Antiepileptic Drugs, 5th Edition

Succinimides

66

Mechanisms of Action

Katherine D. Holland MD, PhD^*

James A. Ferrendelli MD^**

* Staff Physician, Department of Neurology, Cleveland Clinic Foundation, Cleveland, Ohio

** Professor and Chairman, Department of Neurology, University of Texas, Houston Medical School; and Chief, Department of Neurology Service, Hermann Hospital, Houston, Texas

Succinimides, especially ethosuximide (ESM), α-ethyl-α-methyl succinimide, have been used extensively for the treatment of absence (petit mal) seizures. Despite the frequent use of these agents, the sites and mechanisms of action of succinimides are still poorly defined. In this chapter, the available published data on basic pharmacologic actions of succinimides are reviewed. Most of the available literature has focused on ESM; therefore, the main emphasis of this chapter is the mechanism of action of this succinimide. After reviewing its effects on clinical and experimental seizures, its molecular and cellular actions, and its neuronal systems effects, we present a hypothesis explaining its probable mechanism of action.

EFFECTS ON EPILEPTIFORM DISCHARGES

One of the most intriguing facts about ESM is its highly selective effect on clinical and experimental seizures. It completely, or almost completely, controls absence seizures in ~50% of patients with absence epilepsy and reduces the frequency of these seizures in another 40% to 45% of patients (1). The electroencephalographic (EEG) hallmark of absence seizures is generalized 3-Hz spike-and-wave complexes. Succinimides are able to eliminate these discharges in most patients (2,3). ESM is also highly effective at controlling epileptic negative myoclonus (4,5). In contrast, it has no apparent effect against generalized tonic-clonic convulsions or partial seizures. Methsuximide has a broader spectrum of action. It has been reported to be effective in some patients with partial seizures and in children with atypical absence seizures (6,7).

The high degree of therapeutic specificity of ESM in human seizure disorders is reflected by its selective anticonvulsant action against experimental seizures. It is well known that ESM prevents pentylenetetrazol (PTZ; Metrazol) seizures at nontoxic doses in experimental animals. However, it has no effect on maximal electroshock seizures, except at very high, toxic concentrations (8). ESM blocks spontaneous absencelike seizures, which occur in some genetically epilepsy-prone mice (9). It also has been reported to have an anticonvulsant effect on seizures induced by implantation of cobalt into the cerebral cortex (10, 11, 12), systemic administration of γ-hydroxybutyric acid (GHB) (13, 14, 15), application of conjugated estrogen to the brain (16), inhalation of fluorothyl (17) or enflurane (18), barbiturate withdrawal (19), and systemic administration of penicillin, picrotoxin, and a benzodiazepine receptor inverse agonist (20,21). However, it seems to be inactive against allylglycine seizures in photosensitive baboons (22), stroboscopic seizures in epileptic fowl (23), seizures produced by application of aluminum hydroxide (24) to the cerebral cortex, and seizures produced by the systemic administration of bicuculline, N-methyl-D-aspartate, strychnine, or aminophylline (20).

It has become increasingly clear that subcortical brain regions play a crucial role in the propagation of generalized seizures. An important characteristic of anticonvulsant compounds is their ability to depress repetitive impulses in the reticular core. Antiabsence drugs such as ESM have been shown to depress descending reticular inhibitory pathways (25). In addition, there is selective enrichment of ¹⁴C-2-deoxyglucose uptake into the mammillary bodies and their connections during ESM-induced suppression of PTZ seizures (26). The ability of ESM to block PTZ-induced EEG phenomena in rats requires an intact hindbrain (27). This is not required for the action of other antiabsence agents such as valproate. These findings suggest a unique neurosystemic action of ESM that is important for its anti-PTZ, and perhaps antiabsence, activity. However, these observations do not clearly establish the functional anatomy of ESM action, and additional research on this question is needed.

Fromm and coworkers showed that antiabsence drugs preferentially depress central nervous system inhibitory

P.640

pathways that require considerable repetitive stimulation for activation (25,28, 29, 30). These investigators suggest that this accounts for the therapeutic specificity of ESM. In addition, they believe that absence seizures involve paroxysmal activity in inhibitory pathways (25,31). If this is the case, the observation that γ-aminobutyric acid (GABA) responses are reduced by ESM could provide a basis for the antiabsence properties of this drug. However, this theory cannot account for the ability of ESM to protect animals from seizures caused by picrotoxin, whose convulsant actions are thought to result from its ability to antagonize GABA-mediated inhibition noncompetitively.

