Effects of Some Excitatory Amino Acid Antagonists and Drugs Enhancing

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 1997, p. 427–434 0066-4804/97/$04.0010 Copyright q 1997, American Society for Microbiology

Vol. 41, No. 2

Effects of Some Excitatory Amino Acid Antagonists and Drugs Enhancing g-Aminobutyric Acid Neurotransmission on Pefloxacin-Induced Seizures in DBA/2 Mice G. DE SARRO,1 F. NAVA,1 G. CALAPAI,2

AND

A. DE SARRO3*

Chair of Pharmacology, Department of Experimental and Clinical Medicine, School of Medicine, University of Reggio Calabria, Catanzaro,1 and Chair of Pharmacology2 and Chair of Chemotherapy,3 Institute of Pharmacology, School of Medicine, University of Messina, Messina, Italy Received 11 June 1996/Returned for modification 9 September 1996/Accepted 20 November 1996

The behavioral and convulsant effects of pefloxacin (PEFLO), a quinolone derivative, were studied after intraperitoneal (i.p.) administration to Dilute Brown Agouti DBA/2J (DBA/2) mice, a strain genetically susceptible to sound-induced seizures. The anticonvulsant effects of some excitatory amino acid (EAA) antagonists acting at N-methyl-D-aspartate (NMDA) or a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and kainate (KA) receptors and of some compounds enhancing g-aminobutyric acid (GABA)-ergic transmission against seizures induced by PEFLO were also evaluated. The present study demonstrated that both groups of compounds administered i.p. or intracerebroventricularly were able to protect against seizures induced by PEFLO. However, ifenprodil and (6)-a-(chlorophenyl)-4-[(4-fluorophenyl)methyl]-1-piperidineethanol (SL 82.0715), two compounds acting on the polyamine site of the NMDA receptor complex, were unable to provide any protection. The relationship between the different sites of action and the anticonvulsant activities of these derivatives were discussed. Although the main mechanism of PEFLO-induced seizures cannot be easily determined, potential interactions with the receptors of EAA exist. In fact, antagonists of EAA, and in particular, those acting at NMDA receptors, were able to increase the threshold for the seizures or to prevent the seizures induced by PEFLO, while compounds acting at the polyamine site did not provide any protection. The AMPA-KA receptor antagonists were also able to exert anticonvulsant activity, but with minor potency in comparison to those of NMDA antagonists. In addition, the fact that compounds enhancing GABA-ergic neurotransmission were also able to protect the mice against seizures induced by PEFLO suggests an involvement of GABA system. tors (35), we may suppose a possible interaction of quinolones with glutamate receptor binding sites. The genetically epilepsy-prone mouse, or the Dilute Brown Agouti DBA/2J (DBA/2) mouse, has been known since 1947 to be a strain of mice susceptible to audiogenic seizures (21). In fact, mice of this strain undergo an age-dependent (between 16 and 30 days) sequence of generalized convulsions when they are exposed to a loud mixed-frequency sound (12 to 16 kHz; 90 to 120 dB) such as a doorbell (4, 6). The audiogenic seizures in DBA/2 mice is a genetic form of reflex epilepsy; the counterpart in humans may be brain stem or centrencephalic epilepsy, e.g., “absence” or “musicogenic” seizures (4–6, 19, 33). In addition, it has been reported that DBA/2 mice develop resistance to sound-induced seizures with maturity but still have increased susceptibility to seizures resulting from a variety of nonaudiogenic convulsant treatments including chemical and physical stimuli (4, 5, 19). This change in seizure incidence with age is one of the more striking similarities between the human and the mouse forms of epilepsy. Thus, DBA/2 mice have been considered an excellent animal model for the study of certain kinds of human epilepsy and for testing convulsant or anticonvulsant drugs (4, 5, 7). In fact, we have previously demonstrated that the convulsant properties of antibiotics and chemotherapies can be more easily evaluated with this model than with models with other strains of mice (9, 12, 14). This mouse model was chosen for the examination of the effects of PEFLO since it might mirror PEFLO treatment of patients with predisposing epileptic factors. In previous studies we evaluated the proconvulsant effects of some quinolones on seizures induced by imipenem and cefazolin administration to DBA/2

Because of their excellent antibacterial activities, quinolone agents have been widely adopted for use in clinical practice. In particular, pefloxacin (PEFLO) shows a very broad spectrum of activity against gram-positive and gram-negative bacteria and is widely distributed in tissues. This explains why PEFLO is successful against various systemic infections. Although recently developed quinolones are less toxic than earlier compounds, they still produce a very few incidences of various adverse effects on the central nervous system such as dizziness, headache, seizures, and hallucinations (2, 17, 20, 25, 26). We have previously reported the proconvulsant effects of several quinolones when administered together with aminophylline in rats and with b-lactam derivatives in mice (9–12). The convulsant actions of quinolones have been attributed to the inhibition of g-aminobutyric acid (GABA) binding to its receptor (1, 31, 39, 41). However, the concentrations of quinolones able to interact with the GABA system are rather high and varied among the different quinolones tested by a factor of about 100. Thus, it appears questionable whether a specific interaction of quinolones with GABA receptors alone can explain the convulsant activities of these compounds (16). In addition, some in vivo studies indicated that dopamine, opioid, and glutamergic receptors may also be involved in the effects of quinolones on the central nervous system (16, 42). Because quinolones possess some chemical similarities to kynurenic acids, which may be endogenous ligands for excitatory amino acid (EAA) recep* Corresponding author. Mailing address: Institute of Pharmacology, School of Medicine, Piazza XX Settembre 4, 98122 Messina, Italy. Phone: 090/712533. Fax: 090/661029. 427

428

DE SARRO ET AL.

ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. EAA antagonists used in the present study Abbreviationa

Mol wt

Dose range

Pretreatment time (min)

3-[(6)-2-Carboxypiperazin-4-yl]propyl-1-phosphonic acid 3-[(6)-2-Carboxypiperazin-4-yl]propenyl-1-phosphonic acid

CCPb CPPeneb

252.2 268.2

(1)-5-Methyl-10,11-dihydro-5H-dibenzo(a,d)-cyclohepten-5,10-imine maleate or dizolcipine (6)(E)-2-Amino-4-methyl-5-phosphono-3-pentenoate ethyl ester Ifendopril tartrate (6)-a-(Chlorophenyl)-4-(4-fluorophenyl)methyl)-1-piperidineethanol 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)-quinoxaline

MK-801e

337.48

CGP 39551b IFEe SL 82.0715e NBQXf

245.1 400.5 384.32 342

Methylenedioxy-5H-2,3-benzodiazepine hydrochloride 3,5-Dihydro-7,8-dimethoxy-1-phenyl-4H-2,3-benzodiazepin-4-one (2)-2-Amino-7-phosphonic acid Kynurenic acid g-D-Glutamylaminomethylsulfonic acid 6-Cyano-7-nitroquinoxaline-2,3-dione

GYKI 52466f 2,3-BZ-2f AP7b KYNAg g-D-GAMSf CNQXf

329.79 296.15 225.2 189.2 240.2 232.16

2–80 nmol/gc 2–30 nmol/gc; 5–60 pmol/ moused 0.15–0.75 nmol/gc; 100–800 pmol/moused 1–8 nmol/gc 1–2.5 nmol/gc 1–2.6 nmol/gc 7.3–87.7 nmol/gc; 0.33–6.6 nmol/moused 10–100 nmol/gc 16.9–101 nmol/gc 0.1–0.5 nmol/moused 0.5–5 nmol/moused 300–600 pmol/moused 1–10 nmol/moused

30 30 15 30 15 30 30 30 5 5 5 30 15 15 15 15

Compound

a CCP and CPPene were kindly supplied by P. L. Herrling (Sandoz Ltd., Berne, Switzerland), MK-801 (dizocilpine) was kindly supplied by A. Wilkins (Merck Sharp & Dohme, Rome, Italy), CGP 39551 was kindly supplied by M. Schmutz (CIBA-GEIGY, Basel, Switzerland), IFE and SL 82.0715 were kindly supplied by J. Alexander (LERS-Synthelabo, Paris, France), NBQX was kindly supplied by T. Honore9 (Novo Nordish, Malov, Denmark), GYKI 52466 was purchased from RBI (Natick, Mass.), 2,3-BZ-2 was kindly supplied by A. Chimirri (Department of Medicinal Chemistry, University of Messina, Messina, Italy), AP7 and KYNA were purchased from Sigma (St. Louis, Mo.), and g-D-GAMS and CNQX were purchased from Tocris (Buckhurst Hill, United Kingdom). b Competitive NMDA receptor antagonists. c Administered i.p. d Administered i.c.v. e Noncompetitive NMDA receptor antagonists. f AMPA-KA receptor antagonists. g Broad-spectrum EAA antagonist.

mice and found that PEFLO was the most potent of the quinolones tested (9, 12). Thus, with these concepts in mind, in this experimental study we used PEFLO to treat DBA/2 mice with the intention of gaining information on the possible interference of EAA antagonists and of drugs enhancing GABA-ergic transmission with the mechanisms responsible for PEFLO-induced seizures. MATERIALS AND METHODS Subjects. DBA/2 mice (weight, 16 to 24 g; age, 42 to 48 days) were purchased from Charles River (Calco, Como, Italy). Upon arrival at the laboratory, they were housed at 10 mice per cage in a controlled environment (temperature, 22 6 18C; light-dark cycle, lights on from 0800 to 2000 h), with food and water available ad libitum.

TABLE 2. Drugs enhancing GABA-ergic transmission used in the present study Compound

Muscimol g-Vinyl-GABA Baclofen Diazepam Tiagabine hydrochloride

AbbreviaMol wt tiona

MSCb GVGd BACf DIAg TIAh

Range of dose

114.1 0.125–0.5 nmol/mousec 129.18 7.5–15.0 mmol/ge 213.7 11.7–93.6 nmol/ge 284.76 1.75–14 nmol/ge 412 1.2–9.7 nmol/ge

Pretreatment time (min)

15 120 30 30 30

a GVG was kindly supplied by M. Seiler (Merrell-Down, Strasbourg, France), BAC and MSC were purchased from Sigma, TIA was kindly supplied by E. B. Nielsen (Novo-Nordish, Malov, Denmark), and DIA (Valium) was from Roche, Milan, Italy. b GABAA agonist. c Administered i.c.v. d GABA transaminase inhibitor. e Administered i.p. f GABAB agonist. g GABAA-benzodiazepine receptor complex agonist. h GABA uptake inhibitor.

