The presence of a paramagnetic center in a protein increases the linewidths of the protein nuclei and therefore makes more difficult the detection of signals and the collection of the NMR-based constraints necessary to solve the three-dimensional structure. At the same time, paramagnetism-induced relaxation, contact and peudocontact shifts, residual dipolar couplings and cross-correlation effects provide novel constraints that can compensate for the losses in diamagnetic constraints. In addition, being of a different nature, paramagnetism-based constraints can provide information that cannot be possibly obtained otherwise.
In the case of diamagnetic metalloproteins, it should be kept in mind that, even in the absence of paramagnetic line broadening, the very presence of the metal prevents the obtainment of NOE constraints: a metal ion surrounded by liganded protein side chains constitutes a “black hole” across which NOEs are hardly measured. Furthermore, with a few exceptions, metal-protein constraints are not available, so that the metal coordination cage is often ill-determined. In this case, substitution of the diamagnetic metal with a paramagnetic one may provide a dramatic improvement in i) defining the protein ligands and ii) defining the metal coordinates in the structure. The first information can be provided by contact shifts, as they depend on the presence of metal-donor atom coordination bonds, while the second is provided by pseudocontact shifts. Finally, contact shifts can provide information on dihedral angles, in much the same way as 3J coupling measurements in diamagnetic systems.
Another instance where paramagnetic relaxation enhancement may be beneficial is in the study of relatively weak protein-protein and protein-small molecules interaction, where the strong relaxation enhancement permits the detection and the partial characterization of the interaction even in the presence of high molar ratios between unbound and bound forms.
Perhaps the most promising exploitation of paramagnetism in metalloproteins is based on the combination of the various pieces of information derived from the anisotropic magnetic susceptibility tensor (pseudocontact shifts and residual dipolar couplings) to learn about the relative degrees of freedom of one protein domain with respect to another. Broadly speaking, NMR is in principle able to provide information on unstructured or partially structured protein systems, thereby complementing other structural techniques. More and more efforts are dedicated to understand the behavior of unfolded proteins that may be natively lacking tertiary structure. In parallel, there is a continuing interest in understanding the dynamics of proteins that perform their function by changing their structure. The presence of a paramagnetic metal ion helps acquiring information on, e.g., global order parameters of one domain with respect to another, or on the relative population of different conformers. Finally, relatively weak protein-metal ion interactions that may have a functional relevance are best studied when the metal ion is paramagnetic.