Nuclear magnetic resonance (NMR) spectroscopy provides a means of noninvasively observing intracellular sodium. one of the principal ions responsible for the electrical activity of the heart. The NR parameters of resonant frequency, spin density, and the relaxation times T₁ and T₂ yield information regarding the numoer of sodium ions in the sample as well as the interactions between the ions and their environment. However, the difficulty in differentiating between the NMR signals originating from the intracellular and from the extracellular sodium iorns has, until now, precluded quantitative analysis of the above parameters for h~ intracellular sodium. In this study, a shift reagent (dysprosium tripolyphosphate) was used to alter the resonant frequency of the extracellular contributions to the NMR spectrum. Mathematical filtering or presaturation of the extracellular resonance was used to separate more completely the intra- and extracellular sodium NMR signals, thereby allowing for the quantitative investigation of the intracellular sodium in the perfused frog heart.

Sodium NMR spectra were obtained on a Bruker 360 AM spectrometer operating at 95.26 MHz. The results presented here demonstrate the feasibility of using NMR spectroscopy to measure changes in the intracellular sodium levels with pharmacologic and physiologic interventions. 1) The three interventions of adding 10pM ouabain to the perfusate, perfusing with a zero potassium buffer, and replacement of 66% of the sodium in the perfusate with lithium, resulted in the following percent changes in the intracellular sodium levels (mean S.D.):​ ouabain, +(460±60), n=6;​ zero potassium, +(300±30), n=3;​ lithium, -(51±6), n=3. 2) The ability to monitor reversible events was demonstrated by returning to the normal buffer after perfusing with zero potassium;​ the sodium level was shown to return to its initial level within 30 minutes. 3) Smaller changes in the intracellular sodium were observed when the hearts were paced at varying ates. An increase of 45% in the intracellular sodium was observed when changing the pacing rate from 0 to 60 beats per minute, and proportional changes were measured for intermediate pacing rates. These results are comparable to those in the literature which utilize microelectrode measurements. In the experiments reported here, the extracellular calcium was below the normal levels due to binding of the calcium to the shift reagent;​ compensation for this effect is possible and should allow this technique to be extended to further physiologic studies and to mammalian systems.

Motional restriction, or binding, of sodium results in multicomponent NMR relaxation times, primarily affecting T₂ . The relaxation times of the intracellular sodium were measured under both control conditions and after a 5 fold increase in the intracellular sodium (due to oabain exposure). T₁ was measured with the standard inversion recovery technique, yielding a single exponential decay with a time constant of 22.4 ± 3.0 msec (n=5, control) or 24.2 ± 1.5 msec (n=5, ouabain). The T₂ of the intracellular sodium was measured using the standard Hahn echo technique, both with and without the modification of presaturation of the extracellular resonance. The T₂ decay was not well fit by a single exponential;​ a double exponential fit yielded time constants and relative amplitudes of 2.0 1.3 msee (46 8%) and 16.3 4.3 msec (54 ± 8%) for the control hearts (n=5) and 2.1 ± 0.6 msec (43 5%) and 16.8 4.0 msec (57 5%) for the ouabain treated hearts (n=7). The short relaxation times and the biexponential T₂ decays demonstrate that there is some restriction of the sodium ions within the cell relative to ions in free solution, and that these interactions are not affected by a five fold increase in the sodium level. However the relative amplitudes of the different time constants are inconsistent with those which would be expected from a homogeneous pool of nuclei. These results are therefore indicative of restriction and compartmentation of sodium within the cell, i lformation which is not available from microelectrode studies. In addition, the results presented here, along with further studies, may enable the NMR identification of at least a component of the intracellular sodium on the basis of its short T₂ relaxation time, and thus, may eliminate the need for the shift reagent for in-vivo studies. They may also help form the basis for the interpretation of clinically obtained sodium images, in which contrast is determined by the sodium content, T₁, and T₂.