Testing of anticonvulsant activity. We administered PEFLO and anticonvulsant compounds intraperitoneally (i.p.) in a volume of 0.1 ml/10 g of body weight; however, some compounds that do not cross the blood-brain barrier were administered intracerebroventricularly (i.c.v.) under ethyl ether anesthesia in a volume of 5 ml/mouse by the method of De Sarro et al. (13). Briefly, for i.c.v. administration, the injection needle of a Hamilton type 75N syringe fitted with a nylon stop to attain a depth of 3.2 mm was placed perpendicular to the surface of the skull. The coordinates for microinjections were as follows: 1.0 mm lateral from the midline and 1.5 mm posterior from the bregma. Eight to 10 animals were given each dose of the compound being tested. The mice were divided into four groups, as follows: Group 1 consisted of mice injected i.p. with several doses of PEFLO (0.1 to 1.0 mg/g of body weight given i.p.). Group 2 consisted of animals that were pretreated i.p. with several doses of both EAA receptor antagonists and drugs enhancing GABA-ergic transmission and that were then injected with a stated dose of PEFLO (0.78 mg/g, which approximates the i.p. 95% convulsant dose [CD95] for clonus). Group 3 consisted of animals that were pretreated i.c.v. with several doses of both receptor antagonists of EAAs and drugs enhancing GABA-ergic transmission and that were injected i.p. with a stated dose of PEFLO (0.78 mg/g, which approximates the i.p. CD95 for clonus). Group 4 consisted of animals that were pretreated i.p. with a single dose of MK-801, CPPene, CGP 39551, NBQX, 2,3-BZ-2, DIA, and GVG or vehicle at appropriate times before the injection of different doses of PEFLO (0.2 to 2.0 mg/g given i.p.) (see Tables 1 and 2 for definitions of the abbreviations). After the administration of PEFLO the mice were observed for 120 min through a Plexiglass box (40 by 40 by 30 cm), and the occurrence of clonic and tonic seizures or death of animals was scored on the following scale by the

FIG. 1. Site of action of NMDA receptor antagonists.

VOL. 41, 1997

ANTAGONISM OF SEIZURES INDUCED BY PEFLOXICIN

429

TABLE 3. Incidence of seizure phases induced by PEFLO in DBA/2 mice

FIG. 2. Site of action of AMPA-KA receptor antagonists.

method of De Sarro et al. (9): 0, no response; 1, wild running; 2, clonus; 3, tonus; and 4, respiratory arrest. The compounds used and their doses and times of administration are reported in Tables 1 and 2. Figures 1, 2, and 3 show the sites of action of the different compounds tested. Administration of compounds. The compounds were prepared at an appropriate concentration. The compounds administered i.p., PEFLO (pefloxacin mesylate; Rhone-Pulenc Labs, Milan, Italy), CCP, CPPene, MK-801, CGP 39551, IFE, SL 82.0715, NBQX, GVG, BAC, TIA, and DIA, were dissolved in sterile saline. GYKI 52466 and 2,3-BZ-2 were also administered i.p. but were dissolved in 50% dimethyl sulfoxide and 50% sterile saline. The compounds administered i.c.v., CPPene, MK 801, NBQX, AP7, KYNA, CNQX, g-D-GAMS, and MSC, were dissolved in a minimum quantity of 1 N NaOH, and the final volume was made up with sodium phosphate buffer (67 mM). When necessary the pH was adjusted to 7.3 to 7.4 by adding 0.2 N HCl. Electrocortical activity. Electrocortical activity was recorded (eight-channel electrocorticograph machine; OTE Biomedica, Florence, Italy) through four chronically implanted steel-screw electrodes inserted bilaterally onto the frontoparietal area. At least three mice, treated with CCPene, MK-801, DIA, GVG, or 2,3-BZ-2–PEFLO (0.78 mg/g given i.p.) were studied to evaluate changes in ECoG activity. Statistical analysis. Statistical comparisons between control (vehicle 2 PEFLO) and drug-treated groups were made by using Fisher’s exact probability test (incidence of the seizure phases). The percent incidence of each phase of the seizure was determined for each dose of PEFLO administered. CD50s with 95% confidence limits for each phase of the seizure response were estimated by the method of Litchfield and Wilcoxon (24) adapted for analysis with an International Business Machines computer. The CD95 of PEFLO by i.p. administration was also evaluated by the same method (24). The CD95 of PEFLO was used to assess the possible anticonvulsant effects shown by the several compounds tested. The lethality following PEFLO-induced seizures was scored within 120 min, and 50% lethal doses (LD50s) were calculated by the same method (24). The 50% effective doses (ED50s), which were the doses of the compounds tested that were

FIG. 3. GABA synapsis and sites of action of compounds enhancing GABAergic transmission.

No. of mice with indicated episodesa

PEFLO dose (mg/g body weight)

Clonic seizures

Tonic seizures

Death

0.1 0.2 0.4 0.5 0.6 0.8 1.0

0 2 4 7 9 10 10

0 1 3 4 7 9 10

0 0 1 2 5 8 9

a Male DBA/2 mice were randomly assigned to experimental groups, each containing 10 animals, and were injected i.p. with various doses of PEFLO. The data are presented as the number of animals showing seizure phases among the 10 mice tested.

able to suppress seizures in 50% of the animals treated with PEFLO, were also estimated by the method of Litchfield and Wilcoxon (24).

