| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biomedical Sciences, Anatomy and Neurobiology Section, Colorado State University, Fort Collins, Colorado 80523
Submitted 7 August 2003; accepted in final form 9 October 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
50 to 700 ms, depending on the voltage level, were revealed. The INa,S (50–150 pA) was present in both spontaneously active and silent neurons. The neurons exhibited INa,S without visible rundown during 1-h recordings. INa,S had a threshold between –65 and –60 mV and was maximal at about –45 mV. Tetrodotoxin (TTX; 1 µM) completely and reversibly blocked INa,S. Riluzole, an effective blocker of INa,P, inhibited reversibly INa,S with an EC50 of 1–2 µM. Microapplication of 10 µM riluzole during either extracellular or intracellular recording suppressed spontaneous activity in isolated SCN neurons. In the slice preparation, bath application of 20 µM riluzole resulted in decreased firing rate or complete suppression of spontaneous activity in some neurons (9/14) but had no effect on other neurons (5/14). In riluzole-resistant neurons in cell-attached experiments, low-amplitude current spikes were present in 1 µM TTX. We concluded that INa,S is ubiquitously expressed by all SCN neurons and that this current is a necessary but not sufficient depolarizing component of the mechanism for spontaneous firing. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
). Spontaneous firing of SCN neurons is under circadian modulation, and the ionic mechanisms underlying the spontaneous activity of SCN neurons are, however, not well understood. Circadian oscillations of electrical activity are present at the cell level even for completely isolated SCN cells (Honma et al. 1998; Welsh et al. 1995). The well-known role of the fast Na+ current is to generate and propagate action potentials. Many CNS neurons have been found to possess slowly inactivating or "persistent" Na+ current, INa,P (Alzheimer et al. 1993; Bevan and Wilson 1999; Chao and Alzheimer 1995; Del Negro et al. 2002a,b; Hammarström and Gage 1999; Kay et al. 1998; Magistretti et al. 1999; Schwindt and Crill 1977; Taddese and Bean 2002; Urbani and Belluzzi 2000; Uteshev et al. 1995; for review, see Crill 1996; Llinas 1988). Activation of INa,P is in the subthreshold region, and its slowly inactivating or persistent nature suggests that it participates in spontaneous firing of central neurons by contributing to their resting potential and threshold. The question to what extent INa,P undergoes inactivation is still open, although a biophysical analysis of this issue has been done for, at least, layer-II principal neurons of entorhinal cortex (Magistretti and Alonso 1999). SCN neurons are spontaneously active, and it has been suggested that the slowly inactivating component of Na+ current contributes to their spontaneous activity (Pennartz et al. 1997).
The goals of the present investigation were to examine to what extent slowly inactivating Na+ current in SCN neurons (Pennartz et al. 1997) reflects the properties of INa,P and to estimate its contribution to spontaneous firing of SCN neurons. In isolated rat SCN neurons, we studied the properties of this Na+ current (termed here INa,S), which shares the properties of INa,P except that it has a relatively slow inactivation (
i 50–250 ms at –45 mV that corresponds to peak of INa,S). Using riluzole, an effective blocker of regular INa,P and INa,S in SCN neurons, we have estimated the putative contribution of INa,S to spontaneous activity of both isolated SCN neurons and these cells in the slice preparation.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Male adult Harlan Sprague-Dawley rats, 4–6 wk old, were used for all experiments. All protocols were approved by the Animal Care and Use Committee at Colorado State University. Rats were deeply anesthetized with halothane and killed by decapitation. One coronal hypothalamic slice (350 µm) containing both SCNs was prepared using a Vibratome (Lancer) in ice-cold artificial cerebrospinal fluid (ASCF) containing (in mM) 124 NaCl, 3 KCl, 2.5 CaCl2, 2 MgSO4, 1.25 NaH2PO4, 26 NaHCO3, and 10 D-glucose, bubbled with 95% O2-5% CO2 to a final pH of 7.4, osmolarity 295–300 mosM. The slice was made during the day at 09:00–10:00 for recordings at 13:00–16:00. After preparation, the slice was placed in an experimental chamber mounted on the stage of an upright microscope (Axioskop 2, Zeiss) and allowed 2 h to recover. Individual SCN neurons could be clearly distinguished using a water-immersion objective (Olympus x40) with Nomarski differential interference contrast optics and a cooled charge-coupled device camera (Hamamatsu, Japan). During recovery and recording, the slice was superfused at 1.5–2.5 ml/min (34–35°C). Recordings were made from the central region of the SCN. When GABA (100 µM) or TTX (1 µM) were bath applied to evaluate the time required to exchange solutions, complete equilibration of new solutions in the experimental chamber was achieved in 3 min.
