Imipenem

Background & objectives: Epileptic seizures have been reported in patients on imipenem/cilastatin (Imi/Cil) therapy. To investigate contribution of N-methyl-D-aspartate (NMDA) receptors in inducing imipenem/cilastatin (Imi/Cil) seizures, the effects of competitive NMDA antagonist, APV [(±)-2-amino-5-phosphonovaIeric acid], non-competitive NMDA antagonist remacemide [(±)-2-amino-N-(1-methyl-1,2-diphenylethyl)-acetamide], and glycine receptor partial agonist HA-966 [(±)-(3-amino-1-hydroxypyrrolid-2-one)] on electroencephalographic (EEG) activity and behaviour were studied in rats.

Methods: Adult male Wistar albino rats were implanted with electrodes and cannulae were placed into the right lateral ventricle. Animals were divided into five groups: (i) saline (icv)+Imi/Cil (ii) APV (0.2 µmol)+Imi/Cil, (iii) APV (0.4 µmol)+Imi/Cil, (iv) remacemide (100 mg/kg, ip)+Imi/Cil, and (v) HA-966 (200 µg, icv)+Imi/Cil. The drugs were administered 30 min before icv injection of Imi/Cil (100/100 µg), and their effects on incidence of seizures, latencies to EEG changes and convulsions, severity, lethality and time to lethal outcome were studied.

Results: Imi/Cil provoked complete seizure response in all rats and all animals died within 10-18 min after the injection. EEC epileptiform activity preceded behavioral seizures. Clonic-tonic seizures were characterized by continuous bursts of high frequency high amplitude spikes in the EEG. The dose of 0.2 µmol of APV prolonged only the latency to the first EEG changes, while 0.4 µmol dose significantly influenced all seizure parameters. HA-966 increased only the latency to Imi/Cil-induced convulsions, while remacemide had no significant effect on seizure parameters.

Interpretation & conclusion: The results suggested that excitatory neurotransmission contributed to the generation and/or propagation of Imi/Cil-induced seizures in rats, and that the effects of NMDA antagonists depended on a particular binding site within the NMDA receptor complex, and affinity to that site.

Key words APV * EEG * HA-966 * imipenem/cilastatin * NMDA * remacemide * seizures

The epileptic seizures are mainly caused by a disturbance of equilibrium between excitation and inhibition in the central nervous system (CNS). Drugs, which restore this equlibrium by potentiation of inhibition, or by suppression of excitation, may be effective anticonvulsants. Practically all known antagonists of excitatory amino acids exhibit anticonvulsant action. APV [(±)-2-amino-5-phosphonovaleric acid], a competitive antagonist of N-methyl-D-aspartate (NMDA) receptors, blocked chemically-induced seizures1-4, convulsions induced by electrical kindling of amygdala in rats5, audiogenic seizures in DBA/2 mice6 and sound-induced seizures in metaphit (1-1-3-isothiocyanatophenyl-cyclohexyl-piperidine)-treated mice7 and rats8. Remacemide [2-amino-N-(1-methyl-1,2-diphenylethyl)-acetamide], a non-competitive NMDA receptor antagonist, was effective against seizures induced by electroshock, NMDA, kainate, 4-aminopyridine, audiogenic seizures in DBA/2 mice9,10, and decreased spike-wave discharges in absence epilepsy in rats11. HA-966 (3-amino-1-hydroxypyrrolid-2-one), a low-efficacy partial agonist of glycine receptor of the NMDA receptor complex, acts as functional antagonist in v/vo12,13. HA-966 antagonized seizures elicited by NMDA2,N-methyl-D, L-aspartate14 (NMDLA), soundinduced seizures in mice14, and raised the threshold in amygdala kindling epilepsy in rats13.

