Center for Biological Physics, Arizona State University, Tempe, Arizona, USA.
Proteins. 2009 Dec;77(4):796-811. doi: 10.1002/prot.22492.
PDZ domains (PDZs), the most common interaction domain proteins, play critical roles in many cellular processes. PDZs perform their job by binding specific protein partners. However, they are very promiscuous, binding to more than one protein, yet selective at the same time. We examined the binding related dynamics of various PDZs to have insight about their specificity and promiscuity. We used full atomic normal mode analysis and a modified coarse-grained elastic network model to compute the binding related dynamics. In the latter model, we introduced specificity for each single parameter constant and included the solvation effect implicitly. The modified model, referred to as specific-Gaussian Network Model (s-GNM), highlights some interesting differences in the conformational changes of PDZs upon binding to Class I or Class II type peptides. By clustering the residue fluctuation profiles of PDZs, we have shown: (i) binding selectivities can be discriminated from their dynamics, and (ii) the dynamics of different structural regions play critical roles for Class I and Class II specificity. s-GNM is further tested on a dual-specific PDZ which showed only Class I specificity when a point mutation exists on the betaA-betaB loop. We observe that the binding dynamics change consistently in the mutated and wild type structures. In addition, we found that the binding induced fluctuation profiles can be used to discriminate the binding selectivity of homolog structures. These results indicate that s-GNM can be a powerful method to study the changes in binding selectivities for mutant or homolog PDZs.
PDZ 结构域(PDZs)是最常见的相互作用域蛋白,在许多细胞过程中发挥着关键作用。PDZ 通过结合特定的蛋白质伴侣来发挥其作用。然而,它们非常混杂,能与多个蛋白质结合,但同时又具有选择性。我们研究了各种 PDZ 与蛋白质结合的相关动力学,以了解其特异性和混杂性。我们使用全原子正常模式分析和改进的粗粒弹性网络模型来计算与结合相关的动力学。在后者的模型中,我们为每个单一参数常数引入了特异性,并隐含地包含了溶剂化效应。这个改进后的模型,称为特定高斯网络模型(s-GNM),突出了 PDZ 在与 I 类或 II 类肽结合时构象变化的一些有趣差异。通过对 PDZ 残基波动分布进行聚类,我们表明:(i)可以从动力学上区分结合选择性,(ii)不同结构区域的动力学对 I 类和 II 类特异性起着关键作用。s-GNM 进一步应用于一个双特异性 PDZ 上,当 βA-βB 环上存在点突变时,它表现出仅对 I 类的特异性。我们观察到,在突变和野生型结构中,结合动力学都发生了一致的变化。此外,我们发现结合诱导的波动分布可用于区分同源结构的结合选择性。这些结果表明,s-GNM 可以成为研究突变或同源 PDZ 结合选择性变化的一种有力方法。