Brain Research 884 (2000) 31–34
www.elsevier.com / locate / bres
Research report
Chlormethiazole inhibits epileptiform activity by potentiating GABA
Areceptor function
1
*
Ruth M. Empson , Veronica J. Gee, Malcolm J. Sheardown, Nigel R. Newberry
Vernalis Research Ltd, Oakdene Court, 613 Reading Road, Winnersh, Wokingham RG41 5UA, UK
Accepted 15 August 2000
Abstract
Chlormethiazole has sedative, hypnotic, anticonvulsant and neuroprotective properties. Using in vitro grease-gap recordings, we show
21
that it inhibits epileptiform activity in neocortical slices superfused with Mg -free medium (IC50|200 mM). At an antiepileptic concentration (300mM), chlormethiazole potentiated the action of exogenously applied GABA (1 mM) but did not affect responses to the glutamate receptor agonists N-methyl-D-aspartate (10 mM) or L-quisqualic acid (3mM). The GABAA receptor antagonist N-methyl-bicuculline (50mM) reduced chlormethiazole’s potency to inhibit the epileptiform activity. These results indicate that chlormethiazole’s anticonvulsant action is likely mediated by potentiating GABA ergic inhibition rather than by antagonising glutamatergic excitation.A
2000 Elsevier Science B.V. All rights reserved.
Theme: Disorders of the nervous system
Topic: Epilepsy: anticonvulsant drugs
Keywords: GABA receptor; Glutamate receptor; Epilepsy; Stroke
1. Introduction tiate GABAergic function rather than inhibiting gluta-matergic function. However, its anticonvulsant potency has
Chlormethiazole (clomethiazole, Heminevrin , rarely been directly compared to its potency to interact
Zendra ) was first patented in 1938 (Swiss patent 200,248 with amino acid receptor function. Here we succeed in to Hoffman La Roche). It has sedative, hypnotic and relating its anticonvulsant effects to its actions on both anticonvulsant properties and it is neuroprotective in GABAergic and glutamatergic receptor function by using a animal models of stroke [5]. Its main clinical use has been robust in vitro electrophysiological preparation: the cortical as an anticonvulsant, particularly in the treatment of ‘‘wedge’’ of Harrison and Simmonds [10]. We demonstrate ethanol withdrawal syndrome [11,18] and status epilep- that chlormethiazole inhibits epileptiform activity in a ticus [12]. Despite its age, the mechanism responsible for concentration-dependent manner. Because this preparation its anticonvulsant action remains unclear. In a key study responds to quisqualic acid, NMDA and GABA, we were using ionophoretic drug applications onto single neurones able to relate its antiepileptic potency to its actions on in the rat brain in vivo, Gent and Wacey [4], showed that three amino acid receptor-mediated responses. We con-chlormethiazole preferentially potentiated the inhibitory clude that at antiepileptic concentrations chlormethiazole action of GABA — higher ejection currents were needed does not directly inhibit glutamatergic receptor function to depress glutamate-induced excitation. That work has and most likely acts by potentiating GABAA receptor been confirmed and extended [1–3,6,8,9,13,15,17,19]. It is function, as previously suggested.
now generally agreed that chlormethiazole acts to
poten-*Corresponding author. Tel.:144-118-989-9313; fax: 144-118-989- 2. Materials and methods 9300.
E-mail address: n.newberry@vernalis.com (N.R. Newberry).
