Separation of anisotropy and exchange broadening using 15N CSA-15N-1H dipole-dipole relaxation cross-correlation experiments

Based on the measurement of cross-correlation rates between N-15 CSA and N-15-H-1 dipole-dipole relaxation we propose a procedure for separating exchange contributions to transverse relaxation rates (R-2 = 1/T-2) from effects caused by anisotropic rotational diffusion of the protein molecule. This approach determines the influence of anisotropy and chemical exchange processes independently and therefore circumvents difficulties associated with the currently standard use of T-1/T-2 ratios to determine the rotational diffusion tensor. We find from computer simulations that, in the presence of even small amounts of internal flexibility, fitting T-1/T-2 ratios tends to underestimate the anisotropy of overall tumbling. An additional problem exists when the N-H bond vector directions are not distributed homogeneously over the surface of a unit sphere, such as in helix bundles or beta-sheets. Such a case was found in segment 4 of the gelation factor (ABP 120), an F-actin cross-linking protein, in which the diffusion tensor cannot be calculated from T-1/T-2 ratios. The N-15 CSA tensor of the residues for this beta-sheet protein was found to vary even within secondary structure elements, The use of a common value for the whole protein molecule therefore might be an oversimplification. Using our approach it is immediately apparent that no exchange broadening exists for segment 4 although strongly reduced T-2 relaxation times for several residues could be mistaken as indications for exchange processes. (C) 2000 Academic Press.