Sodium currents in medullary neurons isolated from the pre-Bötzinger complex region

Abstract

The pre-Bötzinger complex (preBötC) in the ventrolateral medulla contains interneurons important for respiratory rhythm generation. Voltage-dependent sodium channels mediate transient current (I(NaT)), underlying action potentials, and persistent current (I(NaP)), contributing to repetitive firing, pacemaker properties, and the amplification of synaptic inputs. Voltage-clamp studies of the biophysical properties of these sodium currents were conducted on acutely dissociated preBötC region neurons. Reverse transcription-PCR demonstrated the presence of mRNA for Nav1.1, Nav1.2, and Nav1.6 alpha-subunits in individual neurons. A TTX-sensitive I(NaP) was evoked in all tested neurons by ramp depolarization from -80 to 0 mV. Including a constant in the Boltzmann equation for inactivation by estimating the steady-state fraction of Na+ channels available for inactivation allowed prediction of a window current that did not decay to 0 at voltages positive to -20 mV and closely matched the measured I(NaP). Riluzole (3 microM), a putative I(NaP) antagonist, reduced both I(NaP) and I(NaT) and produced a hyperpolarizing shift in the voltage dependence of steady-state inactivation. The latter decreased the predicted window current by an amount equivalent to the decrease in I(NaP). Riluzole also decreased the inactivation time constant at potentials in which the peak window/persistent currents are generated. Together, these findings imply that I(NaP) and I(NaT) arise from the same channels and that a simple modification of the Hodgkin-Huxley model can satisfactorily account for both currents. In the rostral ventral respiratory group (immediately caudal to preBötC), I(NaP) was also detected, but peak conductance, current density, and input resistance were smaller than in preBötC region cells.

Figure 1.

Retrograde labeling of cranial motoneurons provides landmarks for the preBötzinger complex region. A, A 50 μm parasagittal section at the level of the ventral respiratory column. Cranial motoneurons were retrogradely labeled after a subcutaneous injection of FluoroGold (pseudocolored blue and green in digital image) and subsequent DAB immunolabeling of NK1-receptive neurons and processes (inverted bright-field image placed in the red channel of RGB digital image). B, A fresh-mounted, 300-μm-thick parasagittal section at a level similar to that shown in A. Brief, low-magnification, 4× epifluorescent UV excitation is sufficient to identify the FluoroGold-labeled facial nucleus and compact part of nucleus ambiguus as landmarks for dissection of the region containing the preBötC (square). C, D, Examples of fusiform (C) or round (D) neurons acutely dissociated from the region of the preBötC complex. 7, Facial nucleus; 7n, facial nerve; AmbC, compact part of nucleus ambiguus; LRt, lateral reticular nucleus; Mo5, motor nucleus of the trigeminal nerve.

Figure 2.

Features of transient Na⁺ current in neurons from the preBötC region. A, B, Representative current traces (TTX-subtracted) and protocols used to examine the voltage dependence of activation (A) and inactivation (B; 100 ms prepulses). C, Plots of the peak conductance (○) and the test pulse conductance (•) as a function of voltage. Solid lines are fits to the Boltzmann equation for inactivation and activation. Fitting parameters for this neuron were as follows: V_1/2 = -46.0 mV, k = 6.0 mV (activation; based on the third-order version of the Boltzmann equation); V_1/2 = -68.3 mV, k = 8.5 mV (inactivation; first-order Boltzmann equation).

Figure 3.

Comparison of Na⁺ currents in neurons from the rVRG and preBötC regions. A, Box plots showing transient Na⁺ current amplitude in rVRG (open bars; n = 20) and preBötC (hatched bars; n = 29) neurons. B, Amplitude of persistent Na⁺ current induced by a voltage ramp (50 mV/s) in rVRG neurons (n = 20) compared with preBötC neurons (n = 29). C, Density of I_NaP in neurons from rVRG (open bars; n = 20) preBötC (hatched bars; n = 29) regions. D, R_IN is lower for rVRG (n = 13) compared with preBötC (n = 13) region neurons. E, F, Time constants for activation (E) and inactivation (F). See Materials and Methods for a description of box-plot parameters.

Figure 4.

Na⁺ currents in neurons from the preBötC region do not completely inactivate at more than -20 mV. A, Persistent Na⁺ current was evoked by a ramp from -80 to 0 mV (50 mV/s). The TTX-subtracted trace is illustrated. Inset, Traces before and during TTX (300 nm). B, Representative TTX-subtracted Na⁺ current was elicited at -20 mV by a 70 ms step from a holding potential of -80 mV. In this neuron, the step to -20 mV produced a peak Na⁺ current of -6 nA that decayed to a steady-state level of approximately -75 pA (higher resolution in the inset; note the different scale for time and current), representing 1.2% of the total Na⁺ current. C, Na⁺ channel availability for inactivation at -20 mV. Peak current amplitudes during a 10 mV test pulse were divided by the peak currents measured during prepulses of varying duration. These values were plotted as percentages representing the fraction of remaining Na channels available for inactivation. The data points were subsequently fit with a curve based on the sum of two exponentials (time constants, 13.1 and 1.7 ms).

