Henriques João, Cragnell Carolina, Skepö Marie
Division of Theoretical Chemistry, Lund University , Post Office Box 124, S-221 00 Lund, Sweden.
J Chem Theory Comput. 2015 Jul 14;11(7):3420-31. doi: 10.1021/ct501178z. Epub 2015 Jun 30.
An increasing number of studies using molecular dynamics (MD) simulations of unfolded and intrinsically disordered proteins (IDPs) suggest that current force fields sample conformations that are overly collapsed. Here, we study the applicability of several state-of-the-art MD force fields, of the AMBER and GROMOS variety, for the simulation of Histatin 5, a short (24 residues) cationic salivary IDP with antimicrobial and antifungal properties. The quality of the simulations is assessed in three complementary analyses: (i) protein shape and size comparison with recent experimental small-angle X-ray scattering data; (ii) secondary structure prediction; (iii) energy landscape exploration and conformational class analysis. Our results show that, indeed, standard force fields sample conformations that are too compact, being systematically unable to reproduce experimental evidence such as the scattering function, the shape of the protein as compared with the Kratky plot, and intrapeptide distances obtained through the pair distance distribution function, p(r). The consistency of this deviation suggests that the problem is not mainly due to protein-protein or water-water interactions, whose parametrization varies the most between force fields and water models. In fact, as originally proposed in [ Best et al. J. Chem. Theory Comput. 2014, 10, 5113 - 5124.], balanced protein-water interactions may be the key to solving this problem. Our simulations using this approach produce results in very good agreement with experiment.
越来越多使用未折叠和内在无序蛋白质(IDP)的分子动力学(MD)模拟的研究表明,当前的力场对构象的采样过于紧凑。在这里,我们研究了几种最先进的MD力场(AMBER和GROMOS类型)对组蛋白5模拟的适用性,组蛋白5是一种具有抗菌和抗真菌特性的短(24个残基)阳离子唾液IDP。通过三种互补分析评估模拟的质量:(i)与最近的实验小角X射线散射数据进行蛋白质形状和大小比较;(ii)二级结构预测;(iii)能量景观探索和构象类分析。我们的结果表明,实际上,标准力场对过于紧凑的构象进行采样,系统地无法重现诸如散射函数、与Kratky图相比的蛋白质形状以及通过对距离分布函数p(r)获得的肽内距离等实验证据。这种偏差的一致性表明问题主要不是由于蛋白质-蛋白质或水-水相互作用,其参数化在力场和水模型之间变化最大。事实上,正如[Best等人,《化学理论与计算杂志》,2014年,10,5113 - 5124。]最初提出的那样,平衡的蛋白质-水相互作用可能是解决这个问题的关键。我们使用这种方法的模拟结果与实验非常吻合。