Imipenem is a broad-spectrum antibiotic, which is used for treatment of serious infections in clinics in combination with cilastatin, an inhibitor of enzyme dipeptidase. interestingly, preclinical studies have not revealed the convulsant action of imipenem, and seizures have been reported for the first time in patients on imipenem/cilastatin (Imi/Cil) therapy15,16. In experimental animals Imi/Cil induced seizures after systemic17-19 and icv administration17-20. Cilastatmperse, administered at high doses, did not provoke convulsions17. Convulsant effects of Imi/Cil depend on animal species and route of administration. In DBA/2 and C57 mice Imi/Cil provoked seizures in the form of running, clonus and tonus17,18, while in rats seizures were of limbic type19,20. The convulsant effect of imipenem in mice was due to decrease in inhibition and increase in excitation; GABAA agonist nuiscimol, [alpha]-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate antagonists, competitive and non-competitive NMDA antagonists effectively blocked Imi/Cil-induced seizures in DBA/2 mice18. Binding of [^sup 3^H]GABA to GABA^sub A^ receptors was decreased in the presence of imipenem21.

In order to investigate if NMDA receptors contribute to the development of Imi/Cil-induced epilepsy in rats, the present study was carried out to the effects of APV, remacemide and HA-966 on behavioural and EEG characteristics of Imi/Cil-induced seizures.

Material & Methods

The drugs used in this study were imipenem/cilastatin (Thienam®, Merck Sharp & Dohme B. V., Haarlem, Holland), APV (ICN Pharmaceuticals Inc, Costa Mesa, CA, USA), remacemide (AstraCharnwood, Loughborough, England), and HA-966 (Tocris, Bristol, England).

Adult male Wistar albino rats, weighing about 250 g, were used in experiments (Military Medical Academy Breeding Laboratories, Belgrade, Serbia). The study protocol was approved by the Animal Care Committee of University of Belgrade. The rats were housed individually in transparent plastic cages (50x35x30 cm) under standard conditions (temperature 22± 1°C, humidity 50%, lightdark cycle 12:12h). Food and water were continuously available.

Animals were anaesthetized with sodium pentobarbital (40 mg/kg, ip) and positioned in the stereotaxic apparatus. Cannulae were implanted into the right lateral ventricle (coordinates from bregma: AP=-1.3, L=2.0,4.5 mm deep from the skull surface)22, and electrodes were fixed to the skull with dental acrylic cement.

Rats were allowed to recover from surgery one week before the experiments started. The animals were randomly assigned to the following groups: (i) saline (icv)+Imi/Cil( 100/100 µg, icv, n=8), (H)APV (0.2 µmol, icv)+Imi/Cil (100/100 µg, icv, n=6), (iii) APV (0.4 µmol, icv)+Imi/Cil (100/100 µg, icv, n=8), (iv) remacemide (100 mg/kg, ip)+Imi/Cil (100/100 µg, icv, n=8), and (v) HA-966 (200 µg, icv)+Imi/Cil (100/100 µg, icv, n=6).

The drugs were freshly dissolved in physiological saline before administrations. Remacemide was injected ip, while APV and HA-966 were administered icv 30 min before icv injection of Imi/Cil. The volume of remacemide injection was 0.1 ml. APV, HA-966 and Imi/Cil were applied by a 10 µl Hamilton syringe. APV and HA-966 were applied in a volume of 5 µl, and Imi/Cil in a volume of 10 µl. The rate of injection was 1 µl/5 sec. During the administration the rats were gently restrained by hand.

EEG activity was recorded by means of an 8-channel EEG apparatus (Alvar, France). Analog data were digitized at a sampling rate of 128/sec and after analog to digital conversion the analog EEG data were stored on hard disk. EEG tracings were analyzed visually and power spectral analysis was provided by FFT (Fast Fourier Transformation).

The latency to seizures, latency to the first EEG changes, seizure severity, lethality and time to lethal outcome were measured.

Seizure severity grade was scored on the descriptive rating scale: 0 - normal behavior, 1 - twitching, 2 - fore limb clonus, head nodding, 3 - rearing, and 4 - clonictonic convulsions19.

At the end of experiments, before sacrifying the animals, 5 µl of blue ink was injected icv. Brains were examined to ensure that the dye was distributed throughout the ventricular spaces. The cannulae were found to be placed correctly in the lateral ventricle in all animals.

