The molecular environment of intracellular sodium: 23Na NMR relaxation

The comprehensive approach described in the accompanying paper is illustrated here with the 23Na signal of a concentrated solution of bovine serum albumin (BSA) in saline and the intracellular (Nai) 23Na resonance of a dense suspension of Na(+)-loaded yeast cells. We use frequency shift reagents to discriminate the latter from the extracellular resonance. We find that the Nai signal corresponds to that of an effective single population of Na+ ions exhibiting a single type c spectrum. This is true despite the fact that the yeast protoplasm is too large and too compartmentalized for a given Na+ ion to sample its entirety on the relevant NMR timescale. Our results show clearly that, in addition to the decay of transverse magnetization, the recovery of longitudinal magnetization is biexponential. This is required for a type c spectrum but has not often been detected. The temperature dependence of the relaxation rate constants of the Nai resonance is not consistent with either a simple Debye process or a discrete exchange mechanism connecting two sites in the fast limit. We have fitted the data using an asymmetric continuous distribution of correlation times for the fluctuations of electric field gradients sensed by the Nai nuclei. The analogous distribution function for the Na+ in a 44% (w/w) BSA solution is quite similar to that of the Nai at the same temperature. This suggests that while the macromolecular environment of the Nai ions is quite congested, it is also isotropic on quite a small spatial scale. Also, one can use the correlation time distribution function, obtained from fitting the relaxation data, to calculate a relaxometry curve. This is useful because experimental 23Na relaxometry is difficult. The calculated curve may be a reasonable model for the mostly extracellular 23Na resonance encountered in vivo.