Simultaneous use of solution NMR and X-ray data in REFMAC5 for joint refinement/detection of structural differences.

Simultaneous use of solution NMR and X-ray data in REFMAC5 for joint refinement/detection of structural differences.
复制标题

DOI:
10.1107/s1399004713034160
复制
发表时间:
2014-04
期刊:
Acta crystallographica. Section D, Biological crystallography
影响因子:
--
通讯作者:
Luchinat C
Luchinat C
中科院分区:
其他
文献类型:
--
作者:
Rinaldelli M;Ravera E;Calderone V;Parigi G;Murshudov GN;Luchinat C

文献摘要

被引文献

相似文献

顺磁核磁共振数据(伪接触位移和自取向残余偶极耦合)和反磁残余偶极耦合现在可以在CCP 4的程序REMAC 5中与X射线晶体学数据一起用作结构约束。这些NMR限制可以揭示分子的固态和溶液构象之间的差异,或者在它们不存在的情况下,可以与X射线晶体学数据一起用于结构细化。对CCP 4的程序REFMAC 5进行了修改,以允许同时使用X射线晶体学数据和顺磁NMR数据(伪接触位移和自取向残余偶极耦合)和/或抗磁残余偶极耦合。在REFMAC 5中纳入这些长程NMR限制可以揭示分子的固态和溶液构象之间的差异,或者在它们不存在的情况下,可以与X射线晶体学数据一起用于结构细化。由于NMR和X射线数据是互补的,当单个结构与两组数据一致并且仍然保持合理的“理想”几何形状时,预计衍生的原子模型的可靠性将增加。该程序进行了测试,对五种不同的蛋白质:催化结构域的基质金属蛋白酶1,GB 3,泛素,游离钙调素和钙调素复合肽。在某些情况下,联合细化产生了一个单一的模型与两组观察结果一致,而在其他情况下,它表明,在实验的不确定性之外,在溶液和固体状态下存在不同的蛋白质构象。
Paramagnetic NMR data (pseudocontact shifts and self-orientation residual dipolar couplings) and diamagnetic residual dipolar couplings can now be used in the program REFMAC5 from CCP4 as structural restraints together with X-ray crystallographic data. These NMR restraints can reveal differences between solid state and solution conformations of molecules or, in their absence, can be used together with X-ray crystallographic data for structural refinement. The program REFMAC5 from CCP4 was modified to allow the simultaneous use of X-ray crystallographic data and paramagnetic NMR data (pseudocontact shifts and self-orientation residual dipolar couplings) and/or diamagnetic residual dipolar couplings. Incorporation of these long-range NMR restraints in REFMAC5 can reveal differences between solid-state and solution conformations of molecules or, in their absence, can be used together with X-ray crystallographic data for structural refinement. Since NMR and X-ray data are complementary, when a single structure is consistent with both sets of data and still maintains reasonably ‘ideal’ geometries, the reliability of the derived atomic model is expected to increase. The program was tested on five different proteins: the catalytic domain of matrix metalloproteinase 1, GB3, ubiquitin, free calmodulin and calmodulin complexed with a peptide. In some cases the joint refinement produced a single model consistent with both sets of observations, while in other cases it indicated, outside the experimental uncertainty, the presence of different protein conformations in solution and in the solid state.