In cultured neurons, ESM has no effect on the use-dependent firing of action potentials (32); however, in vivo, ESM is able to depress specific types of repetitive activation. ESM significantly reduces PTZ-induced photic recruitment and photic afterdischarges, and it completely suppresses PTZ-induced spindle activity (33). These rhythmic activities are thought to be an expression of synchronous afterdischarges of the thalamocortical system.

Investigators have suggested that the effect of ESM is mediated either by direct influences on inhibitory cells in the thalamic relay or by actions on ascending reticular activation. The lateral geniculate body has been discounted as the site of action because injections of ESM into the lateral geniculate could not abolish these photic afterdischarges (34). The actions of anticonvulsants on thalamocortical excitability have been examined by recording the cortical response elicited by a pair of stimuli given to the ventral lateral thalamus (31,35,36). These reports show that phenytoin, carbamazepine, and diazepam depressed evoked responses at all frequencies, and ESM and valproate decreased evoked responses at 3 Hz. This provides a basis for the effectiveness of ESM and valproate in controlling absence seizures characterized by 3-Hz spike-and-wave activity. Additionally, Pelligrini and associates (37) studied the effect of ESM on thalamic and cortical neuronal activity in cats. They concluded that ESM acts by disruption of spontaneous intrathalamic synchronizing mechanisms resulting in less efficient thalamocortical impulses necessary for cortical spike-and-wave discharges.

Studies of thalamic neurons reveal that they have tonic and burst-firing properties (38,39). This burst firing may underlie spike-and-wave discharges associated with generalized epilepsy (40). ESM decreases the probability that thalamic neurons will fire in a burst (39,41).

PHYSIOLOGIC EFFECTS

Effects on Excitability or Inhibition

A relationship between ESM and GABA has been suggested by several studies (42,43). Because anticonvulsants such as valproate, phenobarbital, and benzodiazepines potentiate GABA function, several laboratories have examined the possibility that ESM may modify GABAergic neurotransmission. These studies reveal that ESM decreases GABA responses in cultured cortical, hippocampal, thalamic, and spinal cord neurons (32,44, 45, 46). Barnes and Dichter (44) report that 500 µmol/L ESM decreased the mean GABA response by 30%. The GABA receptor-chloride ionophore complex contains at least three distinct binding sites: the GABA site, the benzodiazepine site, and the picrotoxin site. In radioligand binding studies, ESM has no effect on diazepam and GABA binding to rat brain membranes at concentrations <1 mmol/L (47,48). At higher concentrations, ESM substantially inhibits GABA, but not benzodiazepine binding (47). In contrast, the binding of ³⁵S-t-butylbicyclophosphorothionate, the radioligand used to study the picrotoxin receptor, is decreased competitively at concentrations comparable to those used by Barnes and Dichter (47,49). This finding suggests that ESM-induced antagonism of GABA responses results from its action at the picrotoxin site and also implies that the anticonvulsant actions of ESM are not mediated by a postsynaptic enhancement of GABAergic responses.

The role of excitatory amino acids in the pathogenesis of epilepsy is the subject of intense research. Although the relationship between ESM and excitatory amino acids has yet to be fully investigated, some evidence suggests ESM does not work by a blockade of excitatory pathways. For example, ESM does not prevent seizures induced by either kainic acid or N-methyl-D-aspartate, but nonspecific excitatory amino acid antagonists are very potent blockers of these seizure types (50,51). In addition, the anticonvulsant profiles of ESM and certain excitatory amino acid antagonists differ substantially. These data are certainly not conclusive, but they do suggest that ESM does not alter excitatory amino acid function, at least by postsynaptic mechanisms.

Effects on Ion Channels and Transport

Investigators have shown that ESM has no effect on the basal level of calcium ion (Ca²⁺) uptake into synaptosomes (52). Unlike phenytoin and carbamazepine, which inhibit Ca²⁺accumulation into veratridine-depolarized synaptosomes at therapeutic concentrations, ESM is ineffective except at concentrations exceeding 10 mmol/L. Crowder and Bradford (53) confirmed this and additionally showed that ESM has no effect on veratridine-stimulated amino acid neurotransmitter release.

