INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Repetitive electrical stimulation and intense neuronal activity produce dramatic decreases (up to 0.5 mM) of extracellular calcium ([Ca²⁺]_e) in the CNS (Heinemann and Louvel 1983; Heinemann et al. 1977; Krnjevic et al. 1982; Nicholson et al. 1978; Pumain and Heinemann 1985; Somjen 1980). Decreases in [Ca²⁺]_e were also evoked by iontophoretic applications of excitatory amino acids (Heinemann and Pumain 1980). These decreases of [Ca²⁺]_e were largely due to excessive release of excitatory neurotransmitter and/or activation of postsynaptic glutamate receptors.

Decreases of [Ca²⁺]_e are dramatically enhanced during seizure activity (Heinemann and Louvel 1983; Heinemann et al. 1977). In pentetrazol-induced seizure, for example, a decrease of [Ca²⁺]_e by 0.7-1.0 mM was recorded (Heinemann and Louvel 1983; Heinemann et al. 1977). It has been noted that the fall of [Ca²⁺]_e often preceded the onset of seizure events, indicating that the fall of [Ca²⁺]_e might be responsible for initiating the seizure activities. Decrease of [Ca²⁺]_e has also been observed in chronic models of epilepsy including the kindling model and photically induced seizures (Davies and Peterson 1989; Heinemann and Hamon 1986).

Decreases in [Ca²⁺]_e are known to increase neuronal excitability (Hille 1992). The mechanism by which lowering [Ca²⁺]_e enhances neuronal excitability is, however, not fully understood. Lowering [Ca²⁺]_e reduces the shielding of negatively charged groups located at the membrane surface (Hille 1992; Zhou and Jones 1995). By this mechanism, Ca²⁺ may influence the voltage-dependent activation of various ion channels (Hille 1992). Calcium also alters the gating and the permeation properties of several ion channels (Zhou and Jones 1995). In some extreme cases, channel selectivity is lost when Ca²⁺ is reduced to <1 µM. For example, Na⁺ will readily permeate L-type Ca²⁺ channels when [Ca²⁺]_e is lowered to the nanomolar range (Almers and McCleskey 1984; Fukushima and Hagiwara 1985; Hess et al. 1986; Matsuda 1986). It should be noted that, in most of these studies, [Ca²⁺]_e needs to be reduced to an extremely low level (e.g., nanomolar) to have its effect. Such low value is not expected in physiological conditions and rarely seen during seizure activity.

We have recently demonstrated that decreases in [Ca²⁺]_e activate a calcium-sensing nonselective cation (csNSC) channel in cultured mouse hippocampal neurons (Xiong and MacDonald 1999; Xiong et al. 1997). This channel is tonically inhibited by calcium concentration higher than 1.5 mM and activated when [Ca²⁺]_e is lowered. Activation of the csNSC channel is responsible for a sustained membrane depolarization and excitation of neurons (Xiong et al. 1997). As decreases in [Ca²⁺]_e often precede or accompany seizure activities (Heinemann and Louvel 1983; Heinemann et al. 1977), it is expected that activation of the csNSC channel is involved in epileptogenesis. Blocking the csNSC channel is therefore anticipated to have anti-epileptic effect.

Lamotrigine (LTG) is a novel anti-epileptic agent with a broad spectrum of clinical effects (Coulter 1997; Matsuo 1999). Its primary documented cellular mechanism is the blockade of voltage-gated Na⁺ channel (Cheung et al. 1992; Meldrum 1996; Xie et al. 1995). However, this mechanism alone is not sufficient to explain its broad clinical anti-epileptic effect. In the present study, we have studied the effects of LTG on csNSC currents in cultured hippocampal neurons. We report here that, in addition to inhibiting voltage-gated Na⁺ and Ca²⁺ channels, LTG dose dependently depresses csNSC currents in hippocampal neurons. The inhibition of LTG on csNSC channels may partially contribute to its broad spectrum of anti-epileptic effects.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dissociation and culture of mouse hippocampal neurons

Cultures of mouse hippocampal neurons were prepared according to our previously described techniques (MacDonald et al. 1989). The use of mouse brain for neuronal culture was reviewed and approved by the Institutional Animal Care and Use Committee of the University of Toronto and Legacy Clinical Research and Technology Center. Briefly, time-pregnant (E15) mice were anesthetized with halothane followed by cervical dislocation. Fetuses were rapidly removed, and hippocampi were dissected and placed in cold Hanks' solution. The hippocampi were then mechanically dissociated by trituration and plated in 35-mm collagen-coated culture dishes or 25-mm coverslips at densities of ~1 × 10⁶/dish. The cultures were used for electrophysiological recordings 14-21 days after plating.

