Epilepsy Research (2010) 90, 188—198

journa l homepage: www.e lsev ier .com/ locate /ep i lepsyres

Interactions of stiripentol with clobazam andvalproate in the mouse maximalelectroshock-induced seizure model

Jarogniew J. Luszczkia,b,∗, Michal K. Trojnara,c, Neville Ratnarajd,Philip N. Patsalosd, Stanislaw J. Czuczwara,b

a Department of Pathophysiology, Medical University of Lublin, Jaczewskiego 8, PL 20-090 Lublin, Polandb Department of Physiopathology, Institute of Agricultural Medicine, Jaczewskiego 2, PL 20-950 Lublin, Polandc Department of Cardiology, Medical University of Lublin, Jaczewskiego 8, PL 20-090 Lublin, Polandd Pharmacology and Therapeutics Unit, Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, QueenSquare, London WC1N 3BG, United Kingdom

Received 4 October 2009; received in revised form 19 February 2010; accepted 25 April 2010Available online 20 May 2010

KEYWORDSStiripentol;Clobazam;Valproate;Isobolographicanalysis;Maximalelectroshock;Pharmacodynamic/pharmacokineticinteraction

Summary The aim of this study was to characterize the anticonvulsant effects of stiripentol(STP) in combination with clobazam [CLB], and valproate [VPA]) in the mouse maximal elec-troshock (MES)-induced seizure model using the type I isobolographic analysis for parallel andnon-parallel dose—response relationship curves (DRRCs). Potential adverse-effect profiles ofinteractions of STP with CLB and VPA at the fixed-ratio of 1:1 in the MES test with respect tomotor performance, long-term memory and skeletal muscular strength were measured alongwith total brain antiepileptic drug concentrations.

In the mouse MES model, STP administered singly had its DRRC non-parallel to that for CLBand, simultaneously, parallel to that for VPA. With type I isobolography for parallel DRRCs, thecombinations of STP with VPA at three fixed-ratios of 1:3, 1:1 and 3:1 exerted sub-additive(antagonistic) interaction. Isobolography for non-parallel DRRCs revealed that the combinationof STP with CLB at the fixed-ratio of 1:1 produced additive interaction. For all combinations,neither motor coordination, long-term memory nor muscular strength was affected. Total brainantiepileptic drug concentrations revealed bi-direction changes with the most profound beingan 18.6-fold increase in CLB by STP and a 2.3-fold increase in STP by VPA.

∗ Corresponding author at: Department of Pathophysiology, Medical University of Lublin, Jaczewskiego 8, PL 20-090 Lublin, Poland.Tel.: +48 81 718 73 65; fax: +48 81 718 73 64.

E-mail addresses: jluszczki@yahoo.com, jarogniew.luszczki@gmail.com, jarogniew.luszczki@am.lublin.pl (J.J. Luszczki).

0920-1211/$ — see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.eplepsyres.2010.04.006

pharmacokinetic interaction and these data may explain the clinical efficacy seen with this com-bination. In contrast, the antagonism between STP and VPA was surprising since synergism is

s res

®

observed clinically.© 2010 Elsevier B.V. All right

Introduction

Stiripentol (STP) has recently been approved by theEuropean Medicines Agency, through its Orphan DrugDevelopment Programme, for use as adjunctive treatmentwith clobazam (CLB) and valproate (VPA) for the man-agement of refractory generalized tonic—clonic seizuresin patients with severe myoclonic epilepsy in infancy(SMEI; Dravet syndrome) (http://www.emea.europa.eu/humandocs/Humans/EPAR/diacomit/diacomit.htm [revi-sion 5, published 12.02.09]).

STP has been under investigation as an antiepileptic drug(AED) for many years and has reported efficacy in adultpatients with partial seizure with or without secondarygeneralization (Farwell et al., 1993). However, its clinicaldevelopment in adult epilepsy was complicated by the pro-found pharmacokinetic interactions that were associatedwith STP. The metabolism of STP can be enhanced by enzymeinducing AEDs (carbamazepine [CBZ], phenytoin [PHT], phe-nobarbital [PB] and primidone [PRM]) and additionally STPacts as a potent inhibitor of numerous AEDs including CBZ,PHT, PB, VPA and CLB. STP inhibition of CLB includes a par-ticularly more profound inhibition of its pharmacologicallyactive metabolite N-desmethyl clobazam (DMCLB) wherebyserum concentrations increase several fold (Levy et al.,1984; Kerr et al., 1991; Bebin and Bleck, 1994; Tran et al.,1997; Chiron et al., 2000; Cazali et al., 2003; Giraud et al.,2006).

In children with epilepsy adjunctive STP was associatedwith decreased seizure frequency and 3 of 20 childrenbecame seizure free (Perez et al., 1999). This study alsoreported an improvement in patients with SMEI and sub-sequently, it was observed that such patients respondedparticularly well when STP was used in combination withCLB (Chiron et al., 2000; Rey et al., 1999) and VPA (Kassaï etal., 2008; Inoue et al., 2009). Furthermore, good responseshave been observed when STP was administered in combina-tion with clonazepam (CZP) in multifocal epilepsy in infancy(Chiron et al., 1993, 2000), with VPA in refractory epilepsy inchildren (Renard et al., 1993), and with CBZ in CBZ-resistantepilepsies (Loiseau et al., 1990).

STP suppresses seizures in various animal models,including maximal electroshock (MES)-, pentylenetetra-zole (PTZ)-, bicuculline- and strychnine-induced seizures inrodents (Poisson et al., 1984; Shen et al., 1990, 1992). More-over, STP protected mice against cocaine-induced clonicseizures (Gasior et al., 1999) and suppressed spike and wavedischarges in a rat genetic model of petit mal epilepsy(Micheletti et al., 1988) and spike discharges in an aluminagel rhesus monkey model of focal epilepsy (Lockard et al.,

1985).

Previously we reported that the combinations of STPwith CZP, ethosuximide (ETS) and VPA were neutral whilstSTP with PB was favorable with regards to their efficacy

erved.

and adverse-effect profiles in the mouse PTZ-induced clonicseizure model (Luszczki et al., 2006a). With the exceptionof CZP, all other AEDs combinations were complicated byconcurrent pharmacokinetic interactions. Furthermore, thecombination of STP with CBZ exerted a biphasic interaction(antagonistic and synergistic) depending on the fixed-ratiotested in the mouse MES model (Luszczki and Czuczwar,2006).

Because STP is licensed for clinical use in SMEI as adjunc-tive therapy to CLB and VPA, we were particularly keen inthe present study to characterize the interactions betweenthese AED combinations. The mouse MES model was cho-sen because it is considered to be an animal model oftonic—clonic seizures and partial convulsions with or with-out secondary generalization in humans (Löscher et al.,1991; Löscher, 2002a). Adverse effects were ascertainedby the use of the chimney test (a measure of motor per-formance impairment), the step-through passive avoidancetask (a measure of long-term memory deficits), and thegrip-strength test (a measure of skeletal muscular strengthimpairment). Finally, to ascertain whether the observedinteractions were purely pharmacodynamic in nature or thatpharmacokinetic interactions also contributed, total brainSTP, CLB and VPA concentrations were measured.

Materials and methods

Animals and experimental conditions

All experiments were performed on adult male albino Swiss mice(weighing 22—26 g, six-week-old) purchased from licensed breeder(Dr. T. Gorzkowska, Warszawa, Poland). The mice were kept incolony cages with free access to food and tap water under stan-dardized housing conditions (natural light—dark cycle, temperatureof 21 ± 1 ◦C, relative humidity of 55 ± 3%). After 7 days of adapta-tion to laboratory conditions, the animals were randomly assignedto experimental groups consisting of 8 mice. Each mouse was usedonly once. All tests were performed between 9.00 a.m. and 3.00p.m. Procedures involving animals and their care were conductedin accordance with the Guide for the Care and Use of LaboratoryAnimals as adopted and promulgated by the National Institutesof Health. Additionally, all efforts were made to minimize ani-mal suffering and to use only the number of animals necessary toproduce reliable scientific data. The experimental protocols andprocedures described in this manuscript were approved by the LocalEthics Committee at the Medical University of Lublin (License nos.420/2003/446/2003 and 23/2009).