Calcium channels play an important role in neuronal excitability. Originally, three classes of channels that mediate calcium entry into neurons were identified: L (large), T (tiny), and N (neither) channels (54). These channels differ in their magnitude, voltage dependence, activation properties, and pharmacologic profiles. The low-threshold calcium channel (LTCC) is analogous to the T channel and is a transient, low-conductance channel present in thalamic neurons. Experiments by Coulter, Huguenard, and Prince (39,55, 56, 57) showed that therapeutic concentrations of ESM

P.641

reduce the LTCC in thalamic neurons. ESM did not affect the gating properties of the channel; it reduced either the number of LTCC channels or the single-channel conductance. Dimethadione, another antiabsence drug, had the same action. Carbamazepine, phenytoin, and valproate had no effect on the LTCC at clinically relevant concentrations. Convulsant succinimide compounds, such as tetramethylsuccinimide, do not alter thalamic LTCCs. ESM also inhibited high-threshold calcium channels, but to a lesser extent than its effect on the LTCC, and at concentrations significantly higher than the therapeutic range of ESM. Similar effects of ESM on T currents have been demonstrated in sensory neurons (58), but other researchers have failed to find any effects of ESM on T-type calcium currents in numerous other types of neurons (59, 60, 61).

More recently, Leresche and colleagues (41) studied the action of ESM in thalamic neurons. They found that ESM decreased noninactivating sodium ion (Na⁺) currents in thalamocortical neurons at clinically relevant concentrations. It did not alter transient Na⁺ currents. ESM also blocked Ca²⁺-dependent potassium ion (K⁺) channels. In contrast to the work by Coulter et al. (45,53,55, 56, 57), Leresche did not find any effect on the LTCC despite studying various types of thalamic neurons in various strains of rats. The reason for this discrepancy is unclear.

Effects on Voltage Sensitive Receptors

Fohlmeister, Aldelman, and Brennan (62) examined the effects of ESM and valproate on excitable Na⁺- and K⁺-channels in voltage-clamped squid giant axons. These researchers reported that ESM applied to the external surface of the axon reduced Na⁺-current in a voltage-independent manner, reduced maximal K⁺ conductance, and slowed K⁺ channel gating. Internally applied ESM slowed Na⁺ and K⁺ channel gating and reduced the peak conductance of the Na⁺ channel in a voltage-dependent fashion. The significance of these observations in invertebrate neurons as related to the anticonvulsant mechanism of ESM is unclear because of both the species difference and the heroic concentrations of ESM used (60 mmol/L). Other investigators have reported that ESM is unable to inhibit ²H-batrachotoxinin A 20-α-benzoate binding to sodium channels and batrachotoxinin-induced Na⁺ flux in neuroblastoma cells and rat brain synaptosomes at concentrations up to 1 mmol/L (63,64). These data suggest that this sodium channel is not ESM's site of anticonvulsant action.

BIOCHEMICAL EFFECTS

The molecular bases for changes in neuronal excitability involve actions on brain enzyme activity, neuron transmitter processes, and ion channels. The effects of ESM on each of these are discussed.

Effects on Biochemical Systems

Two laboratories have reported that ESM inhibits (Na⁺, K⁺)-adenosine triphosphatase (ATPase) activity but not magnesium ion-ATPase activity in subcellular fractions of cortical tissue (21,65, 66, 67). The data of Gilbert and colleagues suggest that the site of (Na⁺, K⁺)-ATPase inhibition may be restricted to the nerve terminal plasma membrane (21). Unfortunately, these effects were found at ESM concentrations (2.5 and 25 mmol/L) considerably greater than those producing anticonvulsant effects. In addition, although Leznicki and Dymecki (67) reported that ESM inhibits (Na⁺, K⁺-ATPase in brain homogenates, they also found that long-term treatment of animals resulted in an increase in (Na⁺, K⁺)-ATPase activity.

Although ESM has been reported to have little or no direct effect on the GABA- synthesizing enzyme glutamic acid decarboxylase (67), it antagonizes isoniazid-induced inhibition of this enzyme (68). However, single-dose administration of ESM in anticonvulsant doses has no effect on brain GABA concentration (69). Thus, alterations in brain GABA do not contribute to the anticonvulsant effects of ESM. ESM has no effect on the activities of various enzymes involved in the breakdown of neurotransmitters including GABA transaminase, monoamine oxidase, acetylcholinesterase, and arylsulfatase (67).

ESM inhibits NADPH (reduced form of nicotinamide-adenine dinucleotide phosphate)-linked aldehyde reductase in bovine brain (70). This enzyme can convert succinic semialdehyde to GHB and may be the mechanism by which long-term ESM treatment decreases brain GHB levels (71,72). In light of the behavioral and EEG similarities between human absence seizures and administration of exogenous GHB, alterations in endogenous GHB levels could be relevant to the antiabsence actions of ESM (13,14,73). However, this is unable to account for either the anti-PTZ actions of ESM or the ability of ESM to block seizures caused by exogenous GHB. In addition, single-dose administration of ESM produces an increase in brain GHB, possibly by inhibition of GHB dehydrogenase (74), and that this increase coincides with the onset of anticonvulsant effects (72).