Electrophysiology

Whole cell patch-clamp recordings were performed and analyzed as described previously (Xiong et al. 1997). Patch electrodes were constructed from thin-walled borosilicate glass (1.5-mm diam, WPI, Sarasota, FL) on a two-stage puller (PP83, Narishige, Tokyo). The tips of the electrodes were heat-polished on a Narishige microforge (Model MF-83; Scientific Instruments Laboratory, Tokyo) to a final diameter of 1-2 µm. The patch electrodes had resistance between 3 and 5 M when filled with intracellular solution. Whole cell currents were recorded using Axopatch 1-D amplifiers (Axon Instruments, Foster City, CA). Data were filtered at 2 kHz and digitized at 5 kHz using Digidata 1200 DAC units (Axon Instruments). The on-line acquisition was done using pClamp software (version 6, Axon Instruments). During each experiment, a voltage step of 10 mV was applied periodically to monitor the cell capacitance and the access resistance. Recordings in which the access resistance or the capacitance changed by more than 10% during the recordings were not included in data analysis (Xiong et al. 1998).

Ca²⁺ imaging

Ca²⁺-imaging experiments were performed as previously described (Jarvis et al. 1997; Xiong et al. 2000). Cultured hippocampal neurons grown on 25 × 25-mm glass coverslips were washed three times with extracellular solution and incubated with 5 µM fura-2-acetoxymethyl ester for 40-50 min at room temperature. Coverslips with fura-2-loaded cells were then transferred to a perfusion chamber on an inverted microscope (Nikon). Cells were illuminated using a xenon lamp (75 W) and observed with a ×40 UV fluor oil-immersion objective lens (Nikon). Video images were obtained using an intensified CCD camera (PTI IC-110). Digitized images were acquired by averaging four to eight frames at video rates using an image processing board (Axon imaging lightening) in a PC-type computer controlled by Axon Imaging Workbench software (AIW2.1, Axon Instruments). The shutter and filter wheel were also controlled by AIW to allow timed illumination of cells at either 340- or 380-nm excitation wavelengths. Fluorescence of fura-2 was detected at an emission wavelength of 510 nm. Ratio images were analyzed by averaging pixel ratio values in circumscribed regions of cells in the field of view. The values were exported from AIW to Sigmaplot 4.0 and then plotted.

Solutions and chemicals

Standard extracellular solution contained (in mM) 140 NaCl, 5.4 KCl, 1.5 CaCl₂, 25 HEPES, and 33 glucose (pH 7.4 with NaOH; 320-330 mOsm). Tetrodotoxin (TTX) 1 µM was added for voltage-clamp experiment while left out for current-clamp recording to study the firing of the action potentials.

For voltage-clamp experiments, the pipette solution contained (in mM) 140 CsF, 30 CsOH, 10 HEPES, 11 EGTA, 2 tetraethylammonium chloride (TEA), 1 CaCl₂, 4 MgCl₂, and 4 K⁺-ATP (pH 7.3 with CsOH; 310 mOsm). For current-clamp recordings, the pipette solution contained (in mM) 150 KCl, 10 HEPES, 11 EGTA, 2 tetraethylammonium chloride (TEA), 1 CaCl₂, 4 MgCl₂, and 4 K⁺-ATP (pH 7.3 with KOH; 310 mOsm). A multi-barrel perfusion system was employed to achieve a rapid exchange of solutions.