Drugs

The following AEDs were used in this study: STP (Laboratoires

Interactions of stiripentol with clobazam and valproate in the mouse maximal electroshock-induced seizure model 189

In conclusion, the additive interaction between STP and CLB was associated with a concurrent

Biocodex, Gentilly, France), CLB (Frisium ; Sanofi-Aventis, Frank-furt am Main, Germany) and VPA (magnesium salt, ICN-Polfa S.A.,Rzeszow, Poland). All drugs, except for VPA, were suspended in a 1%solution of Tween 80 (Sigma, St. Louis, MO, USA) in saline, whereasVPA was directly dissolved in saline. All drugs were administered

1

bwm3stuscta

M

Tadsv(bGtdtaaTieuh2

I

TpieSelwtsf(CobaLfmisitC—cta[ta

wDdipeta2o(CmtM

wo2awti3Ltmsawbatda

tptapTib

Mc

BaTbwss(Uw((b

90

y intraperitoneal (i.p.) injection in a volume of 0.005 ml/g bodyeight. Fresh drug solutions were prepared on each day of experi-entation and administered as follows: STP —60 min; CLB and VPA —

0 min before seizures and behavioral tests as well as before brainampling for the measurement of AED concentrations. The timeso the peak of maximum anticonvulsant effects for all AEDs weresed as the reference times in all behavioral tests. The route ofystemic (i.p.) administration and these pretreatment times werehosen based upon information about their biological activity fromhe literature (Gasior et al., 1999) and our previous studies (Luszczkind Czuczwar, 2006; Luszczki et al., 2006a,b).

aximal electroshock seizure test

he protective activities of STP, CLB and VPA administered sep-rately were evaluated and expressed as their median effectiveoses (ED50 in mg/kg), protecting 50% of mice against MES-inducedeizures (fixed current intensity of 25 mA and maximum stimulationoltage of 500 V). Electroconvulsions were produced by a current0.2 s stimulus duration) delivered via standard auricular electrodesy a Hugo Sachs generator (Rodent Shocker, Type 221, Freiburg,ermany). The criterion for the occurrence of seizure activity was

he tonic hindlimb extension. The animals were administered withifferent drug doses so as to obtain a variable percentage of pro-ection against MES-induced seizures, allowing the construction ofdose—response relationship curve (DRRCs) for STP, CLB and VPA

dministered alone, according to Litchfield and Wilcoxon (1949).he anticonvulsant activity of the mixture of STP with the stud-

ed AEDs (CLB and VPA) at the fixed-ratios of 1:3, 1:1 and 3:1 wasvaluated and expressed as median effective doses (ED50 mix val-es) against MES-induced seizures. This experimental procedureas been described in detail in our earlier studies (Luszczki et al.,006b, 2009).

sobolographic analysis of interactions

he percent protection of animals against MES-induced seizureser dose of an AED administered alone and the DRRC for eachnvestigated AED in the MES test were fitted using log-probit lin-ar regression analysis according to Litchfield and Wilcoxon (1949).ubsequently, from the respective linear equations the medianffective doses (ED50s) of AEDs administered alone were calcu-ated. To precisely and correctly analyze the experimental dataith isobolography, the test for parallelism of DRRCs for STP and

he investigated AEDs (CLB and VPA) based on the log-probit analy-is was used (Luszczki and Czuczwar, 2004; Luszczki, 2007). The testor parallelism was performed according to Litchfield and Wilcoxon1949), as described in detail in our previous study (Luszczki andzuczwar, 2006). In this test STP had its DRRC non-parallel to thatf CLB and simultaneously, parallel to the DRRC of VPA. Interactionsetween STP and CLB against MES-induced seizures were analyzedccording to the methodology described by Tallarida (2006), anduszczki (2007). Based upon the ED50 values denoted previouslyor the AEDs administered alone, median additive doses of theixture of STP with CLB — i.e., doses of the mixture, which theoret-

cally should protect 50% of the animals tested against MES-inducedeizures (ED50 add) were calculated from two equations of additiv-ty presented by Tallarida (2006). For the lower line of additivityhe equation at a 50% effect for the combination of STP withLB is as follows: y = ED50 CLB − [ED50 CLB/(ED50 STP/x)q/p]; where y

is the dose of CLB; x — is the dose of STP; p and q — areurve-fitting parameters (Hill coefficients) for CLB and STP, respec-

ively. Similarly, for the upper line of additivity the equation at50% effect for the combination of STP with CLB is: y = ED50 CLB

(ED50 STP—x)/ED50 STP]q/p. To calculate the curve-fitting parame-ers (p and q), probits of response for CLB and STP administeredlone were transformed to % effect. It is important to note that

aadps

J.J. Luszczki et al.

hen two drugs produce maximal effect but have non-parallelRRCs, the additivity is represented as an area bounded by twoefined curves (lower and upper isoboles of additivity). The exper-mentally derived ED50 values are statistically different if theiroints are placed outside this region. For supra-additivity (syn-rgy), the experimentally derived ED50 mix points are placed belowhe area bounded by the lower and upper isoboles of additivity,nd for sub-additivity (antagonism) — above this region (Tallarida,006). Proportions of STP and CLB in the mixture were calculatednly for the fixed-ratio combination of 1:1, as recommended earlierLuszczki, 2007; Luszczki et al., 2009), and the mixtures of STP withLB were administered to animals. The evaluation of the experi-entally derived ED50 mix at the fixed-ratio of 1:1 was based upon

he dose of the mixture protecting 50% of animals tested againstES-induced seizures in mice.

Interactions between STP and VPA against MES-induced seizuresere analyzed according to the methodology previously detailed inur earlier studies (Luszczki and Czuczwar, 2004; Luszczki et al.,006a,b). Based on the ED50 values denoted previously for the AEDsdministered alone, the median additive doses of mixtures of STPith VPA (ED50 adds — i.e., doses of the two-drug mixtures, which

heoretically should protect 50% of the animals tested against MES-nduced seizures) for three fixed-ratio combinations of 1:3, 1:1 and:1 were calculated from the equations of additivity presented byoewe (1953), as follows: x/ED50 STP + y/ED50 VPA = 1; where x — ishe dose of STP and y — is the dose of VPA co-administered as aixture that exerts the desired effect (50% effect for ED50). Sub-

equently, proportions of the AEDs in the mixture were calculatednd the respective mixtures of STP with VPA at three fixed-ratiosere administered to animals. The anticonvulsant effects offeredy STP and VPA in combination, at three fixed-ratios of 1:3, 1:1nd 3:1 in the mouse MES model, were evaluated and expressed ashe experimentally derived ED50 mix values, corresponding to theoses of two-drug mixtures, sufficient for the 50% protective effectgainst MES-induced seizures in mice.

Finally, to determine the separate doses of STP and CLB or VPA inhe mixture, the ED50 mix values were multiplied by the respectiveroportions of AEDs (denoted for purely additive mixture). Fur-her details regarding these concepts and all required equationsllowing the calculation of S.E.M. for ED50 add values have beenublished elsewhere (Luszczki, 2007; Luszczki et al., 2006a,b, 2009;allarida, 2006). All detailed calculations required to perform thesobolographic analysis are published as appendices to the papersy Luszczki and Czuczwar (2006) and Luszczki et al. (2009).

easurement of total brain antiepileptic drugoncentrations

rain AED concentrations were determined only in mice that weredministered STP + an AED at the fixed-ratio combination of 1:1.his fixed-ratio combination was chosen because it comprised ofoth AEDs being present at maximally equi-effective doses. Miceere killed by decapitation at times chosen to coincide with that

cheduled for the MES test and whole brains were removed fromkulls, weighed, harvested and homogenized using Abbott buffer2:1 vol/weight; Abbott Laboratories, North Chicago, IL, USA) in anltra-Turrax T8 homogenizer (Staufen, Germany). The homogenatesere centrifuged at 10,000 g for 10 min and the supernatant samples

75 �l) were analyzed by fluorescence polarization immunoassayFPIA) using a TDx analyzer and reagents (VPA) exactly as describedy the manufacturer (Abbott Laboratories, North Chicago, IL, USA).