Isolated neurons
For preparation of isolated SCN neurons, the slice was placed in proteinase K (Sigma, St. Louis MO) in 0.2 mg/ml in PIPES buffer (in mM: 115 NaCl, 5 KCl, 20 PIPES free acid, 1 CaCl2, 4 MgCl2, and 25 D-glucose, pH 7.0; aerated with 100% O2) at 30°C for 5 min, rinsed in PIPES buffer, and placed in trypsin (Sigma Type XI; 1 mg/ml) in PIPES buffer at 30°C for about 30 min. The slice was then rinsed four to five times in PIPES buffer and two regions from both SCNs (500 x 500 µm) were dissected. Neurons were then isolated from each SCN separately by trituration with flame-polished Pasteur pipettes in ice-cold PIPES buffer containing 0.1% DNase. The resulting solution was diluted 1:1 with Neurobasal-A/B-27 (Gibco, Grand Island, NY). Cells were plated on charged plastic culture dishes, incubated at 37°C (95% O2-5% CO2) for 30 min to allow adherence, rinsed, and covered with Neurobasal A/B-27. Neurons were incubated 
72 h prior to electrophysiological recordings that were conducted during the subsequent 5 days. The culture dish was mounted on the stage of an inverted microscope (Eclipse TE300, Nikon) and allowed >1 h to recover. Individual SCN neurons (9–11 µm) could be clearly visualized using a phase contrast objective (Nikon x4 and x40).
Electrophysiological recording
Cell-attached recordings of spontaneous electrical activity from SCN neurons in the slice preparation were obtained at 36°C with a MultiClamp 700A amplifier (Axon Instruments, cutoff 2 kHz), sampled at 10 kHz, and analyzed with pClamp8 software (Axon Instrument). Micropipettes were fabricated from capillary glass (Garner Glass, Claremont, CA;1.2 mm ID, 1.65 mm OD) and filled with solution containing (in mM) 135 NaCl, 10 EGTA, 1 EDTA, and 10 HEPES, (NaOH to pH 7.4 produced [Na+] of 160 mM). The osmolarity was 290–295 mosM, and the electrodes had a resistance of 4–5 M
. The extracellular electrical activity of two SCN neurons was usually recorded simultaneously.
Cell-attached and whole cell recordings from isolated SCN neurons were obtained at room temperature (20–21°C) with an Axopatch-1D patch-clamp amplifier (cutoff 2 kHz), sampled at 10 kHz, digitally filtered at 0.5 kHz for ramp recording, and analyzed with pClamp8 software. Micropipettes were filled with solution containing (in mM) 120 K+-gluconate, 10 HEPES, 1 NaCl, 1 MgCl2, 1 CaCl2, 3 KOH (to pH 7.2–7.4), 5 EGTA, and 2 Na2ATP. Sucrose was added to bring the osmolarity to 300 mosM, and the electrodes had a resistance of 7–8 M. These micropipettes were also used for extracellular cell-attached recording of spontaneous activity of isolated neurons before rupture of the patch membrane. The chamber was superfused with ACSF at 1.5–2.5 ml/min. Offset potential was zeroed just before the electrode contacted the neuronal membrane. All voltage measurements were corrected off-line for the liquid junction potential, which amounted to –15 mV for the K-gluconate-filled micropipette (Neher 1992). The access resistance did not exceed a value of 15 M (typically was 6–12 M) and was not compensated because maximal whole cell current was <150 pA, and thus the voltage error was <2 mV.
Drugs and puff application
The TTX (Sigma), was used as a 1-mM stock solution in distilled water and then diluted to its final concentration in ACSF before the experiment. Riluzole (Sigma) was prepared at 100 mM in DMSO and then diluted to appropriate concentration in ACSF. For puff application of drugs (TTX- or riluzole-containing ACSF) during ACSF perfusion, a plastic tube (160 µm ID) was located near the SCN neurons (200–250 µm) and gravity pressure was used (1 drop/2 s) to apply the drugs. Tests with 1 µM TTX have shown that complete block of fast Na+ current was achieved within 2 s after start of the puff application.
Data analysis
Firing frequency of spontaneously active SCN neurons in the slice preparation, which was averaged for 10 s, was calculated using pClamp version 6.0 (FETCHAN and pSTAT) and plotted with SigmaPlot 2000 version 6.0. Input resistance (Rm) was determined via linear regression applied to the linear portion of the ramp current-voltage (I-V) relationship generated by a slow voltage ramp (100 mV/s; from –85 to –65 mV). In some figures, the linear leak current (IL), was used to extract INa,S by subtracting IL off-line from the total current. In some figures, INa,S was low-pass-filtered (50 Hz; Clampfit 8.1). Values are given as means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Most isolated SCN neurons (90% from >100 recorded) exhibited spontaneous activity when recorded after 3 days of incubation. An irregular bursting or regular firing pattern was observed or the SCN neurons were silent (10%) with episodic action potentials (<1 action potential per min, Fig. 1). By using long-term multielectrode-dish recordings, similar patterns of spontaneous firing have been recorded in isolated and cultured SCN neurons (Honma et al. 1998; Liu and Reppert 2000; Shirakawa et al. 2000; Welsh et al. 1995; see also Walsh et al. 1992). Irregular or bursting activity has been considered a transitional pattern between silence and high-frequency activity during diurnal rhythms (Welsh et al. 1995). In our experiments, the pattern of electrical firing often remained after rupture of the patch membrane (see, for example, Fig. 4, A and B). Voltage-clamp recordings in whole cell mode were obtained from 96 isolated SCN neurons. Recordings with unclamped action potentials were discarded. The neuronal population studied here had a firing rate of 1.2 ± 0.6 Hz (n = 32 neurons; averaging for 5–10 min of recording in cell-attached mode just after seal formation), and input resistance in its linear region between –85 and –65 mV was 3.14 ± 0.34 G (n = 9 neurons).
