Introduction

Top Abstract Introduction Methods Results Discussion References

The therapeutic mechanisms of antiepileptic drugs are of much interest (for a review see Macdonald and Kelly, 1995). Many of these drugs modulate the function of ligand- and voltage-gated ion channels. For example, voltage clamp studies have demonstrated that benzodiazepines increase the open frequency and conductance of the GABA_A receptor/ionophore (Vicini et al., 1987; Rogers et al., 1994; Eghbali et al., 1997). Barbiturates have been shown to increase the GABA_A mean channel open duration (Macdonald et al., 1989; Twyman et al., 1989). These drugs enhance the effects of endogenously released GABA, and potentiate the inhibition it exerts on postsynaptic cells. This is thought to be the primary mechanism by which they prevent seizures. Other antiepileptic drugs such as phenytoin and carbamazepine act by inhibiting voltage-dependent sodium channels. Electrophysiological studies of these compounds have shown that they inhibit rapidly firing action potentials by driving sodium channels into inactivated states from which recovery is slowed (McLean and Macdonald, 1983; Matsuki et al., 1984; McLean and Macdonald, 1986; Schwarz and Grigat, 1989; Kuo and Bean, 1994; Kuo et al., 1997). This is thought to limit neuronal excitability and prevent the excessively rapid discharges which underlie seizures.

In the early 1980s a series of -butyrolactones were synthesized which resembled chemical moieties of the convulsant picrotoxin, a noncompetitive GABA_A antagonist (fig. 1) (Klunk et al., 1982a, b, c, d). It was subsequently found that these compounds had distinct biological activities depending on the identity and location of alkyl substituents on the ring structure. Generally, alkyl groups on the -carbon conferred picrotoxin-like convulsant activity, while -alkyl groups imparted anticonvulsant effects. Subsequent binding and electrophysiological studies lent support to the notion that these compounds, like picrotoxin, act at the GABA_A receptor. They all displaced the picrotoxin analog, ³⁵S-TBPS, and -substituted compounds blocked GABA_A currents, while -substituted compounds potentiated them (Levine et al., 1985; Baker et al., 1988; Zorumski et al., 1989). Study of recombinant GABA_A subunits has demonstrated that anticonvulsant lactones act at a site distinct from the picrotoxin site on the GABA_A receptor (Williams et al., 1997).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Comparison of chemical structures of compounds studied.

Over the past decade, derivatives of the original lactones have been synthesized in which the -substituents, the ring heteroatom, and the size of the ring have been altered (Yoon et al., 1990; Canney et al., 1991; Holland et al., 1995). Extensive electrophysiological studies of the lactone family revealed similar GABA_A potentiation among the chemical variants, as well as the ability to limit epileptiform discharges in hippocampal slice preparations (Ferrendelli et al., 1983). Furthermore, studies of synaptically connected neurons in culture revealed that lactones dramatically prolong the decay of GABAergic inhibitory postsynaptic currents (Holland et al., 1990, 1991).

Most recently, the 3,3-dialkyl/alkyl, benzyl-2-pyrrolidinones and piperidinones were developed and characterized (Reddy et al., 1996,1997). Some exhibited superior anticonvulsant potency against pentylenetetrazole- and maximal electroshock-induced seizures compared to previously described lactone-class drugs. However, electrophysiological studies revealed no commensurate improvement in the potency or efficacy of the drugs' potentiation of GABA_A currents. This suggested that GABA_A receptor modulation might not be the sole mechanism of action. We carried out this study in order to test the hypothesis that 3-BEP modulates voltage-dependent sodium channels.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. 3-BEP, 3,3-diethyl-2-piperidinone, and -ethyl--methyl--thiobutyrolactone were synthesized by previously described methods (Canney et al., 1991; Reddy et al., 1997). The compound, 3-phenyl-2-pyrrolidinone (Nilsson et al., 1992), was prepared according to the method reported for the synthesis of 3,3-dialkyl-2-pyrrolidinones (Reddy et al., 1996). Drug stock solutions were made in DMSO and diluted with extracellular recording solution to their final concentrations. Control and drug solutions contained the same concentration of DMSO. All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO) unless otherwise mentioned.