Data were statistically analysed using Kruskal-Wallis test and Mann-Whitney U test (for latencies to EEG changes, latencies to convulsions, seizure severity, and time to lethal outcome) and Fisher's exact probability test (for number of animals that died).

Results

In the EEG there were no signs of spontaneous epileptiform activity before Imi/Cil (100/100 µg, icv) injection, and all tracings had low spectral powers. All animals treated with Imi/Cil exhibited severe clonic-tonic seizures and died within 10-18 min (13.8±2.5) postinjection. EEG epileptiform activity preceded behavioral convulsions (Table). Clonic-tonic seizures were associated with continuous bursts of high-amplitude spikes occurring at 5-8 per sec, and increase in power spectra (Fig. 1).

Administration of APV produced hypotonia and ataxia in a dose-dependent manner, while in the EEG sharp waves and spikes were recorded (Fig. 2). The only significant effect of APV at a dose of 0.2 µmol was on latency to the first EEG discharges, which appeared later compared to rats treated with saline+Imi/Cil (P

Injection of remacemide (100 mg/kg, ip) elicited sedation, hypotonia, and decreased locomotor activity in animals. The righting reflex was preserved. Remacemide (100 mg/kg, ip) did not affect significantly any parameter of Imi/Cil-induced convulsions (Table). The latency to the first spikes after Imi/Cil injection was shorter in animals pretreated with remacemide, compared to Imi/Cil group (Table). Remacemide did not suppress epileptic activity in the EEG elicited by Imi/Cil (100/100 µg,icv) (Fig.3).

Administration of HA-966 (200 µg, icv) provoked hypotonia and occasionally twitching of head and fore limbs in 3 out of 6 animals. HA-966 significantly (P

Discussion

The anticonvulsant action of APV against Imi/Cilinduced seizures in rats was in agreement with the results of previous studies in rodent models4,6,8. APV inhibited Imi/Cil-induced seizures in a dose-dependent fashion. High dose of APV (0.4 µmol) significantly influenced all parameters of Imi/Cil-induced seizures, but did not abolish convulsions, or lethality. Since APV had anticonvulsant effect against Imi/Cil (100/100 µg, icv)-induced seizures without affecting epileptiform activity, it might be suggested that APV acted as an anticonvulsant rather than an antiepileptic agent in this model of epilepsy. In contrast to this result, APV suppressed both myoclonic jerks and high-voltage spikes elicited by penicillin in rats4. The effective dose of APV against Imi/Cil- induced seizures in rats was higher than in other epilepsy models2,6,8, in which the increase of excitation in the CNS was the main factor provoking seizures. APV inhibited seizures elicited by electrical stimulation of the amygdala in a dose range tested in the present study (0.2 µmol)23. The different potency of APV in various models of epilepsy may be due to a different contribution of excitatory neurotransmission and NMDA receptors in seizure initiation and propagation.

HA-966 (200 µg, icv) significantly increased only the latency to Imi/Cil-induced seizures, while lower doses (10-100 µg, icv) had no significant effect on latency to NMDLA-elicited seizures in mice14. HA-966 (icv) blocked NMDA-induced convulsions with ED50 of 184 µg, but was not effective in antagonizing seizures elicited by kainate and quisqualate in mice2. Higher doses of HA-966 were not used in the present study, since the dose of 200 µg produced proconvulsant effect, which was not reported in studies using lower doses. Generally, all categories of NMDA receptor antagonists at high doses exert proconvulsant effects2,8,13,24.

Remacemide (100 mg/kg, ip) caused no significant change in any parameter of Imi/Cil-induced seizures in rats. In other models remacemide was effective at much lower doses (6-66 mg/kg)10. MK-801 (5-methyl-10, 11-dihydro-5H-dibenzo[a,d]cyclohepten-5, 10-imine), another non-competitive NMDA receptor antagonist, significantly decreased the incidence and severity of Imi/Cil-induced convulsions in DBA/2 mice after both ip and icv application18. In contrast to remacemide, MK-801 has high affinity to PCP (phencyclidine) recognition site of NMDA receptor complex. It is possible that this difference in affinity, besides species difference, contributes to the different potency of non-competitive NMDA antagonists against Imi/Cil-induced convulsions. It has been reported that remacemide decreased epileptiform discharges in absence epilepsy11, and that normalized EEG in epileptic patients25. Adverse behavioural effects of remacemide and HA-966 in our model were in agreement with literature14, 26. Remacemide did not suppress seizures induced by bicuculline, picrotoxine, strychnine and pentylenetetrazole10, in which decrease of inhibition in the CNS plays a key role for generation of seizures, like for imipenem.