|
|
To isolate and study the properties of INa,S, we did not eliminate other ionic currents because we have focused on the contribution of INa,S to the electrical properties of SCN neurons during normal electrical activity with standard ionic media. Experiments were first performed on isolated SCN neurons exhibiting spontaneous activity in which the presence of a persistent or slowly inactivating Na+ current was expected. The protocol used to evoke INa,S was to ramp from a holding potential of –85 mV at a rate of 100 mV/s to different values between –35 and –15 mV and then to ramp backward to the original holding potential. The ascending phase of the voltage-clamp command was used to measure the quasi-steady-state I-V relationship (considered in the following text to be the I-V relationship). The 100 mV/s rate was chosen as a compromise because a faster rate evoked in most neurons an obvious unclamped action potential-generating Na+ current (see for example, Del Negro et al. 2002a, Fig. 4), and a lower rate resulted in a decrease of INa,S amplitude because of its inactivation, which was studied in detail in the experiments described in the following text (Fig. 3, B and C).
|
1-h recordings. Figure 2A shows the slow inward current that was completely and reversibly suppressed by the fast Na+ channel blocker, TTX, (n = 15 neurons). In different neurons, the TTX-sensitive slow inward current had an amplitude of 50–150 pA (95.9 ± 5.5 pA; n = 33 neurons). Interestingly, the slow inward current was always accompanied by a substantial increase in current noise relative to the baseline between –85 and –65 mV (and compare with the current in the presence of TTX). This suggests that the opening of a relatively high-conductance channel generates the slow inward current (Magistretti et al. 1999). Riluzole, a neuroprotective agent with anticonvulsant properties, is an effective blocker of INa,P with EC50 0.5–3 µM in mammalian neurons (Del Negro et al. 2002b; Spadoni et al. 2002; Urbani and Belluzzi 2000), and we tested the effect of 0.1–100 µM riluzole on the slow inward current, INa,S, in isolated SCN neurons. At a dose of 25 µM, riluzole completely and reversibly (Fig. 2B) inhibited the slow inward current and the accompanying current noise (n = 17 neurons). To further characterize the observed inhibition of the slow inward current, a dose-response curve was constructed. The EC50 of riluzole-induced inhibition was between 1 and 2 µM (Fig. 2C). Thus INa,S in SCN neurons was directed inward between –65 and –30 mV, blocked by TTX and riluzole at appropriate concentrations, exhibited threshold properties and substantial noise typical for INa,P; and thus INa,S possessed the main features of INa,P documented on other neurons.
|
To assess the role of INa,S recorded during depolarizing ramps in the properties of spontaneously firing neurons, we studied INa,S in voltage-clamp experiments that were conducted just before and after recording of spontaneous activity of the same SCN neuron in current-clamp mode. These two types of data were plotted on the same voltage axis, and one could see that action potential threshold and the development of INa,S were in register during the depolarizing ramp between spontaneously occurring action potentials (Fig. 3A; n = 6 neurons).
From experiments on other neurons, the amplitude of INa,P depends on the speed of the voltage ramp (Agrawal et al. 2001; Del Negro et al. 2002a). To characterize INa,S in SCN neurons and estimate the contribution of this current to the depolarizing ramp under current-clamp conditions, we studied the dependence of INa,S on ramp speed. The multispeed voltage-ramp protocol revealed that slower ramp speeds progressively attenuated the amplitude of the slow inward current (Fig. 3B), reflecting the slow inactivation kinetics of INa,S (similar to INa,P in other neurons) (Agrawal et al. 2001; Del Negro et al. 2002a; Fleidervish and Gutnick 1996). The INa,S amplitude, measured at the peak of the I-V relationship (Fig. 3B) and plotted as a function of the inverse of the ramp slope (Fig. 3C), demonstrated a decay that approximated a single exponential. Magistretti and Alonso (1999) described a biexponential decay, but we observed a monoexponential decay in all experiments with the clear absence of partially unclamped action-potential-generating fast Na+ current near the INa,S maximum. The slope constant of INa,S-ramp slope dependence was 137 ± 6 mV/s and I0 (see Fig. 3C, legend) was 28.3 ± 1.9 pA (n = 6 neurons).