Cell culture. Hippocampi were dissected from 1-day-old Sprague-Dawley rat pups, cut into pieces (<1 mm) and placed in 3 ml Leibovitz's L-15 medium containing 1 mg/ml papain and 0.2 mg/ml bovine serum albumin for 20 min at 37°C. The hippocampi were triturated with Pasteur pipettes and the suspension was centrifuged through 2 ml of medium containing 10 mg/ml trypsin inhibitor and 10 mg/ml bovine serum albumin. The cells were then resuspended in growth medium containing 90% Eagle's minimal essential medium (without glutamine), 10% NuSerum (Collaborative Biomedical Products of Becton Dickinson, Franklin Lakes, NJ), 20 U/ml penicillin and 20 µg/ml streptomycin. They were then plated onto a monolayer of cortical glial cells in 35-mm culture dishes (2.5 × 10⁵ cells/dish) which had been precoated with poly-L-lysine. For the study of synaptic currents, neurons were plated onto dishes containing isolated glial `microislands` which were formed by spraying the dishes with collagen droplets using a microatomizer (Thomas Scientific, Swedesboro, NJ) according to previously published methods (Mennerick and Zorumski, 1994).

Electrophysiology. Experiments were carried out at room temperature (about 25°C) on the stage of an inverted microscope (Nikon, Melville, NJ) using whole-cell patch clamp techniques. In order to optimize voltage clamp conditions for sodium currents, recordings were made from neurons less than thirty hours in culture, when they were still electrically compact (Yang et al., 1993). For activation and inactivation I-V plots, cells were used during the first 6 hr in culture. The growth medium was removed and replaced with an extracellular recording solution containing (in mM): 140 NaCl, 20 TEA-Cl, 3 KCl, 10 HEPES, 5.5 glucose, 1 CaCl₂, 1 MgCl₂, 0.1 CdCl₂, .1 picrotoxinin (pH adjusted to 7.3 with 1 N NaOH). Electrodes were made from 1.2 O.D./0.68 I.D. (mm) borosilicate glass (W.P.I.) and had resistances of 4 to 6 M when containing a solution of (in mM): 110 CsCl, 20 NaCl, 10 TEA-Cl, 10 HEPES, 1.1 EGTA, 5.5 glucose, 2 MgATP (pH adjusted to 7.2 with 1 N CsOH). Recordings were carried out using an Axopatch-1B amplifier (Axon Instruments, Foster City, CA). Pipette and whole cell capacitance as well as series resistance were corrected by the compensation circuitry on the amplifier. Final series resistance values were typically 8 to 10 M, and compensation of 80% or more was possible without significant oscillation. Voltage clamped neurons were stepped to command potentials and the resulting currents were digitized at 10 kHz using an A/D converter and a PC running commercial software (pClamp6, Axon Instruments). Test solutions were applied via gravity-driven, solenoid valve-controlled, microperfusion pipettes (approximately 200 µm diameter) positioned 100 to 150 µm from the cell under observation.

Data analysis. All currents were filtered at 10 kHz. When determining peak sodium current values and generating I-V plots, data were corrected by subtracting the leak current recorded from hyperpolarizing pulses or the residual currents after complete sodium channel inactivation. Any drug interaction with the GABA_A receptor and subsequent alteration of membrane resistance was eliminated by including 100 µM picrotoxinin in the extracellular solution. Concentration-response curves were fit (Origin, Microcal, Northampton, MA) to a logistic equation: 1/(1+([drug]/IC₅₀)ⁿ) where IC₅₀ is the concentration of half-maximal block and n represents the Hill coefficient. Steady state inactivation curves were fit to the Boltzman equation: 1/(1+exp[ (V-V₅₀)/k]) where V₅₀ is the half-inactivated potential and k represents the slope factor. Recovery from inactivation was fit to the monoexponential function: 1-[y₀+A₁exp(-t/)]. Individual sodium current decays were fit (pClamp6) to a monoexponential function. All statistical values are presented as mean ± S.E.

