Epilepsy Research (2012) 101, 217—227

jou rn al h om epa ge: www.elsev ier .com/ locate /ep i lepsyres

5-HT3 receptor mediates the dose-dependent effectsof citalopram on pentylenetetrazole-induced clonicseizure in mice: Involvement of nitric oxide

Borna Payandemehra,b, Arash Bahremanda, Reza Rahimiana, Pouya Ziai a,Afsaneh Amouzegara, Mohammad Sharifzadehb, Ahmad Reza Dehpoura,∗

a Department of Pharmacology, School of Medicine, Tehran University of Medical Sciences, Tehran, Iranb Department of Pharmacology and Toxicology, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran

Received 21 August 2011; received in revised form 21 March 2012; accepted 7 April 2012Available online 9 May 2012

KEYWORDSCitalopram;Pentylenetetrazole;Clonic seizurethreshold;Mice

Summary Citalopram is a selective serotonin reuptake inhibitor (SSRI), widely used in thetreatment of depressive disorders. It has been shown that citalopram affects seizure suscep-tibility. Although the exact mechanism of these effects are not yet fully understood, recentdata suggest that 5HT3 receptors and nitric oxide (NO) might participate in the central effectsof SSRIs. In this study in a mouse model of clonic seizure induced by pentylenetetrazole weinvestigated whether 5-HT3 receptors are involved in the effects of citalopram on seizurethreshold. In our experiments, citalopram at lower doses (0.5 and 1 mg/kg, i.p) significantlyincreased the seizure threshold and at higher doses (≥25 mg/kg) showed proconvulsive effects.Moreover, mCPBG (a 5-HT3 receptor agonist) at low and non-effective doses augmented whilenon-effective doses of tropisetron prevented the anticonvulsive properties of citalopram.

On the other hand, Low doses of nitric oxide synthase inhibitors L-NAME and 7-NI alone or incombination with lower doses of 5-HT3 receptor agonist enhanced the anticonvulsive propertyof citalopram, while L-arginine (NO precursor) alone or in combination with tropisetron blockedthe protective effect of citalopram.

In summary, our findings demonstrate that 5-HT3 receptor mediates the anticonvulsant prop-erties of low doses of citalopram, whereas it seems that the proconvulsive effect is mostly

mediated through the NO pathway and can be totally blocked by NOS inhibitors. This couldpropose a new approach toward finding the mechanism of citalopram activity, curtailing theadverse effects of citalopram anof citalopram overdose.© 2012 Elsevier B.V. All rights re

Abbreviations: NO, nitric oxide; NOS, nitric oxide synthase; mCPBG, mester; 7-NI, 7-nitroindazole.

∗ Corresponding author at: Department of Pharmacology, School of MedTehran, Iran. Tel.: +98 21 8897 3652; fax: +98 21 6640 2569.

E-mail address: dehpour@yahoo.com (A.R. Dehpour).

0920-1211/$ — see front matter © 2012 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.eplepsyres.2012.04.004

d perhaps managing the convulsions as a vicious consequence

served.

eta-Chlorophenylbiguanidine; L-NAME, NG-nitro-L-arginine methyl

icine, Tehran University of Medical Sciences, P.O. Box 13145-784

2

I

Cm(ppeFsiWpsms

d5tS2ep1sgem

aeupf1srgesaIpPaB

ttlo

dUo5wahc

M

A

MIswl2w1isoiGmta

D

TPN(C(fiPawacip

samteCapgtre

S

T3(Loscher, 2002). The needle was then secured to the tail

18

ntroduction

italopram is a selective serotonin reuptake inhibitor (SSRI)ainly used in the treatment of major depressive disorder

Baumann, 1992; Price, 1999). In addition to its antide-ressant features, some anticonvulsant and antiepilepticroperties have been reported for citalopram in differ-nt animal models and clinical trials (Kabuto et al., 1994;avale et al., 2003; Clinckers et al., 2004a). In contrast,eizure is an adverse effect of high doses of citalopram whenngested accidentally or intentionally (Fisher et al., 2002;aring et al., 2008). In fact, these biphasic dose-dependentroperties of antidepressants including SSRIs have been theubject of many studies (Loscher, 2009). However, no clearechanism for these paradoxical responses of citalopram on

eizures has been yet established.Regarding the possible pathways involved in SSRIs effects;

ifferent subtypes of serotonin receptors including 5HT1A,HT2 and 5HT3 receptors have received much attention forheir potential contribution to many central responses ofSRIs (Eison and Mullins, 1995; Stahl, 1998; Keltner et al.,002). It is now obvious that many SSRIs including citalopramxert some of their pharmacological effects such as antide-ressant activity via serotonin type 3 receptors (5-HT3) (Fan,994; Redrobe and Bourin, 1997; Choi et al., 2003). Moreoverome important adverse effects of SSRIs including hyper-lycemia (Carvalho et al., 2004), sexual dysfunction (Nelsont al., 1997) and nausea (Limebeer et al., 2009) might beediated through 5-HT3 receptor activation.5-HT3 receptor is the only ligand-gated ion channel

mong seven known classes of serotonin receptors (Derkacht al., 1989). This receptor plays an important role in reg-lating communication between cells in the central anderipheral nervous systems and is the target of many dif-erent therapeutic agents and abused drugs (Derkach et al.,989; Chameau and van Hooft, 2006). Recently, the pos-ible participation of nitric oxide (NO) in many functionalesponses of 5-HT3 receptor activation has been investi-ated. For instance, NO contributes to the pressor effectlicited by 5-HT3 receptor stimulation in the nucleus tractusolitarii (Sevoz-Couche et al., 2002) and neurogenic relax-tions of proximal colon in guinea pig (Riad et al., 1994).n this regard, lately, we have shown that 5-HT3 receptorlays an important role in seizure susceptibility of mice inTZ-induced seizure model and investigate role of NO as

modulator in this phenomenon (Gholipour et al., 2010;ahremand et al., 2011).

The exact mechanisms by which serotonin 5-HT3 recep-or interacts with nitrergic system in neuronal cells is nototally understood, though different possible mechanismsike modulatory effect on cytosolic Ca2+ activity due to entryf extracellular Ca2+ have been proposed (Reiser, 1990).