The results of this study, together with that of a previous study in mice18, indicated the importance of NMDA receptors in inducing Imi/Cil seizures, but has not ruled out contribution of other mechanisms in provoking the seizures. It has been reported that GABA-ergic system influenced Imi/Cil-elicited seizures18, while the roles of cholinergic, dopaminergic, opioid and peptidergic systems are to be determined.

Acknowledgment

This research was supported by a grant No. 1943 from the Republic Ministry of Science (Serbia). APV was a gift from ICN Pharmaceuticals Inc., HA-966 from Tocris, and remacemide from Astra Charnwood.

References

1. Meldrum BS, Croucher MJ, Czuczwar SJ, Collins JF, Curry K, Joseph M, et al. A comparison of the anticonvulsant potency of (±)2-amino-5-phosphonopentanoic acid and (±)2-amino-7-phosphonoheptanoic acid. Neuroscience 1983; 9 : 925-30.

2. Koek W, Colpaert FC. Selective blockade of N-methyl-D-aspartate (NMDA)-induced convulsions by NMDA antagonist and putative glycine antagonists: relationship with phencyclidine-like behavioral effects. J Pharmacoi Exp Ther 1990; 252:349-57.

3. Czuczwar SJ, Frey HH, Loscher W. Antagonism of N-methyl-D, L-aspartic acid-induced convulsions by antiepileptic drugs and other agents. Eur J Pharmacol 1985; 108 : 273-80.

4. Loeb C, Patrone A, Besio G, Balestrino M, Mainardi P. The excitatory amino acid antagonist amino-phosphono-valeric acid (APV) provides protection against penicillin-induced epileptic activity in the rat. Epilepsy Res 1990; 6 : 249-51.

5. Croucher MJ, Cotterell KL, Bradford HF. Competitive NMDA receptor antagonists raise electrically kindled generalized seizure thresholds. Neurochem Res 1992; 17 : 409-13.

6. Croucher MJ, Collins JF, Meldrum BS. Anticonvulsant action of excitatory amino acid antagonists. Science 1982; 276 : 899-901.

7. Debler EA, Lipovac MN, Lajtha A, Zlokovic BV, Dunlop DS, Jacobson AE, et al. Metaphit-induced audiogenic seizures in mice: I. Pharmacologie characterization. Epilepsia 1993; 34 : 201-10.

8. Zivanovic D, Susie V, Stanojlovic O. Inhibition of metaphitinduced audiogenic seizures by APV in rats. Physiol Res 1999; 48: 149-56.

9. Stagnitto ML, Palmer GC, Ordy JM, Griffith RC, Napier JJ, Becker CN, et al. Preclinical profile of remacemide: a novel anticonvulsant effective against maximal electroshock seizures in mice. Epilepsy Res 1990; 7 : 11-28.

10. Davies JA. Remacemide hydrochloride: a novel antiepileptic agent. Gen Pharmacol 1997; 28 : 499-502.

11. Van Luijtelaar ELJM, Coenen AML. Effects of remacemide and its metabolite FPL 12495 on spike-wave discharges, electroencephalogram and behaviour in rats with absence epilepsy. Neuropharmacology 1995; 34 : 419-25.

12. Chiamulera C, Costa S, Reggiani A. Effect of NMDA - and strychnine-insensitive glycine site antagonists on NMDA mediated convulsions and learning. Psychopharmacology (Berl) 1990; 102: 551-2.

13. Rundfeldt C, Wlaz P, Loscher W. Anticonvulsant activity of antagonists and partial agonists for the NMDA receptor-associated glycine site in the kindling model of epilepsy. Brain Res 1994; 653: 125-30.