	Results

Top Abstract Introduction Methods Results Discussion References

3-BEP blocks voltage-dependent sodium currents. To observe 3-BEP effects on sodium currents, solutions were used that blocked all other voltage-gated conductances. Under these conditions, all current evoked by a depolarizing step from -80 mV was blocked by 1 µM tetrodotoxin (data not shown). Due to the rapid time course and large size of voltage-gated sodium currents in cultured rat hippocampal neurons, we carried out experiments on neurons that had been in culture less than 30 hr to ensure adequate space and time clamp. The voltage dependence of sodium current activation and inactivation was studied in cells less than 6 hr in culture. During this time, cultures contain neurons which are small (approximately 25 µm diameter) and nearly spherical, since they have not yet extended elaborate processes (fig. 2). Currents evoked by depolarizing voltage steps began activation simultaneously (fig. 4a). Records exhibited smooth activation and inactivation phases. Furthermore, peak currents and inactivation time courses changed incrementally as a function of 5 mV changes in step voltage values. We therefore judged that our voltage clamp conditions were acceptable.

View larger version (104K): [in this window] [in a new window]   Fig. 2.   Photomicrograph (120X magnification) of hippocampal neurons at 6 hr in culture. Arrows indicate cells that have not yet extended elaborate processes and typically afford voltage clamp conditions that allow the recording of sodium currents. The scale bar represents 100 µm.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   3-BEP inhibition of sodium currents. a, Sample recordings from a neuron voltage clamped at -80 and -50 mV. Currents evoked by depolarization to -10 mV were recorded in control and 600 µM 3-BEP solutions. The drug had a larger effect at -50 mV. The scale bar represents 400 pA and 5 msec. b, Concentration responses for sodium current inhibition at different holding potentials. The ratio of drug to control current is plotted. Each point represents the mean ± standard error of at least five cells. The solid lines are the results of logistic equation curve fits. The extrapolated curves eventually reach 0 (complete block).

3-BEP effects on the sodium activation curve. To study the dependence of 3-BEP blockade on activation voltage, sodium currents were elicited by depolarizing cells from -80 mV to a range of potentials in 5 mV increments. 600 µM 3-BEP elicited similar blockade at all potentials for which sodium current was observed (fig. 4). Therefore, 3-BEP inhibits voltage-dependent sodium current in a manner which is independent of activation voltage. Examining the decays of current evoked by steps to -10 mV revealed that 3-BEP accelerated the rate of inactivation. Under control conditions, current inactivation was well fit to a first order exponential decay with  = 0.771 ± 0.044 msec while in the presence of 3-BEP  = 0.678 ± 0.049 msec (n = 10 cells, P < .01, paired t test).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   3-BEP effects on the sodium activation curve. a, Sample sodium currents from a neuron in control, 600 µM 3-BEP and wash conditions. Currents shown resulted from depolarizations from a holding potential of -80 mV to potentials of -40 through +20 mV in 5 mV increments. The scale bar represents 2 msec and 500 pA. b, Current-voltage relationship for all neurons tested. Neurons were depolarized from -80 mV to Vmem. Peak currents were normalized to those evoked by depolarization to 0 mV in control solution. Each point represents the mean ± S.E. of 10 or more neurons.

3-BEP shifts the inactivation curve. The effects of 3-BEP on steady state inactivation were assessed with a prepulse protocol. Neurons were subjected to 2-sec prepulses to voltages ranging from -130 to -20 mV followed by a test pulse to -10 mV (fig. 5a). The currents elicited by the test pulse were normalized to those after a -130 mV prepulse in control solution, and then plotted vs. prepulse voltage (fig. 5b). The inactivation V₅₀ was -65.3 ± 1.7 mV in control solution and -72.0 ± 1.3 mV in the presence of 600 µM 3-BEP. This represented a statistically significant difference (n = 11 cells, P < .001, paired t test). Thus, in addition to decreasing peak currents at hyperpolarized prepulse potentials, 3-BEP alters the voltage-dependence of steady state inactivation. These data are consistent with a higher 3-BEP affinity for inactivated channels than for those at rest (fig. 5c).