Brain STP concentrations were determined by HPLC using an

utomated Gilson (Anachem) HPLC system comprising of a 234utosampler, two 306 pumps and a UV 155 variable wave lengthetector set at 215 nm. The mobile phase was constituted ashosphate (50 mM) and acetonitrile (25:75) and chromatographiceparation was achieved by use of an ESA RP-C18 3 �m column

e m

ctmwitlda

G

TtacMFmttTrlgeni

S

Tltssamwttav1tS(ucttmtfP

S

Interactions of stiripentol with clobazam and valproate in th

(ESA Analytical Ltd.). Brain homogenate samples were prepared foranalysis as follows: 50 �l brain homogenate were pipetted into a 1.5-ml plastic tube to which was added 100 �l acetonitrile and vortexmixed for 1 min. Subsequently, samples were centrifuged for 5 min,and 90 �l of the supernatant were transferred into an autosamplervial from which 10 �l were injected automatically into the column.Quantitation was achieved by use of chromatographic peak areasand these were linearly related over the range 0.1—25 �g/ml STP.The within-batch and between-batch precision was <4% and <6%respectively.

Brain concentrations of CLB and its active metabolite DMCLBwere determined by HPLC using an automated Spetra System(Thermo Fisher Scientific) HPLC system. The system comprised ofa Spetra System AS3000 autosampler, Spetra System P4000 pumpsand Spetra System UV 2000 variable wave length detector set at215 nm. Mobile phase was constituted as phosphate (45 mmol) andacetonitrile in a ratio of 50:50. Chromatographic separation wasachieved using a LiChrosphere 60 RP-Select B 5 �m column (VWRInternational). Brain homogenate samples and calibration standardswere prepared for analysis as follows: 300 �l brain homogenate werepipetted into a 2-ml plastic tube to which were added 300 �l ofphosphate buffer and 50 �l of internal standard. The sample wasvortex mixed for 1 min and centrifuged for 5 min. Then the samplesand standards were extracted and eluted using Isolute 50 mg C181 ml solid phase extraction cartridge columns (International Sor-bent Technology) using VacMaster sample processing station (JonesChromatography). The eluent was concentrated using a Gyro-Vapevaporator. The concentrated residue was reconstituted in 50 �lmobile phase and transferred into an autosampler vial, from which10 �l were injected automatically into the column. Quantitationwas achieved by use of chromatographic peak areas and thesewere linearly related over the range 0.01—6 �g/ml CLB and DMCLB.The within-batch and between-batch precision was <10% and <14%respectively. Total brain concentrations were expressed in �g/ml ofbrain supernatants as means ± S.D. of at least eight separate brainpreparations.

Chimney test

The effects of STP, CLB and VPA administered alone and in com-bination (administered at doses corresponding to their ED50 mixvalues at the fixed-ratio of 1:1 from the MES-induced seizure test)on motor coordination impairment were quantified with the chim-ney test of Boissier et al. (1960). In the chimney test, animalshad to climb backwards up the plastic tube (3 cm inner diameterand 30 cm length). Motor impairment was indicated by the inabilityof the animals to climb backward up the transparent tube within60 s. Data were presented as a percentage of animals that failed toperform the chimney test. This experimental procedure has beendescribed in detail in our earlier studies (Luszczki and Czuczwar,2004; Luszczki et al., 2006a,b).

Step-through passive avoidance task

On the first day before training, each animal received eitherthe studied AEDs administered alone or the respective combina-tion of STP with CLB and VPA, at doses corresponding to theirED50 mix values at the fixed-ratio of 1:1 from the MES-inducedseizure test. Subsequently, animals were placed in an illumi-nated box (10 cm × 13 cm × 15 cm) connected to a larger darkbox (25 cm × 20 cm × 15 cm) equipped with an electric grid floor.Entrance of animals to the dark box was punished by an adequate

electric footshock (0.6 mA for 2 s). The animals that did not enterthe dark compartment were excluded from subsequent experimen-tation. On the following day (24 h later), the pre-trained animalsdid not receive any treatment and were placed again into the illu-minated box and observed up to 180 s. Mice that avoided the dark

MaAGD

ouse maximal electroshock-induced seizure model 191

ompartment for 180 s were considered to remember the task. Theime that the mice took to enter the dark box was noted and theedian latencies (retention times) with 25th and 75th percentilesere calculated. The step-through passive avoidance task gives

nformation about ability to acquire the task (learning) and to recallhe task (retrieval). Therefore, it may be regarded as a measure ofong-term memory (Venault et al., 1986). This experimental proce-ure has been described in detail in our earlier studies (Luszczki etl., 2003b, 2005a).

rip-strength test

he effects of STP, CLB and VPA administered alone and in combina-ion (administered at doses corresponding to their ED50 mix valuest the fixed-ratio of 1:1 from the MES-induced seizure test) on mus-ular strength in mice were quantified by the grip-strength test ofeyer et al. (1979). The grip-strength apparatus (BioSeb, Chaville,rance) comprised a wire grid (8 cm × 8 cm) connected to an iso-etric force transducer (dynamometer). The mice were lifted by

he tails so that their forepaws could grasp the grid. The mice werehen gently pulled backward by the tail until the grid was released.he maximal force exerted by the mouse before losing grip wasecorded. The mean of 3 measurements for each animal was calcu-ated and subsequently, the mean maximal force of 8 animals perroup was determined. The skeletal muscular strength in mice wasxpressed in N (newtons) as means ± S.E.M. of at least 8 determi-ations. This experimental procedure has been described in detailn our earlier study (Luszczki et al., 2009).

tatistics

he ED50 and ED50 mix values (with their respective 95% confidenceimits) for STP, CLB and VPA administered alone or in combina-ion at the fixed-ratios of 1:3, 1:1 and 3:1 in the MES-inducedeizure test were calculated by computer-assisted log-probit analy-is according to Litchfield and Wilcoxon (1949). In the isobolographicnalysis for non-parallel DRRCs, the experimentally derived ED50

ix value for the mixture of STP with CLB at the fixed-ratio of 1:1as statistically compared with its respective theoretically addi-

ive ED50 add values using the unpaired Student’s t-test, accordingo the method described by Tallarida (2006). In the isobolographicnalysis for parallel DRRCs, the experimentally derived ED50 mixalues for the mixture of STP with VPA at three fixed-ratios of:3, 1:1 and 3:1 were statistically compared with their respec-ive theoretically additive ED50 add values using the unpairedtudent’s t-test, according to the method described by Tallarida2000). Total brain AED concentrations were statistically analyzedsing the unpaired Student’s t-test. Qualitative variables from thehimney test were compared using the Fisher’s exact probabilityest. Median retention times obtained in the passive avoidanceask were statistically evaluated using Kruskal—Wallis nonpara-etric ANOVA. Mean values of skeletal muscular strength from

he grip-strength test were verified with one-way ANOVA. Dif-erences among values were considered statistically significant if< 0.05.

oftware used

icrosoft’s Excel spreadsheet was used to perform calculationsnd to graphically illustrate the results in form of isobolograms.ll statistical tests were performed using commercially availableraphPad Prism version 4.0 for Windows (GraphPad Software, Saniego, CA, USA).

192 J.J. Luszczki et al.

Figure 1 Log-probit dose-response relationship curve (DRRC)analysis of stiripentol (STP) and clobazam (CLB) administeredalone and in combination against maximal electroshock (MES)-induced seizures. Doses of STP and CLB administered alone andin combination at the fixed-ratio of 1:1 were transformed to log-arithms, whereas their protective anticonvulsant effects weretransformed to probits according to Litchfield and Wilcoxon(1949). Linear regression equations of DRRCs for STP and CLBadministered alone and in combinations are shown whereby y —is the probit of response, x — is the logarithm (to the base 10) ofan AED dose or a dose of the mixture of STP with CLB and r2 —ctt

R

Aam

SdaaSt

Table 2 Isobolographic analysis of interaction (for non-parallel DRRCs) between stiripentol (STP) and clobazam(CLB) at the fixed-ratio of 1:1 against maximal electroshock(MES)-induced seizures.