1 Unanaesthetised male Sprague–Dawley rats (80–120 g, Present address: School of Biological Sciences, Royal Holloway
Uni-versity of London, Egham, Surrey TW20 0EX, UK. Harlan–Olac) were killed by rapid decapitation (Home
32 R.M. Empson et al. / Brain Research 884 (2000) 31 –34
Office-approved Schedule 1 method). Coronal sections of has shown that these responses are mediated by AMPA, forelimb / frontal neocortex (500-mm-thick) were cut in NMDA and GABAA receptors, respectively [10,14,16]. ice-cold artificial cerebrospinal fluid (aCSF) using an When studied together, the agonists were applied alternate-Oxford Vibratome, trimmed to¯1–1.5-mm-wide pieces of ly (see Fig. 1). Chlormethiazole was superfused for $30 tissue and placed in a two-compartment chamber with the min and the agonists reapplied in its presence. The drug ventral part of the cortex protruding through a greased-gap sources were as follows: chlormethiazole (Tocris), NMDA in the Perspex partition dividing the two chambers. The (Sigma), L-quisqualate (Tocris), GABA (Sigma),
N-chamber containing the dorsal part of the neocortex (¯0.5 methyl-bicuculline (Sigma), tetrodotoxin (TTX,
Calbioch-ml) was continuously perfused (2–2.5 ml / min) with aCSF em), (6)-APV (Tocris and Sigma). The salts were Analar while the aCSF in the chamber containing the corpus grade. Data are expressed as mean6S.E.M. Statistical callosum was static. The experiments were performed at comparisons were made with the unpaired Student’s t-test. room temperature. The superfusing aqueous aCSF initially
contained, in mM, NaCl 135, KCl 3, NaH PO2 4 1.25,
MgCl 1, CaCl 2, glucose 5 and NaHCO 26 and was2 2 3 3. Results
pre-equilibrated with 95% O / 5% CO . Twenty minutes2 2
21
after setting up the slices the superfusate was switched to In the ‘‘Mg -free’’ superfusate, the spontaneously one without MgCl2 which led to the appearance of occurring epileptiform events occurred at a rate of spontaneous, synaptically-mediated, depolarising, epilep- 1.760.17 times per minute (n533). These events were tiform discharges — sometimes called paroxysmal dis- mediated in part by NMDA receptors as they were charges. The frequency of these events was counted over antagonized by (6)-APV with an IC50 of 12 mM (not 10 min periods. In some experiments, 0.5mM tetrodotoxin shown, cf. [9].). Chlormethiazole reduced the frequency of was included in the MgCl - free medium to prevent the2 these events in a concentration-dependent manner (IC50|
synaptically driven spontaneous events. Depolarising re- 200mM, Fig. 1B). At 300mM, it had no significant effect sponses were induced by superfusing the glutamate agon- on the depolarisations induced by either quisqualate or ists, L-quisqualic acid (3 mM) and N-methyl-D-aspartate NMDA (Fig. 1A). In contrast, at this concentration it
(NMDA, 10 mM), and g-aminobutyric acid (GABA, 1 significantly increased responses to superfused GABA by mM) for 1 min periods every 20 min. Previous research 160614% over control values (n57, e.g. Fig. 1A). We
R.M. Empson et al. / Brain Research 884 (2000) 31 –34 33
therefore considered that chlormethiazole may be inhib- duces NMDA-induced responses indirectly [3,6], but the possibility that chlormethiazole acted at a novel site on the iting synaptic activity in this slice by potentiating
NMDA receptor could not be excluded. In our experi-GABA ergic inhibition. The superfusion of the experi-GABAA A
ments, we clearly demonstrated that chlormethiazole re-receptor antagonist N-methyl-bicuculline (50 mM)
in-duces glutamate receptor function indirectly by potentiat-creased the frequency of epileptiform events by on average
ing GABA ergic neurotransmission. 40% suggesting that there was indeed a GABA ergicA A
In conclusion, our results support the hypothesis that inhibitory tone in the slice. Furthermore, chlormethiazole’s
chlormethiazole exerts its anticonvulsant actions predomi-inhibition of spontaneous epileptiform activity was
an-nantly via a potentiation of GABA ergic inhibition rather tagonised by this compound with the IC50 for chlor- A
than by its weak antagonism of glutamate receptor activa-methiazole being shifted to.1 mM (Fig. 1B).
tion. An action of chlormethiazole on glutamate receptor
function was demonstrable at the highest concentration tested (1 mM) when chlormethiazole did reduce the effects
References
of quisqualate (although not significantly) and particularly, NMDA (Fig. 1C). This effect was probably mediated
[1] J.I. Addae, T.W. Stone, Effects of anticonvulsants on responses to
indirectly since it was not seen if synaptic inhibition was
excitatory amino acids applied topically to rat cerebral cortex, Gen.
prevented by the combined application of the sodium
Pharmacol. 19 (1988) 455–462.
channel blocker tetrodotoxin (0.5 mM) and the GABAA [2] A.J. Cross, J.M. Sterling, T.N. Robinson, D.M. Bowen, P.T. Francis,
receptor antagonist N-methyl-bicuculline (50 mM) (Fig. A.R. Green, The modulation by chlormethiazole of the GABA -A
1C). receptor complex in rat brain, Br. J. Pharmacol. 98 (1989) 284–290.