View larger version (14K): [in this window] [in a new window]   Fig. 5.   3-BEP effects on steady state inactivation. a, Neurons were subjected to 2-sec steps to Vprepulse ranging from -130 to -20 mV, followed by a depolarizing test pulse to -10 mV. b, Inactivation curves for all neurons tested. Test pulse current peaks were normalized to those elicited after a -130 mV prepulse in control solution. Each point represents the mean ± S.E. of 11 neurons. The solid lines represent the results of Boltzman equation curve fits. Control V50 = -64.2 mV, 3-BEP V50 = -71.5 mV. c, Kinetic scheme of sodium channel behavior. At steady state, sodium channels are distributed between resting (R) and inactivated (I) conformations in a voltage-dependent equilibrium. R* and I* represent drug-bound resting (blocked) and drug-bound inactivated conformations, respectively.

3-BEP slows recovery from inactivation. Recovery from inactivation was observed by depolarizing neurons from -80 to -10 mV for a 100-msec prepulse, allowing them to recover at -80 mV for a variable period of time, and then subjecting them to a test pulse to -10 mV (fig. 6a). The ratio of test pulse current to prepulse current was examined as a function of recovery time. In the presence of 600 µM 3-BEP, recovery was markedly slowed ( = 12.8 ± 1.3 msec, n = 8) compared to the time course of recovery in control solution ( = 4.9 ± .7 msec, n = 9) (fig. 6b). Again, this represented a statistically significant difference (P < .001, Student's t test). Therefore, in addition to altering steady state sodium channel inactivation, 3-BEP slows the transition from inactivated to resting state.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   BEP effects on recovery from inactivation. a, Recovery was studied by depolarizing neurons to -10 mV for 100 msec, then allowing them to recover at -80 mV for a variable length of time (t) before depolarizing again. A sample current trace is shown. Scale bar represents 50 msec and 500 pA. b, The ratio of test pulse currents to prepulse currents plotted vs. recovery time in control and 600 µM 3-BEP solutions. Each point represents the mean ± S.E. of eight or more neurons. The solid lines are the results of first order exponential equation curve fits.

GABA_A receptor and sodium channel modulation occur at comparable concentrations. Because previous research has demonstrated that lactone convulsant and anticonvulsant activity was determined by drug effects on the GABA_A receptor (Klunk et al., 1982a), we wondered whether 3-BEP inhibited sodium currents at concentrations that potentiate GABA currents. Autaptic GABAergic inhibitory postsynaptic currents are prolonged at 3-BEP concentrations greater than 100 µM (fig. 7a). Consistent with this observation, peak currents elicited by the exogenous application of 3 µM GABA to neurons held at -60 mV are increased by 3-BEP. A concentration-response curve for this activity is presented here and superimposed on the sodium blockade relationship for neurons held at -60 mV (fig. 7b). Both sets of data are normalized to the drug's maximal effect to aid in comparison. The apparent IC₅₀ for sodium current inhibition is 487 µM while the GABA_A potentiation EC₅₀ is 575 µM. Therefore we conclude that sodium channel and GABA_A modulations occur at comparable concentrations.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Comparison of 3-BEP inhibition of sodium currents and modulation of GABAA receptors. a, Example of an autaptic IPSC in the presence of 0, 100, 300 and 600 µM 3-BEP. b, Comparison of 3-BEP concentration-response relationships for sodium channel blockade and GABAA current potentiation. Each dataset is normalized to 3-BEP's apparent maximal effects; 100% sodium block and 220% potentiation of 3 µM GABA peak current. Each point represents the mean ± S.E. of five or more neurons. Solid lines are the results of logistic equation curve fits.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Many of the 3,3-dialkyl- and 3-alkyl-3-benzyl- substituted 2-pyrrolidinones and 2-piperidinones are better anticonvulsants than any previously reported `lactone family' drugs. Furthermore, several of them, including 3-BEP, have higher potency in vivo than the clinically used anticonvulsants, ethosuximide and valproic acid (Reddy et al., 1996, 1997). This stresses the importance of understanding the mechanisms by which they prevent seizures. As with previously studied lactone drugs, these most recent derivatives were thought to exert their biological effects by modulating the GABA_A receptor in a manner similar to the benzodiazepines and barbiturates. However, they did not differ substantially when assayed for this effect; as with older compounds they did not demonstrate GABA_A potentiation until they reached concentrations in the hundreds of micromolar to low millimolar range. We set out to determine whether these newer compounds demonstrated additional activities which might explain their superiority in vivo.