STP + CLB n STP CLB

ED50 mix 122.0 ± 11.2 16 115.5 6.5aED50 add 93.4 ± 19.9 44 88.4 5.0bED50 add 199.9 ± 35.4 44 189.2 10.7

Data are presented as median effective doses (ED50 valuesin mg/kg ± S.E.M.) for two-drug mixtures, determined eitherexperimentally (ED50 mix) or theoretically calculated (ED50 add)from the equations of additivity (Tallarida, 2006), protecting 50%of the animals against MES-induced seizures. The actual dosesof STP and CLB that comprised the mixtures at the fixed-ratio of1:1 for the ED50 mix and ED50 add values are presented in sepa-rate columns. n — total number of animals used at those doseswhose expected anticonvulsant effects ranged 16—84% (i.e., 4—6probits). Total number of animals were determined either exper-imentally (nmix) or theoretically from the equation of additivity(nadd = n STP + n CLB—4). Statistical evaluation of data was per-formed with unpaired Student’s t-test according to Tallarida(2000).

a ED50 add value calculated from the equation for the lower

(bpf(Ett

oefficient of determination. Test for parallelism revealed thathe experimentally determined DRRC for STP was non-parallelo that for CLB. For more details see Table 1.

esults

nticonvulsant effects of stiripentol and clobazamdministered separately and in combination in theouse MES model

TP administered alone (i.p., 60 min before the test) at

oses ranging between 250 and 325 mg/kg produced a clearnticonvulsant effect that increased from 25% to 87.5%gainst MES-induced seizures. The equation of DRRC forTP (y = 15.204 × −32.151; Fig. 1) allowed the determina-ion of the ED50 value for STP, which was 277.7 ± 12.1 mg/kg

(fiaf1

Table 1 Anticonvulsant effects of stiripentol (STP), clobazam (Celectroshock (MES)-induced seizures in mice.

Drug ED50 n

STP 277.7 ± 12.1 24CLB 15.6 ± 2.8 24VPA 261.0 ± 11.7 16

aTest for parallelismSTP vs. CLB S.R. = 1.61STP vs. VPA S.R. = 1.02

S.R. > f ratio S.R., the examined two DRRCs are non-parallelS.R. < f ratio S.R., the examined two DRRCs are parallel

Results are presented as median effective doses (ED50 values in mg/kgat doses whose expected anticonvulsant effects ranged 4—6 probits (16of p and q1 or q2 values; S.R. — slope function ratio for the respectivSCLB, SSTP and SVPA — are slopes for the AEDs administered alone; f ratiocombinations.

line of additivityb ED50 add value calculated from the equation for the upper

line of additivity

Table 1). Similarly, CLB administered singly (i.p., 30 minefore the test) at doses ranging between 10 and 30 mg/kgroduced a definite antiseizure activity that increasedrom 25% to 87.5%. The equation of DRRC for CLBy = 3.679x + 0.607; Fig. 1) allowed the determination of theD50 value for CLB which was 15.6 ± 2.8 mg/kg (Table 1). Theest for parallelism of DRRCs between STP and CLB revealedhat the AEDs had their DRRCs non-parallel to one another

Table 1; Fig. 1). The combination of STP with CLB at thexed-ratio of 1:1 exerted antiseizure activity in the MES testnd the experimentally derived ED50 mix value from the DRRCor the mixture of both AEDs (y = 9.221x − 15.599; Fig. 1) was22.0 ± 11.2 mg/kg (Table 2).

LB), and valproate (VPA) administered singly against maximal

CFP p/q

7.34 (p) —2.46 (q1) 2.9912.18 (q2) 0.60

f ratio S.R. = 1.26 Non-parallelf ratio S.R. = 1.07 Parallel

± S.E.M.) of STP, CLB and VPA. n — total number of animals used% and 84%); CFP — (q and p) curve-fitting parameters; p/q — ratioe two-drug combinations (i.e., SCLB/SSTP and SSTP/SVPA), where:S.R. — factor for slope function ratio for the respective two-drug

Interactions of stiripentol with clobazam and valproate in the m

Figure 2 Log-probit dose-response relationship curve (DRRC)analysis of stiripentol (STP) and valproate (VPA) administeredalone and in combination against maximal electroshock (MES)-induced seizures. Doses of STP and VPA administered alone andin combination at the fixed-ratios of 1:3, 1:1 and 3:1 were trans-formed to logarithms, whereas their protective anticonvulsanteffects were transformed to probits according to Litchfield andWilcoxon (1949). Linear regression equations of DRRCs for STPand VPA administered alone and in combinations are shownwhereby y — is the probit of response, x — is the logarithm(to the base 10) of an AED dose or a dose of the mixture of STPwith VPA and r2 — coefficient of determination. Test for paral-

aMtFcsvA(tacA(

Is

TDfitfw91TE

Is

TrVtvwas 323.9 ± 21.0 mg/kg which was significantly more than

lelism revealed that the experimentally determined DRRC forSTP was parallel to that for VPA when administered alone. Formore details see Table 1.

Anticonvulsant effects of stiripentol and valproateadministered separately and in combination in themouse MES model

VPA administered singly (i.p., 30 min before the test) atdoses ranging between 225 and 300 mg/kg produced a clearantiseizure effect that increased from 12.5% to 87.5% inthe mouse MES model. The equation of DRRC for VPA(y = 18.082x − 38.699; Fig. 2) allowed the determination of

the ED50 value for VPA and was 261.0 ± 11.7 mg/kg (Table 1).The test for parallelism of DRRCs between STP and VPArevealed that the AEDs had their DRRCs parallel to oneanother (Table 1; Fig. 2). The combination of STP with VPA

tTEfi

Table 3 Isobolographic analysis of interactions (for parallel DRmaximal electroshock (MES)-induced seizure model.

Combination FR ED50 mix nmix STP

STP + VPA 1:3 323.9 ± 21.0** 16 84.91:1 346.9 ± 27.2*** 16 178.83:1 324.6 ± 23.8* 8 247.2

Data are presented as median effective doses (ED50 values in mg/kgseizures. The ED50 values were either experimentally determined from(ED50 add) from the equation of additivity (Loewe, 1953). STP — dose ofevaluation of data was performed using unpaired Student’s t-test. FRanimals used at those doses whose expected anticonvulsant effects wedrugs (nmix) and theoretically calculated (nadd = n STP + n VPA − 4) from t

* P < 0.05 vs. the respective ED50 add values.** P < 0.01 vs. the respective ED50 add values.

*** P < 0.001 vs. the respective ED50 add values.

ouse maximal electroshock-induced seizure model 193

t the fixed-ratio of 1:3 exerted antiseizure activity in theES test and the experimentally derived ED50 mix value from

he DRRC for the mixture of both AEDs (y = 12.544x − 26.489;ig. 2) was 323.9 ± 21.0 mg/kg (Table 3). Similarly, STPombined with VPA at the fixed-ratio of 1:1 produced anti-eizure activity in the MES test and thus, the ED50 mix

alue calculated from the DRRC for the mixture of bothEDs (y = 10.361x − 21.320; Fig. 2) was 346.9 ± 27.2 mg/kgTable 3). The last tested combination of STP with VPA inhe mouse MES model at the fixed-ratio of 3:1 exerted alsodefinite anticonvulsant effect and the ED50 mix value cal-

ulated directly from the DRRC for the mixture of bothEDs (y = 15.712x − 34.460; Fig. 2) was 324.6 ± 23.8 mg/kgTable 3).

sobolographic analysis of interaction betweentiripentol and clobazam in the mouse MES model

he isobolographic analysis of interaction for non-parallelRRCs revealed that the mixture of STP with CLB at thexed-ratio of 1:1 exerted additive interaction in the MESest (Fig. 3). The experimentally derived ED50 mix valueor this fixed-ratio combination was 122.0 ± 11.2 mg/kg,hereas the additively calculated ED50 add values were3.4 ± 19.9 mg/kg (for the lower ED50 add) and99.9 ± 35.4 mg/kg (for the upper ED50 add; Table 2).hus, the ED50 mix value did not significantly differ from theD50 add values (Table 2; Fig. 3).