In the present study we have investigated the dose-ependent effects of citalopram on seizure susceptibility.sing a selective 5-HT3 receptor agonist (mCPBG) and antag-nist (Tropisetron), we also examined the potential role of-HT3 receptors. Finally, involvement of 5-HT1A receptorsas assessed by means of WAY-100635, a selective 5-HT1A

ntagonist. We examined the modulatory effects of NO andow this contributed to the dose-dependent properties ofitalopram.

bft

B. Payandemehr et al.

aterials and methods

nimals

ale NMRI mice weighing 23—30 g (Razi Institute, Karadj,ran) were used in the study. The animals were housed intandard polycarbonate cages in groups of 4—5 and theyere in a temperature-controlled room (22 C) with 12 h

ight/12 h dark cycle. Animals were acclimated at least days before experiments with free access to food andater. The experiments were conducted between 09:00 and5:00. All procedures were carried out in accordance withnstitutional guidelines for animal care and use and pos-ible measures were undertaken to minimize the numberf animals used and also to minimize animals’ discomfortncluding immediate euthanasia after acute experiments.roups consisted of at least eight animals and each ani-al was used only once. Additionally, all efforts were made

o reduce animal suffering and to use only the number ofnimals necessary to produce reliable scientific data.

rugs

he following drugs were used throughout the study:entylenetetrazole (PTZ) (Sigma, UK), L-arginine (L-ARG),G-L-arginine methyl ester (L-NAME), 7-nitroindazole7-NI) (Sigma, St Louis, MO, USA), WAY-100635, m-hlorophenylbiguanidine (mCPBG) and TropisetronSigma—Aldrich, UK). Citalopram was a generous giftrom Bakhtar shimi (Kermanshah, Iran). WAY-100635 admin-stered subcutaneously (s.c) and all other drugs, exceptTZ, were administered intraperitoneally (i.p), injected in

volume of 10 ml/kg of the body weight of the mice. 7-NIas suspended in a 1% aqueous solution of tween 80 andll other drugs were dissolved in normal saline. To provokelonic seizures, PTZ was infused intravenously (0.5%, i.v.)nto the lateral tail vein of mouse (see Section ‘‘Seizurearadigms’’).

The doses were chosen based on previously publishedtudies (Bahremand et al., 2010; Gholipour et al., 2010)nd pilot experiments. In experiments with sequential treat-ents, the intervals between NOS inhibitors or L-ARG and

he agonist or antagonist of 5HT3 were 15 min so theffective blockade of enzymes by inhibitors were allowed.italopram was injected 15 min after administration of 5HT3

gonist or antagonist or 5HT1A antagonist i.e. 30 min beforeerforming the test. As such, tropisetron or WAY-100635 wasiven enough time to blockade the receptors before injectshe citalopram and also mCPBG to exerts its effect. Theesults are suggestive that selected time intervals allowedffective activation or blockade of receptors.

eizure paradigms

he clonic seizure threshold was determined by inserting a0-gauge dental needle into the lateral tail vein of mouse

y a narrow piece of adhesive tape. With mouse movingreely, the PTZ solution (0.5%) was slowly infused into theail vein at a constant rate of 1 ml/ min using an infusion

f citalopram 219

Figure 1 Effect of administration of different doses ofcitalopram (0.1, 0.5, 1, 5, 10, 25 and 50 mg/kg, i.p.) on PTZ-induced seizure threshold in mice. Citalopram was administered30 min before determination of PTZ seizure threshold. Dataare expressed as mean ± S.E.M of seizure threshold in eachgroup. Each group consisted of 9 mice. *P < 0.05, **P < 0.01 and***P < 0.001 compared with saline control group.

sseo

n(ndbii

R

Et

FipPdt

psad

5-HT3 and nitric oxide modulate the anticonvulsant effect o

pump (NE 1000, new era pump system, Inc), which was con-nected to the dental needle by polyethylene tubing. Infusionwas halted when general clonus (forelimb clonus followed byfull clonus of the body) was observed. Minimal dose of PTZ(mg/kg of mice weight) needed to induce general clonus wasrecorded as an index of clonic seizure threshold. As such,seizure threshold is dependent on PTZ dose administeredand time-related.

Statistical analysis

Data of seizure thresholds are expressed as mean and stan-dard error of the mean (S.E.M.) of clonic seizure thresholdsin each experimental group. The one-way analysis of vari-ance (ANOVA) followed by Tukey post hoc for multiplecomparisons was employed. Analyses were confirmed byusing non-parametric Kruskal—Wallis test with Dunn posthoc. A value of p < 0.05 was considered the significance levelbetween the groups.

Experiments

In experiment 1, animals received an intraperitoneal injec-tion of different doses of citalopram (0.1, 0.5, 1, 5, 10,25 and 50 mg/kg) 30 min before the determination of thethreshold of clonic seizures induced by intravenous admin-istration of PTZ solution. Control animals received the samevolume of isotonic saline solution. The doses and time pointwere chosen according to pilot experiments. In experiment2, mCPBG, a 5HT3 agonist, was injected in different doses(1, 5 and 10 mg/kg) 45 min before determination of seizurethreshold and 15 min before the administration of differ-ent doses of citalopram (0.1, 10, 25 mg/kg). Suitable controlgroup received saline with the same time intervals.

In experiment 3, animals received acute intraperitonealadministration of tropisetron, a 5HT3 antagonist, (0.5, 1,2, 5, and 10 mg/kg) or saline 45 min before determinationof PTZ-induced clonic seizure threshold. Chosen doses oftropisetron (1 and 2 mg/kg) was administered before citalo-pram (1, 10 and 25 mg/kg). Suitable control group receivedsaline with the same time intervals.

Experiment 4 examined the effects of the NO precursorL-arginine on different doses of citalopram with or withoutthe pre-treatment with selective doses of tropisetron.

L-Arginine was administered 15 min before thetropisetron or saline and 60 minutes before PTZ. Rele-vant control groups received different combinations ofsaline and/or citalopram and/or tropisetron before thetest.

Experiment 5 investigates the effect of L-NAME (1,5 mg/kg) on different doses of citalopram with or withoutthe pre-treatment with selective doses of mCPBG.