Studies in genetic absence epilepsy in rats from Strasbourg (GAERS) have shown that endogenously released nitric oxide can suppress generalized spike-and-wave discharges (75). ESM has been shown to promote nitric oxide release in GAERS animals and also to suppress the epileptiform discharges in these animals. The molecular biochemical basis for this increase in nitric oxide release is unknown.

Neuroprotective Effects

Potential neuroprotective effects of ESM have not been studied. However, from information about its lack of action at excitatory amino acid receptors (50,51), and a study

P.642

showing that ESM can release nitric oxide (75), a compound associated with neuronal injury, it is unlikely that ESM has significant neuroprotective properties (76).

Effects at Synapse-Drug Receptor Interactions

Several laboratories have demonstrated that GHB produces seizures in experimental animals. These seizures resemble human absence seizures behaviorally, electrically, and pharmacologically (13, 14, 15,72). Although the mechanism by which GHB produces seizures remains obscure, specific high-affinity GHB binding sites associated with the GABA-mediated chloride ionophore have been identified (77,78). It has also been established that GHB is capable of blocking flow through dopaminergic neurotransmitter systems (19). ESM is highly effective in preventing GHB-induced seizures. However, ESM is not able to compete with ³H-GHB for binding to rat brain (77). This finding suggests that the anti-GHB and antiabsence effects of this drug are not the result of direct action at the putative GHB binding site. ESM inhibits depolarization-evoked release of GHB from hippocampal slices (79). Although this may explain the rise in brain GHB concentration after single-dose ESM administration, it cannot account for the ability of ESM to block seizures produced by exogenously supplied GHB.

Fluphenazine, a dopamine receptor antagonist, and α-methylparatyrosine, an inhibitor of dopamine synthesis, can block the protective effects of ESM in the GHB seizure model (80). This finding indicates that the anticonvulsant action of ESM, at least in the GHB model, may be related to some effect on dopaminergic neurotransmission, possibly by augmentation of dopamine-mediated inhibition in the central nervous system. The involvement of dopaminergic systems in the action of ESM is supported by the following observations. First, L-DOPA, a dopamine precursor, prevents cobalt-induced seizures that are also prevented by ESM (12). Second, cortical spikes produced by the topical application of penicillin are prevented by systemic L-DOPA and by topical dopamine but not by topical norepinephrine (81). As mentioned earlier, ESM prevents seizures produced by systemic administration of penicillin (82). Finally, dopamine receptor agonists decrease the duration of spike-and-wave discharges in rats with spontaneous absencelike seizures (83). ESM is also able to abolish seizures in this form of epilepsy (84). However, these results must be interpreted cautiously because, at present, ESM has not been shown directly to alter dopamine-mediated processes either in vivo or in vitro.

SUMMARY OF “UNIFIED MECHANISM”

Any complete description of the mechanisms of action of an antiepileptic drug would require a full understanding of the pathophysiologic mechanisms of epilepsy and an explanation of how the drug modifies these to prevent seizures. Because the pathophysiologic mechanisms of epilepsy are still incompletely understood, one can only speculate about the mechanisms of action of most antiepileptic drugs.

Anticonvulsant drugs are widely believed to act by (a) direct modification of membrane function in excitable cells, (b) alteration of chemically mediated neurotransmission, and/or (c) alteration in the activity of ion channels. There is no evidence that ESM indirectly and nonspecifically alters membrane structure, thereby disrupting ionic channels. ESM is highly water soluble, so it is unlikely that much of it inserts into cellular membranes that have a high lipid content. Furthermore, it has none of the properties of general anesthetics that are thought to exert their effects by a direct action on cellular membranes.

The action of ESM does not appear to be directly related to its known actions on neurotransmitter processes. Indirect evidence indicates that it may deplete excitatory neurotransmitter stores mediating the spinal monosynaptic reflex. This is thought to occur by an increase in fractional release per stimulus without resultant increase in synthesis. Although a similar effect in brain could selectively depress repetitive impulses, thereby preventing seizures, it is not likely because the increased release of excitatory neurotransmitters on initial impulses could be enough to potentiate seizure activity. In addition, direct measurements of neurotransmitters indicate that, in many systems, synthesis can more than compensate for increased release even at the highest firing rates attainable. A more tenable explanation would be that ESM may increase the influence of inhibitory neurotransmitters. The suggested depressant effects on corticofugal inhibition of the spinal trigeminal nucleus may well be a result of some action on neuronal pathways subserved by inhibitory neurotransmitters. However, present evidence suggests that ESM does not significantly increase, but rather diminishes, GABA-mediated inhibitory processes. Possibly the anticonvulsant effect of ESM may involve dopamine-mediated neurotransmission, but this is also uncertain. Other still unidentified, inhibitory neurotransmitter systems may be responsible for or may have a role in ESM mechanisms of action.

