frequency among others) that are critical to each cell’s response to stimulus. Given that there is substantial cell-to-cell heterogeneity, even in clonal cell lines, static measurements at fixed time points cannot reveal the mechanisms of dynamic information processing. We hypothesize that components of the same signaling pathway are deterministically linked to one another in a single cell, even though there is substantial heterogeneity between cells. Here, we propose to multiplex expression of live-cell fluorescent reporters for up- and down-stream components of the same signaling pathway in the same cell, and correlate time-varying signals from live-cell microscopy data. Using a hybrid of quantitative imaging, microfluidics and computational techniques we will extract time-varying data from 100s-1000s of single cells in each experimental condition, and compare them across several different cell lines. We will also compare cellular responses across different inflammatory factors that share signaling modules and converge on the NF-κB transcriptional system, such as TNF, LPS or IL-1 among others. Using a rich single-cell dataset, we will use transfer entropy to measure mutual information between features of time-varying signals in the same pathway, and infer mechanisms of signal transduction in addition to correlations with cell fate. Data from live-cell experiments will be incorporated into mechanistic models to formalize our understanding of how information is relayed through the signaling network into transcription, and suggest perturbations to test predicted mechanisms. We anticipate that increasingly accurate models may lead to non-intuitive strategies to manipulate decisions in single cells. Through a detailed understanding of how dynamic molecular signals encode, process, and decode information, we have the potential to understand biological problems that are deeply rooted in disease, and use this knowledge to rationally design therapies that impact cell fate decisions.