L-NAME administered intraperitoneally 60 min beforeseizure induction, mCPBG and citalopram injected sequen-tially 45 min and 30 min before PTZ induced clonic seizurethreshold. Control animals were used for each part.

Experiment 6 investigates the effect of different doses

of 7-NI (15, 30 mg/kg) on different doses of citalo-pram with or without the pre-treatment with selectivedoses of mCPBG. 7-NI administered intraperitoneally 60 minbefore seizure induction, mCPBG and citalopram injected

msav

equentially 45 min and 30 min before PTZ induced cloniceizure threshold. Relevant control groups received differ-nt combinations of vehicle and/or drug based on each partf experiment.

In experiment 7, animals received acute subcuta-eous administration of WAY-100635, a 5HT1A antagonist,0.3 and 0.6 mg/kg) or saline 45 min before determi-ation of PTZ-induced clonic seizure threshold. Chosenoses of WAY-100635 (0.3 and 0.6 mg/kg) was administeredefore citalopram (1, 10 and 25 mg/kg). Correspond-ng control group received saline at the same timentervals.

esults

ffect of different doses of citalopram on seizurehreshold

ig. 1 illustrates the dose-dependent effect of acutentraperitoneal administration of different doses of citalo-ram (0.1, 0.5, 1, 5, 10, 25, and 50 mg/kg, i.p) onTZ-induced clonic seizure threshold. The seizure thresholdetermination was done 30 min after citalopram administra-ion.

One-way ANOVA revealed a significant effect for citalo-ram (F[7,60] = 20.302, P < 0.001, n = 9) and post hoc analysishowed a significant anticonvulsant effect for citalopramt doses of 0.5 or 1 mg/kg and proconvulsant effect atoses ≥25 mg/kg compared with saline-treated control ani-als. Based on this experiment the doses 0.1 mg/kg was

elected as a sub-effective dose of anticonvulsant activity

nd 10 mg/kg was chosen as a sub-effective dose of procon-ulsant activity.

220 B. Payandemehr et al.

Figure 2 Effects of administration of different doses ofmCPBG, 5HT3 receptor agonist (1, 5, 10, mg/kg, i.p.) on PTZ-induced seizure threshold in mice alone (a) or before theadministration of pro (25 mg/kg) and anticonvulsant (0.1 and10 mg/kg) doses of citalopram (b). mCPBG was administered15 min before administration of citalopram. Data are expressedas mean ± S.E.M of seizure threshold in each group and the num-ber of animals in each group was 9 in this experiments (n = 9).*P < 0.05, **P < 0.01 and ***P < 0.001 compared with saline controlgt

Eae

F5scec(n

nbisHp(c

Eac

F(ib

Figure 3 Effects of different doses of tropisetron, specific5HT3 receptor antagonist (0.5, 1, 2, 5 and 10 mg/kg, i.p.)alone (a) and in combination with pro (25 mg/kg) and anticon-vulsant (1 and 10 mg/kg) doses of citalopram in PTZ-inducedseizure threshold. Tropisetron was administered 15 min beforei.p. injection of citalopram and 45 min before determinationof PTZ-induced seizure threshold. Each group consisted of 9mice (n = 9). Data are expressed as mean ± S.E.M. of seizurethreshold in each group. *P < 0.05 ***P < 0.001 compared withsr

Cwa(

ot(epdes[shp

Ep

Fasi

roup. ##P < 0.01, In comparison with corresponding citalopramreated group.

ffect of acute administration of m-CPBG (5-HT3gonist) on anticonvulsant and proconvulsantffects of citalopram

ig. 2a shows the effect of different doses of mCPBG (1,, 10 mg/kg, i.p.) on the threshold of PTZ-induced cloniceizure. mCPBG was injected 45 min before PTZ-inducedlonic seizure threshold determination. Comparison of theffect of different doses of mCPBG with saline-treatedontrols show an anticonvulsant effect at higher dose10 mg/kg) for this agent alone (F[3,33] = 6.410, P < 0.05,

= 9).As shown in Fig. 2b pre-treatment with low and per se

on-effective doses of mCPBG (1 and 5 mg/kg, i.p.) 15 minefore administration of citalopram (0.1 and 10 mg/kg,.p.) significantly increased the anticonvulsant effect ofub-effective citalopram in a dose dependent manner.owever, it seems that these doses of mCPBG could nototently inhibit the procovulsant effect of citalopram25 mg/kg, i.p.) compared with related citalopram/salineontrol (F[9,73] = 13.740, P < 0.001, n = 9).

ffect of acute administration of tropisetron onnticonvulsant and proconvulsant effects ofitalopram

ig. 3a shows the effect of different doses of tropisetron0.5, 1, 2, 5, 10 mg/kg, i.p.) on the threshold of PTZ-nduced clonic seizure. Tropisetron was injected 45 minefore PTZ-induced clonic seizure threshold determination.

ocip

aline/saline control group. ##P < 0.01, in comparison with cor-esponding citalopram treated group.

omparison of the effect of different doses of tropisetronith saline-treated controls show a proconvulsant alter-tion at the doses of 5 mg/kg or higher for this agent aloneF[5,49] = 9.967, P < 0.001, n = 9).