All experiments were performed at room temperature (2224°C). Data are expressed as means ± SE. Student's t-test was employed for the analysis of statistical significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LTG inhibits low-Ca²⁺-induced membrane depolarization and excitation

The effect of LTG on membrane potential of cultured hippocampal neurons was first studied using the current-clamp configuration. In the absence of TTX, the majority of neurons displayed spontaneous firing of action potentials at resting potentials (58 ± 3 mV, n = 5). The frequency of the firing of action potentials was largely dependent on membrane potentials. For example, hyperpolarization by injecting a small outward current reduced or eliminated firing while depolarization by injecting inward current increased the frequency of action potentials. As described previously (Xiong et al. 1997), lowering [Ca²⁺]_e induced a sustained depolarization of membrane potential and an increase in the action potential firing rate (Fig. 1). In five neurons, reduction in [Ca²⁺]_e from 1.5 to 0.5 mM depolarized the membrane potential by 13.6 ± 1.8 mV (n = 5) and increased frequency of action potentials from 0.45 ± 0.2 Hz to 10.5 ± 1.2 Hz (P < 0.01; Fig. 1A). We have previously demonstrated that this excitation of neurons in response to decreased [Ca²⁺]_e was due to the activation of csNSC channels (Xiong et al. 1997). To study the effect of LTG on csNSC-mediated membrane depolarization and the excitation of hippocampal neurons, LTG was co-applied with 0.5 mM Ca²⁺. As shown in Fig. 1, addition of LTG (300 µM) significantly decreased the amplitude of depolarization induced by 0.5 mM Ca²⁺ (4.8 ± 1.2 mV with LTG compared with 13.6 ± 1.8 mV without, P < 0.01, n = 5). The frequency of action potentials was also reduced by LTG from 10.5 ± 1.2 to 2 ± 0.7 Hz (P < 0.01, n = 4).

View larger version (30K): [in this window] [in a new window]   Fig. 1. The effect of lamotrigine (LTG) on calcium-sensing nonselective cation (csNSC) channel-mediated membrane depolarization and excitation of hippocampal neurons. A: the membrane potential of a cultured hippocampal neuron was recorded using the whole cell current-clamp configuration (KCl in the recording pipette). Decreasing Ca2+ from 1.5 to 0.5 mM for a period of 5 s depolarized and strongly excited the cell. Addition of LTG (300 µM) inhibited the depolarization and excitation induced by lowering the [Ca2+]e. B: in a different neuron, [Ca2+]e was first decreased to 0.75 mM and the membrane potential depolarized to 48 mV. Application of LTG (300 µM) induced membrane hyperpolarization and eliminated the firing of action potentials.

Figure 1B shows another example of the effect of LTG on low [Ca²⁺]_e-induced excitation. In this recording, the cell was first perfused with a lower Ca²⁺ (0.75 mM) solution to activate csNSC channels. This resulted in a depolarized membrane potential at 48 mV with high firing rate of action potentials. After stabilization of membrane potential, 300 µM LTG was applied to the cell for 5 s. LTG not only inhibited the firing of action potentials but also caused membrane hyperpolarization as expected from an inhibition of csNSC channels.

LTG has been shown to block voltage-gated Na⁺ channels (Xie et al. 1995). It may be argued that the effect of LTG on membrane potentials is simply due to its effect on Na⁺ channels. To address this, we have repeated the experiment using Na⁺ channel blocker TTX. TTX (0.5 µM) completely eliminated the firing of action potentials. However, unlike LTG, TTX had no effect on lowering [Ca²⁺]_e induced membrane depolarization (n = 3, not shown).

LTG dose dependently inhibits csNSC currents

To better characterize the effect of LTG on csNSC channel-mediated response, we then employed the voltage-clamp configuration to directly study the effect of LTG on csNSC currents. Cells were voltage clamped at a holding potential of 60 mV, and the currents through csNSC channels were activated by step reductions of [Ca²⁺]_e. As shown in Fig. 2, reduction of [Ca²⁺]_e from 1.5 to 0.2 mM induced a slowly activating sustained inward current as reported previously (Xiong et al. 1997). Addition of LTG in the low-Ca²⁺ solution caused a concentration-dependent inhibition of the inward currents with an IC₅₀ of 171 ± 26 µM and Hill coefficient of 1.1 ± 0.2 (n = 6). The threshold concentration required for LTG block of csNSC current is ~10 µM, close to its therapeutic plasma level.

View larger version (17K): [in this window] [in a new window]   Fig. 2. Dose-dependent inhibition of csNSC current by LTG in cultured hippocampal neuron. csNSC current was activated by a 2-s step reduction of [Ca2+]e from 1.5 to 0.2 mM. Holding potential was 60 mV. Addition of LTG in 0.2 mM Ca2+ solution dose dependently inhibited the csNSC current with an IC50 of 171 ± 25.8 µM and Hill coefficient of 1.1 ± 0.2 (n = 6).