RESULTS Effects of systemic administration of PEFLO. Systemic (i.p.) administration of PEFLO (0.1 to 1.0 mg/g) consistently resulted in clonic-tonic seizures which appeared within 14 to 20 min following the administration of the drug to the mice and which lasted from 25 to 90 min, depending on the dose. After administration of the lowest doses of PEFLO, mice showed an increase in locomotor activity, rare episodes of wild running or escape response, nodding, and clonus of the forelimbs and hind limbs. With higher doses (0.6 and 0.8 mg/g) tonus of the forelimbs and hind limbs was also observed, and in some animals the tonic extension of the hind limbs was followed by respiratory arrest and death (Table 3). The CD50 of PEFLO for producing clonus was 0.35 mg/g (0.26 to 0.47 mg/g), while that for triggering tonus was 0.44 mg/g (0.34 to 0.56 mg/g) (Fig. 4 and Table 4). Furthermore, PEFLO induced lethality, with a LD50 of 0.62 mg/g (0.53 to 0.74 mg/g). Effects of systemic administration of EAA antagonists on seizures induced by PEFLO. The i.p. administration of the various EAA antagonists, MK-801 (0.05, 0.1, 0.2, and 0.3 mg/g, equivalent to 0.15, 0.30, 0.60, and 0.75 nmol/g, respectively), CPP (2.0, 5.0, 10.0, and 20.0 mg/g, equivalent to 8, 20, 40, and 80 nmol/g, respectively), CPPene (2.0, 4.0, 6.0, and 8.0 mg/g, equivalent to 7.9, 19.8, 39.6, and 79.3 nmol/g, respectively), CGP 39551 (0.25, 0.5, 1.0, and 2.0 mg/g, equivalent to 1, 2, 4,

FIG. 4. Dose-response curves illustrating the effects of the i.p. pretreatment with CPPene (8 mg/g), GVG (1.5 mg/g), 2,3-BZ-2 (30 mg/g), and NBQX (30 mg/g) on seizures induced by i.p. administration of PEFLO in DBA/2 mice. Ordinate, percentage of response of clonic seizures; abscissa, doses of PEFLO. å, saline 1 PEFLO; F, CPPene 1 PEFLO; ■, GVG 1 PEFLO; ç, 2,3-BZ-2 1 PEFLO; }, NBQX 1 PEFLO.

430

DE SARRO ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 4. Convulsant doses of PEFLO after pretreatment with vehicle and the largest doses of some EAA antagonists and DIA or GVG administered i.p. to DBA/2 micea Compound

Vehicle 1 PEFLO CGP 39551 1 PEFLO CPPene 1 PEFLO NBQX 1 PEFLO 2,3-BZ-2 1 PEFLO MK-801 1 PEFLO DIA 1 PEFLO GVG 1 PEFLO

CD50 (mg/g [95% confidence limits]) Clonic phase

Tonic phase

0.35 (0.26–0.47) 1.23 (1.02–1.48)b 1.35 (1.21–1.51)b 1.02 (0.90–1.16)b 1.17 (0.92–1.49)b 1.12 (0.95–1.32)b 1.43 (1.18–1.73)b 1.57 (1.28–1.92)b

0.44 (0.34–0.56) 1.33 (1.12–1.58)b 1.48 (1.27–1.72)b 1.27 (0.94–1.71)b 1.24 (1.04–1.48)b 1.18 (0.84–1.66)b 1.64 (1.36–1.98)b 1.82 (1.25–2.65)b

a PEFLO was administered at 0.2 to 2.0 mg/g i.p. All data were calculated by the method of Litchfield and Wilcoxon (24). b Significant differences (P , 0.05) between CD50s for groups treated with vehicle-PEFLO and groups treated with EAA antagonists-PEFLO and DIA- or GVG-PEFLO are denoted.

and 8 nmol/g, respectively), NBQX (10, 20, and 30 mg/g, equivalent to 29.2, 58.4, and 87.7 nmol/g, respectively), GYKI 52466 (10, 20, and 30 mg/g, equivalent to 33, 66, and 100 nmol/g, respectively), 2,3-BZ-2 (10, 20, and 30 mg/g, equivalent to 34, 67, and 101 nmol/g, respectively), IFE (0.6, 0.8, and 1.0 mg/g, equivalent to 1.5, 2, and 2.5 nmol/g, respectively), SL 82.0715 (0.6, 0.8, and 1.0 mg/g, equivalent to 1.6, 2.1, and 2.6 nmol/g, respectively), at different times (Table 1) before the administration of a single dose of PEFLO (0.78 mg/g; approximately the CD95) was able to prevent the clonic and/or tonic components of PEFLO-induced convulsions (Table 5). At the doses used, IFE and SL 82.0715, however, were unable to protect against the tonic and clonic components of PEFLO-induced seizures. Following the systemic administration of EAA antagonists, the order of the potencies of the antagonists in protecting the mice against PEFLO-induced convulsions were as follows: MK-801 . CGP 39551 . CPPene . CPP . GYKI 52466 . NBQX . 2,3-BZ-2 (Table 5). In addition, pretreatment with CGP 39551 (2 mg/g, equivalent to 8 nmol/g), MK801 (0.3 mg/g, equivalent to 0.75 nmol/g), NBQX (30 mg/g, equivalent to 87.7 nmol/g), 2,3-BZ-2 (30 mg/g, equivalent to 101 nmol/g), and CPPene (8 mg/g, equivalent to 30 nmol/g) was able to increase the threshold for seizures induced by different

