Computational Modeling of Cell Migration in 3D Matrices

描述（由申请人提供）：我们的总体目标是构建和测试预测计算模型，以了解不同的细胞内信号如何整合以控制三维环境中细胞对线索的迁移反应。我们的技术方法包括创新组合的几个特点：{1}在相关生物大分子基质中对单个细胞迁移参数进行3D定量跟踪；{2}三维迁移过程中关键细胞内信号的定量生化测量；{3}通过受体特性的遗传操纵和关键分子开关的药理抑制来调节这些信号；{4}计算模拟了这些信号与相应迁移响应的定量关系；和{5}的非直觉模型预测的实验测试。我们关注表皮生长因子受体家族（ErbB）信号诱导的信号和反应。在肿瘤进展到侵袭和转移过程中，ErbB家族受体配体结合下游产生的信号强烈影响许多细胞类型的迁移反应行为，包括癌细胞。这种典型的“线索-信号-反应”系统在器官发生和组织再生过程中具有重要的生理作用，当异常时可导致肿瘤侵袭和传播的病理。因此，了解信号控制是如何定量地发挥作用的，对于基础科学和治疗应用都具有广泛的相关和有用的意义。虽然已经确定了ErbB信号网络中的许多单独的组成部分，但将这些不同的信号通路与迁移反应行为综合起来的定量模型现在才出现。在真正代表肿瘤侵袭和传播障碍的3D环境中，很少有关于信号控制迁移的基础工作。我们的工作密切结合了计算建模和专用的定量实验测量，在3D矩阵中ErbB家族诱导的细胞迁移和信号网络活动。我们的建模方法将不关注更普遍追求的“线索信号”方面（即从配体/受体线索产生的信号），而是关注严重不足的“信号响应”方面。虽然很明显，ErbB受体下游的多个信号通路在调节迁移中发挥重要作用，但尚不清楚多通路网络如何定量整合以产生观察到的表型行为。这个问题将使用称为决策树分析的统计建模框架来解决，该框架定义了一个控制层次结构，将信号的逻辑组合与所有线索条件下的迁移行为反应联系起来。我们的目标是通过三维矩阵应用决策树模型来预测ErbB受体信号传导对上皮细胞和癌细胞迁移的影响。公共卫生相关性：我们的目标是构建和测试预测计算模型，以了解细胞内信号如何整合以控制细胞在三维环境中对线索的迁移反应。我们关注表皮生长因子受体家族（ErbB）信号诱导的信号和反应，与组织再生和肿瘤侵袭有关。我们解决了ErbB家族受体激活下游的多途径信号网络如何定量整合以产生观察到的迁移行为的问题，使用被称为决策树分析的统计建模框架，该框架定义了一个控制层次，将信号的逻辑组合与所有线索条件下的迁移反应联系起来。

DESCRIPTION (provided by applicant): Our overall objective is to construct and test predictive computational models for how diverse intracellular signals integrate to govern cell migration responses to cues in three-dimensional environments. Our technical approach comprises several features in innovative combination: {1} 3D quantitative tracking of individual cell migration parameters within relevant biomacromolecular matrices; {2} quantitative biochemical measurements of key intracellular signals during this 3D migration; {3} modulation of these signals by genetic manipulation of receptor properties and by pharmacological inhibition of key molecular switches; {4} computational modeling of the quantitative relationship of these signals to the consequent migration response; and {5} experimental test of non-intuitive model predictions. We focus on signals and responses induced by the epidermal growth factor receptor family (ErbB) cues. Signals generated downstream of ErbB family receptor ligand binding strongly influence migration response behavior of many cell types, including carcinoma cells during tumor progression to invasion and metastasis. This paradigmatic "cue-signal- response" system is important physiologically during organogenesis and tissue regeneration, and when aberrant enables the pathology of tumor invasion and dissemination. Thus, understanding how the signaling control is exerted quantitatively should have broadly relevant and useful implications for both basic science and therapeutic applications. Although a multitude of individual components in the ErbB signaling network have been identified, quantitative models integratively relating these divergent signaling pathways to migration response behavior are only now emerging. Very little fundamental work on signaling governing migration has been performed in 3D environments that truly represent the barriers to tumor invasion and dissemination. Our work intimately integrates computational modeling with dedicated quantitative experimental measurement of ErbB family-induced cell migration and signaling network activity within 3D matrices. Our modeling approach will focus not on the more commonly-pursued "cue-signal" facet (that of signal generation from ligand/receptor cues, but instead the sorely under-addressed "signal-response" facet. Although it is clear that multiple signaling pathways downstream of ErbB receptors can play significant roles in regulating migration, what is not understood is how multi-pathway networks quantitatively integrate to yield the observed phenotypic behavior. This question will be addressed using the statistical modeling framework known as Decision Tree analysis, which defines a control hierarchy relating logical combinations of signals to the migration behavioral response across all cue conditions. Our goal for the proposed grant is to apply decision tree modeling to prediction of effects of ErbB receptor signaling on epithelial and carcinoma cell migration through three-dimensional matrices. PUBLIC HEALTH RELEVANCE: Our goal is to construct and test predictive computational models for how intracellular signals integrate to govern cell migration responses to cues in three-dimensional environments. We focus on signals and responses induced by the epidermal growth factor receptor family (ErbB) cues, with relevance to tissue regeneration and tumor invasiveness. We address the question of how multi-pathway signaling networks downstream of ErbB family receptor activation quantitatively integrate to yield observed migration behavior, using the statistical modeling framework known as Decision Tree analysis which defines a control hierarchy relating logical combinations of signals to the migration response across all cue conditions.