Experimental models suggest that thalamic neurons play an important role in the generation of thalamocortical rhythmicity that underlies the 3-Hz spike-and-wave discharges seen during absence seizures (85), and calcium currents are involved in the production of low-threshold calcium spikes involved in the generation of sleep spindles. Studies on the kinetic properties of calcium currents in thalamocortical relay neurons suggest that the T current or LTCC is necessary and sufficient to generate the low-threshold calcium spikes produced in thalamic relay neurons. Although multiple studies have shown that ESM can decrease the bursting of these thalamocortical neurons, the molecular mechanism is unclear. Some researchers have

P.643

shown that ESM can effect T currents, whereas others have not been able to reproduce this finding. An alternative hypothesis is that ESM blocks spike-and-wave discharges and burst firing in thalamocortical neurons by inhibition of a noninactivating sodium current in thalamic neurons.

REFERENCES

1. Browne TR, Dreifuss FE, Dyken PR, et al. Ethosuximide in the treatment of absence (petit mal) seizures. Neurology1975;25: 515-524.
2. Browne TR, Dreifuss FE, Penry JK, et al. Clinical and EEG estimates of absence seizure frequency. Arch Neurol1983;40: 469-472.
3. Blomquist HK, Zetterlaund B. Evaluation of treatment in typical absence seizures: the roles of long-term EEG monitoring and ethosuximide. Acta Paediatr Scand1985;74:409-415.
4. Oguni H, Uehara T, Tanaka T, et al. Dramatic effect of ethosuximide on epileptic negative myoclonus: implications for the neurophysiological mechanism. Neuropediatrics1998;29:29-34.
5. Capovilla G, Beccaria F, Veggiotti P, et al. Ethosuximide is effective in the treatment of epileptic negative myoclonus in childhood partial epilepsy. J Child Neurol1999;14:395-400.
6. Wilder BJ, Buchanan RB. Methsuximide for refractory complex partial seizures. Neurology1981;31:741-744.
7. Tennison MB, Greenwood RS, Miles MV. Methsuximide for intractable childhood seizures. Pediatrics1991;87:186-189.
8. Chen A, Weston JK, Bratton AC Jr. Anticonvulsant activity and ethosuximide. Epilepsia1963;XX:66-76.
9. Heller AH, Dichter MA, Sidman RL. Anticonvulsant sensitivity of absence seizures in the tottering mutant mouse. Epilepsia1983;25:25-34.
10. Dow RC, Forfar JC, McQueen JK. The effects of some anticonvulsant activity on cobalt-induced epilepsy. Epilepsia1973;14: 203-212.
11. Kastner I, Klingberg F, Muller M. Zur Wirkung des Ethosuximids auf die Kobalt-in-duzierte “Epilepsie” der Ratte. Arch Int Pharmacodyn Ther1970;186:220-226.
12. Scuvee-Moreau J, Lepot M, Brotchi J, et al. Action of phenytoin, ethosuximide, and of the carbidopa-L-dopa association in semichronic cobalt-induced epilepsy in the rat.Arch Int Pharmacodyn Ther1977;230:92-99.
13. Godschalk M, Dzoljîc MR, Bonta IL. Antagonism of gamma-hydroxybutyrate-induced hypersynchronization in the ECoG of the rat by anti-petit mal drugs. Neurosci Lett1976;3:145-150.
14. Snead OC. Gammahydroxybutyrate in the monkey. II. Effect of chronic oral anticonvulsant drugs. Neurology1978;28:643-648.
15. Snead OC. Gammahydroxybutyrate in the monkey. III. Effect of intravenous anticonvulsant drugs. Neurology1978;28: 1173-1178.
16. Julien RM, Fowler GW, Danielson MG. The effect of antiepileptic drugs on estrogen-induced electrographic spike-wave discharges. J Pharm Exp Ther1975;193:647-656.
17. Adler MW. The effect of single and multiple lesions of limbic system on cerebral excitability. Psychopharmacologia1972;24: 218-230.
18. Schettine A, Wilder BJ. Effects of anticonvulsant drugs on enflurane cortical dysrhythmias. Anesth Analg1974;53:951-962.