However, pre-treatment with low and sub-effective dosesf tropisetron (1 and 2 mg/kg, i.p.) 15 min before adminis-ration of an effective anticonvulsant dose of citalopram1 mg/kg, i.p.) significantly inhibited the anticonvulsantffect of citalopram (Fig. 3b) in comparison with citalo-ram 1 mg/kg/saline control group. Also these non-effectiveoses of tropisetron could unmasked a proconvulsantffect of citalopram (10 mg/kg, i.p.) and decreased theeizure threshold significantly and dose dependently (F9,74] = 9.395, P < 0.001, n = 9) (Fig. 3b). Finally, as it ishown in the Fig. 3b, potent dose of tropisetron (2 mg/kg)ardly alter the proconvulsive effect of higher dose of citalo-ram (25 mg/kg).

ffect of L-arginine on anticonvulsant androconvulsant effects of citalopram

ig. 4a shows the effect of different doses of L-ARG (10nd 30 mg/kg, i.p.) on the threshold of PTZ-induced cloniceizure threshold. L-ARG was injected 60 min before PTZ-nduced clonic seizure threshold determination. Comparison

f the effect of different doses of L-ARG with saline-treatedontrols using one-way ANOVA failed to show an alterationn seizure threshold for this agent alone (F [2,9] = 1.367,ost hoc P > 0.05, n = 8). However, as seen in Fig. 4b,

5-HT3 and nitric oxide modulate the anticonvulsant effect of citalopram 221

Figure 4 Low doses of L-ARG, NOS (nitric oxide synthase) pre-cursor (10 and 30 mg/kg) could not alter the seizure thresholdalone (a). Pre-treatment of mice with these low and ineffec-tive doses of L-ARG enhanced the proconvulsant properties anddecreased the seizure threshold of the animals treated withcitalopram (10 and 25 mg/kg) (b). Combinations of L-ARG andtropisetron, 5HT3 receptor antagonist (1 and 2 mg/kg) had nosignificant effect on modulating the seizure threshold of micetreated with higher dose of citalopram (25 mg/kg) but showeda very potent proconvulsant effect on lower dose (10 mg/kg) ofcitalopram (c). L-ARG or saline was administered 15 min beforei.p. injection of tropisetron and 60 min before determination ofPTZ seizure threshold. Each group consisted of 8 mice (n = 8).Data are expressed as mean ± S.E.M. of seizure threshold ineach group. *P < 0.05 **P < 0.01 and ***P < 0.001 compared withcorresponding saline control group; #P < 0.05 and ##P < 0.01 com-

Figure 5 Low doses of L-NAME, NOS (nitric oxide synthase)inhibitor (1 and 5 mg/kg) could not alter the seizure thresholdalone (a). Pre-treatment of mice with these low and ineffectivedoses of L-NAME enhanced the anticonvulsant properties andincreased seizure threshold of the animals treated with citalo-pram (0.1, 10 and 25 mg/kg) (b). Combinations of L-NAME andmCPBG, 5HT3 receptor agonist (1 and 5 mg/kg) had no signifi-cant effect on modulating the seizure threshold of mice treatedwith higher dose of citalopram (25 mg/kg) but enhanced theanticonvulsant effect on lower dose (0.1 and10 mg/kg) of citalo-pram (c). L-NAME or saline was administered 15 min before i.p.injection of mCPBG and 60 min before determination of PTZseizure threshold. Each group consisted of 8 mice (n = 8). Dataare expressed as mean ± S.E.M. of seizure threshold in eachgroup. ***P < 0.001 compared with corresponding saline controlgroup; #P < 0.05 and ##P < 0.01 compared with correspondingcitalopram treated group. % P < 0.05 and %% P < 0.01. In compar-i

riP

idanL

pcP

idla

pared with corresponding citalopram treated group. % P < 0.05and %% P < 0.01 In comparison with corresponding Citalopramand L-ARG treated group.

pre-treatment with these low and per se non-effective dosesof L-ARG (10 and 30 mg/kg, i.p.) 30 min before adminis-tration of citalopram (10 mg/kg, i.p.) unmasked a potentproconvulsant effect and decreased the seizure thresholdsignificantly for proconvulsant dose of citalopram (25 mg/kg,i.p.) respectively and in a dose dependent manner. Compar-ison of the effect of pre-treatment with different doses ofL-ARG in comparison with corresponding citalopram/salinecontrol groups (F(8,54) = 17.163, post hoc P < 0.001, n = 8)shows that L-ARG has made the citalopram treated micemore susceptible to convulsion.

Finally, one-way ANOVA analysis followed by the Tukeypost hoc demonstrated that, these doses of L-ARG in com-bination of potent doses of tropisetron could barely changethe seizure threshold of citalopram in comparison with L-ARG/citalopram (F[14,86] = 13.373, post hoc P < 0.001, n = 8)corresponding groups as shown in Fig. 4c.

Effect of L-NAME on dose-dependent effects ofcitalopram

Fig. 5a & b illustrates the effect of pre-treatment with dif-ferent doses of L-NAME, a nonspecific NOS inhibitor (1 and5 mg/kg, i.p.), alone (a) and in combination with selectivedoses of citalopram (0.1 mg/kg 10 or 25 mg/kg). Analysis

nomp

son with corresponding citalopram and L-NAME treated group.

evealed that L-NAME, at relatively lower doses, did notnfluence seizure threshold alone (F (2, 27) = 0.554, post hoc

> 0.05, n = 8).As shown in Fig. 5b, low doses of L-NAME (1 and 5 mg/kg,

p), when administered 30 min before the sub-effectiveoses of citalopram (0.1 mg/kg and 10), unmasked a potentnticonvulsant effect and increased seizure threshold sig-ificantly and dose dependently, however these doses of-NAME could inhibit the proconvulsant effects of citalo-ram 25 mg/kg significantly in comparison with relateditalopram/saline control group (F(10, 81) = 14.322, post hoc

< 0.001, n = 8).As shown in Fig. 5c, low doses of L-NAME (1 and 5 mg/kg,

p), when administered 30 min before the sub-effectiveose of citalopram (0.1 mg/kg and 10) in combination withower doses of mCPBG (1 and 5 mg/kg) unmasked a potentnticonvulsant effect and increased seizure threshold sig-ificantly and dose dependently, however regarding the

ne-way ANOVA followed by post hoc analysis, it seems thatCPBG could not inhibit the proconvulsant effects of citalo-ram (25 mg/kg) significantly in comparison with related

