School of Physics and Electronics, Shandong Normal University, Jinan, Shandong250014, China.
J Phys Chem B. 2023 Feb 16;127(6):1312-1324. doi: 10.1021/acs.jpcb.2c07811. Epub 2023 Feb 3.
Protein-DNA binding mechanisms in a complex manner are essential for understanding many biological processes. Over the past decades, numerous experiments and calculations have analyzed the specificity of protein-DNA binding. However, the accuracy of binding free energy prediction for multi-base DNA systems still needs to be improved. Fis is a DNA-binding protein that regulates various transcription and recombination reactions. In the present work, we tested several methods of predict binding free energy based on this system to find a favorable prediction scheme and explore the binding mechanism of Fis protein and DNA. Two solvent models (explicit and implicit solvent models) were chosen for the dynamics process, and the predicted binding free energy was more accurate under the explicit solvent model. When different Poisson-Boltzmann/Generalized Born (PB/GB) models were tested for DNA force fields (BSC1 and OL15), it was found that the binding free energy predicted by the selected OL15 force field performed better and the correlation between predicted and experimental values was improved with the increasing interior dielectric constant (Dk). Finally, using Dk = 8, the GB model combined with interaction entropy (IE), which was calculated for entropic contribution (GB_IE_8), was screened out for the binding free energy prediction and analysis of the Fis-DNA system, and the validity of the method was further verified by testing the Cren7-DNA system. By performing conformational analysis of the minor groove, it was found that mutation of the DNA central sequence A/T to C/G and deletion of the guanine 2-amino group would change the minor groove width and thus affect the formation of the major groove, altering the interaction and atomic contact between the protein and the major groove, thus changing the binding affinity of Fis and DNA. Hopefully, the series of tests in this work can shed some light on the related studies of protein and DNA systems.
蛋白质与 DNA 之间的结合机制对于理解许多生物学过程至关重要。在过去的几十年中,已经有大量的实验和计算分析了蛋白质与 DNA 结合的特异性。然而,对于多碱基 DNA 系统的结合自由能预测的准确性仍有待提高。Fis 是一种调节多种转录和重组反应的 DNA 结合蛋白。在本工作中,我们针对该体系测试了几种预测结合自由能的方法,以找到有利的预测方案,并探索 Fis 蛋白与 DNA 的结合机制。选择了两种溶剂模型(显式溶剂模型和隐式溶剂模型)进行动力学过程,显式溶剂模型预测的结合自由能更为准确。当测试不同的泊松-玻尔兹曼/广义 Born(PB/GB)模型用于 DNA 力场(BSC1 和 OL15)时,发现选择的 OL15 力场预测的结合自由能更好,并且随着内部介电常数(Dk)的增加,预测值与实验值之间的相关性得到了提高。最后,使用 Dk=8,选择了结合相互作用熵(IE)的 GB 模型(GB_IE_8)进行 Fis-DNA 体系的结合自由能预测和分析,并且通过测试 Cren7-DNA 体系进一步验证了该方法的有效性。通过对小沟构象进行分析,发现 DNA 中心序列 A/T 突变为 C/G 以及鸟嘌呤 2-氨基的缺失会改变小沟宽度,从而影响大沟的形成,改变蛋白质与大沟的相互作用和原子接触,进而改变 Fis 与 DNA 的结合亲和力。希望本工作中的一系列测试能够为蛋白质与 DNA 体系的相关研究提供一些启示。