19. Roth RH, Walters JR, Aghajanian GK. Effect of impulse flow on the release and synthesis of dopamine in the rat striatum. In: Udsin E, Snyder SH, eds. Frontiers in catecholamine research.Oxford: Pergamon Press, 1973:567-574.
20. Ferrendelli JA, Holland KD, McKeon AC, et al. Comparison of the anticonvulsant activities of ethosuximide, valproate, and a new anticonvulsant, thiobutyrolactone. Epilepsia1989;30: 617-622.
21. Gilbert JC, Wyllie MG. The effects of the anticonvulsant ethosuximide on adenosine triphosphatase activities of synaptosomes prepared from rat cerebral cortex. Br J Pharmacol1974;52: 139P-140P.
22. Meldrum BS, Horton RW, Toseland PA. A primate model for testing anticonvulsant agents. Mol Pharmacol1975;14:347-356.
23. Davis HL, Johnson DD, Crawford RD. Epileptiform seizures in domestic fowl. IX. Implication of the absence of anticonvulsant activity on ethosuximide in a pharmacological model of epilepsy. Can J Physiol Pharmacol1978;56:893-896.
24. Lockard JS, Levy RH, Congdon WC, et al. Efficacy testing of valproic acid compared to ethosuximide in monkey model. II. Seizure, EEG, and diurnal variation. Epilepsia1977;18:205-224.
25. Fromm GH, Terrence CF. Effect of antiepileptic drugs on the brainstem. In: Fromm GH, Faingold DL, Browning RA, et al., eds. Epilepsy and the reticular formation: the role of the reticular core in convulsive seizures.New York: Alan R. Liss, 1987: 119-136.
26. Mirski MA, Ferrendelli JA. Selective metabolic activation of the mammillary bodies and their connections during ethosuximide induced suppression of pentylenetetrazol seizures. Epilepsia1986; 27:194-203.
27. Mares P, Pohl M, Kubova H, et al. Is the site of action of ethosuximide in the hindbrain? Physiol Res1994;43:51-56.
28. Fromm GH, Glass JD, Chattha AS, et al.. Effect of anticonvulsant drugs on inhibitory and excitatory pathways. Epilepsia1981; 22:65-73.
29. Fromm GH, Glass JD, Chattha AS, et al. Antiabsence drugs and inhibitory pathways. Neurology1980;30:126-131.
30. Shibuya T, Fromm GH, Terrence CF. Differential effect of ethosuximide and of electrical stimulation on inhibitory and excitatory mechanisms. Epilepsy Res1987;1:35-39.
31. Englander RN, Johnson RN, Brickley JJ, et al. Ethosuximide and bicuculline inhibition in petit mal epilepsy. Neurol Neurochir Psychiatr1977;18:265-275.
32. McLean MJ, MacDonald RL. Sodium valproate, but not ethosuximide, produces use- and voltage-dependent limitation of high frequency repetitive firing of action potentials of mouse central neurons in cell culture. J Pharmacol Exp Ther1986;237: 1001-1011.
33. Wenzel J, Krueger E, Mueller M. Hemmung pentylenetetrazolinduzierter hypersynchroner Aktivität im thalamokortikalen System durch Ethosuximid. Acta Biol Med Germ1971;26:567-572.
34. Kastner I, Rougerie A. Photisch Ausgelöste potentialfolgen nach lokaler Aplikation von Ethosuximid ins Corpus geniculatum laterale. Acta Biol Med Germ1978;37:677-679.
35. Englander RN, Johnson RN, Brickley JJ, et al. Effects of antiepileptic drugs on thalamocortical excitability. Neurology1977;27:1134-1139.
36. Nowack WJ, Johnson RN, Englander RN, et al. Effects of valproate and ethosuximide on thalamocortical excitability. Neurology1979;29:96-99.
37. Pelligrini A, Dossi RC, Dal Pos F, et al. Ethosuximide alters intrathalamic and thalamocortical synchronizing mechanisms: a possible explanation of its antiabsence effect.Brain Res1989;497: 344-360.
38. Huguenard JR, Prince DA. A novel type of T-current underlies prolonged Ca²⁺dependent burst firing in GABAergic neurons of rat thalamic reticular nucleus. J Neurosci1992;12:3804-3817.
39. Huguenard JR, Prince DA. Intrathalmic rhythmicity studied in vitro:nominal T-current modulation causes robust antioscillatory effects. J Neurosci 1994;14:5485-5502.