222 B. Payandemehr et al.

Figure 6 Low doses of 7-NI, NOS (nitric oxide synthase)inhibitor (15 and 30 mg/kg) could not alter the seizure thresholdalone (a). Pre-treatment of mice with these low and inef-fective doses of 7-NI enhanced the anticonvulsant propertiesand increased seizure threshold of the animals treated withcitalopram (0.1 and 25 mg/kg) (b). Combinations of 7-NI andmCPBG, 5HT3 receptor agonist (1 and 5 mg/kg) had no signifi-cant effect on modulating the seizure threshold of mice treatedwith higher dose of citalopram (25 mg/kg) but enhanced theanticonvulsant effect on lower dose (0.1 mg/kg) of citalopram(c). 7-NI or saline was administered 15 min before i.p. injec-tion of mCPBG and 60 min before determination of PTZ seizurethreshold. Each group consisted of 8 mice (n = 8). Data areexpressed as mean ± S.E.M. of seizure threshold in each group.*P < 0.05 ***P < 0.001 compared with corresponding saline con-trol group; #P < 0.05 and ##P < 0.01 compared with correspondingcs

ch

Ec

Atnhswcedcsc

io

Figure 7 Effects of different doses of WAY 100635, specific5HT1A receptor antagonist (0.3, and 0.6 mg/kg, s.c.) alone (a)and in combination with pro (25 mg/kg) and anticonvulsant(1 and 10 mg/kg) doses of citalopram in PTZ-induced seizurethreshold. WAY 100635 had no significant effect on the anticon-vulsant (b) and proconvulsant effects of citalopram (c). WAY100635 was administered 15 min before i.p. injection of citalo-pram and 45 min before determination of PTZ-induced seizurethreshold. Each group consisted of 8 mice (n = 8). Data aree*

osdfisc

Eac

F(cioton

dbo

italopram treated group. % P < 0.05 in comparison with corre-ponding citalopram and 7-NI treated group.

italopram/L-NAME control group (F(14,116) = 21.802, postoc P < 0.001, n = 8). The full data are shown in Table 1.

ffect of 7-NI on dose-dependent effects ofitalopram

s shown in Fig. 6a, different doses of the preferen-ial neuronal NOS inhibitor 7-NI (15 and 30 mg/kg, ip) didot alter seizure threshold alone, (F (2, 9) = 1.457, postoc P > 0.05, n = 8). However, as seen in Fig. 6b, analysishows that these low doses of 7-NI (15 and 30 mg/kg, ip),hen administered 30 min before the sub-effective dose ofitalopram (0.1 mg/kg), unmasked a potent anticonvulsantffect and increased seizure threshold significantly and doseependently, also these doses of 7-NI could inhibit the pro-onvulsant effects of higher dose of citalopram (25 mg/kg)ignificantly in comparison with related citalopram/saline

ontrol group (F(8,59) = 16.178, post hoc P < 0.001, n = 8).

As shown in Fig. 6c, low doses of 7-NI (15 and 30 mg/kg,p), when administered 30 min before the sub-effective dosef citalopram (0.1 mg/kg) in combination with lower doses

cwpW

xpressed as mean ± S.E.M. of seizure threshold in each group.P < 0.05 ***P < 0.001 compared with saline/saline control group.

f mCPBG (1 and 5 mg/kg) unmasked a potent anticonvul-ant effect and increased seizure threshold significantly andose dependently, however regarding the one-way ANOVAollowed by post hoc analysis, it seems that mCPBG could notnhibit the proconvulsant effects of citalopram (25 mg/kg)ignificantly in comparison with related citalopram/7-NIontrol group (F(14,99) = 18.373, post hoc P < 0.001, n = 8).

ffect of acute administration of WAY 100635 onnticonvulsant and proconvulsant effects ofitalopram

ig. 7a illustrates the effect of different doses of WAY 1006350.3, 0.6 mg/kg, s.c.) on the threshold of PTZ-inducedlonic seizure. WAY 100635 was injected 45 min before PTZ-nduced clonic seizure threshold determination. Comparisonf mice receiving WAY 100635 at different doses with saline-reated controls showed a proconvulsant effect at the dosef 0.6 mg/kg for this agent alone (F [2,22] = 4.292, P < 0.05,

= 8).However, pre-treatment with effective and sub-effective

oses of WAY 100635 (0.3 and 0.6 mg/kg, s.c.) 15 minefore administration of an effective anticonvulsant dosef citalopram (1 mg/kg, i.p.) failed to influence the anti-

onvulsant effect of citalopram (Fig. 7b) in comparisonith citalopram 1 mg/kg control group (F [2. 15] = .597,ost hoc P > 0.05, n = 8). In addition, the same doses ofAY 100635 could not significantly alter the threshold seen

f cit

pvaembm

5raj2bS2(tspb2Hdnc2resdaFrp

l1t1oawast2eR

g(nI(Nd1

5-HT3 and nitric oxide modulate the anticonvulsant effect o

with the sub-effective dose of citalopram (10 mg/kg, i.p.)(F [7,46] = 21.524, P < 0.001, n = 8). Finally, as shown in theFig. 7c, WAY 100635 (0.3, 0.6 mg/kg) did not alter the pro-convulsive effect of higher dose of citalopram (25 mg/kg).For more clarity; actual means ± SD for each experiment ispresented in Table 1.

Discussion

The PTZ-induced clonic seizures primarily model a typeof forebrain regulated seizures that are associated withincreased activity in major epileptogenic centers of fore-brain such as amygdala and piriform cortex (Swinyard andKupferberg, 1985). This model is widely used to predictthe efficacy of different drugs and treatments against non-convulsive absence or myoclonic seizures in humans using astandard paradigm (Loscher et al., 1991).

In our study, citalopram at low doses (0.5and 1 mg/kg)decreased and at higher doses (25and 50 mg/kg) increasedthe seizure susceptibility induced by PTZ (Figure 1).The biphasic dose-dependent properties of SSRIs includ-ing citalopram are a well-known phenomenon (Loscher,2009). In one study, the anticonvulsant effects of cen-trally administered lower concentrations of 5-HT againstpilocarpine-induced seizures in rats has been demonstrated,while the higher concentrations worsened seizure outcome(Clinckers et al., 2004b). Growing body of evidence supportsthe importance of serotonergic deficit in epileptogenesis(Bagdy et al., 2007) and protective characteristics of 5-HT in the central nervous system (CNS) against epilepsy(Mainardi et al., 2008). As a result, citalopram like otherSSRIs shows some benefits in the treatment of depressivepatients with epilepsy (Hovorka et al., 2000; Specchio et al.,2004). For instance, citalopram, given to non-depressedpatients with poorly controlled epilepsy, improved the symp-toms and reduced the seizure frequency (Favale et al.,2003). In contrast, ingestion of high doses of citalopram hasbeen associated with seizure occurrence in patients espe-cially when other predisposing factors coexist (Fisher et al.,2002; Waring et al., 2008). In fact, generalized seizureshappen in 2—6% of patients who present to hospital afterdeliberate citalopram overdose (Isbister et al., 2004; Kellyet al., 2004).