sobolographic analysis of interaction betweentiripentol and valproate in the mouse MES model

he isobolographic analysis of interaction for parallel DRRCsevealed that all three fixed-ratio combinations of STP withPA at 1:3, 1:1, and 3:1 exerted sub-additive (antagonis-ic) interaction in the MES test (Table 3; Fig. 4). The ED50 mix

alue for the mixture of STP with VPA at the fixed-ratio of 1:3

he corresponding ED50 add of 265.2 ± 9.3 mg/kg (P < 0.01;able 3; Fig. 4). Similarly, the experimentally derivedD50 mix value for the combination of STP with VPA at thexed-ratio of 1:1 was 346.9 ± 27.2 mg/kg and considerably

RCs) of stiripentol (STP) with valproate (VPA) in the mouse

VPA ED50 add nadd STP VPA

239.1 265.2 ± 9.3 36 69.4 195.8168.1 269.3 ± 8.4 36 138.8 130.577.5 273.5 ± 9.6 36 208.2 65.3

± S.E.M.) protecting 50% of animals tested against MES-inducedthe mixture of two AEDs (ED50 mix) or theoretically calculated

STP in the mixture; VPA — dose of VPA in the mixture. Statistical— fixed-ratio of drug dose combinations; n — total number of

re ranged 4—6 probits, denoted for the experimental mixture ofhe equation of additivity.

194 J.J. Luszczki et al.

Figure 3 Isobologram showing interactions between stiripen-tol (STP) and clobazam (CLB) against maximal electroshock(MES)-induced seizures. The solid lines on the X- and Y-axesrepresent the S.E.M. for STP and CLB administered alone. (©)depicts the experimentally derived ED50 mixs (±S.E.M.), whereED50 = median effective dose; (�) depicts the theoretically cal-culated ED50 adds (±S.E.M.) for total dose expressed as theproportion of STP and CLB that produced 50% protection ofanimals against MES-induced seizures (S.E.M. values are pre-sented as horizontal and vertical error bars). The lower andupper isoboles of additivity represent the curves connecting theED50 values for STP and CLB administered alone. The dotted linestarting from the point (0; 0) corresponds to the fixed-ratio of1:1 for the combination of STP with CLB. The diagonal dashedline connects the ED50 for STP and CLB on the X- and Y-axes.The points A′ and A′′ depict the theoretically calculated ED50 add

values for both lower and upper isoboles of additivity. The pointM represents the experimentally derived ED50 mix value for totaldose of the mixture expressed as proportions of STP and CLB thatproduced a 50% anticonvulsant effect (50% isobole). The sum ofX- and Y-coordinates, for each point placed on the isobologram,corresponds to the respective ED50 values. The point S reflectsthe ED50 add value denoted theoretically from the Loewe’s equa-tion for the fixed-ratio combination of 1:1. The experimentallyderived ED50 mix value is placed between the points S and A′,within the area of additivity bounded by two isoboles of addi-ta1

h2Soswo

B

TwcoCo

Figure 4 Isobolograms showing interactions betweenstiripentol (STP) and valproate (VPA) against maximal elec-troshock (MES)-induced seizures. The solid lines on the X- andY-axes represent the S.E.M. for STP and VPA administeredalone. (©) depicts the experimentally derived ED50 mixs(±S.E.M.), where ED50 = median effective dose; (�) depictsthe theoretically calculated ED50 adds (±S.E.M.) for total doseexpressed as the proportion of STP and VPA that produced 50%protection of animals against MES-induced seizures (S.E.M.values are presented as horizontal and vertical error bars).The straight line connecting the ED50 values of STP and VPArepresents the theoretical line of additivity for a continuum ofdifferent fixed dose ratios. The dotted lines starting from thepoint (0; 0) correspond to the fixed-ratios of 1:3, 1:1, and 3:1for the combination of STP with VPA. The points M1, M2, and M3reflect the ED50 mix values at the fixed-ratios of 1:3, 1:1 and 3:1,respectively. Similarly, the points A1, A2, and A3 illustrate theED50 add values at the same fixed-ratios. The experimental ED50

mix values of the mixture of STP with VPA for all the fixed-ratiosare placed significantly above the theoretical line of additivity,indicating a sub-additive (antagonistic) interaction. X- andY-coordinates for all points are: M1 (84.8; 239.1), M2 (178.8;1at

c(p

Etlig

Wc1o

ivity, indicating additive interaction between STP and CLB. X-nd Y-coordinates for all points are: A′ (88.4; 5.0), A′′ (189.2;0.7), S (138.8; 7.8), and M (115.5; 6.5).

igher than that for the corresponding ED50 add, which was69.3 ± 8.4 mg/kg (P < 0.001; Table 3; Fig. 4). The mixture ofTP with VPA at the fixed-ratio of 3:1 sub-additively (antag-nistically) protected the animals against MES-inducedeizures, and the ED50 mix value was 324.6 ± 23.8 mg/kg,hich was significantly elevated compared to the ED50 add

f 273.5 ± 9.6 mg/kg (P < 0.05; Table 3; Fig. 4).

rain antiepileptic drug concentrations

otal brain concentrations revealed bi-directional changes

hen STP was co-administered with CLB and VPA at dosesorresponding to the ED50 mix values at the fixed-ratiof 1:1 from the MES test (Table 4). Thus, STP increasedLB (P < 0.001) brain concentrations whilst concentrationf DMCLB and VPA were decreased. In contrast, STP brain

((dih

68.1), M3 (247.2; 77.5), A1 (69.4; 195.8), A2 (138.8; 130.5),nd A3 (208.2; 65.3). *P < 0.05, **P < 0.01 and ***P < 0.001 vs.he respective ED50 add values.

oncentrations were increased by CLB (P < 0.05) and VPAP < 0.001). The increase in brain CLB concentrations wasarticularly profound being almost 19-fold (Table 4).

ffects of stiripentol, clobazam, valproate andheir combination on motor performance,ong-term memory and skeletal muscular strengthn the chimney, passive avoidance andrip-strength tests

hen STP and CLB and VPA were co-administered at dosesorresponding to their ED50 mix values at the fixed-ratio of:1 from the MES-induced seizure test, motor performancef animals as assessed by the chimney test was unaffected

Table 5). Furthermore, none of the combinations studiedSTP with CLB and VPA) impaired long-term memory asetermined in the passive avoidance test (Table 5). Sim-larly, STP concomitantly administered with CLB and VPAad no significant impact on skeletal muscular strength of

Interactions of stiripentol with clobazam and valproate in the mouse maximal electroshock-induced seizure model 195

Table 4 Total brain concentrations of clobazam (CLB), N-desmethyl clobazam (DMCLB), stiripentol (STP), and valproate (VPA)administered singly or in combination.

Treatment (mg/kg) Total brain concentration (�g/ml)a

CLB (6.5) + vehicle 0.05 ± 0.03CLB (6.5) + STP (115.5) 0.93 ± 0.07*** (↑ 18.6-fold)DMCLB + vehicle 0.45 ± 0.26DMCLB + STP (115.5) 0.28 ± 0.07 (↓ 38%)STP (115.5) + vehicle 0.22 ± 0.08STP (115.5) + CLB (6.5) 0.37 ± 0.17* (↑ 68%)VPA (168.1) + vehicle 112.7 ± 15.3VPA (168.1) + STP (178.8) 90.3 ± 13.9 (↓ 20%)STP (178.8) + vehicle 3.24 ± 0.27STP (178.8) + VPA (168.1) 7.58 ± 0.66*** (↑ 2.3-fold)

Data are presented as means ± S.D. of eight determinations.a Concentrations relate to that antiepileptic drug shown first in the respective treatment column. Statistical evaluation was performed

by use of the unpaired Student’s t-test. ↑ and ↓ represent an increase and decrease in brain concentrations compared to their respective

eaatsIefios

itaaat1ar

control groups.* P < 0.05 vs. the respective control group.

*** P < 0.001 vs. the respective control group.

the animals as assessed by the grip-strength test (Table 5).Moreover, it was found that the control (vehicle-treated)mice and those receiving AEDs alone (at doses correspond-ing to their ED50 values from the mouse MES model) did notshow any significant signs of impaired motor coordination,long-term memory or muscular skeletal strength, as assessedin the chimney, passive avoidance and grip-strength tests,respectively (Table 5).

