Multiscale kinetic analysis of Cu/Zn efflux in Halobacterium sp. NRC-1

描述（由申请人提供）：所有细胞类型都能忍受环境的变化，因此，进化出了对复杂刺激做出反应的非凡方式。重要的是，独立的细胞功能可以直接或间接地结合起来处理这种环境扰动。准确预测细胞行为需要知道这些功能是如何通过基因调控来协调的。此外，了解动态相声如何影响这些反应将为了解不良疾病和药物副作用的原因和适当治疗提供重要见解。根据兴趣的大小，这样做的方法有很大的不同。在只有少数基因的小系统中，模型通常是具有很高动力学精度的正式化学方程。另外，由于参数数据有限、高维和非线性，统计方法对于预测更大、更复杂的全局网络更可行。因此，定量的动态预测在大尺度上通常是不可用的，在底层成分的微小变化如何影响全球动态方面造成了不确定性。我将使用一种多尺度方法来解决这个问题，该方法将一个小调控子回路的动态模型集成到一个统计推断、预测的全球基因调控模型中。我将研究H. salinarum NRC-1的铜（Cu）外排途径。基于其调控结构，该子电路在Cu特异性外排泵yvgX的表达中具有潜在的双稳态。然而，这种双稳定性并不容易被观察到。值得注意的是，H. salinarum NRC-1也使用多特异性外排泵zntA来输出Cu，这可能解释了双稳定性的缺乏。我假设zntA调控使用许多与yvgX相同的控制点（转录因子，金属伴侣复合物等），这种调控共享提供了强大/快速的表达动态，但以调控串扰为代价。具体来说，来自zntA的额外反馈迫使瞬态yvgX在高Cu浓度下表达为高“开”状态，在低/稳态量下表达为低“关”状态，避免了对细胞有害的双稳态。然而，zntA对其他金属离子，即锌（Zn）的敏感性产生的串扰破坏了Cu的稳态，迫使系统更接近双稳态。此外，通过结合小型和大型建模技术，我应该能够预测系统范围内的基因表达扰动作为这种串扰的结果。

DESCRIPTION (provided by applicant): All cell types endure changes to their environments, and thus, have evolved remarkable ways to react to complex stimuli. Importantly, separate cell functions may combine directly or indirectly to handle such environmental perturbations. Accurate prediction of cellular behavior requires knowing how these functions are coordinated by gene regulation. Moreover, understanding how dynamic cross-talk affects these responses will provide important insights into the causes and appropriate treatments for adverse disease and drug side effects. The ways to do so vary greatly based on scales of interest. In small systems of few genes, models are typically formal chemical equations that offer great dynamical accuracy. Alternately, statistical approaches are more feasible to predict larger, more complex, global networks due to limited parameter data, high dimensionality, and nonlinearity. Hence, quantitative dynamic predictions are typically unavailable at the large scale, creating uncertainty in how small changes to underlying components affect global dynamics. I will address this issue using a multiscale approach that integrates a dynamic model of a small regulatory subcircuit into a statistically inferred, predictive, model of global gene regulation. I will study the copper (Cu) efflux pathway in H. salinarum NRC-1. Based on its regulatory architecture, the subcircuit is potentially bistable in expression of the Cu specific efflux pump yvgX. Yet, such bistability is not readily observed. Notably, H. salinarum NRC-1 also uses a multispecific efflux pump, zntA, to export Cu, which may explain the lack of bistability. I hypothesize that zntA regulation uses many of the same control-points (transcription factors, metallochaperone complexes, etc.) as yvgX and such regulatory sharing offers robust/rapid expression dynamics, but at the cost of regulatory cross-talk. Specifically, additional feedback from zntA forces transient yvgX expression to a high "ON" state in high Cu concentrations and a low "OFF" state in low/homeostatic amounts, avoiding bistability which is detrimental to the cell. However, cross-talk from zntA's sensitivity to other metal ions, namely zinc (Zn), disrupts Cu homeostasis, forcing the system closer to bistability. Moreover, by combining small and large scale modeling techniques, I should be able to predict system wide gene expression perturbations as a result of this cross-talk.