Sequence-dependent conformational energy of DNA derived from molecular dynamics simulations:: Toward understanding the indirect readout mechanism in protein-DNA recognition
Sequence-dependent conformational energy of DNA derived from molecular dynamics simulations:: Toward understanding the indirect readout mechanism in protein-DNA recognition
复制标题
DOI：
10.1021/ja053241l
复制
发表时间：
2005-11-23
影响因子：
15
通讯作者：
Sarai, A
中科院分区：
文献类型：
--
作者：
Araúzo-Bravo, MJ;Fujii, S;Sarai, A
DNA构象的序列依赖性在其被蛋白质和配体识别中起着至关重要的作用。为了阐明序列和构象之间的关系，有必要量化DNA的构象能和特异性。在这里，我们做了一个系统的分析十二聚体DNA结构，包括所有136个独特的四核苷酸序列的中心，通过分子动力学模拟。采用一个简化的构象模型，用六个参数来描述相邻碱基对的几何结构和沿这些坐标的调和势沿着，我们从模拟的许多轨迹中估计了中心碱基对台阶的平衡构象参数和平均力的调和势。这使我们能够估计任何给定DNA序列和结构的构象能和特异性。我们测试了我们的方法，通过使用序列结构线程来估计构象能和Z分数作为许多B-DNA和A-DNA晶体结构的特异性的量度。两种结构的平均Z分数均为负值，表明模拟的平均力的潜力能够预测晶体结构的序列特异性，并且它可用于研究两种类型DNA的序列特异性。我们还估计了构象能和Z分数在DNA中的位置分布，并表明它们具有很强的位置依赖性。这种分析使我们能够识别出导致特异性的特定构象。所呈现的结果将提供一个深入了解DNA序列识别的蛋白质和配体的机制。
Sequence dependence of DNA conformation plays a crucial role in its recognition by proteins and ligands. To clarify the relationship between sequence and conformation, it is necessary to quantify the conformational energy and specificity of DNA. Here, we make a systematic analysis of dodecamer DNA structures including all the 136 unique tetranucleotide sequences at the center by molecular dynamics simulations. Using a simplified conformational model with six parameters to describe the geometry of adjacent base pairs and harmonic potentials along these coordinates, we estimated the equilibrium conformational parameters and the harmonic potentials of mean force for the central base-pair steps from many trajectories of the simulations. This enabled us to estimate the conformational energy and the specificity for any given DNA sequence and structure. We tested our method by using sequence-structure threading to estimate the conformational energy and the Z-score as a measure of specificity for many B-DNA and A-DNA crystal structures. The average Z-scores were negative for both kinds of structures, indicating that the potential of mean force from the simulation is capable of predicting sequence specificity for the crystal structures and that it may be used to study the sequence specificity of both types of DNA. We also estimated the positional distribution of conformational energy and Z-score within DNA and showed that they are strongly position dependent. This analysis enabled us to identify particular conformations responsible for the specificity. The presented results will provide an insight into the mechanisms of DNA sequence recognition by proteins and ligands.