Of Bits and Bugs - On the Use of Bioinformatics and a Bacterial Crystal Structure to Solve a Eukaryotic Repeat-Protein Structure
Of Bits and Bugs - On the Use of Bioinformatics and a Bacterial Crystal Structure to Solve a Eukaryotic Repeat-Protein Structure
复制标题
DOI：
10.1371/journal.pone.0013402
复制
发表时间：
2010-10-14
期刊：
影响因子：
3.7
通讯作者：
Niessing, Dierk
中科院分区：
文献类型：
--
作者：
Graebsch, Almut;Roche, Stephane;Niessing, Dierk
Pur-alpha是一种参与细胞周期控制、转录和神经元功能的核酸结合蛋白。最初，无法预测Pur-alpha的三维结构。然而，最近，我们解决了来自果蝇Drosophila melanogaster的Pur-α的X射线结构，并表明它包含所谓的PUR结构域。在这里，我们解释了我们如何利用生物信息学工具结合细菌同系物的X射线结构测定，以获得衍射晶体和高分辨率的果蝇Pur-alpha结构。首先，我们使用灵敏的远程同源性检测的方法，找到三个重复的区域在Pur-alpha。我们意识到，我们缺乏对这些重复如何相互作用形成球状结构域的理解，这是结晶和结构测定的一个主要问题。利用我们关于重复基序的信息，我们随后鉴定了一种仅含有一个重复的远距离细菌同源物。我们确定了细菌的晶体结构，发现其中两个重复序列相互作用形成一个球状结构域。基于这种细菌结构，我们计算了真核蛋白质的计算模型。该模型使我们能够设计一个可结晶的片段，并确定果蝇Pur-alpha的结构。成功的关键是细菌蛋白的单个重复序列自组装成球状结构域，这告诉我们在真核蛋白结晶试验中要包括的重复序列的数量和边界。这项研究表明，一个遥远的原核蛋白质的结构域排列简单，可以指导真核结晶构建体的设计。由于许多真核蛋白质含有多个重复序列或重复结构域，因此这种方法可能对一系列蛋白质的结构研究具有指导意义。
Pur-alpha is a nucleic acid-binding protein involved in cell cycle control, transcription, and neuronal function. Initially no prediction of the three-dimensional structure of Pur-alpha was possible. However, recently we solved the X-ray structure of Pur-alpha from the fruitfly Drosophila melanogaster and showed that it contains a so-called PUR domain. Here we explain how we exploited bioinformatics tools in combination with X-ray structure determination of a bacterial homolog to obtain diffracting crystals and the high-resolution structure of Drosophila Pur-alpha. First, we used sensitive methods for remote-homology detection to find three repetitive regions in Pur-alpha. We realized that our lack of understanding how these repeats interact to form a globular domain was a major problem for crystallization and structure determination. With our information on the repeat motifs we then identified a distant bacterial homolog that contains only one repeat. We determined the bacterial crystal structure and found that two of the repeats interact to form a globular domain. Based on this bacterial structure, we calculated a computational model of the eukaryotic protein. The model allowed us to design a crystallizable fragment and to determine the structure of Drosophila Pur-alpha. Key for success was the fact that single repeats of the bacterial protein self-assembled into a globular domain, instructing us on the number and boundaries of repeats to be included for crystallization trials with the eukaryotic protein. This study demonstrates that the simpler structural domain arrangement of a distant prokaryotic protein can guide the design of eukaryotic crystallization constructs. Since many eukaryotic proteins contain multiple repeats or repeating domains, this approach might be instructive for structural studies of a range of proteins.