Elevated intracellular Na+ concentrations in developing spinal neurons

Abstract

Over 25 years ago it was first reported that intracellular chloride levels (Cl^-_in ) were higher in developing neurons than in maturity. This finding has had significant implications for understanding the excitability of developing networks and recognizing the underlying causes of hyperexcitability associated with disease and neural injury. While there is some evidence that intracellular sodium levels (Na⁺_in ) change during the development of non-neural cells, it has largely been assumed that Na⁺_in is the same in developing and mature neurons. Here, using the sodium indicator SBFI, we test this idea and find that Na⁺_in is significantly higher in embryonic spinal motoneurons and interneurons than in maturity. We find that Na⁺_in reaches ~ 60 mM in mid-embryonic development and is then reduced to ~ 30 mM in late embryonic development. By retrogradely labeling motoneurons with SBFI we can reliably follow Na⁺_in levels in vitro for hours. Bursts of spiking activity, and blocking voltage-gated sodium channels did not influence observed motoneuron sodium levels. On the other hand, Na⁺_in was reduced by blocking the Na⁺ -K⁺ -2Cl^- cotransporter NKCC1, and was highly sensitive to changes in external Na⁺ and a blocker of the Na⁺ /K⁺ ATPase. Our findings suggest that the Na⁺ gradient is weaker in embryonic neuronal development and strengthens in maturity in a manner similar to that of Cl^- .

Keywords: SBFI; ATPase; chick embryo; ionic gradient; sodium.

Figure 1

Average action potential peak of E10 spinal motoneurons indicates intracellular Na⁺ may be higher than expected. A–B) Example of a train of action potentials recorded whole cell from an E10 (A) or E14 (B) spinal motoneuron. APs were evoked by 1.2X stimulus threshold in both groups. C) Scatter plot of action potential peaks showing that peaks are significantly more depolarized at E14 compared to peaks at E10 (p=0.01; unpaired t Test; E10 n=13 cells and E14 n=10 cells).

Figure 2

Stable SBFI-labeling of motoneurons in the embryonic chick spinal cord. A–C) Examples of SBFI-labeled cells from a chick spinal cord at E10 (ventral view). Cells were imaged using 340nm (A) and 380nm (B) illumination. C) Image of enlarged area boxed in B. Pink circles indicate regions of interest (ROIs) drawn around motoneuron somas. ROIs used for background subtractions were taken from non-labeled side of the cord. D–F) Graph of the measured intensities when illuminated with 340 (D) and 380 (E), or ratios (340nm/380nm, F) from 4 motoneurons over 4 hours. The color of each cell corresponds to the same cell across panels. In F, the average of 10 motoneurons from the same cord shown in black - inset represents the average ratio for 4 cords (34 cells) over 3 hours (ANOVA p=0.8, Bonferroni’s p=1).

Figure 3

SBFI measures similarly high concentrations of Na⁺_in at E10 by calibrating cells individually or through population averages. A) Bar chart shows that SBFI measures of Na⁺_in suggest a value of approximately 40mM in spinal motoneurons whether calibrations were carried out for each cell separately (individually) or by taking average values for calibration parameters across the populations (average). ns = no significant difference. B) Relationship between averaged calibration ratios and solutions containing different concentrations of Na⁺ from 6 cords. The calibration curve was calculated using the displayed equation.

Figure 4

Na⁺_in for E10 motoneurons during increases in temperature and expression of SNA. A) SBFI measurements suggest that motoneuron Na⁺_in is temperature-dependent (p=0.009). B) Examples of SBFI-labeled cells from a chick spinal cord at E10 (ventral view) comparing illumination at 340nm and 380nm at 23 °C and 27°C. C) Graphs of the measured intensities with illumination at 340nm (top) and 380nm (bottom) from 4 motoneurons at 23 °C and 27 °C. The color of each cell corresponds to the same cell across graphs. D) Example of an episode of SNA recorded from the VLF at E10 (top). Scatter plot of intracellular Na⁺ from 4 cells measured before, during and after the episode pictured (grey dots). Black dots represent the average of 10 cells from the same cord. Below is a bar chart showing no change in the average intracellular Na⁺ measured with SBFI before, during, and after spontaneous episodes measured from 4 cords (n= 40 cells, ANOVA p = 0.93, Bonferroni’s p = 1).

Figure 5

Motoneuron Na⁺_in during development and following reductions in extracellular Na⁺ and Na⁺-K⁺-ATPase function. Scatter graph indicating 340/380 ratios (left axis) and SBFI-measured Na⁺_in (right axis) for individual cells in control conditions (black dots), following 0mM Na solution perfusion (red dots), and following ouabain bath application (blue dots). In control conditions E6 n=60 cells/6 cords, at E8 n=60 cells/7 cords, at E10 n=96 cells/11 cords, at E12 n=65 cells/7 cords, at E14 n=68 cells/7 cords, and at E16 n=81 cells/9 cords. Bonferroni’s: * indicates p<0.001; ANOVA p<0.001. In the presence of 0mM Na⁺ solution all cells approach the ratio of 0.35 which corresponds to 0mM intracellular Na⁺ At E6 n= 57 cells/6 cords, E8 n= 49 cells/5 cords, E10 n= 80 cells/8 cords, E12 n= 73 cells/8 cords, E14 n= 68 cells/7 cords, and at E16 n= 62 cells/7 cords: * indicates p<0.001 compared to average in normal Na⁺ solution. In ouabain, E6 n=40 cells/4 cords, E8 n=38 cells/4 cords, E10 n=59 cells/6 cords, E12 n=40 cells/4 cords, E14 n=58 cells/6 cords, and at E16 n=50 cells/5 cords. One way ANOVA with post-hoc Bonferroni test for multiple comparisons: * indicates p<0.001 compared to average in normal Na⁺ solution.

Figure 6

Na⁺ entry pathways at E10 and E14. SBFI measurements show that blocking NKCC1 with 10µM bumetanide increased Na⁺_in in E10, but not E14 motoneurons. SBFI motoneuron measurements show that blockade of voltage-gated Na⁺ channels with 1µM TTX, in the presence of bumetanide, had no effect on Na⁺_in at E10 or E14. At E10 n=40 cells/4 cords, E14 n=40 cells /4 cords. One way ANOVA with post-hoc Bonferroni test for multible comparisions: * indicates p<0.001 compared to average in normal Tyrode’s solution and ns = no significant difference.