The landmark of INa,P is its clockwise hysteresis that can be recorded in response to triangular (ascending and then descending) voltage-ramp commands (Powers and Binder 2003). Hysteresis was observed in all neurons studied (n = 53 neurons) and was present when the descending ramp was begun from membrane potentials more positive than the maximum of INa,S (Fig. 3D). It seems unlikely that the hysteresis was due to activation of K+ outward current, although a small contribution of voltage-activated outward current above –45 mV obviously takes place. Indeed, we observed small outward current during TTX (Fig. 2A) or riluzole (Fig. 2B) applications in this region of membrane potential with the same ramp protocol (see also Fig. 3B). Besides, we observed hysteresis of INa,S in isolated SCN neurons at conditions when outward K+ currents were completely eliminated (not shown here). A dendritic origin of hysteresis (Powers and Binder 2003) was also excluded because of the complete absence of dendrites in these acutely isolated neurons. Considering the dependence of INa,S on ramp speed (Fig. 3, B and C), hysteresis was probably due to inactivation of INa,S. Furthermore, comparing the basic slope constant for cortical neurons (20 mV/s in Magistretti and Alonso 1999) and for SCN neurons (137 mV/s, Fig. 3C), we should expect inactivation of INa,S approximately sevenfold faster for SCN neurons in comparison with INa,P for cortical neurons.
To address this issue directly, membrane currents were recorded when the ascending voltage-ramp command was changed to constant levels at different phases of development of INa,S. This protocol allowed us to avoid activation of fast Na+ current. As shown in Fig. 3E, the leak-subtracted INa,S revealed significant time-dependent inactivation when voltage was clamped at a constant level. A noninactivating component ranged between 4 and 20% (8.5 ± 2.3 pA; n = 6 neurons) of maximal value of INa,S, and it possibly represented the actual "persistent" Na+ current in SCN neurons. The time course of inactivation of INa,S could be fitted by a monoexponential function. The time constant of inactivation of the main component of INa,S depended on the holding potential and decreased with depolarization (Fig. 3F, n = 5 neurons), varying between 50 and 700 ms. The data from the latter experiment suggest that inactivation was partially responsible for the sharp decrease in INa,S amplitude after the maximal value was achieved (Figs. 2, A and B, and 3, A, B, and D).
INa,S in silent SCN neurons
From the experiments described above (Fig. 3A), INa,S should be an essential component of the mechanism for spontaneous activity of SCN neurons. In light of this, it seemed interesting to determine whether silent neurons exhibit this current or whether it is an attribute of spontaneously active neurons only. Five silent SCN neurons were studied, and representative results are shown in Fig. 4. To determine whether a neuron was actually silent, its extracellular activity in cell-attached mode was first recorded (Fig. 4A). In whole cell mode, these neurons had a resting potential of approximately –80 mV, and depolarizing pulses of appropriate amplitude in current clamp evoked action potentials (Fig. 4B). In all silent neurons studied, the regular ramp protocol revealed an INa,S (Fig. 4C) that exhibited all of the properties of INa,S in spontaneously active neurons (not shown here), including riluzole sensitivity (Fig. 4D). However, in contrast with spontaneously active cells, the I-V relationship of silent neurons intersected the zero-current axis in the subthreshold region (Fig. 4C), and the potential of intersection corresponded as expected to the resting potential of the neuron recorded in current-clamp mode (Fig. 4B). Thus INa,S was present in silent neurons, but it alone was not sufficient to produce net inward current on the total I-V relationship of SCN neurons in the subthreshold region (Fig. 4C).
Effect of riluzole on isolated neurons
Riluzole has been used previously to estimate the contribution of INa,P to the generation of the respiratory rhythm in mammals (Del Negro et al. 2002b; Koizumi and Smith 2002). To estimate the contribution of INa,S to spontaneous activity of isolated SCN neurons, puff applications of 5–100 µM riluzole-containing ACSF were delivered to the recorded neurons. Because the spontaneous activity of isolated SCN neurons was mostly irregular, an application of 1–3 s was delivered in a repetitive manner at equal intervals (2–10 min) that depended on the recovery from riluzole after the first application. The time required for washout appeared to reflect primarily the location of the neuron in the dish and the variability in the flow of ACSF rather than the individual properties of different SCN neurons, although this issue was not actually studied. Figure 5A shows representative traces that demonstrate suppression of neuron firing in cell-attached mode after application of 10 µM riluzole. This concentration of drug produced 75%-inhibition of INa,S in SCN neurons (Fig. 2C) and a 10 mV shift of steady-state inactivation of fast Na+ current (Urbani and Belluzzi 2000). The presence of single spikes on the background of suppression of firing suggests that the main target for riluzole is INa,S and not a steady-state inactivation of fast Na+ current. Whole cell experiments (n = 5 neurons) showed that the suppressing effect of riluzole on spontaneous firing was due predominantly to hyperpolarization, which obviously was associated with inhibition of INa,S (Fig. 5B). In all cases (n = 16 neurons), application of riluzole resulted in transient suppression of spontaneous activity in both cell-attached and whole cell recordings. In most neurons, application of drug did not exert any effect on the parameters of the action potential (amplitude and duration), but in some cases (5 of 22 applications of riluzole in whole cell recording), riluzole decreased action-potential amplitude (compare, for example, Fig. 5B, top vs. bottom). This effect was considered to be due to the deteriorating state of the neuron after the long-lasting recording (see DISCUSSION for details). We did not find any effect of riluzole application on the linear portion of leakage current in any of the neurons studied or on resting potential of silent SCN neurons in whole cell experiments (n = 4 neurons).