The mechanism involved in the dual effect of citalo-pram on seizure susceptibility is not yet fully understood. Ina recent review by Loscher, the apparent dose-dependenteffect of antidepressants on seizure threshold was docu-mented: low but clinically relevant doses of antidepressantsappear to exert an anticonvulsant effect in a variety of ani-mal seizure models via an increase in noradrenaline or 5-HTsynaptic levels, while proconvulsant activity may be seen atsupra-therapeutic doses (Loscher, 2009).

In our study, a 5-HT3 receptor agonist, mCPBG (10 mg/kg)showed anticonvulsant effect, while a 5HT3 receptor antago-nist, tropisetron (5 mg/kg and higher) decreased the seizurethreshold (Figs. 2a and 3a, respectively). These are in linewith previous publishing which report the same results for

other 5-HT3 agonists and antagonists (Grant et al., 1994;Gholipour et al., 2010). As shown in Figs. 2 and 3, tropisetroncompletely inhibited the anticonvulsant effect of citalo-pram against seizures induced by PTZ, whereas m-CPBG

1

Na

alopram 223

otentiated this citalopram effect. Conversely, the procon-ulsant effects of citalopram in these models barely changedfter pre-treatment with m-CPBG and tropisetron. Given thenhancement of the anticonvulsant effect of citalopram by-CPBG and its full blockade by tropisetron, it is plausi-le that the mentioned action of citalopram at low doses isediated through the 5-HT3 receptor.The role of various serotonin receptor subtypes especially

-HT1 and 5-HT3 receptors is documented in some centralesponses of SSRIs. The involvement of 5-HT1A subtypes incute and chronic effects of citalopram has been the sub-ect of many studies (Arborelius et al., 1995; Papp et al.,002; Pruus et al., 2002). Activation of 5-HT1A receptors haseen reported to suppress (Pericic and Svob Strac, 2007;algado-Commissariat and Alkadhi, 1997; Pericic et al.,005), potentiate (Jakus et al., 2003) or not to affectLöscher and Czuczwar, 1985) seizures. It has been shownhat stimulation of 5-HT1A receptor can exert anticonvul-ant effects (Clinckers et al., 2004a,b) and citalopramrotection against pilocarpine-induced motor seizure haseen attributed to this receptor subtype (Clinckers et al.,004a). However, in the present study, using selective 5-T1A antagonist, WAY 100635, we demonstrated that doseependent effects of citalopram on seizure threshold mightot be mediated through 5-HT1A receptors, a finding thatonforms to some previous study (Pericic and Svob Strac,007). These conflicting results might be due to disparity inoutes of administration or difreneces in the seizure modelmployed in various studies. It is noteworthy that eacheizure paradigm possesses unique characteristics and thatifferent drugs have differential effects on various settingsnd on various seizure components (Loscher et al., 1991).or example, even the different animal species have beeneported to produce different results using the same seizurearadigms (Kirkby et al., 1996).

5-HT3 receptor has been implicated in antidepressant-ike effect of citalopram in animals (Redrobe and Bourin,997). Using patch clamp technique, the role of 5-HT3 recep-or in therapeutic effects of SSRIs is now more clear (Fan,994; Choi et al., 2003). Our results, in accordance withther studies, imply that 5-HT3 receptor plays an indispens-ble role in alteration of seizure susceptibility. Recently,e have shown that the enhancement of 5-HT3 receptorctivity resulted in anticonvulsant effect in the PTZ-inducedeizure model and selective antagonism at the 5-HT3 recep-or yielded proconvulsive effects in mice (Gholipour et al.,010). Of note, seizure has been reported as an adverseffect of ondansetron (Sargent et al., 1993; Sharma andaina, 2001).

Nitric oxide (NO) is regarded as both neuronal messen-er and/or neurotransmitter in the central nervous systemBredt et al., 1990). NO is generated from L-arginine by theitric oxide synthase (NOS) family (Snyder and Bredt, 1991).n the CNS, NO is mostly produced by the neuronal NOSnNOS), a Ca-dependent enzyme (Snyder and Bredt, 1991).O is also a known modulator of seizure susceptibility withiverse anticonvulsant (Starr and Starr, 1993; Tsuda et al.,997) or proconvulsant (De Sarro et al., 1991; Osonoe et al.,

994) properties in different seizure settings.

Our results demonstrate that L-NAME (a non-specificOS inhibitor) and 7-NI (preferentially nNOS inhibitor)t independently non-effective doses augmented the

2 B. Payandemehr et al.

aL

heoNeb2seNts2moiaptdooasidaTso

e(cilBigGmcoaplspntteaaaCefisp

Figure 8 Citalopram inhibit the reuptake of serotonin insynapses so it could be effect on 5HT3 receptors, this Fig pro-posed mechanism for the interaction of 5-HT3 and NO system inmodulation of seizure threshold in GABAergic interneurons. 5-HN

‘aNpdpo(2dcacu

eimpwapvmsauasroaaaoC

24

nticonvulsive property of citalopram at lower doses while,-ARG enhanced the proconvulsive effect of citalopram atigher doses. Involvement of NO pathways in the centralffects of citalopram and other SSRIs has been the focusf many studies (Kumar et al., 2010; Singh et al., 2003).OS inhibitors significantly potentiated the protectiveffect of citalopram against immobilization stress-inducedehavioural and biochemical alterations (Kumar et al.,010). L-NAME augments the effect of various antidepres-ants including citalopram in different paradigms (Harkint al., 2004; Ulak et al., 2008). It is known that SSRIs diminishO levels in the CNS which has been proposed as a feature ofheir antidepressant activity, suggesting a crosstalk betweenerotonergic and NO systems (Luo and Tan, 2001; Crespi,010). Interestingly in our study, the NOS inhibitors andCPBG show additive effect to elevate the seizure thresh-

ld in mice (Fig. 5c and 6c). On the other hand, tropisetronn combination with L-ARG revealed a strong proconvulsivection in citalopram-treated mice. These data implies thearticipation of both 5-HT3 and NO pathways on the cen-ral effects of citalopram especially in lower anticonvulsiveoses. However, this was not the case for the higher dosesf citalopram. The combination of 5-HT3 agonist and antag-nist and NO pathway modulators proved to be as potents NO modulators alone when used with higher proconvul-ive doses of citalopram. Thus, it appears that 5-HT3 is notnvolved in the neurotoxic effect of citalopram in higheroses and this effect is more likely mediated by NO pathwaylone rendering the animals more susceptible to seizure.hese data simply indicate the difference in the character oferotonergic and NO pathways to mediate the central effectf citalopram in present study.

