Chen Zhongjing, Krause Gerd, Reif Bernd
Forschungsinstitut für Molekulare Pharmakologie (FMP), Robert-Rössle-Str. 10, D-13125 Berlin, Germany.
J Mol Biol. 2005 Dec 9;354(4):760-76. doi: 10.1016/j.jmb.2005.09.055. Epub 2005 Oct 5.
Polymerization of the soluble beta-amyloid peptide into highly ordered fibrils is hypothesized to be a causative event in the development of Alzheimer's disease. Understanding the interactions of Abeta with inhibitors on an atomic level is fundamental for the development of diagnostics and therapeutic approaches, and can provide, in addition, important indirect information of the amyloid fibril structure. We have shown recently that trRDCs can be measured in solution state NMR for peptide ligands binding weakly to amyloid fibrils. We present here the structures for two inhibitor peptides, LPFFD and DPFFL, and their structural models bound to fibrillar Abeta(14-23) and Abeta(1-40) based on transferred nuclear Overhauser effect (trNOE) and transferred residual dipolar coupling (trRDC) data. In a first step, the inhibitor peptide structure is calculated on the basis of trNOE data; the trRDC data are then validated on the basis of the trNOE-derived structure using the program PALES. The orientation of the peptide inhibitors with respect to Abeta fibrils is obtained from trRDC data, assuming that Abeta fibrils orient such that the fibril axis is aligned in parallel with the magnetic field. The trRDC-derived alignment tensor of the peptide ligand is then used as a restraint for molecular dynamics docking studies. We find that the structure with the lowest rmsd value is in agreement with a model in which the inhibitor peptide binds to the long side of an amyloid fibril. Especially, we detect interactions involving the hydrophobic core, residues K16 and E22/D23 of the Abeta sequence. Structural differences are observed for binding of the inhibitor peptide to Abeta14-23 and Abeta1-40 fibrils, respectively, indicating different fibril structure. We expect this approach to be useful in the rational design of amyloid ligands with improved binding characteristics.
可溶性β-淀粉样肽聚合成高度有序的纤维被认为是阿尔茨海默病发展过程中的一个致病事件。在原子水平上理解β-淀粉样蛋白(Aβ)与抑制剂的相互作用是开发诊断和治疗方法的基础,此外,还可以提供有关淀粉样纤维结构的重要间接信息。我们最近表明,可以在溶液状态核磁共振中测量与淀粉样纤维弱结合的肽配体的弛豫分散交叉峰(trRDC)。我们在此展示了两种抑制剂肽LPFFD和DPFFL的结构,以及基于转移核Overhauser效应(trNOE)和转移剩余偶极耦合(trRDC)数据与纤维状Aβ(14 - 23)和Aβ(1 - 40)结合的结构模型。第一步,基于trNOE数据计算抑制剂肽的结构;然后使用PALES程序基于trNOE衍生的结构验证trRDC数据。假设Aβ纤维的取向使得纤维轴与磁场平行,从trRDC数据获得肽抑制剂相对于Aβ纤维的取向。然后将肽配体的trRDC衍生的取向张量用作分子动力学对接研究的约束。我们发现均方根偏差(rmsd)值最低的结构与抑制剂肽结合到淀粉样纤维长边的模型一致。特别是,我们检测到涉及疏水核心、Aβ序列中K16和E22 / D23残基的相互作用。分别观察到抑制剂肽与Aβ14 - 23和Aβ1 - 40纤维结合的结构差异,表明纤维结构不同。我们期望这种方法在合理设计具有改善结合特性的淀粉样配体方面有用。