Measuring and Predicting PDZ Domain-Peptide Interactions on a Genome-wide Scale

描述（由申请人提供）：在众多的蛋白质相互作用结构域中，PDZ结构域是最常见的结构域之一。它们通常与其他相互作用模块结合，在建立和维持细胞极性、指导蛋白质运输和协调突触信号传导方面发挥重要作用。在PDZ结构域敲除小鼠中观察到的严重的神经元和发育表型以及它们在人类先天性疾病中的含义强调了它们的重要性。PDZ结构域功能的巨大多样性表现在它们的丰度上：在小鼠基因组中至少编码246个PDZ结构域。为了理解它们各自的作用，首先有必要以广泛和生物学相关的方式定义它们的识别特性，并构建准确预测它们在哺乳动物蛋白质组中的相互作用的模型。本文提出了一种结合实验和计算的方法来检测、测量、预测和干扰PDZ结构域介导的蛋白质-蛋白质相互作用，并研究它们的生物学相关性。在Aim 1中，提出了一种使用蛋白质微阵列和高通量荧光极化来训练统计模型的策略，该模型可以在蛋白质组范围内预测PDZ结构域-肽的相互作用。实验最初将集中在小鼠PDZ域，但随后将扩展到包括人类域。这些工作所产生的实验数据和预测模型应该对研究含PDZ结构域的蛋白质的生物界有价值。在目标2和目标3中，将更详细地研究新型相互作用的生物学相关性。初步实验发现了PDZ结构域和β -连环蛋白之间的几种新的相互作用。在Aim 2中，提出了实验来验证¿-catenin在形成或维持紧密连接中起功能作用的假设。初步实验还发现了PDZ结构域与多药耐药蛋白之间的许多新的相互作用。在Aim 3中，提出了实验来确定含有PDZ结构域的蛋白如何调节这些药物转运体的细胞表面表达、亚细胞定位和功能。了解多重耐药蛋白是如何被调控的，可能会提供新的方法来对抗常见的耐药机制。最后，在Aim 4中，提出了将预测建模与蛋白质微阵列技术相结合的计划，以设计具有定制选择性的肽和PDZ结构域。作为概念验证，选择性肽将被设计用于与β -连环蛋白或多药耐药蛋白相互作用的PDZ结构域，PDZK1的无功能突变体将被制备来分析其在调节血液胆固醇水平中的作用。总之，我们希望我们的高通量实验所强调的新型相互作用和由这些数据得出的预测模型将被生物界广泛采用，作为研究PDZ结构域介导的蛋白质-蛋白质相互作用的强大资源。

DESCRIPTION (provided by applicant): Among the myriad protein interaction domains, PDZ domains are one of the most frequently encountered. They are often found in combination with other interaction modules and play an important role in establishing and maintaining cell polarity, in directing protein trafficking, and in coordinating synaptic signaling. Their importance is underscored by the severe neuronal and developmental phenotypes observed in PDZ domain knockout mice and by their implication in human congenital diseases. The enormous diversity of PDZ domain function is manifest in their abundance: there are at least 246 PDZ domains encoded in the mouse genome. To understand their individual roles, it is necessary first to define their recognition properties in a broad and biologically relevant fashion and to construct models that accurately predict their interactions across the mammalian proteome. Here, a combined experimental and computational approach is presented to detect, measure, predict, and perturb PDZ domain-mediated protein-protein interactions and to investigate their biological relevance. In Aim 1, a strategy is presented that uses protein microarrays and high-throughput fluorescence polarization to train a statistical model that predicts PDZ domain-peptide interactions on a proteome-wide scale. Experiments will initially focus on mouse PDZ domains, but will subsequently be extended to include human domains as well. Both the experimental data and predictive model arising from these efforts should prove valuable to the biological community that studies PDZ domain-containing proteins. In Aims 2 and 3, the biological relevance of novel interactions will be investigated in greater detail. Preliminary experiments have uncovered several novel interactions between PDZ domains and beta-catenin. In Aim 2, experiments are proposed to test the hypothesis that ¿-catenin plays a functional role in forming or maintaining tight junctions. Preliminary experiments have also uncovered numerous novel interactions between PDZ domains and multidrug resistance proteins. In Aim 3, experiments are proposed to determine how PDZ domain-containing proteins modulate the cell-surface expression, subcellular localization, and function of these drug transporters. Understanding how multidrug resistance proteins are regulated may provide new ways to counteract common mechanisms of drug resistance. Finally, in Aim 4 a plan is presented to combine predictive modeling with protein microarray technology to design peptides and PDZ domains with tailored selectivities. As proof-of-concept, selective peptides will be designed to target PDZ domains that interact with beta-catenin or multidrug resistance proteins, and nonfunctional mutants of PDZK1 will be prepared to dissect its role in regulating blood cholesterol levels. Overall, it is our hope that both the novel interactions highlighted by our high-throughput experiments and the predictive models resulting from these data will be broadly adopted by the biological community as a powerful resource to study PDZ domain-mediated protein-protein interactions.