|
To estimate the contribution of INa,S to spontaneous activity of intact SCN neurons, bath application of 20 µM riluzole on electrical firing of neurons was studied in the slice preparation. Recordings were usually from two neurons simultaneously. From 14 neurons tested, riluzole partially or completely suppressed activity of nine cells and exerted a weak effect or no effect on five cells. Representative effects of riluzole on the firing frequency of two neurons is presented in Fig. 6A. The effect of riluzole was reversible within 30 min after replacement of solution. Analysis of the data suggests that if 4 riluzole-resistant neurons (Fig. 6B, 
) were excluded, the suppressing effect of riluzole was inversely correlated to the average firing rate (Fig. 6B, P < 0.001). We hypothesized that riluzole-resistant electrical firing and TTX-resistant membrane-potential oscillations (Kononenko and Dudek 2002; Pennartz et al. 2002; N. I. Kononenko, unpublished results) are manifestations of the same mechanism. Thus SCN neurons, whose activity was completely or partially riluzole-resistant, would thus be expected to exhibit TTX-resistant electrical oscillations. In these neurons, riluzole- and TTX-resistant mechanisms hypothetically operate together with INa,S to generate spontaneous electrical activity. To test this hypothesis, we studied the effect of 1 µM TTX on the electrical activity of neurons whose activity was not affected or was inhibited by no more than 50% by riluzole. Indeed, five of five tested riluzole-resistant SCN neurons (if activity was recorded simultaneously from 2 cells, riluzole-sensitive cells were excluded) exhibited TTX-resistant electrical oscillations (Fig. 6C).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Comparison with other currents
As described in the present work, INa,S shares virtually all of the properties of the well-known INa,P in other mammalian neurons. These properties include 1) activation threshold using a ramp protocol (30–120 mV/s) at –65 to –60 mV, and a peak at approximately –45 mV, 2) TTX sensitivity (Agrawal et al. 2001; Alzheimer et al. 1993; Chao and Alzheimer 1995; Del Negro et al. 2002a,b; Fleidervish and Gutnick 1996; Magistretti and Alonso 1999; Taddese and Bean 2002; Urbani and Belluzzi 2000), 3) suppression by riluzole with EC50 0.5–3 µM (Del Negro et al. 2002b; Spadoni et al. 2002; Urbani and Belluzzi 2000), and 4) a substantial increase in current noise accompanying the current activation (Agrawal et al. 2001; Chao and Alzheimer 1995; Del Negro et al. 2002b; Rybak et al. 2003; Spadoni et al. 2002). The only apparent difference between the current studied in our present experiments and INa,P is its inactivation. This inactivation was most clearly expressed when INa,S achieved its maximum with ascending ramp commands and then membrane potential was clamped at a constant level (Fig. 3E). As was expected from comparing slope constants of the current-ramp slope dependences for cortical (Magistretti and Alonso 1999) and SCN neurons (Fig. 3C), direct measurement of time constants of inactivation yielded approximately sevenfold faster value for SCN neurons (50–700 ms) against 3.4–6.8 s for cortical neurons. Similar if not identical to INa,S, the slowly inactivating component of Na+ current has been observed in SCN neurons (Pennartz et al. 1997) and cerebellar Purkinje cells (Kay et al. 1998) using a step-command protocol.
Properties of INa,S
Our attention focused on the electrical properties that could be directly connected with the spontaneous firing activity of SCN neurons. The threshold for INa,S corresponds to the pacemaker potential in spontaneously active SCN neurons. During a current-clamp recording of spontaneous activity, inactivation of K+ current and the corresponding decrease of the hyperpolarizing afterpotential depolarizes the membrane potential and achieves threshold for INa,S (Fig. 3A). Estimation of even the minimal voltage speed during the depolarizing ramp in a spontaneously active neuron (i.e., in the middle of the interspike interval) yields a value of 60 mV/s, which should produce an INa,S that is 65 pA (Fig. 3C). The voltage speed during the hyperpolarizing afterpotential and the corresponding INa,S are much larger. Such inward current is sufficient to evoke an appropriate depolarizing driving force in SCN neurons whose linear input resistance in the slice preparation is 1 G (Pennartz et al. 1997). Thus our recordings of INa,S under voltage-clamp conditions reveal its characteristics at the relevant physiological conditions during current-clamp recordings.
