Chen Tsai-Chen, Hsiao Chih-Lun, Huang Shing-Jong, Huang Jie-Rong
Institute of Biochemistry and Molecular Biology, National Yang-Ming University, No. 155 Section 2 Linong Street, Taipei, Taiwan.
Protein Pept Lett. 2016;23(11):967-975. doi: 10.2174/0929866523666160920100045.
In NMR experiments, the chemical shift is typically the first parameter measured and is a source of structural information for biomolecules. Indeed, secondary chemical shifts, the difference between the measured chemical shifts and those expected for a randomly oriented sequence of peptides (the "random coil"), are correlated with the secondary structure of proteins; secondary shift analysis is thereby a standard approach in structural biology. For intrinsically disordered or denatured proteins furthermore, secondary chemical shifts reveal the propensity of particular segments to form different secondary structures. However, because the atoms in unfolded proteins all have very similar chemical environments, the chemical shifts measured for a certain atom type vary less than in globular proteins. Since chemical shifts can be measured precisely, the secondary chemical shifts calculated for an unfolded system depend mainly on the particular random coil chemical shift database chosen as a point of reference. Certain databases correct the random coil shift for a given residue based on its neighbors in the amino acid sequence. However, these corrections are typically derived from the analysis of model peptides; there have been relatively few direct and systematic studies of the effect of neighboring residues for specific amino acid sequences in disordered proteins. For the study reported here, we used the intrinsically disordered C-terminal domain of TDP-43, which has a highly repetitive amino-acid sequence, as a model system. We assigned the chemical shifts of this protein at low pH in urea. Our results demonstrate that the identity of the nearest neighbors is decisive in determining the value of the chemical shift for atoms in a random coil arrangement. Based on these observations, we also outline a possible approach to construct a random-coil library of chemical shifts that comprises all possible arrangement of tripeptides from a manageable number of polypeptides.
在核磁共振实验中，化学位移通常是首先测量的参数，并且是生物分子结构信息的来源。实际上，二级化学位移，即测得的化学位移与肽的随机取向序列（“无规卷曲”）预期的化学位移之间的差异，与蛋白质的二级结构相关；因此，二级位移分析是结构生物学中的标准方法。此外，对于内在无序或变性的蛋白质，二级化学位移揭示了特定片段形成不同二级结构的倾向。然而，由于未折叠蛋白质中的原子都具有非常相似的化学环境，所以对于某一原子类型测得的化学位移变化比球状蛋白质中的要小。由于化学位移可以精确测量，为未折叠系统计算的二级化学位移主要取决于所选作参考点的特定无规卷曲化学位移数据库。某些数据库会根据给定残基在氨基酸序列中的相邻残基来校正其无规卷曲位移。然而，这些校正通常源自对模型肽的分析；对于无序蛋白质中特定氨基酸序列的相邻残基效应，直接和系统的研究相对较少。对于此处报道的研究，我们使用了TDP - 43的内在无序C末端结构域，其具有高度重复的氨基酸序列，作为模型系统。我们在低pH值的尿素溶液中确定了该蛋白质的化学位移。我们的结果表明，最近邻残基的身份对于确定无规卷曲排列中原子的化学位移值起决定性作用。基于这些观察结果，我们还概述了一种可能的方法来构建一个化学位移的无规卷曲文库，该文库包含来自数量可控的多肽的所有可能的三肽排列。