doses of PEFLO given i.p. (Fig. 4 and Table 4). The highest doses of all compounds studied, with the exception of IFE (1 mg/g, equivalent to 2.5 nmol/g) and SL 82.0715 (1 mg/g, equivalent to 2.6 nmol/g), were able to significantly reduce the incidence of the wild running phase. Effects of systemic administration of compounds enhancing GABA-ergic neurotransmission on seizures induced by PEFLO. i.p. pretreatment with DIA (0.5, 1, and 2 mg/g, equivalent to 1.76, 3.51, and 7 nmol/g, respectively), TIA (1, 2, and 4 mg/g, equivalent to 2.4, 4.8 and 9.7 nmol/g, respectively), and GVG (1,250, 1,500, and 2,000 mg/g, equivalent to 9.7, 11.6, and 15.5 mmol/g, respectively) at different times (Table 2) before PEFLO administration (0.78 mg/g given i.p.; approximately the CD95) was able to significantly reduce the occurrence of seizures induced by PEFLO. Whereas BAC at the lowest dose (2.5 mg/g, equivalent to 11.7 nmol/g) was unable to significantly reduce the tonic and clonic components of PEFLO-induced seizures, the largest doses of BAC (5, 10, and 20 mg/g, equivalent to 32.4, 46.8, and 93.6 nmol/g, respectively) significantly antagonized both the clonic and tonic seizures, but some animals lost the righting reflex (Table 5). The order of the potencies of the compounds described above in protecting against PEFLO-induced convulsions was DIA . TIA . BAC . GVG (Table 5). In addition, pretreatment with GVG (1,500 mg/g, equivalent to 11.6 mmol/g) and DIA (2 mg/g, equivalent to 7 nmol/g) was able to increase the threshold for seizures induced by PEFLO (Fig. 4 and Table 4). i.c.v. administration of anticonvulsants. The previous i.c.v. administration of the EAA antagonists AP7 (0.3, 0.4, and 0.5 nmol/mouse), CPPene (0.02, 0.04, and 0.06 nmol/mouse), MK801 (0.2, 0.4 and 0.8 nmol/mouse), g-D-GAMS (0.4, 0.5, and 0.6/mouse), KYNA (1.2, 2.5, and 5 nmol/mouse), CNQX (3.3, 6.6, and 10.0 nmol/mouse), NBQX (1.0, 3.3, and 6.6 nmol/ mouse), and the GABAA agonist MSC (0.25, 0375, and 0.5 nmol/mouse) was able to antagonize the seizures induced by PEFLO (0.78 mg/g given i.p.). The relative potencies of these compounds in protecting against PEFLO-induced convulsions were as follows: CPPene . MK-801 . MSC . AP7 . g-DGAMS . NBQX . KYNA . CNQX (Table 6). Electrocortical activity. In the animals used to study electrocortical activity, we observed that the electrocorticographic epileptic discharges appeared within 12 to 20 min in mice pretreated i.p. with vehicle-PEFLO (0.78 mg/g). Epileptic dis-

TABLE 5. ED50s of some EAA antagonists and drugs enhancing GABA-ergic transmission administered i.p. against seizures induced by PEFLO in DBA/2 micea Compound

Dose range (nmol/g i.p.)

CPP CPPene CGP 39551 MK-801 GYKI 52466 2,3-BZ-2 NBQX IFE SL 82.0715 DIA TIA BAC GVG

2–80 2–30 1–8 0.15–0.75 10–100 16.9–101 7.3–87.7 1–2.5 1–2.6 1.75–14 1.2–9.7 11.7–93.6 9,700–15,500

ED50 (mg/g [95% confidence limits]) Clonic phase

19.98 (9.29–42.96) 7.68 (4.63–12.75) 2.08 (1.1–3.93) 0.27 (0.15–0.49) 26.77 (14.63–48.99) 41.97 (26.07–67.57) 37.89 (19.43–73.88) .2.5 (NAb) .2.6 (NAb) 1.93 (1.16–3.22) 2.5 (1.31–4.77) 33.08 (19.23–56.9) 9,529 (7,875–11,530)

Tonic phase

7.53 (4.21–13.48) 4.66 (3.17–6.85) 0.82 (0.33–2.05) 0.15 (0.09–0.25) 17.43 (8.98–33.81) 30.99 (18.23–52.68) 27.60 (15.08–50.51) .2.5 (NAb) .2.6 (NAb) 1.12 (0.63–1.99) 1.8 (0.87–3.71) 22.74 (12.63–40.93) 8,159 (6,475–10,280)

a PEFLO was administered at 0.78 mg/g i.p. All data are for antagonism of clonic and tonic seizures and were calculated by the method of Litchfield and Wilcoxon (24). b NA, not active.

VOL. 41, 1997


TABLE 6. ED50s of some EAA antagonists and MSC, a GABAA agonist, administered i.c.v. against seizures induced by PEFLO in DBA/2 micea Compound

AP7 CPPene KYNA MK-801 g-D-GAMS NBQX CNQX MSC

ED50 (mg/g [95% confidence limit])

Dose range (nmol/mouse)

Clonic phase

Tonic phase

0.3–0.5 0.02–0.06 1.2–5 0.2–0.8 0.4–0.6 1.0–6.6 3.3–10 0.125–0.5

0.25 (0.17–0.38) 0.013 (0.004–0.043) 1.74 (0.93–3.28) 0.23 (0.15–0.35) 0.45 (0.36–0.55) 1.68 (0.93–3.05) 4.21 (2.35–7.57) 0.25 (0.15–0.38)

0.20 (0.13–0.29) 0.008 (0.005–0.013) 1.16 (0.58–2.32) 0.17 (0.11–0.26) 0.37 (0.30–0.44) 0.64 (0.35–1.18) 2.49 (1.35–4.61) 0.19 (0.13–0.27)

a PEFLO was administered at 0.78 mg/g i.p. All data are for antagonism of clonic and tonic seizures and were calculated by the method of Litchfield and Wilcoxon (24).

charges did not appear in animals pretreated with CPPene (8 mg/g given i.p)–PEFLO or DIA (1 mg/g given i.p.)–PEFLO or were less evident in those animals pretreated with MK-801 (0.3 mg/g given i.p.)–PEFLO, 2,3-BZ-2 (20 mg/g given i.p.)–PEFLO, or GVG (1.5 mg/g given i.p.)–PEFLO (Fig. 5, 6, and 7).