Figure 5.

Conditioning pulse trains reduce the amount of transient and persistent Na⁺ currents at -20 mV in neurons from the preBötC region. Entry into inactivation was determined using a 500 ms prepulse to -20 mV. Inactivation was measured with a 10 ms test pulse to 10 mV after a 20 ms recovery interval at -80 mV (○). The conditioning trains consisted of 15 20 ms pulses separated by 20 ms intervals at -80 mV (•). Insets, Transient Na⁺ current with (•) and without (○) conditioning trains at higher resolution either during beginning of prepulse at -20 mV (left inset) or during test pulse at 10 mV (right inset). The effect of conditioning trains was also studied on persistent Na⁺ current measured at the end of prepulse (middle inset; note different current scale). Note that the conditioning trains and step pulse reduced the amplitude of I_NaT similarly (∼50%). The step pulse similarly reduced I_NaP (∼60%).

Figure 6.

Persistent and window currents may arise from the same Na⁺ channels in preBötC region neurons. A, Voltage dependence of the conductance underlying persistent current compared with that of window conductance (solid line) for the same cell. Both conductances were normalized to the same maximum values and are approximately coincident. B, Box plot illustrating the offsets in peak conductance of window minus persistent conductances in neurons from the preBötC region (n = 20). Note that the median and lower bounds for the interquartile range both equal zero.

Figure 7.

Riluzole reduces transient and persistent Na⁺ currents in preBötC region neurons. A, Riluzole (3 μm) reduced transient Na⁺ current (∼70%) evoked by a step to -20 from -80 mV in a representative neuron. B, Concentration-dependent reduction by riluzole of Na⁺ current during steps from -80 to -20 mV. Data were fit to the Langmuir isotherm with an EC₅₀ value of 2.4 μm. Concentrations of riluzole <100 nm were not used; accordingly, current inhibition by lower concentrations represents values predicted by the theoretical fit of the curve. Control (○) and riluzole (•; 3 μm) influence the voltage dependence of activation and inactivation. Solid lines are fits to the Boltzmann equation. Note that riluzole reduced availability at hyperpolarized potentials and shifted the steady-state inactivation curve to the left. D, Relative window conductance for control and riluzole computed from C. E, Riluzole (3 μm) reduced the I_NaP (∼70% at -45 mV; ∼80% at -20 mV). F, Effect of riluzole on I_NaP activation in one cell. Na⁺ conductance was derived from TTX-subtracted current in A. Solid lines are fits to the Boltzmann equation.

Figure 8.

The effect of riluzole on transient Na⁺ currents evoked by a 50 ms pulse to -20 mV in a preBötC region neuron is dependent on holding potential. A-D, Representative Na⁺ currents evoked in control and with riluzole (3 μm) by a step from holding potentials of -100 (A), -80 (B), -60 (C), and -55 (D) mV. Insets, Currents at the end of 50 ms steps at higher resolution. Note the enhanced effect of riluzole on transient and persistent Na⁺ currents with steps initiated from depolarized holding potentials. E, Summary of the reduction of transient (□) and persistent (▪) Na⁺ current by riluzole (3 μm; n = 11). F, Riluzole reduced the inactivation time course at potentials more negative than -30 mV. Currents were evoked by 50 ms step depolarizations from -45 to 0 mV (5 mV increments) from a holding potential of -80 mV. Decaying phases of current traces were fit by a monoexponential function, and the time constant in the absence (○) and presence (•) of riluzole was plotted as a function of the test potentials. RIL, Riluzole; CTRL, control.

Figure 9.

Summary of Na⁺ channel α-subunit mRNA detection in neurons from preBötC and rVRG regions. A, Example of single-cell RT-PCR on six neurons (1 per column) from the preBötC region. The leftmost column is a 100 bp ladder. Top, Nav1.1 (353 bp). Bottom, Nav1.6 (256 bp). B, α-Subunits were detected in both the preBötC and rVRG regions. Three left histogram pairs show the detection rate for individual α-subunits; three right pairs show the rate of codetection when two α-subunits were simultaneously examined in individual neurons. Note the higher frequency of Nav1.1 in preBötC region neurons and Nav1.2 in rVRG neurons. ^*p < 0.05 between preBötC and rVRG. The number of cells analyzed is as follows: from preBötC region, 127 Nav1.1, 37 Nav1.2, 116 Nav1.6, 37 Nav1.1 and Nav1.2, 37 Nav1.2 and Nav1.6, 116 Nav1.1 and Nav1.6; from rVRG region, 45 Nav1.1, 44 Nav1.2, 59 Nav1.6, 30 Nav1.1 and Nav1.2, 24 Nav1.2 and Nav1.6, 15 Nav1.1 and Nav1.6.