School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane, QLD 4072, Australia.
J Mol Biol. 2012 Mar 9;416(5):668-77. doi: 10.1016/j.jmb.2011.12.056. Epub 2012 Jan 18.
Paramagnetic metal ions generate pseudocontact shifts (PCSs) in nuclear magnetic resonance spectra that are manifested as easily measurable changes in chemical shifts. Metals can be incorporated into proteins through metal binding tags, and PCS data constitute powerful long-range restraints on the positions of nuclear spins relative to the coordinate system of the magnetic susceptibility anisotropy tensor (Δχ-tensor) of the metal ion. We show that three-dimensional structures of proteins can reliably be determined using PCS data from a single metal binding site combined with backbone chemical shifts. The program PCS-ROSETTA automatically determines the Δχ-tensor and metal position from the PCS data during the structure calculations, without any prior knowledge of the protein structure. The program can determine structures accurately for proteins of up to 150 residues, offering a powerful new approach to protein structure determination that relies exclusively on readily measurable backbone chemical shifts and easily discriminates between correctly and incorrectly folded conformations.
顺磁金属离子在核磁共振谱中产生赝接触位移（PCSs），表现为化学位移的可轻易测量的变化。金属可以通过金属结合标签整合到蛋白质中，PCSs 数据构成了相对于顺磁各向异性张量（Δχ张量）的磁场灵敏度坐标系中核自旋位置的强大长程约束。我们表明，使用单个金属结合位点的 PCS 数据以及骨架化学位移，可以可靠地确定蛋白质的三维结构。在结构计算过程中，程序 PCS-ROSETTA 会自动从 PCS 数据中确定 Δχ 张量和金属位置，而无需任何蛋白质结构的先验知识。该程序可以准确地确定多达 150 个残基的蛋白质结构，为仅依赖于可轻易测量的骨架化学位移且能准确区分正确和错误折叠构象的蛋白质结构确定提供了一种强大的新方法。
