What can really be learned from dielectric spectroscopy of protein solutions?: A case study of ribonuclease A

We report on a dielectric relaxation study of aqueous solutions of ribonuclease A at 298.15 K as a function of protein concentration between 0.5 and 6 wt % in the MHz/GHz frequency range. The spectra can be decomposed into five modes of Debye type diffusive behavior. In agreement with the standard interpretation, we assign the two dominant modes at low and high frequency (beta-relaxation and gamma-relaxation, respectively) to protein tumbling and bulk water relaxation. We observe three further modes (delta1-delta3) between beta- and gamma-relaxation, in contrast to a bimodal delta-dispersion frequently reported. We attribute the high frequency part (delta3) near 40 ps to hydration water reorientation, which, in the notion of other authors, corresponds to loosely bound water. We argue that the existence of tightly bound water, often deduced from the low frequency part in the nanosecond regime (delta1), is inconsistent with a highly mobile hydration layer observed by NMR techniques and molecular dynamics (MD) simulations. On the same grounds, we reject hydration water-bulk water exchange as a mechanism for delta-dispersion. In accordance with MD simulations, we assume that protein-water cross-correlations drive the nanosecond (delta1) process. We also discuss the role of intraprotein motions, which may contribute near 500 MHz (delta2). We discuss the meaning of the hydrodynamic radius and of the hydration numbers in light of the high mobility of hydration waters. We show that because of protein-protein interactions, the effective dipole moment of the protein decreases with increasing protein concentration.