[3] A.J. Cross, M.F. Snape, A.R. Green, Chlormethiazole antagonizes seizures induced by N-methyl-DL-aspartate without interacting with the NMDA receptor complex, Psychopharmacology 112 (1993)
4. Discussion 403–406.
[4] J.P. Gent, T.A. Wacey, The effects of chlormethiazole on single unit
The principal finding of this study is that chlor- activity in rat brain; interactions with inhibitory and excitatory neurotransmitters, Br. J. Pharmacol. 80 (1983) 439–444.
methiazole inhibits epileptiform activity in neocortical
[5] A.R. Green, Chlormethiazole (Zendra) in acute ischaemic stroke:
slices in vitro. At concentrations of the drug that inhibit
basic pharmacology and biochemistry and clinical efficacy,
Phar-epileptiform activity, it potentiates the action of GABA but macol. Ther. 80 (1998) 123–147.
has no significant effect on the actions of two glutamate [6] A.R. Green, A. Misra, K.E. Hewitt, M.F. Snape, A.J. Cross, An
agonists. Additionally, a GABAA receptor antagonist in- investigation of the possible interactions of clomethiazole with glutamate and ion channel sites as an explanation of its
neuroprotec-hibits its ‘‘anti-epileptic’’ action. This is the first
demon-tive activity, Pharmacol. Toxicol. 83 (1998) 90–94.
stration of all these actions in an in vitro model of
[7] A.R. Green, T.K. Murray, A. Misra, M.F. Snape, J.A. Jones, A.J.
epilepsy. The value of using an in vitro rather than an in Cross, The metabolism of chlomethiazole in gerbils and the neuro-vivo preparation has allowed the direct comparison of protective and sedative activity of the metabolites, Br. J. Pharmacol.
three amino acid neurotransmitter functions in the same 129 (2000) 95–100.
[8] T.G. Hales, J.J. Lambert, Modulation of GABA and glycine
preparation and removed any confounding interactive A
receptors by chlormethiazole, Eur. J. Pharmacol. 210 (1992) 239–
effects of the anaesthetic on the actions of chlormethiazole.
246.
Moreover the actions of chlormethiazole to potentiate [9] N.L. Harrison, M.A. Simmonds, Two distinct interactions of barbi-GABA responses and inhibit the epileptiform activity turates and chlormethiazole with the GABAA receptor complex in
occurred at concentrations similar to those observed to be rat cuneate nucleus in vitro, Br. J. Pharmacol. 80 (1983) 387–394. [10] N.L. Harrison, M.A. Simmonds, Quantitative studies on some
neuroprotective in a gerbil model of ischaemia,¯200mM
antagonists of N-methyl-D-aspartate in slices of rat cerebral cortex,
[7]. This suggests that the GABA potentiation mechanism
Br. J. Pharmacol. 84 (1985) 381–391.
may also be relevant to the therapeutic, neuroprotective [11] S.K. Majumdar, Chlormethiazole: current status in the treatment of actions of chlormethiazole. the acute ethanol withdrawal syndrome, Drug Alcohol Depend. 27
The action of chlormethiazole on neocortical GABAA (1991) 201–207.
[12] P.J. Martin, P.A. Millac, Status epilepticus: management and
out-receptor function was consistent with previous studies
come of 107 episodes, Seizure 3 (1994) 107–113.
showing that it binds to a site on this receptor that is
[13] E.J. Moody, P. Skolnick, Chlormethiazole: neurochemical actions at
distinct from the barbiturate and benzodiazepine sites the gamma-aminobutyric acid receptor complex, Eur. J. Pharmacol. [2,8,9,13,19]. We confirmed the weak antagonist action on 164 (1989) 153–158.
quisqualate-induced responses on this preparation [17] and [14] I. Phillips, K.F. Martin, K.S.J. Thompson, D.J. Heal, GABA-evoked depolarisations in the rat cortical wedge: involvement of GABA
the depression of NMDA–induced responses as seen in A 2
receptors and HCO3 ions, Brain Res. 798 (1998) 330–332.
vivo [3,6,15]. However, both of the effects we observed
[15] S. Pilip, E.M. Urbanska, S.J. Cruczwar, Z. Kleinrok, W.A. Turski,
were likely to be indirect as blocking GABAergic synaptic Chlormethiazole anticonvulsant efficacy diminished by N-methyl-D -transmission prevented them. It has been suggested from aspartate but not kainate in mice, Eur. J. Pharmacol. 345 (1998)
34 R.M. Empson et al. / Brain Research 884 (2000) 31 –34
[16] M.J. Sheardown, A new and specific non-NMDA receptor antago- [18] M. Skrobalova, J. Styk, The use of hemineurine (Astra) in alcoholic nist, FG 9065, blocks L-AP4-evoked depolarisation in rat cerebral delirium, Acta Nerv. Super. (Praha) 7 (1965) 254–255.
cortex, Eur. J. Pharmacol. 148 (1988) 471–474. [19] Y. Zhong, M.A. Simmonds, Interactions between loreclazole, chlor-[17] M.A. Simmonds, A.L. Horne, Antagonism of excitatory amino acids methiazole and pentobarbitone at GABA(A) receptors: functional