431

DISCUSSION The mechanisms of the convulsant activities of quinolones are largely unresolved, despite suggestions from in vitro data that quinolones may be acting through inhibition of GABA binding to its receptor sites (1, 31, 39, 41). In particular, since the epileptogenic activities of some quinolones were suppressed by compounds that enhance GABA-ergic neurotransmission, i.e., GVG and TIA, or by MSC and DIA, which are agonists for the GABAA-benzodiazepine receptor complex (22), while it was influenced at neurotoxic doses by baclofen, a GABAB receptor agonist, the epileptogenic activities of quinolones likely involve the GABAA-benzodiazepine receptor complex. This conclusion is in agreement with the results of previous studies on the interaction of quinolones with the benzodiazepine-GABA receptor complex (1, 31, 39, 41). In addition, Enginar and Eroglu (18) demonstrated that the quinolone ofloxacin is able to reduce the threshold of convulsions induced by pentylentetrazole, which is believed to block the actions of GABA through its effects at the chloride ionophore coupled to the GABAA receptor complex, and therefore, it has been considered a specific DIA antagonist (29, 36, 37). Concerning the anticonvulsant potencies of the different compounds tested, differences exist among the groups. The most important finding of this study was that N-methyl-D-aspartate (NMDA) antagonists such as CPP, CGP 39551, CPPene, AP7, and MK-801 blocked seizures induced by i.p. administration of

FIG. 5. Electrocorticographic patterns after saline–PEFLO (0.78 mg/g given i.p.) administration (A, B, C, and D) and GVG (1.5 mg/g given i.p.)–PEFLO (0.78 mg/g given i.p.) administration (A1, B1, C1 and D1) in DBA/2 mice. rCx; right cortex; 1Cx, left cortex. (A and A1) Electrocorticographic recordings 15 min after the injection of saline or GVG. (B and B1) Electrocorticographic recordings 20 min after PEFLO injection. (C and C1) Electrocorticographic recordings 60 min after PEFLO injection. (D and D1) Electrocorticographic recordings 120 min after PEFLO injection.

432

DE SARRO ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 6. Electrocorticographic patterns after CPPene (8 mg/g given i.p.)–PEFLO (0.78 mg/g given i.p.) administration (A, B, C, and D) and MK-801 (0.3 mg/g given i.p.)–PEFLO (0.78 mg/g given i.p.) administration (A1, B1, C1, and D1) in DBA/2 mice. rCx, right cortex; 1Cx, left cortex. (A and A1) Electrocorticographic recordings 15 min after the injection of CPPene or MK-801. (B and B1) Electrocorticographic recordings 20 min after PEFLO injection. (C and C1) Electrocorticographic recordings 60 min after PEFLO injection. (D and D1) Electrocorticographic recordings 120 min after PEFLO injection.

PEFLO with potencies higher than or equal to those of the GABAA agonist MSC, the benzodiazepine DIA, and the inhibitor of GABA uptake TIA. g-D-GAMS, GYKI 52466, 2,3BZ-2, and NBQX, which are preferential a-amino-3-hydroxy5-methyl-4-isoxazolepropionic acid (AMPA)-kainate (KA) receptor antagonists (8, 14, 23, 34), also antagonized the seizures induced by PEFLO, but with minor potencies compared with those of the NMDA antagonists. The doses of AMPA-KA antagonists used are in the range of those which have been shown to block excitatory neurotransmission at non-NMDA receptors, thus demonstrating that AMPA-KA-dependent mechanisms are activated in the course of PEFLO-induced seizures as well. The broad-spectrum antagonist KYNA was also found to be a potent anticonvulsant, and a trend similar to that seen for KYNA was seen following the systemic administration of some selective NMDA receptor antagonists. It is remarkable that the doses of MK-801, CPPene, CGP 39551, and CPP required to block the clonic and tonic components of PEFLO-induced convulsions were less than or similar to those used in other experimental models to antagonize clonic-tonic seizures in mice (27, 38), while the doses of drugs enhancing GABA-ergic transmission are usually higher than those able to block audiogenic seizures in DBA/2 mice (4–6). IFE and SL 82.0715, two compounds that act on the polyamine site of the NMDA receptor complex (3, 30), were unable to protect against seizures induced by PEFLO, suggesting that the polyamine site does not play a principal role in the genesis of seizures induced by PEFLO. The results of this study demonstrate that the excitation mediated by dicarboxylic amino acids

plays a crucial role in the pathogenesis of seizures caused by PEFLO in mice, similar to that observed following pretreatment with EAA antagonists in seizures induced by bicuculline, pilocarpine, and imipenem (14, 15, 28, 40). It is most likely that excessive activation of EAA receptors occurs secondarily to or concomitantly with the impairment of the inhibitory GABAergic neurotransmission caused by PEFLO and is essential for the propagation of seizures. The NMDA receptors seem to play a particularly pivotal role in PEFLO-induced seizures, since NMDA antagonists were potent anticonvulsants in the present study. This observation is in line with those from a previous study showing that the proconvulsive activities of some quinolones may be antagonized by EAA antagonists (42). The present study also demonstrated that the seizure intensity was consistently reduced in animals pretreated with most EAA antagonists or the latter phases of seizures failed to occur. In addition, a different incidence of seizures was observed in mice after pretreatment with EAA antagonists. A quantitative comparison of the anticonvulsant efficacies of the EAAs in antagonizing PEFLO-induced seizures and audiogenic seizures in DBA/2 mice seems to demonstrate some similarities among the EAAs (e.g., the anticonvulsant potencies of CPP, CPPene, CGP 39551, and NBQX). However, IFE and SL 82.0715 showed no or weak anticonvulsant activity or less of a protective effect against PEFLO-induced seizures, while they were potent anticonvulsants against audiogenic seizures (3). The possibility that PEFLO possesses agonistic or modulatory properties at receptors activated by EAAs might be an alternative explanation to the one involving GABA-ergic

