Both approaches outlined above provide a solution to Eqs. (1-3) by reducing the number of necessary dynamics-related parameters to three and by using certain constant values for the dipolar interaction and 15N CSA, also assuming Rex=0 (in the SD mapping). The dynamic picture derived from such analyses could be inaccurate if these parameters vary from one site to another. For example, the site-specific variations in 15N CSA become important for the analysis of relaxation data at higher fields when the CSA contribution becomes comparable to that from the dipolar interaction, and for TROSY applications . A more accurate approach to protein dynamics should allow independent derivation of protein spectral densities, site-specific 15N CSA and dipolar interactions, as well as Rex values. An approach has been suggested recently based on relaxation measurements at multiple fields (spectrometers) [6, 19, 20]. This method is a natural extension of previous approaches [37, 40] in that all major components in Eqs. (1-5), and not only SDs, are treated as fitting parameters. The approach is based on the fact that some terms in these equations are either field-independent [d, 8, and J(0)] or scale with field B0 in a known way (c x B0, Rex x B2). Therefore, when combining data measured at several fields, the number of experimental parameters increases faster than the number of unknowns, so that already a complete set of three-field data theoretically could allow determination of almost all unknowns in Eqs. (1-3) . This could provide a useful tool for protein dynamics analysis free from the oversimplifying assumptions. The utility of this approach was demonstrated for the full characterization of the spectral densities, 15N CSA tensor, and Rex in ubiquitin, unbiased by the assumption of constant CSA [19, 20]. Similar approaches were recently used by other groups [41, 42]. The necessity to use multiple-field data for accurate analysis of protein motions in multidomain systems was recently demonstrated in Ref. .
Was this article helpful?