Recently, NO contribution to some of peripheral (Riadt al., 1994; Sevoz-Couche et al., 2002) and central effectsGholipour et al., 2010) of 5-HT3 receptor has been dis-overed. It is shown that the activation of 5-HT3 receptorn hippocampus and cortex hyperpolarizes the interneuronseading to an increase in seizure threshold (Morales andloom, 1997; Chameau and van Hooft, 2006). Some studies

ndicate that 5-HT3 stimulation directly activates GABAer-ic interneurons mediating this inhibitory effect (Ropert anduy, 1991; Morales and Bloom, 1997). A proposed mechanismay be activation of the 5-HT3 receptor as an inward Ca2+

onducting ion channel, resulting in early depolarizationf inhibitory interneurons (Derkach et al., 1989; Chameaund van Hooft, 2006). This may explain the anticonvulsiveroperties reported for citalopram as well in our study. Simi-arly, it is conceivable that blocking the baseline endogenoustimulation of 5-HT3 receptors by tropisetron may induceroconvulsive properties. In this regard, the 5-HT3 ago-ist mchlorophenylbiguanidine (mCPBG) has been showno facilitate GABA release in a synaptic button prepara-ion of hippocampal CA1 pyramidal neurons (Katsurabayashit al., 2003). As proposed by some studies, cytosolic Ca2+

nd NO are possible mechanisms by which the serotonergicnd NO system interconnect (Reiser, 1990). 5-HT3 receptorctivation could directly affect NOS by altering cytosolica2+ levels. Briefly, 5-HT3 activation may have two opposite

ffects on the inhibitory interneuron: one toward increasedring and subsequent GABA release, which is anticonvul-ive, and the other, towards increased NOS activity androconvulsive consequences. Hence, one could expect a

itff

T3 R, 5-HT3 receptor; NMDA R, NMDA receptor; nNOS, neuronalO synthase; ↑, increase.

‘masked’’ anticonvulsant effect from pure 5- HT3 agonists,t least at lower doses. Concomitant NO depletion by L-AME could ‘‘unmask’’ and potentiate this anticonvulsiveathway (Fig. 8). This hypothesis may explain the dose-ependent and paradoxical effects of citalopram in seizurearadigm. Given the reported dose dependent effects forther SSRIs [and selective noradrenalin reuptake inhibitorsSNRIs)] in different paradigms (Oke et al., 2001; Stahl et al.,005; Borowicz et al., 2010), they may also exhibit similarual effects in PTZ-induced seizure. However, as each singleompound can show an individual behaviour, more studiesre needed to delineate whether this observed property ofitalopram in PTZ-evoked seizure is a class effect or else anique trait not shared by other congeners.

In summary, we demonstrated that citalopram couldxert both anticonvulsant and proconvulsant effect depend-ng on doses administered in the PTZ-induced clonic seizureodel; then verified that the anticonvulsant effect of citalo-ram is mediated through the 5-HT3 receptor. In addition,e investigated the modulatory role of nitric oxide in medi-ting the central effect of citalopram especially at higherroconvulsive doses. Finally, we showed that the procon-ulsive properties of citalopram at higher doses are notediated through 5-HT3 receptors. These findings first may

hed more light on the mechanisms of action of citalopramgainst seizures at lower doses which are not completelynderstood yet; second, might offer the 5HT3 receptor as

good candidate to improve the effects or handle theide effects of citalopram in patients by its modulatoryole. Moreover, our results warrants further investigationf the advantages and disadvantages of citalopram alonend in combination with NOS inhibitors as a potential ther-py in depressive patients with epilepsy. Nonetheless, thentidepressant properties of SSRIs generally have a delayednset, requiring at least 1—2 weeks of therapy to appear.onsequently, due to the adaptive mechanisms occurring

n the biologic system following chronic treatment with

hese agents, they may demonstrate a different behaviourrom that predicted by their acute administration. There-ore, investigating the effects of long-term treatment with

5-HT3 and nitric oxide modulate the anticonvulsant effect of citalopram 225

Table 1 The actual mean of each experimental group shown in this table, S.D, standard deviation; N, number of animals.