Discussion

Results presented in this study indicate that STP in a dose-dependent manner suppressed tonic hind limb extension(seizure activity) in mice and this finding is consistent withthe results presented earlier by Poisson et al. (1984), whofound that STP is effective against electrically inducedseizures in rodents. Previously, we reported that STP dosedependently elevated the threshold for electroconvulsionsin mice (Luszczki et al., 2007), and when combined with CBZ

it exerted a biphasic interaction (synergistic and antagonis-tic) in the mouse MES model (Luszczki and Czuczwar, 2006).The synergistic (supra-additive) interaction was observedfor the combination of STP with CBZ at the fixed-ratio of1:3, whereas the two-drug mixtures at the fixed-ratio of 3:1

mdrr

Table 5 Effects of stiripentol (STP), clobazam (CLB), valproatechimney test, on retention time in the passive avoidance task, mu

Treatment (mg/kg) Motor performance (%)

Vehicle 100STP (277.7) + vehicle 100CLB (15.6) + vehicle 100VPA (261.0) + vehicle 100STP (115.5) + CLB (6.5) 100STP (178.8) + VPA (168.1) 100

Results (n = 8) are presented as: (1) percentage of animals without impairwith 25th and 75th percentiles in parentheses); (3) mean grip strength (performed with nonparametric Kruskal—Wallis ANOVA, by one-way ANO

xerted a sub-additive (antagonistic) interaction (Luszczkind Czuczwar, 2006). Additionally, with type I isobolographicnalysis we observed that STP exerted additive interac-ion with CZP, ETS, PB and VPA in the mouse PTZ-inducedeizure model (Luszczki et al., 2006a). In contrast, the typeisobolographic analysis of interaction in this study providedvidence that the combination of STP with VPA at threexed-ratios of 1:3, 1:1 and 3:1 exerted sub-additive (antag-nistic) interaction in terms of suppression of MES-inducedeizures in mice.

The mechanism of action of STP, CLB and VPA may explainn part the pharmacodynamic interactions observed withhese combinations. The primary mode of action of STP is vian effect GABA inhibitory neurotransmission in the brain — itcts as a unique positive allosteric modulator (weak partialgonist) (Fisher, 2009); inhibits GABA metabolism throughhe blockade of GABA-transaminase activity (Poisson et al.,984); reduces synaptosomal uptake of GABA (Wegmann etl., 1978) and increases the mean open duration of GABAA

eceptor-dependent chloride channels by a barbiturate-like

echanism (Quilichini et al., 2006). With respect to VPA, therug blocks low-threshold T-type calcium channels in neu-ons, potentiates GABA-ergic transmission in specific brainegions and it increases synthesis of GABA by activating

(VPA), and their combination on motor performance in thescular strength in the grip-strength test.

Retention time (s) Grip strength (N)

180 (180; 180) 86.5 ± 5.5180 (165.8; 180) 86.3 ± 5.0180 (180; 180) 85.9 ± 5.2175 (145.5; 180) 87.2 ± 5.7180 (165; 180) 85.0 ± 5.2

165.5 (135.5; 180) 85.2 ± 5.5

ment of motor coordination; (2) median retention times (seconds;newtons ± S.E.M.). For the three datasets statistical analysis wasVA and the Fisher’s exact probability test, respectively.

1

tlretatrerc(oppats

mAriIo2tmAi

ipcdcScCTbwliVb1aoawAcic

waciSw

ttScp

baiim

bctcmdcmfsccetanPwcapb2mP(ncduaoot

wtwcpp

C

P

96

he GABA-synthesizing enzyme glutamic acid decarboxy-ase (Löscher, 2002b). VPA increases the potassium-inducedelease of GABA in cortical neurons (Löscher, 2002b; Waldent al., 1993). The drug reduces the level and release ofhe excitatory amino acid aspartate in the brain and itttenuates neuronal excitation by the reduction of activa-ion of NMDA receptors (Rogawski and Porter, 1990). Withegards to CLB, a 1,5-benzodiazepine, it enhances GABA-rgic activity by binding to the � subunit of the GABAA

eceptor and increasing the frequency of chloride channelonductance by allosteric activation of the GABAA receptorNg and Collins, 2007). Moreover, CLB increases expressionf glutamate transporter protein 1 (GLT1) and GABA trans-orter protein 3 (GAT3) in the brain (Doi et al., 2005). Atresent, it is difficult to unequivocally ascertain which of thebove-mentioned mechanisms of action are responsible forhe pharmacodynamic interactions observed in the presenttudy.

The fact that STP combined with VPA in the mouse MESodel exerted antagonistic interaction and that the sameED combination in the mouse PTZ-induced seizure modelesulted in additive interaction, it can be concluded thatnteractions between drugs is seizure model dependent.ndeed we reported on similar effects for the combinationf loreclezole (LCZ) with PB and VPA (Luszczki et al., 2005b,006b). Thus the combination of LCZ with PB and VPA athe fixed-ratio of 1:1 was supra-additive (synergistic) in theouse MES model (Luszczki et al., 2006b), whereas the sameED combinations at the fixed-ratio of 1:1 exerted additive

nteraction in the mouse PTZ model (Luszczki et al., 2005b).A major reason why STP has been abandoned for develop-

ng and therefore licensing in adult epilepsy is because of itsrofound propensity to interact. Nevertheless, despite thisharacteristic it has achieved a license for SMEI (Dravet syn-rome) — a difficult to treat syndromic epilepsy — and manyhildren have become seizure free when prescribed STP.TP is not only a potent hepatic enzyme inhibitor (inhibitsytochrome P450 isoenzymes: CYP1A2, CYP3A4, CYP1A2,YP3A4, CYP2D6, 2CYPC9 and CYP2C19; Kerr et al., 1991;ran et al., 1997) but additionally its metabolism is inducedy enzymes inducing AEDs (e.g. CBZ, PHT, PB and PRM). Thushilst blood levels of PHT, CBZ and its primary pharmaco-

ogically active metabolite CBZ-epoxide, PB, PRM, CLB andts primary pharmacologically active metabolite DMCLB andPA are increased by STP, blood levels of STP are decreasedy CBZ, PHT, PB and PRM (Levy et al., 1984; Kerr et al.,991; Bebin and Bleck, 1994; Tran et al., 1997; Chiron etl., 2000; Cazali et al., 2003; Giraud et al., 2006). Becausef the interaction potential of STP, in the present study welso measured brain concentrations of all AEDs including STPhen administered at the fixed-ratio of 1:1 (Table 4). BrainED concentrations were determined instead of plasma con-entrations because the brain represents the biophase (i.e.,n the central compartment where the AEDs exert their anti-onvulsant effects).

The most profound changes in brain AED concentrationere those associated with STP in combination with CLB

nd VPA and these data confirm what has been reportedlinically in blood. Namely there was a 18.6-fold increasen brain CLB concentrations and a 2.3-fold increase in brainTP concentrations and the latter observation is exactly thate reported earlier (Luszczki et al., 2006b). In addition

tUsDP

J.J. Luszczki et al.

o these interactions other interactions were observed andhose that were statistically significant include increases inTP concentrations by CLB (Table 4). These pharmacokinetichanges invariably mask both the additive and antagonisticharmacodynamic interactions observed in this study.