P.644

40. Inoue M, Duysens J, Vossen JMH, et al. Thalamic multiple-unit activity underlying spike-wave discarges in anesthetized rats. Brain Res1993;612:35-40.
41. Leresche N, Parri H, Erdemli G, et al. On the action of the antiabsence drug ethosuximide in the rat and cat thalamus. J Neurosci1998;18:4842-4845.
42. Klunk WE, Covey DF, Ferrendelli JA. Structure-activity relationships of alkyl substituted γ-butyrolactones and succinimides. Mol Pharmacol1982;22:444-450.
43. Klunk WE, Kalman BL, Ferrendelli JA, et al. Computer-assisted modeling of the picrotoxinin and γ-butyrolactone receptor site. Mol Pharmacol1983;23:511-518.
44. Barnes DM, Dichter MA. Effects of ethosuximide and tetramethylsuccinimide on cultured cortical neurons. Neurology1984; 34:620-625.
45. Coulter, DA, Huguenard JR, Price DA. Differential effects of petit mal anticonvulsants and convulsants on thalamic neurones: GABA current blockade. Br J Pharmacol1990;100:807-813.
46. Holland KD, Ferrendelli JA, Covey DF, et al. Physiological regulation of the picrotoxin receptor by γ-butyrolactones and γ-thiobutyrolactones in cultured hippocampal neurons.J Neurosci1990;10:1719-1727.
47. Holland KD, McKeon AC, Covey DF, et al. Binding interactions of convulsant and anticonvulsant γ-butyrolactones and γ-thiobutyrolactones with the picrotoxin receptor. J Pharmacol Exp Ther1990;254:578-583.
48. Skerritt JH, Johnston GAR. Interactions of some anesthetic, convulsant, and anticonvulsant drugs at GABA-benzodiazepine receptor-ionophore complexes in rat brain synaptosomal membranes. Neurochem Res1983;8:1351-1362.
49. Pitkänen A, Saano V, Tuomisto L, et al. Effect of anticonvulsant drugs on (³⁵S)t-butylbicyclophosphorothionate binding in vitroand ex vivo. Pharmacol Toxicol 1987;61:103-106.
50. Clifford DB, Lothman EW, Dodson WE, et al. Effect of anticonvulsant drugs on kainic acid-induced epileptiform activity. Exp Neurol1982;76:156-167.
51. Czuczwar SJ, Frey H-H, Löscher W. N-Methyl-D,L-aspartic acid-induced convulsions in mice and their blockade by antiepileptic drugs and other agents. In: Nisticô G, Morselli PL, Lloyd KG, et al., eds. Neurotransmitters, seizures, and epilepsy III.New York: Raven Press, 1986:235-246.
52. Sohn RS, Ferrendelli JA. Anticonvulsant drug mechanisms phenytoin, phenobarbital, and ethosuximide and Ca²⁺flux in isolated presynaptic endings. Arch Neurol 1976;33:626-629.
53. Crowder JM, Bradford HF. Common anticonvulsants inhibit Ca²⁺uptake and amino acid neurotransmitter release in vitro. Epilepsia 1987;28:378-382.
54. Nowycky MC, Fox AP, Tsien RW. More types of neuronal calcium channels with different agonist sensitivity. Nature1985; 316:440-443.
55. Coulter, DA, Huguenard JR, Price DA. Characterization of ethosuximide reduction of low-threshold calcium current in thalamic neurons. Ann Neurol1989;25:582-593.
56. Coulter, DA, Huguenard JR, Price DA. Specific petit mal anticonvulsants reduce calcium currents in thalamic neurons. Neurosci Lett1989;98:74-78.
57. Coulter, DA, Huguenard JR, Price DA. Differential effects of petit mal anticonvulsants and convulsants on thalamic neurones: calcium current blockade. Br J Pharmacol1990;100:800-806.
58. Kostyuk PG, Molokanova EA, Pronchuck NF, et al. Different action of ethosuximide on low- and high-threshold calcium currents in rat sensory neurons. Neuroscience1992;51:755-758.
59. Pfrieger FW, Veselovsky NS, Gottmann K, et al. Pharmacological characterization of calcium currents and synaptic transmission between thalamic neurons in vitro. J Neurosci1992;12: 4347-4357.
60. Sayer RJ, Brown AM, Schwindt PC. Calcium currents in acutely isolated human neocortical neurons. J Neurophysiol1993;69: 1596-1606.
61. Thompson SM, Wong RS. Development of calcium current subtypes in isolated rat hippocampal pyramidal cells. J Physiol (Lond)1991;439:671-689.
62. Fohlmeister JF, Adelman WJ, Brennan JJ. Excitable channel currents and gating times in the presence of anticonvulsant ethosuximide and valproate. J Pharmacol Exp Ther1984;230:75-81.
63. Willow M, Catterall WA. Inhibition of binding of [³H]batrachotoxinin A 20-α-benzoate to sodium channels by the anticonvulsant drugs diphenylhydantoin and carbamazepine.Mol Pharmacol1982;22:267-635.
64. Willow M, Kuenzel EA, Catterall WA. Inhibition of voltage-sensitive sodium channels in neuroblastoma cells and synaptosomes by the anticonvulsant drugs diphenylhydantoin and carbamazepine. Mol Pharmacol1984;25:228-234.
65. Gilbert JC, Buchan P, Scott AK. Effects of anticonvulsant drugs on monosaccharide transport and membrane ATPase activities of cerebral cortex. In: Harris P, Mawdsley C, eds. Epilepsy.Edinburgh: Churchill Livingstone, 1974:98-104.
66. Gilbert JC, Scott AK, Wyllie MG. Effects of ethosuximide on adenosine triphosphatase activities of some subcellular fractions prepared from rat cerebral cortex. Br J Pharmacol1974;50: 452P-453P.
67. Leznicki A, Dymecki J. The effect of certain anticonvulsants in vitroand in vivo on enzyme activities in rat brain. Neurol Neurochir Pol 1974;24:413-419.
68. Löscher W, Frey H-H. Effects of convulsant and anticonvulsant agents on level and metabolism of γ-aminobutyric acid in mouse brain. Naunyn Schmiedebergs Arch Pharmacol1977;296; 263-269.
69. Lin-Mitchell E, Chweh AY, Swinyard EA. Effect of ethosuximide alone and in combination with γ-aminobutyric acid receptor agonists on brain γ-aminobutyric acid concentration, anticonvulsant activity and neurotoxicity in mice. J Pharmacol Exp Ther1986;237:486-489.
70. Erwin VG, Deitrich RA. Inhibition of bovine brain aldehyde reductase by anticonvulsant compounds. Biochem Pharmacol1973;2:2615-2624.
71. Snead OC, Bearden LJ, Pergram V. Effect of acute and chronic anticonvulsant administration on endogenous γ-hydroxybutyrate in rat brain. Neuropharmacology1980;19:47-52.
72. Tabakoff B, von Wartburg JP. Separation of aldehyde reductases and alcohol dehydrogenase from brain by affinity chromatography: metabolism of succinic semialdehyde and ethanol. Biochem Biophys Res Commun1975;63:957-966.
73. Winters WD, Spooner CE. A neurophysiological comparison of gamma-hydroxybutyrate with pentobarbital in cats. Electroencephalogr Clin Neurophysiol1965;18:287-296.
74. Hechler V, Ratomponirina C, Maitre M. Gamma-hydroxbutyrate conversion into GABA induces displacement of GABAB binding that is blocked by valproate and ethosuximide. J Pharmacol Exp Ther1996;281:753-760.
75. Faradji H, Rousset C, Deilly G, et al. Sleep and epilepsy: a key role for nitric oxide? Epilepsia2000;41:794-801.
76. Matsui T, Nagafuji T, Kumanishi T, et al. Role of nitric oxide in pathogenesis underlying ischemic cerebral damage. Cell Mol Neurobiol1999;19:177-189.
77. Snead OC, Liu C-C. Gamma-hydroxybutyric acid binding sites in rat and human brain synaptosomal membranes. Biochem Pharmacol1984;33:2587-2590.
78. Snead OC, Nichols AC. γ-Hydroxybutyric acid binding sites: evidence for coupling to a chloride anion channel. Neuropharmaology1987;26:1519-1523.