VOL. 41, 1997


433

FIG. 7. Electrocorticographic patterns after DIA (1.0 mg/g given i.p.)–PEFLO (0.78 mg/g given i.p.) administration (A, B, C, and D) and 2,3-BZ-2 (20 mg/g given i.p.)–PEFLO (0.78 mg/g given i.p.) administration (A1, B1, C1, and D1) in DBA/2 mice. rCx, right cortex; 1Cx: left cortex. (A and A1) Electrocorticographic recordings 15 min after the injection of DIA or 2,3-BZ-2. (B and B1) Electrocorticographic recordings 20 min after PEFLO injection. (C and C1) Electrocorticographic recordings 60 min after PEFLO injection. (D and D1) Electrocorticographic recordings 120 min after PEFLO injection.

transmission, which requires experimental confirmation. Since previous studies have indicated that systemic administration of quinolones to experimental animals or humans may sometimes produce proconvulsant or epileptogenic effects (9–12, 17, 18, 20, 25, 32, 42), we conclude that physicians should consider the possible epileptogenic activity of PEFLO when treating patients with predisposing epileptic factors.

7.

8.

9.

ACKNOWLEDGMENTS Financial support from the Italian Ministry of University and Scientific and Technological Research (MURST) is gratefully acknowledged. We thank Antonino Giacopello for skillful technical assistance. REFERENCES 1. Akahane, K., M. Sekiguchi, T. Une, and Y. Osada. 1989. Structure-epileptogenicity relationship of quinolones with special reference to their interaction with g-aminobutyric acid receptor sites. Antimicrob. Agents Chemother. 33:1704–1707. 2. Arcieri, G., H. E. Griffith, G. Gruenwaldt, A. Heyd, B. O’Brien, N. Becker, and R. August. 1987. Ciprofloxacin: an update on clinical experience. Am. J. Med. 82:381–386. 3. Carter, C., J. P. Rivy, and B. Scatton. 1989. Ifenprodil and SL-82.0715 are antagonists at the polyamine site of the N-methyl-D-aspartate (NMDA) receptor. Eur. J. Pharmacol. 164:611–612. 4. Chapman, A. G., M. J. Croucher, and B. S. Meldrum. 1984. Evaluation of anticonvulsant drugs in DBA/2 mice with sound-induced seizures. Arzneim.Forsch./Drug Res. 34:1261–1264. 5. Chapman, A. G., G. B. De Sarro, M. Premachandra, and B. S. Meldrum. 1987. Bidirectional effect of b-carbolines in reflex epilepsy. Brain Res. Bull. 19:337–346. 6. Chapman, A. G., and B. S. Meldrum. 1985. Epilepsy prone mice: genetically

10.

11.

12.

13.

14.

15.

16.

determined sound-induced seizures, p. 197–208. In P. C. Jobe and H. E. Laird (ed.), Neurotransmitters and epilepsy. The Humana Press, Clifton, N.J. Chapman, A. G., and B. S. Meldrum. 1989. Non-competitive N-methyl-Daspartate antagonists protect against sound-induced seizures in DBA/2 mice. Eur. J. Pharmacol. 166:201–211. Chapman, A. G., S. E. Smith, and B. S. Meldrum. 1991. The anticonvulsant effect of the non-NMDA antagonists, NBQX and GYKI 52466 in mice. Epilepsy Res. 9:92–96. De Sarro, A., D. Ammendola, and G. B. De Sarro. 1994. Effects of some quinolones on imipenem-induced seizures in DBA/2 mice. Gen. Pharmacol. 25:369–379. De Sarro, A., and G. B. De Sarro. 1991. Responsiveness of genetically epilepsy-prone rats to aminophylline-induced seizures and interactions with quinolones. Neuropharmacology 30:169–176. De Sarro, A., G. R. Trimarchi, D. Ammendola, and G. De Sarro. 1992. Repeated treatment with quinolones potentiates the seizures induced by aminophylline in genetically epilepsy-prone rats. Gen. Pharmacol. 23:853– 859. De Sarro, A., M. Zappala, A. Chimirri, S. Grasso, and G. B. De Sarro. 1993. Quinolones potentiate cefazolin-induced seizures in DBA/2 mice. Antimicrob. Agents Chemother. 37:1497–1503. De Sarro, G. B., C. Ascioti, E. Donato di Paola, M. J. Vidal, and A. De Sarro. 1992. Effects of antiepileptic drugs, calcium channel blockers and other compounds on seizures induced by activation of voltage-dependent L calcium channel in DBA/2 mice. Gen. Pharmacol. 23:1205–1216. De Sarro, G. B., D. Ammendola, F. Nava, and A. De Sarro. 1995. Effects of some excitatory amino acid antagonists on imipenem-induced seizures in DBA/2 mice. Brain Res. 671:131–140. De Sarro, G. B., B. S. Meldrum, A. De Sarro, and S. Patel. 1992. Excitatory neurotransmitters in the lateral habenula and peduncolopontine nucleus of rats modulate limbic seizures induced by pilocarpine. Brain Res. 591:209– 222. Dimpfel, W., M. Spuler, A. Dalhoff, W. Hoffman, and G. Schluter. 1991. Hippocampal activity in the presence of quinolones and fenbufen in vitro.

434

DE SARRO ET AL.