Because STP is 99% bound to plasma albumin, proteininding displacement could be a further contributing mech-nism for the observed changes in brain concentrations. VPAs also significantly bound to plasma proteins (88—92%) ands well known to be associated with protein binding displace-ent interactions (Patsalos et al., 2008).With regards to adverse effects of the various AED com-

inations, despite the substantial increases in brain AEDoncentrations (up to ∼19-fold), results from the step-hrough passive avoidance task indicate that none of theombinations were associated with an affect long-termemory. Furthermore, there were no effects on motor coor-ination and skeletal muscular strength as assessed by thehimney and grip-strength tests, respectively. These dataight suggest that these tests of adverse effects may there-

ore not be sensitive, however this is not the case. In atudy of tiagabine (TGB) co-administered with VPA, a signifi-ant impairment in motor coordination, as determined in thehimney test, was observed (Luszczki et al., 2003a). Theseffects were associated with a significant (93%) increase inotal VPA brain concentrations (Luszczki et al., 2003a). Instudy of WIN 55,212-2 mesylate (a non-specific cannabi-

oid CB1 and CB2 receptor agonist) in combination withB and VPA, a significant impairment in motor coordinationas observed despite no substantial increase in drug brainoncentrations (Luszczki and Czuczwar, 2009). Similarly,dverse effects have been documented in the step-throughassive avoidance task (TGB with gabapentin [GBP] and viga-atrin [VGB] with CZP and VPA) (Luszczki et al., 2003b,005a). Finally, in the grip-strength test, WIN 55,212-2esylate in combination with four classical AEDs (CBZ,HT, PB and VPA) significantly reduced muscular strengthLuszczki and Czuczwar, 2009). Moreover, these effects wereot associated with a significant effect on total AED brainoncentrations (Luszczki and Czuczwar, 2009). The above-escribed data clearly indicate that the experimental testssed in the present study were sensitive for measurement ofnimal behavior and adverse effect changes that might haveccurred. Consequently, it can be concluded that the lackf adverse effects when STP was combined with VPA or CLBestifies to a low toxic potential of these drug combinations.

In conclusion, STP and CLB in combination are associatedith pharmacodynamic additivity and this feature along with

heir concurrent pharmacokinetic interaction may explainhy clinically this combination is so effective in SMEI. Inontrast, STP in combination with VPA was associated withharmacodynamic antagonism which was additionally com-licated by pharmacokinetic interactions.

onflict of interest

rofessor Philip N. Patsalos has received speaker’s or consul-

ancy fees and/or research grants from Sanofi Aventis andCB Pharma. Professor Stanislaw J. Czuczwar has receivedupport from UCB Pharma and Sanofi-Aventis as a speaker.r Neville Ratnaraj has received a travel grant from UCBharma. The other authors have no disclosures to declare.

e m

L

L

L

L

L

L

L

L

L

L

L

L

L

L

L

Interactions of stiripentol with clobazam and valproate in th

Acknowledgments

This study was supported in part by a grant (KBN2P05D05126) from the State Committee for ScientificResearch (Warszawa, Poland). The authors are grateful forthe generous gifts of stiripentol from Laboratoires Biocodex(Gentilly, France) and valproate magnesium from ICN PolfaS.A. (Rzeszow, Poland). Professor J.J. Luszczki is a Recipi-ent of the Fellowship for Leading Young Researchers fromthe Ministry of Science and Higher Education (Warszawa,Poland).

References

Bebin, M., Bleck, T.P., 1994. New anticonvulsant drugs. Focus onflunarizine, fosphenytoin, midazolam and stiripentol. Drugs 48,153—171.

Boissier, J.R., Tardy, J., Diverres, J.C., 1960. Une nouvelle méth-ode simple pour explorer l’action tranquilisante: le test de lacheminée. Med. Exp. (Basel) 3, 81—84.

Cazali, N., Tran, A., Treluyer, J.M., Rey, E., d’Athis, P., Vincent, J.,Pons, G., 2003. Inhibitory effect of stiripentol on carbamazepineand saquinavir metabolism in human. Br. J. Clin. Pharmacol. 56,526—536.

Chiron, C., Coppola, G., Plouin, P., Robain, O., Renard, F., 1993.Multifocal epilepsy in infancy: value of stiripentol—clonazepamcombination. Epilepsia 34 (Suppl. 6), 73.

Chiron, C., Marchand, M.C., Tran, A., Rey, E., d’Athis, P., Vincent,J., Dulac, O., Pons, G., 2000. Stiripentol in severe myoclonicepilepsy in infancy: a randomised placebo-controlled syndrome-dedicated trial, STICLO study group. Lancet 356, 1638—1642.

Doi, T., Ueda, Y., Tokumaru, J., Willmore, L.J., 2005. Molecular reg-ulation of glutamate and GABA transporter proteins by clobazamduring epileptogenesis in Fe(++ + )-induced epileptic rats. BrainRes. Mol. Brain Res. 142, 91—96.

Farwell, J.R., Anderson, G.D., Kerr, B.M., Tor, J.A., Levy, R.H., 1993.Stiripentol in atypical absence seizures in children: an open trial.Epilepsia 34, 305—311.

Fisher, J.L., 2009. The anti-convulsant stiripentol acts directly onthe GABA(A) receptor as a positive allosteric modulator. Neu-ropharmacology 56, 190—197.

Gasior, M., Ungard, J.T., Witkin, J.M., 1999. Preclinical eval-uation of newly approved and potential antiepileptic drugsagainst cocaine-induced seizures. J. Pharmacol. Exp. Ther. 290,1148—1156.

Giraud, C., Treluyer, J.M., Rey, E., Chiron, C., Vincent, J., Pons, G.,Tran, A., 2006. In vitro and in vivo inhibitory effect of stiripentolon clobazam metabolism. Drug Metab. Dispos. 34, 608—611.

Inoue, Y., Ohtsuka, Y., Oguni, H., Tohyama, J., Baba, H., Fukushima,K., Ohtani, H., Takahashi, Y., Ikeda, S., 2009. Stiripentol openstudy in Japanese patients with Dravet syndrome. Epilepsia 50,2362—2368.

Kassaï, B., Chiron, C., Augier, S., Cucherat, M., Rey, E., Gueyffier,F., Guerrini, R., Vincent, J., Dulac, O., Pons, G., 2008. Severemyoclonic epilepsy in infancy: a systematic review and a meta-analysis of individual patient data. Epilepsia 49, 343—348.

Kerr, B.M., Martinez-Lage, J.M., Viteri, C., Tor, J., Eddy, A.C., Levy,R.H., 1991. Carbamazepine dose requirements during stiripentoltherapy: influence of cytochrome P-450 inhibition by stiripentol.Epilepsia 32, 267—274.

Levy, R.H., Loiseau, P., Guyot, M., Blehaut, H.M., Tor, J., More-

land, T.A., 1984. Stiripentol kinetics in epilepsy: nonlinearityand interactions. Clin. Pharmacol. Ther. 36, 661—669.

Litchfield, J.T., Wilcoxon, F., 1949. A simplified method of eval-uating dose-effect experiments. J. Pharmacol. Exp. Ther. 96,99—113.

L

ouse maximal electroshock-induced seizure model 197

ockard, J.S., Levy, R.H., Rhodes, P.H., Moore, D.F., 1985. Stiripen-tol in acute/chronic efficacy tests in monkey model. Epilepsia26, 704—712.

oewe, S., 1953. The problem of synergism and antagonism of com-bined drugs. Arzneimittelforschung 3, 285—290.

oiseau, P., Levy, R.H., Houin, G., Rascol, O., Dordain, G.,1990. Randomized double-blind, parallel, multicenter trialof stiripentol added to carbamazepine in the treatment ofcarbamazepine-resistant epilepsies: an interim analysis. Epilep-sia 31, 618—619.

öscher, W., 2002a. Animal models of epilepsy for the developmentof antiepileptogenic and disease-modifying drugs. A compari-son of the pharmacology of kindling and post-status epilepticusmodels of temporal lobe epilepsy. Epilepsy Res. 50, 105—123.

öscher, W., 2002b. Basic pharmacology of valproate: a review after35 years of clinical use for the treatment of epilepsy. CNS Drugs16, 669—694.