Voltage-independent inhibition of csNSC current by LTG

We then tested whether the effect of LTG on the csNSC current depends on the membrane potential. Neurons were voltage clamped at different membrane potentials ranging from 60 to +40 mV, and the currents were activated by a step reduction of [Ca²⁺]_e from 1.5 to 0.2 mM. As shown previously (Xiong et al. 1997), currents through csNSC channels display a linear current-voltage (I-V) relationship with reversal potential of ~0 mV (Fig. 3). Co-application of LTG (300 µM) with 0.2 mM Ca²⁺ reduced the currents at both negative and positive membrane potentials to a similar extent, demonstrating a voltage-independent inhibition of csNSC channels by LTG (31 ± 3.0% inhibition at 60 mV, 34 ± 3.4% at 40 mV, and 29 ± 2.9% at +40 mV, n = 4-5, P > 0.05). Because LTG has been shown to inhibit N- and P-type Ca²⁺ channels, we have repeated the experiment in the presence of -conotoxin MVIIC, a toxin that blocks both N- and P-type Ca²⁺ channels (McDonough et al. 1996; Waterman 1997). In the presence of 1 µM -conotoxin MVIIC, LTG blocked the csNSC current to a similar degree as in the absence of -conotoxin MVIIC (35 ± 6.1% at 60 mV and 29 ± 3.6 at +40 mV, n = 4), indicating that the effect of LTG on csNSC currents has nothing to do with its effect on N- and P-type Ca²⁺ channels.

View larger version (21K): [in this window] [in a new window]   Fig. 3. The effect of LTG on csNSC current does not depend on membrane potentials. csNSC current was activated by a 2-s step reduction of [Ca2+]e from 1.5 to 0.2 mM with membrane potential held at different values ranging from 60 to +40 mV. LTG inhibited the csNSC current at both negative and positive potentials to a similar extent, demonstrating a voltage-independent effect (31 ± 3.0% inhibition at 60 mV and 29 ± 2.9% at +40 mV, n = 4-5, P > 0.05).

LTG does not affect the potency of Ca²⁺ block of csNSC current

We have shown previously that Ca²⁺ ion is an effective endogenous blocker of the csNSC channels (Xiong et al. 1997). To study the mechanism underlying LTG inhibition of csNSC currents, we have explored the possibility whether the effect of LTG was due to an increase in the potency of the Ca²⁺ blockade. The dose-inhibition curves of Ca²⁺ block of csNSC currents were therefore constructed in the absence and presence of LTG (300 µM). As shown in Fig. 4, LTG affected neither the shape nor the IC₅₀ value of the Ca²⁺ dose-inhibition relationship (IC₅₀ for Ca²⁺ block of csNSC current: 145 ± 18 µM in the absence vs. 136 ± 10 µM in the presence of LTG, n = 5, P > 0.05), indicating that the effect of LTG on csNSC currents was not due to an increase in the potency of Ca²⁺ blockade.

View larger version (15K): [in this window] [in a new window]   Fig. 4. Effect of LTG on dose-inhibition relationship of Ca2+ block of csNSC current. Dose-inhibition curves of Ca2+ block on sNSC currents was constructed in the absence and in the presence of LTG (300 µM). The IC50 values for Ca2+ block were 145 ± 18 µM in the absence and 136 ± 10 µM in the presence of LTG (n = 5, P > 0.05).

It has been shown that the effect of LTG on Na⁺ channels depends on the frequency of stimulation (use-dependent block) (Kuo and Lu 1997; Xie et al. 1995). This mechanism selectively dampens pathologic activation of Na⁺ channels without interacting with normal Na⁺ channel function. We have performed a similar experiment to see whether the effect of LTG on csNSC currents also depends on the frequency of channel activation. csNSC currents were activated repeatedly by step reductions of [Ca²⁺]_e with intervals between two activations as short as 1.0 s (Fig. 5). Unlike its effect on voltage-gated Na⁺ channels, the inhibition of LTG on csNSC currents was not enhanced with repeated stimulation, indicating a lack of use dependence of the effect of LTG on csNSC channels (Fig. 5).