Like INa,P in other mammalian neurons, the amplitude of INa,S in SCN cells exhibited dependence on ramp-speed. Agrawal et al. (2001) and Del Negro et al. (2002a) considered the ramp-speed dependence for INa,P as an indirect measure of inactivation. In our experiments, ramp-speed dependence of INa,S was described perfectly by a single exponential (Fig. 3C), whereas Magistretti and Alonso (1999) reported biexponential decay; they suggested that the existence of a faster exponential component may be due to the presence of Na+ current components that are kinetically intermediate between fast Na+ current and the persistent Na+ current. Our precaution of discarding recordings with unclamped action-potential-generating fast Na+ current possibly excluded the faster component from our analysis. Alternatively, different neuronal populations may express different forms of slowly inactivating Na currents.
A defining property of INa,S was the hysteresis observed at the ascending and descending components of the ramp protocol (Fig. 3D). The simplest interpretation of this phenomenon, consistent with the ramp-speed dependence of INa,S (Fig. 3, B and C), is that the hysteresis reflects inactivation of current at membrane depolarization. Indeed, clamping of the membrane potential at the peak of INa,S revealed its inactivation (Fig. 3E) with a time constant (i) of 50–250 ms at its maximum. Clamping before the maximum current (i.e., at more negative membrane potentials) produced an increase of i in all neurons studied (Fig. 3, F and E). It is still uncertain whether the increase of i in this case is due to hyperpolarization of the membrane or to the decreased amplitude of INa,S itself, and further studies using a voltage-step protocol are needed. Overall, our data are consistent with the interpretation that INa,S undergoes inactivation, and its decrease after reaching the maximum (i.e., at approximately –45 mV) is due to inactivation rather than a decrease in electrical driving force. Previously, the time constant of inactivation for slowly inactivating Na+ current in SCN neurons was reported as 10 ms (Pennartz et al. 1997), which is much faster that in our experiments (50–250 ms). One explanation could be an additional fast Na+ current component in SCN neurons, as suggested by Magistretti and Alonso (1999). Alternatively, Pennartz et al. (1997) used a step protocol for the recording of slowly inactivating Na+ current, and contamination of total current produced by fast Na+ current could not be excluded. Based on our experiments, we hypothesize a physiological role of inactivation of INa,S during sustained electrical firing of SCN neurons: after achievement of threshold for action-potential generation, inactivation avoids large (tens or hundreds of pA) depolarizing current that could prevent complete repolarization of the membrane; this would be sufficient to deinactivate the fast Na+ current for subsequent action-potential generation.
Previous attempts to find a correlation between the pattern of electrical activity and the properties of INa,P have been undertaken in the pre-Bötzinger complex inspiratory neurons (Del Negro et al. 2002a). In light of this, it was interesting to determine whether spontaneously active and silent SCN neurons express different levels of INa,P. Is expression of INa,P sufficient for pacemaker activity? Our experiments did not reveal any significant differences in the properties of INa,P between spontaneously active and silent neurons (Fig. 4C). Also, we did not find any significant differences in input resistance between spontaneously active and silent neurons (not shown here). For silent neurons, only an upward parallel shift of the I-V relationship was observed, with a corresponding intersection of the I-V relationship with the zero-current axis in the subthreshold region. Thus to explain this observation, one can hypothesize that an additional depolarizing force besides INa,S, needs to be expressed in spontaneously active SCN neurons.
Effect of riluzole on spontaneous firing
The blockage of INa,P by riluzole has successfully been used to estimate directly the participation of INa,P in the generation of the respiratory rhythm in mammals (Del Negro et al. 2002b; Koizumi and Smith 2002). Although riluzole at a concentration that is sufficient to significantly suppress INa,P (10 µM) slightly shifts the steady-state inactivation of fast transient sodium current (Urbani and Belluzzi 2000), it did not exert a visible effect on the amplitude of action potentials (Del Negro et al. 2002b). In our experiments, puff application of riluzole (5 µM) resulted in hyperpolarization of the membrane and a transient suppression of spontaneous activity in all neurons studied (Fig. 5). Together with this observation, riluzole did not hyperpolarize the membrane of silent neurons. This is in accordance with the suggestion that the target for riluzole is the INa,P, and the threshold of INa,P is more positive than the resting potential of silent neurons (Fig. 4, C and D). Thus these results provide additional evidence that INa,P participates in spontaneous electrical firing in isolated SCN neurons.
In some cases, we observed a suppressing effect of riluzole on the action-potential amplitude, especially during long-lasting recordings and after a few applications of drug (Fig. 5B, top and bottom). Our present interpretation of this result is that INa,S contributes to both the apparent neuronal leakage current and the general Na+ current. If a neuron is healthy (i.e., fast INa is much greater than leakage current), the effect of riluzole on action-potential amplitude is negligible. After prolonged whole cell recording, the leakage current could increase while fast INa decreased. In these cases, leakage and fast INa become comparable, and riluzole could then decrease action-potential amplitude.