P.645

79. Vayer P, Charlie B, Mandel P, et al. Effect of anticonvulsant drugs on γ-hydroxybutyrate release from hippocampal slices: inhibition by valproate and ethosuximide. J Neurochem1987;49:1022-1024.
80. Klunk WE, Ferrendelli JA. Reversal of the anticonvulsant action of ethosuximide by drugs that diminish CNS dopaminergic neurotransmission. Neurology1980;30:421.
81. Kobayashi K, Shirakabe T, Kishikawa H, et al. Catecholamine levels in penicillin-induced epileptic focus of the cat cerebral cortex. Acta Neurochir Suppl (Wien)1976;23:93-100.
82. Guberman AG, Gloor P, Sherwin AL. Response of generalized penicillin epilepsy in the cat to ethosuximide and diphenylhdantoin. Neurology1975;25:758-764.
83. Warter J-M, Vergnes M, Depaulis A, et al. Effects of drugs affecting dopaminergic neurotransmission in rats with spontaneous petit mal-like seizures. Neuropharmacology1988;27:269-274.
84. Micheletti G, Vergnes M, Marescaux C, et al. Antiepileptic drug evaluation in a new animal model: spontaneous petit malepilepsy in the rat. Arzneimittelforschung1985;35:473-475.
85. Gloor P, Fariello RG. Generalized epilepsy: some of its cellular mechanisms differ from those of focal epilepsy. Trends Neurosci1988;11:63-68.