öscher, W., Fassbender, C.P., Nolting, B., 1991. The role of tech-nical, biological and pharmacological factors in the laboratoryevaluation of anticonvulsant drugs. II. Maximal electroshockseizure models. Epilepsy Res. 8, 79—94.

uszczki, J.J., 2007. Isobolographic analysis of interaction betweendrugs with nonparallel dose—response relationship curves: apractical application. Naunyn-Schmiedebergs Arch. Pharmacol.375, 105—114.

uszczki, J.J., Czuczwar, S.J., 2004. Isobolographic profile of inter-actions between tiagabine and gabapentin: a preclinical study.Naunyn-Schmiedebergs Arch. Pharmacol. 369, 434—446.

uszczki, J.J., Czuczwar, S.J., 2006. Biphasic characteristic ofinteractions between stiripentol and carbamazepine in themouse maximal electroshock-induced seizure model: a three-dimensional isobolographic analysis. Naunyn-SchmiedebergsArch. Pharmacol. 374, 51—64.

uszczki, J.J., Czuczwar, S.J., 2009. Effect of WIN 55,212-2mesylate—–a highly potent cannabinoid CB1 and CB2 receptoragonist on the protective action of carbamazepine, pheny-toin, phenobarbital and valproate in the mouse maximalelectroshock-induced seizure model. Epilepsia 50 (Suppl. 11),365 [Abstract from the 2009 Annual Meeting of the AmericanEpilepsy Society].

uszczki, J.J., Swiader, M., Czuczwar, M., Kis, J., Czuczwar, S.J.,2003a. Interactions of tiagabine with some antiepileptics in themaximal electroshock in mice. Pharmacol. Biochem. Behav. 75,319—327.

uszczki, J.J., Swiader, M., Parada-Turska, J., Czuczwar, S.J.,2003b. Tiagabine synergistically interacts with gabapentin in theelectroconvulsive threshold test in mice. Neuropsychopharma-cology 28, 1817—1830.

uszczki, J.J., Wojcik-Cwikla, J., Andres, M.M., Czuczwar, S.J.,2005a. Pharmacological and behavioral characteristics of inter-actions between vigabatrin and conventional antiepileptic drugsin pentylenetetrazole-induced seizures in mice: an isobolo-graphic analysis. Neuropsychopharmacology 30, 958—973.

uszczki, J.J., Ratnaraj, N., Patsalos, P.N., Czuczwar, S.J.,2005b. Pharmacodynamic and/or pharmacokinetic characteris-tics of interactions between loreclezole and four conventionalantiepileptic drugs in pentylenetetrazole-induced seizures inmice: an isobolographic analysis. Epilepsy Behav. 7, 639—651.

uszczki, J.J., Ratnaraj, N., Patsalos, P.N., Czuczwar, S.J., 2006a.Characterization of the anticonvulsant, behavioral and pharma-cokinetic interaction profiles of stiripentol in combination withclonazepam, ethosuximide, phenobarbital, and valproate usingisobolographic analysis. Epilepsia 47, 1841—1854.

uszczki, J.J., Ratnaraj, N., Patsalos, P.N., Czuczwar, S.J., 2006b.Isobolographic analysis of interactions between loreclezoleand conventional antiepileptic drugs in the mouse maximalelectroshock-induced seizure model. Naunyn-SchmiedebergsArch. Pharmacol. 373, 169—181.

1

L

L

M

M

N

P

P

P

Q

R

R

R

S

S

T

T

T

V

W

98

uszczki, J.J., Dudra-Jastrzebska, M., Andres-Mach, M., Czernecki,R., Barcicka-Klosowska, B., Misiuta-Krzesinska, M., Zwolinski,J., Filip, D., 2007. Stiripentol in a dose-dependent manner ele-vates the threshold for maximal electroshock-induced seizuresin mice. J. Pre-Clin. Clin. Res. 1, 155—157.

uszczki, J.J., Antkiewicz-Michaluk, L., Czuczwar, S.J., 2009.Isobolographic analysis of interactions between 1-methyl-1,2,3,4-tetrahydroisoquinoline and four conventionalantiepileptic drugs in the mouse maximal electroshock-inducedseizure model. Eur. J. Pharmacol. 602, 298—305.

eyer, O.A., Tilson, H.A., Byrd, W.C., Riley, M.T., 1979. A methodfor the routine assessment of fore- and hindlimb grip strengthof rats and mice. Neurobehav. Toxicol. 1, 233—236.

icheletti, G., Vargnes, M., Lannes, B., Tor, J., Marescaux, A.,Depaulis, A., Warter, J.M., 1988. Effect of stiripentol on petitmal-like epilepsy in Wistar rats. Epilepsia 29, 709.

g, Y.T., Collins, S.D., 2007. Clobazam. Neurotherapeutics 4,138—144.

atsalos, P.N., Berry, D.J., Bourgeois, B.F.D., Cloyd, J.C., Glauser,T.A., Johannessen, S.I., Leppik, I.E., Tomson, T., Perucca, E.,2008. Antiepileptic drugs—best practice guidelines for therapeu-tic drug monitoring: a position paper by the Subcommission onTherapeutic Drug Monitoring, ILAE Commission on TherapeuticStrategies. Epilepsia 49, 1239—1276.

erez, J., Chiron, C., Musial, C., Rey, E., Blehaut, H., d’Athis,P., Vincent, J., Dulac, O., 1999. Stiripentol: efficacy andtolerability in children with epilepsy. Epilepsia 40, 1618—1626.

oisson, M., Huguet, F., Savattier, A., Bakri-Logeais, F., Narcisse,G., 1984. A new type of anticonvulsant, stiripentol. Pharmaco-logical profile and neurochemical study. Arzneimittelforschung34, 199—204.

uilichini, P.P., Chiron, C., Ben-Ari, Y., Gozlan, H., 2006. Stiripentol,

a putative antiepileptic drug, enhances the duration of openingof GABA-receptor channels. Epilepsia 47, 704—716.

enard, F., Musial, C., Tor, J., Dulac, O., 1993. Single-blind add-ontrial of stiripentol in epileptic children. Epilepsia 34 (Suppl. 6),74.

W

J.J. Luszczki et al.

ey, E., Tran, A., d’Athis, P., Chiron, C., Dulac, O., Vincent,J., Pons, G., 1999. Stiripentol potentiates clobazam in child-hood epilepsy: a pharmacological study. Epilepsia 40 (Suppl. 7),112—113.

ogawski, M.A., Porter, R.J., 1990. Antiepileptic drugs: pharmaco-logical mechanisms and clinical efficacy with consideration ofpromising developmental stage compounds. Pharmacol. Rev. 42,223—286.

hen, D.D., Levy, R.H., Moor, M.J., Savitch, J.L., 1990. Efficacy ofstiripentol in the intravenous pentylenetetrazol infusion seizuremodel in the rat. Epilepsy Res. 7, 40—48.

hen, D.D., Levy, R.H., Savitch, J.L., Boddy, A.V., Tombret, F.,Lepage, F., 1992. Comparative anticonvulsant potency and phar-macokinetics of (+)- and (−)-enantiomers of stiripentol. EpilepsyRes. 12, 29—36.

allarida, R.J., 2000. Drug Synergism and Dose-Effect Data Analysis.Chapman & Hall/CRC, Boca Raton, USA.

allarida, R.J., 2006. An overview of drug combination analysis withisobolograms. J. Pharmacol. Exp. Ther. 319, 1—7.

ran, A., Rey, E., Pons, G., Rousseau, M., d’Athis, P., Olive, G.,Mather, G.G., Bishop, F.E., Wurden, C.J., Labroo, R., Trager,W.F., Kunze, K.L., Thummel, K.E., Vincent, J.C., Gillardin,J.M., Lepage, F., Levy, R.H., 1997. Influence of stiripentol oncytochrome P450-mediated metabolic pathways in humans: invitro and in vivo comparison and calculation of in vivo inhibitionconstants. Clin. Pharmacol. Ther. 62, 490—504.

enault, P., Chapouthier, G., de Carvalho, L.P., Simiand, J., Morre,M., Dodd, R.H., Rossier, J., 1986. Benzodiazepine impairs andbeta-carboline enhances performance in learning and memorytasks. Nature 321, 864—866.

alden, J., Altrup, U., Reith, H., Speckmann, E.J., 1993. Effects ofvalproate on early and late potassium currents of single neurons.Eur. Neuropsychopharmacol. 3, 137—141.

egmann, R., Ilies, A., Aurousseau, M., 1978. Pharmaco-cellularenzymology of the mechanism of action of stiripentol incardiazol-induced epilepsy. III. Protein, nucleoprotein, lipid, andproteoglycan metabolism. Cell. Mol. Biol. Incl. Cyto Enzymol. 23,455—480.