View larger version (21K): [in this window] [in a new window]   Fig. 5. Lack of use dependence of LTG inhibition on csNSC current. A: example traces showing the csNSC current activated by repetitive reduction of [Ca2+]e from 1.5 to 0.2 mM in the absence and presence of 300 µM LTG. B: summary data showing the amplitude of csNSC current induced by repetitive reduction of [Ca2+]e both in the absence and presence of LTG.

Effects of other anti-epileptic agents on csNSC currents

Several studies have shown that anti-epileptic agents phenytoin and carbamazepine share similar pharmacological properties with LTG as Na⁺ channel blockers (Kuo and Lu 1997). It has been suggested that these agents bind to a common site located on the external side of neuronal Na⁺ channels (Kuo 1998). We therefore tested the possibility that phenytoin and carbamazepine also inhibit csNSC currents. Similar to LTG, carbamazepine (300 µM) depressed the amplitude of csNSC current by ~20% (n = 5) (Fig. 6). However, with concentrations as high as 1 mM, phenytoin had no effect on the current (not shown), indicating that the common motif responsible for blocking Na⁺ current is not responsible for the effect of LTG on csNSC channels. Other anti-epileptic agents including ethosuximide, felbamate, topiramate, and valproic acid had no effect on csNSC current with concentration as high as 1 mM.

View larger version (18K): [in this window] [in a new window]   Fig. 6. Effects of other anti-epileptic agents on csNSC current. Current was activated by a 2-s step reduction of [Ca2+]e from 1.5 to 0.2 mM. HP = 60 mV. Carbamazapine (300 µM) slightly inhibited the csNSC current, while felbamate and ethosuximide had no effect.

LTG inhibits csNSC-mediated increase in [Ca²⁺]_i

Fluorescent Ca²⁺-imaging technique was also used to study changes of intracellular Ca²⁺ concentration mediated by lowering [Ca²⁺]_e and the effect of LTG on the Ca²⁺ response. In a total of 58 cultured hippocampal neurons tested, ~30% of neurons responded to a step decrease of [Ca²⁺]_e (from 1.5 to 0.5 mM) with a detectable increase in intracellular Ca²⁺ concentration. The reason that the rest of neurons did not respond is not clear. One possibility is that the sensitivity of imaging system we used is not high enough to detect small changes. Nevertheless, for those neurons which responded, repeated increase in [Ca²⁺]_i without desensitization could be evoked with a decrease of [Ca²⁺]_e from 1.5 to 0.5 mM. We tested the effect of LTG only on the neurons that responded to low [Ca²⁺]_e. In 10 neurons tested for LTG, decrease of [Ca²⁺]_e from 1.5 to 0.5 mM induced an increase in 340/380 ratio from 0.4 to 1.1 (n = 10). LTG (300 µM) did not affect the baseline Ca²⁺ but reduced the low [Ca²⁺]_e induced increase in 340/380 ratio by 30 ± 5% (Fig. 7, P < 0.05, n = 10).

View larger version (35K): [in this window] [in a new window]   Fig. 7. Effect of LTG on intracellular Ca2+ response induced by lowering the [Ca2+]e. Step reduction of [Ca2+]e from 1.5 to 0.5 mM (for 5 s) increased the concentration of intracellular Ca2+. LTG (300 µM) did not affect the baseline of intracellular Ca2+ concentration but reduced the Ca2+ increase induced by lowering [Ca2+]e.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, we demonstrated that LTG decreases the excitability of cultured hippocampal neurons in the presence of reduced [Ca²⁺]_e, a condition that is commonly observed in CNS during seizure activities. We further demonstrated that the effect of LTG on membrane excitability was partially due to its inhibition of csNSC currents we previously described (Xiong et al. 1997). As activation of csNSC current is suggested to be involved in the generation of seizure activity (Xiong et al. 1997), we conclude that the inhibition of csNSC current by LTG contributes partially to its broad spectrum of anti-epileptic activity.

The mechanism of how LTG inhibits the csNSC current is not clear. As Ca²⁺ is an effective blocker of csNSC channels (Xiong et al. 1997), one possibility is that LTG increases the potency of Ca²⁺ blockade. We therefore compared the Ca²⁺ block of csNSC current in the absence and presence of LTG. The IC₅₀ value of Ca²⁺ block of csNSC current in the presence of LTG was the same as that without LTG, indicating that the inhibition of csNSC current by LTG was not due to an increase in the potency of Ca²⁺ block.