Although a suppressing effect of riluzole (20 µM) was also generally observed on the electrical firing of SCN neurons in the slice preparation, there were quantitative differences (Fig. 6A). First, some neurons were insensitive to riluzole. Second, the effectiveness of riluzole on the riluzole-sensitive neurons depended on their firing rate: neurons possessing higher firing rate were less sensitive to riluzole. A TTX-resistant mechanism for oscillations in membrane potential has been shown for SCN neurons (Kononenko and Dudek 2002; Pennartz et al. 2002; N. I. Kononenko, unpublished data), and this mechanism alone could permit riluzole-resistant firing. Experiments with TTX application on riluzole-insensitive (completely or partially) neurons in slice preparations suggested that INa,S alone is not absolutely required for spontaneous firing (Fig. 6C). TTX application suppressed action-potential generation, but low-amplitude oscillations in membrane potential were still present in whole cell recordings (Kononenko and Dudek 2002; Pennartz et al. 2002). At present, there are two different hypotheses regarding TTX-resistant mechanisms of pacemaker oscillations in SCN neurons; further studies are needed to determine whether L-type Ca2+-channels (Pennartz et al. 2002) or subthreshold voltage-dependent cation (SVC) channels (Kononenko and Dudek 2002; N. I. Kononenko, unpublished results) underlie pacemaker oscillations.
Putative clinical effect of riluzole
At the recommended daily dose, the maximal plasma level of riluzole ranges between 0.9 and 1.6 µM (Urbani and Belluzzi 2000; based on Le Liboux et al. 1997). The EC50 for INa,S in SCN neurons is 1–2 µM (Fig. 2C), and 10 µM produced complete suppression of spontaneous firing (Fig. 5). Based on the role of SCN neuronal firing in circadian behavior of mammals, riluzole could alter circadian rhythms at clinical levels.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
GRANTS
This research was supported by National Institute of Mental Health Grant MH-59995.
|
FOOTNOTES |
|---|
Address reprint requests and other correspondence to: F. Edward Dudek (E-mail: ed.dudek{at}colostate.edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Alzheimer C, Schwindt PC, and Crill WE. Postnatal development of a persistent Na+ current in pyramidal neurons from rat sensorimotor cortex. J Neurophysiol 69: 290–292, 1993.
Bevan MD and Wilson CJ. Mechanisms underlying spontaneous oscillation and rhythmic firing in rat subthalamic neurons. J Neurosci 19: 7617–7628, 1999.
Chao TI and Alzheimer C. Effects of phenytoin on the persistent Na+ current of mammalian CNS neurons. Neuroreport 6: 1778–1780, 1995.[ISI][Medline]
Crill WE. Persistent sodium current in mammalian central neurons. Annu Rev Physiol 58: 349–362, 1996.[CrossRef][ISI][Medline]
Del Negro CA, Koshiya N, Butera RJ Jr, and Smith JC. Persistent sodium current, membrane properties and bursting behavior of pre-Bötzinger complex inspiratory neurons in vitro. J Neurophysiol 88: 2242–2250, 2002a.
Del Negro CA, Morgado-Valle C, and Feldman JL. Respiratory rhythm: an emergent network property? Neuron 34: 821–830, 2002b.[CrossRef][ISI][Medline]
Fleidervish IA and Gutnick MJ. Kinetics of slow inactivation of persistent sodium current in layer V neurons of mouse neocortical slices. J Neurophysiol 76: 2125–2130, 1996.
Gillette MU and Tischkau SA. Suprachiasmatic nucleus: the brain's circadian clock. Recent Prog Horm Res 54: 33–58, 1999.[Medline]
Hammarström AKM and Gage PW. Nitric oxide increases persistent sodium current in rat hippocampal neurons. J Physiol 520: 451–461, 1999.
Honma S, Shirakawa T, Katsuno Y, Namihira M, and Honma K. Circadian periods of single suprachiasmatic neurons in rats. Neurosci Lett 250: 157–160, 1998.[CrossRef][ISI][Medline]
Kay AR, Sugimori M, and Llinas R. Kinetic and stochastic properties of a persistent sodium current in mature guinea pig cerebellar Purkinje cells. J Neurophysiol 80: 1167–1179, 1998.
Koizumi H and Smith JC. Perturbations of respiratory pattern and rhythm in vitro by block of persistent sodium current (INaP). Soc Neurosci Abstr 28: 173.6, 2002.
Kononenko NI and Dudek FE. Stationary potential-dependent TTX-resistant single channels underlying spontaneous firing of SCN neurons. Soc Neurosci Abstr 28: 834.5, 2002.
Le Liboux A, Lefebvre P, Le Roux Y, Truffinet P, Aubeneau M, Kirkesseli S, and Montay G. Single- and multiple-dose pharmacokinetics of riluzole in white subjects. J Clin Pharmacol 37: 820–827, 1997.
Liu C and Reppert SM. GABA synchronizes clock cells within the suprachiasmatic circadian clock. Neuron 25: 123–128, 2000.[CrossRef][ISI][Medline]
Llinas RR. The intrinsic electrophysiological properties of mammalian neurons: insights into central nervous system function. Science 242: 1654–1664, 1988.
Magistretti J and Alonso A. Biophysical properties and slow voltage-dependent inactivation of a sustained sodium current in entorhinal cortex layer-II principal neurons: a whole-cell and single-channel study. J Gen Physiol 114: 491–509, 1999.