14. Singh L, Donald AE, Foster AC, Hutson PH, Iversen LL, Iversen SD, et al. Enantiomers of HA-966 (3-amino-1-hydroxypyrrolid-2-one) exhibit distinct central nervous system effects: (+)-HA-966 is a selective glycine/N-methyl-D-aspartate receptor antagonist, but (-)-HA-966 isapotenty-butyrolactone-like sedative. Proc Natl Acad Sci USA 1990; 87 : 347-51.

15. Barza M. Imipenem: first of a new class of beta-lactam antibiotics. Ann Intern Med 1985; 103 : 552-60.

16. Leo RJ, Ballow CH. Seizure activity associated with imipenem use: clinical case reports and review of the literature. DICP 1991; 25: 351-4.

17. De Sarro A, Imperatore C, Mastroeni P, De Sarro G. Comparative convulsant potencies of two carbapenem derivatives in C57 and DBA/2 mice. J Pharm Pharmacol 1995; 47 : 292-6.

18. de Sarro G, Ammendola D, Nava F, De Sarro A. Effects of some excitatory amino acid antagonists on imipenem-induced seizures in DBA/2 mice. Brain Res 1995; 677 : 131-40.

19. Kamei C, Kitazumi K, Tsujimoto S, Yoshida T, Tasaka K. Comparative study of certain antibiotics on epileptogenic property, including (1Rpi, 5S, 6S)-2-[(6,7-dihydro-5H-pyrazolo[1,2-a][1,2,4] triazolium-6-yl)] thio-6-[(R)-1-hydroxyethyl]-1-methyl-carbapenem-3-carboxylate (LJC10627), acarbapenem antibiotic with broad antimicrobial spectrum. J Pharmacobio-Dyn 1991; 14 : 509-17.

20. Hikida M, Masukawa Y, Nishiki K, Inomata N. Low neurotoxicity of LJC 10627, a novel 1[beta]-methyl carbapenem antibiotic: inhibition of [gamma]-aminobutyric acid^sub A^, benzodiazepine, and glycine receptor binding in relation to lack of central nervous system toxicity in rats. Antimicrob Agents Chemother 1993; 37 : 199-202.

21. Williams PD, Bennett DB, Comerski CR. Animal model for evaluating the convulsive liability of [beta]-lactam antibiotics. Antimicrob Agents Chemother 1988; 32 : 758-60.

22. Paxinos G, Watson C. The rat brain in stercotaxic coordinates. London: Academic Press; 1982.

23. Cain DP, Desborough KA, McItrick DJ. Retardation of amygdala kindling by antagonism of NMD-aspartate and muscarinic cholinergic receptors: evidence for the summation of excitatory mechanisms in kindling. Exp Neural 1988; 100 : 179-87.

24. Klockgether T, Turski L, Schwarz M, Sontag KH, Lehmann J. Paradoxical convulsant action of a novel non-competitive N-memyl-D-aspartate (NMDA) antagonist, tiletamine. Brain Res 1988; 461 : 343-8.

25. Owen L, Cresswell P, Gifford C, McDade G, Mawer G. Influence of remacemide on EEG in chronic epilepsy. Epilepsy Res 1992; 1 (Suppl A) 7.

26. Garske GE, Palmer GC, Napier JJ, Griffith RC, Freedman LR, Harish EW, et al. Preclinical profile of the anticonvulsant remacemide and its enantiomers in the rat. Epilepsy Res 1991; 9 : 161-74.

DraganaZivanovic, Olivera Stanojlovic & Veselinka Susic

Institute of Physiology, School of Medicine, University of Belgrade, Belgrade, Serbia

Received April 8, 2003

Reprint requests : Dr Dragana Zivanovic, Institute of Physiology, School of Medicine

Visegradska 26/11, 11000 Belgrade, Serbia

e-mail : zdragana@afrodita.rcub.bg.ac.yu

Copyright Indian Council of Medical Research Feb 2004
Provided by ProQuest Information and Learning Company. All rights Reserved