Solid-state NMR spectroscopy of a paramagnetic protein:: Assignment and study of human dimeric oxidized CuII-ZnII superoxide dismutase (SOD)
Solid-state NMR spectroscopy of a paramagnetic protein:: Assignment and study of human dimeric oxidized CuII-ZnII superoxide dismutase (SOD)
复制标题
DOI：
10.1002/anie.200603093
复制
发表时间：
2007-01-01
影响因子：
16.6
通讯作者：
Emsley, Lyndon
中科院分区：
文献类型：
--
作者：
Pintacuda, Guido;Giraud, Nicolas;Emsley, Lyndon
顺磁系统的溶液核磁共振研究已经很发达顺磁中心源于许多过渡金属离子的固有特征——不成对电子。顺磁中心以几种方式改变核磁共振谱的外观，最明显的是改变化学位移和增加弛豫速率。一方面，这些变化构成了对这些系统中金属电子结构的独特而直接的探测另一方面，由于顺磁效应以一种明确的方式依赖于分子的结构，它们在确定分子几何形状时提供了各种结构限制因此，顺磁现象提供了有关电子结构、蛋白质-蛋白质或蛋白质-核酸对接、配体结合、溶剂定位和多结构域蛋白质灵活性的信息。然而，顺磁分子的核磁共振研究经常受到同样的超细效应的阻碍，主要是由于顺磁增强的核弛豫，[1]破坏了核磁共振实验的获得。近年来，固体中的魔角旋转（MAS） 13C核磁共振波谱技术取得了重大进展，从而为各种分配策略和大分子的三维结构确定铺平了道路然而，晶体中顺磁相互作用的较高复杂性至今限制了顺磁分子的核磁共振研究，[5]和顺磁蛋白几乎没有被固态核磁共振研究过在这里，我们提出了一个均匀标记的蛋白质含有顺磁性金属中心的第一个详细的研究。我们记录了核磁共振相关谱，并在固态下得到了完整的分配。与解的情况相反，我们注意到相关的观测受顺磁弛豫的影响最小，并且我们观察到大部分的贡献来自靠近金属中心的原子核。作为基准，我们研究了人体铜锌超氧化物歧化酶（SOD）的氧化形式。SOD是一种二聚体酶，分子量约为32kDa，含有两个相同的对称单元，共153个残基过去对SOD进行了深入的研究，获得了该酶和几种突变体的几种x射线[8]和核磁共振溶液结构[9]。然而，由于所谓的II型铜中心的电子T1弛豫时间较长，传统的溶液核磁共振研究几乎无法获得氧化形式，距离金属离子小于12的质子的1H核磁共振共振扩大到无法检测由于顺磁展宽取决于观察到的原子核的回旋磁比的平方，13C或15N自旋接近金属离子比1H自旋更容易被核磁共振探测到例如，在SOD的单体类似物中，通过使用专为直接获取13C而设计的新仪器和实验装置，溶液状态核磁共振实验检测并分配了与金属离子接近5的13C自旋特别是，由于纵向弛豫受顺磁中心的影响较小，基于纵向磁化转移的序列（例如，13C-13C NOESY）已被证明在研究靠近金属中心的原子的共振方面特别有效。[10,11]固态核磁共振特别适合后一种方法，因为碳-13的获取是常规的，并且基于偶极相互作用的转移方案可以在短延迟内实现定向、高效的同核（13C-13C）和异核（1H-13C, 1H-15N, 13C-15N）转移。
Solution NMR studies of paramagnetic systems are welldeveloped.[1] Paramagnetic centers originate from unpaired electrons that are intrinsic features of many transition-metal ions. A paramagnetic center changes the appearance of the NMR spectrum in several ways, most obviously by altering chemical shifts and increasing relaxation rates. On one side, these changes constitute a unique, direct probe of the electronic structure of the metals in these systems.[2] On the other hand, as the paramagnetic effects depend in a welldefined manner on the structure of the molecule, they provide a variety of structural restraints in the determination of the molecular geometry.[3] Paramagnetic phenomena thus provide information on electronic structure, protein–protein or protein–nucleic acid docking, ligand binding, solvent mapping, and the flexibility of multidomain proteins. However, NMR studies of paramagnetic molecules are often hindered by the very same large hyperfine effects, and mostly by the paramagnetically enhanced nuclear relaxation,[1] which undermines the acquisition of the NMR experiments. Magic angle spinning (MAS) 13C NMR spectroscopy in the solid state has recently made significant advances, thereby paving the way to a variety of assignment strategies and 3D structure determinations of macromolecules.[4] Nevertheless, the higher complexity of the paramagnetic interactions in crystals has so far limited the NMR study of paramagnetic molecules,[5] and paramagnetic proteins have barely been studied by solid-state NMR.[6] Herein, we present the first detailed investigation of a uniformly labeled protein containing a paramagnetic metal center. We have recorded NMR correlation spectra and obtained the complete assignment in the solid state. Contrary to the solution case, we notably find that the observation of correlations is minimally affected by paramagnetic relaxation, and we observe most of the contributions from nuclei that are close to the metal center. As a benchmark, we studied the oxidized form of human copper–zinc superoxide dismutase (SOD). SOD is a dimeric enzyme of about 32kDa containing two identical and symmetric units of 153 residues.[7] SOD has been thoroughly studied in the past, with several X-ray [8] and NMR solution structures [9] available of the enzyme and of several mutants. Owing to the long electronic T1 relaxation times of the socalled type II copper center, however, the oxidized form is virtually inaccessible to traditional solution NMR studies, with 1H NMR resonances of protons that are closer than 12 from the metal ion broadened beyond detection.[10] Since the paramagnetic broadening depends on the square of the gyromagnetic ratio of the observed nucleus, 13C or 15N spins that are close to the metal ion are easier to detect by NMR than 1H spins.[11] For example, in the case of a monomeric analogue of SOD, 13C spins that are as close as 5 to the metal ion have been detected and assigned by solution-state NMR experiments by using a new instrumental and experimental setup tailored for 13C direct acquisition.[12] In particular, as longitudinal relaxation is less influenced by the paramagnetic center, sequences based on longitudinal magnetization transfers (for example, 13C–13C NOESY) have proven particularly effective in studying resonances of atoms that are close to metal centers.[10, 11] Solid-state NMR is particularly suited to these latter approaches, since carbon-13 acquisition is routine, and transfer schemes based on dipolar interactions are available that allow directed, highly efficient homonuclear (13C–13C) and heteronuclear (1H–13C, 1H–15N, 13C–15N) transfers over short delays …