Magistretti J, Ragsdale DS, and Alonso A. High conductance sustained single-channel activity responsible for the low-threshold persistent Na+ current in entorhinal cortex neurons. J Neurosci 19: 7334–7341, 1999.
Neher E. Correction for liquid junction potentials in patch clamp experiments. Methods Enzimol 207: 123–131, 1992.
Pennartz CM, Bierlaagh MA, and Geurtsen AM. Cellular mechanisms underlying spontaneous firing in rat suprachiasmatic nucleus: involvement of a slowly inactivating component of sodium current. J Neurophysiol 78: 1811–1825, 1997.
Pennartz CM, de Jeu MT, Bos NP, Schaap J, and Geurtsen AM. Diurnal modulation of pacemaker potentials and calcium current in the mammalian circadian clock. Nature 416: 286–290, 2002.[CrossRef][Medline]
Powers RK and Binder MD. Persistent sodium and calcium currents in rat hypoglossal motoneurons. J Neurophysiol 89: 615–624, 2003.
Rybak IA, Ptak K, Shevtsova NA, and McCrimmon DR. Sodium currents in neurons from the rostroventrolateral medulla of the rat. J Neurophysiol 90: 1635–1642, 2003.
Schwindt P and Crill WE. A persistent negative resistance in cat lumbar motoneurons. Brain Res 120: 173–178, 1977.[CrossRef][ISI][Medline]
Shirakawa T, Honma S, Katsuno Y, Oguchi H, and Honma KI. Synchronization of circadian firing rhythms in cultured rat suprachiasmatic neurons. Eur J Neurosci 12: 2833–2838, 2000.[CrossRef][ISI][Medline]
Spadoni F, Hainsworth AH, Mercuri NB, Caputi L, Martella G, Lavaroni F, Bernardi G, and Stefani A. Lamotrigine derivatives and riluzole inhibit INa,P in cortical neurons. Neuroreport 13: 1167–1170, 2002.[CrossRef][ISI][Medline]
Taddese A and Bean BP. Subthreshold sodium current from rapidly inactivating sodium channels drives spontaneous firing of tuberomammillary neurons. Neuron 33: 587–600, 2002.[CrossRef][ISI][Medline]
Urbani A and Belluzzi O. Riluzole inhibits the persistent sodium current in mammalian CNS neurons. Eur J Neurosci 12: 3567–3574, 2000.[CrossRef][ISI][Medline]
Uteshev V, Stevens DR, and Haas HL. A persistent sodium current in acutely isolated histaminergic neurons from rat hypothalamus. Neuroscience 66: 143–149, 1995.[CrossRef][ISI][Medline]
Walsh IB, van den Berg RJ, Marani E, and Rietveld WJ. Spontaneous and stimulated firing in cultured rat suprachiasmatic neurons. Brain Res 588: 120–131, 1992.[CrossRef][ISI][Medline]
Welsh DK, Logothetis DE, Meister M, and Reppert SM. Individual neurons dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron 14: 697–706, 1995.[CrossRef][ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  Choon Kiat Sim and D. B. Forger Modeling the Electrophysiology of Suprachiasmatic Nucleus Neurons J Biol Rhythms, October 1, 2007; 22(5): 445 - 453. [Abstract] [PDF]
|
|
|
|
 |
|
 S. J. Kuhlman Biological Rhythms Workshop IB: Neurophysiology of SCN Pacemaker Function Cold Spring Harb Symp Quant Biol, January 1, 2007; 72(0): 21 - 33. [Abstract] [PDF]
|
|
|
|
 |
|
 M. A. Castro-Alamancos, P. Rigas, and Y. Tawara-Hirata Resonance (~10 Hz) of excitatory networks in motor cortex: effects of voltage-dependent ion channel blockers J. Physiol., January 1, 2007; 578(1): 173 - 191. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Kuhlman and D. G. McMahon Encoding the Ins and Outs of Circadian Pacemaking. J Biol Rhythms, December 1, 2006; 21(6): 470 - 481. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Paydarfar, D. B. Forger, and J. R. Clay Noisy Inputs and the Induction of On-Off Switching Behavior in a Neuronal Pacemaker J Neurophysiol, December 1, 2006; 96(6): 3338 - 3348. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. V. S. Faustino and D. F. Donnelly An important functional role of persistent Na+ current in carotid body hypoxia transduction J Appl Physiol, October 1, 2006; 101(4): 1076 - 1084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Kuo, R. H. Lee, L. Zhang, and C. J. Heckman Essential role of the persistent sodium current in spike initiation during slowly rising inputs in mouse spinal neurones J. Physiol., August 1, 2006; 574(3): 819 - 834. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. K. Purvis and R. J. Butera Ionic Current Model of a Hypoglossal Motoneuron J Neurophysiol, February 1, 2005; 93(2): 723 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
on the graph) were fitted by linear regression (P < 0.001). 
, application of 1 µM TTX-containing ACSF. Bottom: fragments at high temporal and amplitude resolution show current spikes before (left) and after (right) TTX application.