Antimicrob. Agents Chemother. 35:1142–1146. 17. Eng, R. H., A. N. Munsif, B. G. Yangco, S. M. Smith, and H. Chmel. 1989. Seizure propensity with pefloxacin. Arch. Intern. Med. 149:1881–1883. 18. Enginar, N., and L. Eroglu. 1991. The effect of ofloxacin and ciprofloxacin on pentylenetetrazol-induced convulsions in mice. Pharmacol. Biochem. Behav. 39:587–589. 19. Engstrom, F. L., and D. M. Woodbury. 1988. Seizure susceptibility in DBA/2 and C57 mice: the effects of various convulsants. Epilepsia 29:389–395. 20. Fass, R. J. 1987. Adverse reactions associated with quinolones. Quinolones Bull. 3:5–6. 21. Hall, C. S. 1947. Genetic differences in fatal audiogenic seizures between two inbred strains of house mice. J. Hered. 38:2–6. 22. Holland, K. D., A. C. McKeon, D. J. Canney, D. F. Covey, and J. A. Ferrendelli. 1992. Relative anticonvulsant effects of GABAmimetic and GABA modulatory agents. Epilepsia 33:981–986. 23. Honore*, T., S. N. Davies, J. Drejer, E. J. Fletcher, P. Jacobsen, D. Lodge, and F. E. Nielsen. 1988. Quinoxalinediones: potent competitive non-NMDA glutamate receptor antagonists. Science 241:701–703. 24. Litchfield, J. T., and F. Wilcoxon. 1949. A simplified method of evaluating dose-effects experiments. J. Pharm. Exp. Ther. 96:99–113. 25. Lucet, I. C., H. Tilly, G. Lerebours, I. L. Gres, and H. Piguet. 1988. Neurological toxicity related to pefloxacin. J. Antimicrob. Chemother. 21:811–812. 26. Neu, H. C. 1988. Clinical use of the quinolones. Lancet 8571:1319. 27. Nevins, M. E., and S. M. Arnolde. 1989. A comparison of the anticonvulsant effects of competitive and non-competitive antagonists of the N-methyl-Daspartate receptor. Brain Res. 503:1–4. 28. Patel, S., G. B. De Sarro, and B. S. Meldrum. 1988. Regulation of seizure threshold by excitatory amino acids in the striatum and entopeduncular nucleus of rats. Neuroscience 23:837–850. 29. Ramanjaneyulu, R., and M. K. Ticku. 1984. Interaction of pentamethylenetetrazole and tetrazole analogues with the picrotoxinin site of the benzodiazepineGABA receptor-ionophore complex. Eur. J. Pharmacol. 98:337–345. 30. Reynolds, I. J., and R. J. Miller. 1989. Ifenprodil is a novel type of N-methylD-aspartate receptor antagonist: interaction with polyamines. Mol. Pharmacol. 36:758–765.

ANTIMICROB. AGENTS CHEMOTHER. 31. Segev, S., M. Rehavi, and E. Rubinstein. 1988. Quinolones, theophylline, and diclofenac interactions with the g-aminobutyric acid receptor. Antimicrob. Agents Chemother. 32:1624–1626. 32. Semel, J. D., and N. Allen. 1991. Seizures in patients simultaneously receiving theophylline and pefloxacin or ciprofloxacin or metronidazole. South. Med. J. 84:465–468. 33. Seyfried, T. N., and G. H. Glaser. 1985. A review of mouse mutants as genetic models of epilepsy. Epilepsia 26:143–150. 34. Sheardown, M. J., E. O. Nielsen, A. J. Hansen, P. Jacobsen, and T. Honore*. 1990. 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F) quinoxaline: a neuroprotectant for cerebral ischemia. Science 247:571–574. 35. Stone, T. W. 1982. Quinolinic acid and kynurenic acid. Trends Pharmacol. Sci. 5:215. 36. Syapin, P. J., and D. W. Rickman. 1981. Benzodiazepine receptor increase following repeated pentylenetetrazole injections. Eur. J. Pharmacol. 72:117– 120. 37. Ticku, M. K., and S. K. Rastagi. 1986. Convulsant/anticonvulsant drugs and GABAergic transmission, p. 163–177. In G. Nistico9, P. L. Morselli, K. E. Lloyd, R. G. Fariello, and J. Engel, (ed.), Neurotransmitters, seizures and epilepsy III. Raven Press, New York, N.Y. 38. Troupin, A. S., J. R. Mendius, F. Cheng, and M. W. Risinger. 1986. MK-801, p. 191–201. In B. S. Meldrum and R. J. Porter (ed.), New anticonvulsant drugs. Libbey, London, United Kingdom. 39. Tsuji, A., H. Sato, Y. Kume, J. Tamai, E. Okezaki, O. Nagata, and H. Kato. 1988. Inhibitory effects of quinolone antibacterial agents on gamma-aminobutyric acid binding to receptor sites in rat brain membranes. Antimicrob. Agents Chemother. 32:190–194. 40. Turski, W. A., E. Urbanska, M. Dziki, J. Parada-Turska, and C. Ikonomidou. 1990. Excitatory amino acid antagonists protect mice against seizures induced by bicuculline. Brain Res. 514:131–134. 41. Unseld, E., G. Ziegler, A. Gemeinhardt, U. Janssen, and U. Klotz. 1990. Possible interaction of fluoroquinolones with the benzodiazepine-GABAAreceptor complex. Br. J. Cl. Pharmacol. 30:63–70. 42. Williams, P. D., and D. R. Helton. 1991. The proconvulsive activity of quinolone antibiotics in an animal model. Toxicol. Lett. 58:23–28.