Eva de Alba

Residual Dipolar Couplings in Protein Structure Determination

Each magnetic nucleus behaves like a magnetic dipole able to create a local magnetic field at the position of nearby nuclei. In the presence of an external magnetic field, the local field modifies the original Larmor frequency of the affected nucleus. Such an interaction is called the dipole-dipole interaction or dipolar coupling. Its magnitude depends on, among other factors, the distance between the interacting nuclei and the angle that the internuclear vector forms with the magnetic field. Through this angular dependence it is possible to relate the position of the two interacting nuclei with respect to an arbitrary axis system of reference. Therefore, dipolar couplings can be used to obtain structural information. In liquid samples, which usually provide high-resolution nuclear magnetic resonance (NMR) spectra, the internuclear vector moves isotropically and the dipolar coupling averages to zero. In the solid state, where this vector has a fixed orientation, the dipole-dipole interactions are numerous and strong, broadening NMR signals such that structural information at high resolution cannot be obtained. An intermediate situation is achieved by partially restricting molecular tumbling of liquid samples. The alignment of a fraction of molecules in the presence of the magnetic field allows the measurement of dipolar couplings. Because they are scaled down owing to partial alignment, we refer to them as residual dipolar couplings (RDCs). The structural information obtained from RDCs has impacted enormously traditional protein structure determination based on nuclear Overhauser effect-derived distance restraints. Methods to measure RDCs and their application to protein structure determination are illustrated.

Simple multidimensional NMR experiments to obtain different types of one-bond dipolar couplings simultaneously

In order to measure residual dipolar couplings, the molecule under study has to be partially oriented in the presence of the magnetic field. It has been observed that some protein samples are not stable under the conditions imposed by the orienting media. If different types of dipolar couplings are measured sequentially, their values will not agree with a unique alignment tensor that is changing slowly over time. This could bias the structure calculation. It would be more appropriate to obtain different types of dipolar couplings simultaneously, such that all the data correspond to one effective alignment tensor. We describe here a general NMR strategy designed to do so, that can be adapted to various existing pulse sequences.