We have now confirmed the hypothesis that 3-BEP inhibits voltage-dependent sodium currents. This activity was not shared by a well studied thiobutyrolactone anticonvulsant, suggesting a functional difference between 3-BEP and older members of the lactone family of drugs. Sodium current blockade was not seen with two other structurally similar compounds, suggesting the existence of a chemical determinant for sodium channel interactions. 3-BEP appears to utilize a mechanism common to other known sodium channel blocking anticonvulsants: the modulation of inactivation kinetics. However, the ability of 3-BEP to inhibit sodium currents in cells held at hyperpolarized prepulse potentials (fig. 5b) represents a subtle difference between it and phenytoin and carbamazepine, which bind and exert their effects only on depolarization (Kuo and Bean, 1994; Kuo et al., 1997).

Our data are consistent with a model of sodium channel modulation that includes different 3-BEP affinities for resting and inactivated channels. At steady state, sodium channels are distributed between resting and inactivated conformations in a voltage-dependent equilibrium (fig. 5c). At hyperpolarized potentials nearly all sodium channels are in their resting conformation (R). 3-BEP has sufficient affinity for resting channels to bind a fraction and reduce peak currents elicited by depolarizing steps from these hyperpolarized holding potentials (fig. 5b). As neurons are subjected to depolarized prepulses, more sodium channels assume an inactivated conformation (I). 3-BEP's enhanced affinity and binding to inactivated channels (itself a `neutral' modulation since inactivated channels are, by definition, already closed) decreases the fraction of channels in the unbound, inactivated state. As a result, a proportionate decrease in the fraction of channels in resting conformation occurs, causing an additional reduction in peak current. This process is evidenced by the change in steady state inactivation V₅₀. This scheme is similar to those presented by Kuo and Bean (1994) and Kuo et al. (1997) for carbamazepine and phenytoin.

Our comparison of the 3-BEP concentration-response relationships for sodium current inhibition and GABA_A receptor modulation revealed that these effects occur at comparable concentrations. Despite its relatively high IC₅₀ and EC₅₀ values for these activities, 3-BEP is a potent anticonvulsant; its ED₅₀s for pentylenetetrazole- and maximal electroshock-induced seizures in mice are 61 and 70 mg/kg, respectively (Reddy et al., 1997). To date, GABA_A receptor modulation has represented the most likely mechanism by which 3-BEP and other members of the lactone family prevent seizures. Thus, if we consider 3-BEP's GABA_A potentiation physiologically relevant, we must now also accept its inhibition of sodium currents as an important activity. This implies that 3-BEP may prevent seizures by both augmenting GABAergic inhibition and decreasing intrinsic neuronal excitability by blocking sodium channels. Furthermore, the relationship among potentiation of GABA_A currents, blockade of sodium currents and anticonvulsant activity is likely quite complex. Even a small effect on membrane voltage might prevent the generation of paroxysmal events. For anticonvulsant efficacy, it may not be necessary to reach brain extracellular space concentrations equal to the EC₅₀ for GABA current enhancement or the IC₅₀ for sodium current block.

The fact that 3-BEP exhibits this activity is significant for several reasons. First, the blockade of sodium channels is widely accepted to be the relevant mechanism of action of certain clinically useful antiepileptic medications. Second, it may begin to explain why it exhibits a higher potency than the previous generation of lactone drugs. Finally, given the structural similarity of different voltage-gated ion channels, it is possible that 3-BEP and related compounds have additional effects on potassium or calcium channels which may contribute to their anticonvulsant mechanisms. We realize that other factors besides the in vitro electrophysiological properties of these drugs may influence their in vivo effects. For instance, issues of pharmacokinetics and drug metabolism may be very important. Despite this, a better understanding of the full range of activities exhibited by 3-BEP and related drugs may offer insight into their beneficial and toxic effects in vivo.