Phenytoin and carbamazepine share similar pharmacological properties as LTG in blocking the voltage-gated Na⁺ channel (Kuo and Lu 1997). It was suggested that these agents bind to a common site located on the external side of Na⁺ channels (Kuo 1998). We have compared the effect of LTG with that of phenytoin and carbamazepine. Similar to LTG, carbamazepine slightly depressed the csNSC current. However, with concentrations as high as 1 mM, phenytoin had no effect. This data suggest that the common motif responsible for blocking Na⁺ channels is not responsible for the effect of LTG on csNSC channels.

The effect of LTG on Na⁺ channels is voltage and activity dependent (Kuo and Lu 1997; Xie et al. 1995). This is due to its selective interaction with the inactivated state of the channel. Unlike its effect on Na⁺ channel, the effect of LTG on csNSC current does not depend on the voltage or the frequency of stimulation. This indicates that LTG would inhibit low-[Ca²⁺]_e-induced membrane depolarization and the excitation of neurons regardless of membrane potential and the frequency of action potential. This effect will likely reset the membrane potential to a lower level away from the threshold of Na⁺ channel activation hence reducing the overall excitability of neurons. This effect of LTG, combining with its direct inhibition of Na⁺ channels, would be effective in suppressing abnormal neuronal excitability in the presence of reduced [Ca²⁺]_e.

In ~30% of neurons, activation of csNSC channels induced an increase in [Ca²⁺]_i. Because this experiment was performed on neurons that were not voltage-clamped, whether the increase of Ca²⁺ was due to Ca²⁺ entry directly from csNSC channels or indirectly through voltage-gated Ca²⁺ channels is not clear. Further studies will be carried out in the presence of blockers of voltage-gated Ca²⁺ channels. Alternatively, we will use simultaneous patch-clamp and Ca²⁺-imaging technique to determine whether Ca²⁺ entry is directly through csNSC channels.

Current frontline anti-epileptic drugs fall into several cellular mechanistic categories. Drugs effective in control of partial and generalized tonic-clonic seizures are use- and voltage-dependent blockers of Na⁺ channels. Examples include phenytoin, carbamazepine, valproic acid, and lamotrigine (Coulter 1997; Macdonald and Greenfield 1997; Meldrum 1996). These agents selectively dampen pathologic activation of Na⁺ channels, without interacting with normal Na⁺ channel function. Drugs effective in control of generalized absence seizures likely block low-threshold T-type calcium currents. Examples include ethosuximide, trimethadione, and methsuximide (Coulter et al. 1989a,b). Agents that augment function of GABA_A receptors, e.g., diazepam and clonazepam, have broad-spectrum anti-epileptic effects (Coulter 1997; Robertson 1986).

LTG has a broad spectrum of clinical effects against various types of epilepsy. It is effective against both partial and generalized seizures, including absence seizures (Messenheimer 1995; Yuen 1994). Furthermore, LTG has also been used for the treatment of bipolar disorder (Calabrese and Gajwani 2000; Engle and Heck 2000) and pain (Eisenberg et al. 1998; Simpson et al. 2000; Zakrzewska et al. 1997). The primarily documented cellular mechanism of action is Na⁺ channel block, a mechanism shared by other anti-epileptic agents including phenytoin and carbamazepine. Unlike LTG, however, phenytoin and carbamazepine are ineffective against the absence seizure (Coulter 1997).

It is expected that the effects of LTG on other ion channels may combine with its Na⁺ channel blocking actions to account for its broad clinical efficacy. Recent studies have shown that in addition to blocking Na⁺ channel, LTG inhibits high-threshold voltage-gated Ca²⁺ channel (Grunze et al. 1998b; Lees and Leach 1993; Stefani et al. 1997; Wang et al. 1996). It has also been shown that, in hippocampal slices, LTG positively modulates a transient outward K⁺ current (Grunze et al. 1998a,b). Our present study demonstrated that LTG also concentration dependently depresses the csNSC current. This effect of LTG may partially contribute to its broad spectrum of anti-epileptic effects and its effect on affective disorders and pain. Our data also suggest that the csNSC channel may be a potential therapeutic target for epileptic seizures.