Ensemble approach for NMR structure refinement against 1H paramagnetic relaxation enhancement data arising from a flexible paramagnetic group attached to a macromolecule
Ensemble approach for NMR structure refinement against 1H paramagnetic relaxation enhancement data arising from a flexible paramagnetic group attached to a macromolecule
复制标题
DOI：
10.1021/ja031580d
复制
发表时间：
2004-05-12
影响因子：
15
通讯作者：
Clore, GM
中科院分区：
文献类型：
--
作者：
Iwahara, J;Schwieters, CD;Clore, GM
H-1原子核的顺磁驰豫增强(Pre)测量通过提供独特的长程(10-35埃)距离信息，有可能在大分子的核磁共振结构确定中发挥重要作用。最近在蛋白质和核酸上的特定位置共价连接顺磁性基团的方法学进展使PRE能够应用于各种生物大分子。然而，由于人工引入的顺磁性基团暴露在溶剂中，并通过几个可自由旋转的键与大分子相连，它们本质上是灵活的。这使得传统的使用单点表示的H-1-PRE的反计算不准确，从而严重限制了H-1-PRE作为结构精细化工具的用途。为了绕过这些限制，我们开发了一个理论框架和计算策略，用来准确地反算大分子上柔性顺磁性基团产生的H-1-PreS。在该方案中，基于模拟退火法计算中顺磁群的多结构表示，采用改进的所罗门-Bloembergen方程计算H-1-Pre，其中引入了“无模型”形式。使用几个在DNA中三个不同位置含有DT-EDTA-Mn2+的SRY/DNA络合物，研究了H-1-Pre的系综计算方法，允许从2D通键关联实验中以直接的方式获得一个包含435个实验主链和侧链H-1-PreS的大型数据集。采用完全交叉验证的计算表明，系综表示提供了一种在结构精化中准确利用来自柔性顺磁性基团的主链和侧链H-1-Pre数据的方法。H-1-Pre精化的结果与以前获得的核磁共振约束相结合，表明可以很容易地在精度方面获得显著的提高。这在大分子复合体的情况下尤其重要，其中从核Overhauser增强数据得出的分子间翻译限制可能是有限的。
Paramagnetic relaxation enhancement (PRE) measurements on H-1 nuclei have the potential to play an important role in NMR structure determination of macromolecules by providing unique long-range (10-35 Angstrom) distance information. Recent methodological advances for covalently attaching paramagnetic groups at specific sites on both proteins and nucleic acids have permitted the application of the PRE to various biological macromolecules. However, because artificially introduced paramagnetic groups are exposed to solvent and linked to the macromolecule by several freely rotatable bonds, they are intrinsically flexible. This renders conventional back-calculation of the H-1-PRE using a single-point representation inaccurate, thereby severely limiting the utility of the H-1-PRE as a tool for structure refinement. To circumvent these limitations, we have developed a theoretical framework and computational strategy with which to accurately back-calculate H-1-PREs arising from flexible paramagnetic groups attached to macromolecules. In this scheme, the H-1-PRE is calculated using a modified Solomon-Bloembergen equation incorporating a "model-free" formalism, based on a multiple-structure representation of the paramagnetic group in simulated annealing calculations. The ensemble approach for H-1-PRE back-calculation was examined using several SRY/DNA complexes incorporating dT-EDTA-Mn2+ at three distinct sites in the DNA, permitting a large data set comprising 435 experimental backbone and side-chain H-1-PREs to be obtained in a straightforward manner from 2D through-bond correlation experiments. Calculations employing complete cross-validation demonstrate that the ensemble representation provides a means to accurately utilize backbone and side-chain H-1-PRE data arising from a flexible paramagnetic group in structure refinement. The results of H-1-PRE based refinement, in conjunction with previously obtained NMR restraints, indicate that significant gains in accuracy can be readily obtained. This is particularly significant in the case of macromolecular complexes where intermolecular translational restraints derived from nuclear Overhauser enhancement data may be limited.