Group Treatment (mg/kg) Mean ± S.D N

01 Saline-control 42.55 ± 0.90 902 Citalopram 0.1 45.77 ± 2.22 903 Citalopram 0.5 52.31 ± 3.61 904 Citalopram 1 55.51 ± 2.20 905 Citalopram 5 45.68 ± 4.20 906 Citalopram 10 40.55 ± 1.95 907 Citalopram 25 35.44 ± 1.00 908 Citalopram 50 31.83 ± 1.72 909 mCPBG 1 46.44 ± 3.45 910 mCPBG 5 47.85 ± 2.24 911 mCPBG 10 50.75 ± 2.20 912 mCPBG 1 + Citalopram0.1 44.31 ± 2.48 913 mCPBG 5 + Citalopram0.1 54.16 ± 1.60 914 mCPBG 5 + Citalopram10 46.49 ± 2.66 915 mCPBG 5 + Citalopram25 40.18 ± 1.98 916 Tropisetron 0.5 41.51 ± 1.98 917 Tropisetron 1 40.81 ± 3.20 918 Tropisetron 2 37.44 ± 2.13 919 Tropisetron 5 35.35 ± 1.78 920 Tropisetron 10 32.08 ± 1.99 921 Tropisetron 2 + Citalopram1 45.50 ± 2.49 922 Tropisetron 2 + Citalopram10 35.09 ± 2.37 923 Tropisetron 1 + Citalopram25 34.96 ± 2.07 924 Tropisetron 2 + Citalopram25 33.95 ± 2.33 925 L-arg 10 44.18 ± 1.92 826 L-arg 30 42.31 ± 2.88 827 L-arg 10 + Citalopram10 37.56 ± 1.78 828 L-arg 30 + Citalopram10 36.16 ± 2.30 829 L-arg 10 + Citalopram25 35.30 ± 2.50 830 L-arg 30 + Citalopram25 33.12 ± 2.05 831 L-arg 10 + Tropisetron1 + Citalopram10 34.63 ± 2.80 832 L-arg 10 + Tropisetron2 + Citalopram10 33.56 ± 2.18 833 L-arg 30 + Tropisetron2 + Citalopram10 32.32 ± 2.54 834 L-arg 10 + Tropisetron1 + Citalopram25 35.70 ± 2.47 835 L-arg 10 + Tropisetron2 + Citalopram25 35.01 ± 2.36 836 L-arg 30 + Tropisetron2 + Citalopram25 33.23 ± 2.81 837 L-NAME 1 44.71 ± 3.32 838 L-NAME 5 42.80 ± 2.83 839 L-NAME 1 + Citalopram0.1 46.63 ± 2.36 840 L-NAME 5 + Citalopram0.1 50.21 ± 2.30 841 L-NAME 5 + Citalopram10 46.94 ± 2.98 842 L-NAME 1 + Citalopram25 40.07 ± 3.21 843 L-NAME 5 + Citalopram25 41.46 ± 4.05 844 L-NAME 1 + mCPBG1 + Citalopram0.1 52.93 ± 2.80 845 L-NAME 5 + mCPBG1 + Citalopram0.1 54.28 ± 2.19 846 L-NAME 1 + mCPBG5 + Citalopram10 57.04 ± 2.55 847 L-NAME 1 + mCPBG1 + Citalopram25 40.30 ± 2.48 848 L-NAME 5 + mCPBG1 + Citalopram25 42.27 ± 2.37 849 L-NAME 1 + mCPBG5 + Citalopram25 42.01 ± 2.80 850 7-NI 15 44.18 ± 1.92 851 7-NI 30 42.31 ± 2.90 852 7-NI 15 + Citalopram0.1 49.56 ± 1.78 853 7-NI 30 + Citalopram0.1 54.16 ± 2.71 854 7-NI 15 + Citalopram25 42.30 ± 3.07 855 7-NI 30 + Citalopram25 45.12 ± 3.35 856 7-NI 15 + mCPBG1 + Citalopram0.1 51.63 ± 2.90 857 7-NI 30 + mCPBG1 + Citalopram0.1 55.56 ± 2.21 858 7-NI 15 + mCPBG5 + Citalopram0.1 57.32 ± 3.14 8

226 B. Payandemehr et al.

Table 1 (Continued)

Group Treatment (mg/kg) Mean ± S.D N

59 7-NI 15 + mCPBG1 + Citalopram25 36.70 ± 2.35 860 7-NI 30 + mCPBG1 + Citalopram25 41.01 ± 2.63 861 7-NI 15 + mCPBG5 + Citalopram25 45.23 ± 3.17 862 WAY100635 0.3 41.06 ± 2.66 863 WAY100635 0.6 37.40 ± 1.53 864 WAY100635 0.3 + Citalopram1 54.71 ± 0.98 865 WAY100635 0.6 + Citalopram1 52.78 ± 2.25 866 WAY100635 0.3 + Citalopram10 42.52 ± 1.20 867 WAY100635 0.3 + Citalopram25 37.31 ± 3.01 8

cai

R

A

B

B

B

B

B

B

C

C

C

C

C

C

D

D

E

F

F

F

G

G

H

H

I

68 WAY100635 0.6 + Citalopram25

italopram on experimental seizure is of relevance beforeny conclusion can be drawn in terms of its possible benefitn concurrent depression and epilepsy in human.

eferences

rborelius, L., Nomikos, G.G., Grillner, P., Hertel, P., Höök, B.B.,Hacksell, U., Svensson, T.H., 1995. 5-HT 1A receptor antagonistsincrease the activity of serotonergic cells in the dorsal raphenucleus in rats treated acutely or chronically with citalopram.Naunyn Schmiedebergs. Arch. Pharmacol. 352, 157—165.

agdy, G., Kecskemeti, V., Riba, P., Jakus, R., 2007. Serotonin andepilepsy. J. Neurochem. 100, 857—873.

ahremand, A., Nasrabady, S.E., Ziai, P., Rahimian, R., Hedayat,T., Payandemehr, B., Dehpour, A.R., 2010. Involvement of nitricoxide-cGMP pathway in the anticonvulsant effects of lithiumchloride on PTZ-induced seizure in mice. Epilepsy Res. 89,295—302.

ahremand, A., Payandemehr, B., Rahimian, R., Ziai, P., Pour-mand, N., Loloee, S., Ebrahimi, A., Ghasemi, A., Fakhfouri,G., Ghasemi, M., 2011. The role of 5-HT3 receptors in theadditive anticonvulsant effects of citalopram and morphine onpentylenetetrazole-induced clonic seizures in mice. EpilepsyBehav. 21, 122—127.

aumann, P., 1992. Clinical pharmacokinetics of citalopram andother selective serotonergic reuptake inhibitors (SSRI). Int. Clin.Psychopharmacol. 6 (Suppl. 5), 13—20.

orowicz, K.K., Furmanek-Karwowska, K., Morawska, M., Luszczki,J.J., Czuczwar, S.J., 2010. Effect of acute and chronic treat-ment with milnacipran potentiates the anticonvulsant activityof conventional antiepileptic drugs in the maximal electroshock-induced seizures in mice. Psychopharmacology (Berl) 207,661—669.

redt, D.S., Hwang, P.M., Snyder, S.H., 1990. Localization of nitricoxide synthase indicating a neural role for nitric oxide. Nature347, 768—770.

arvalho, F., Barros, D., Silva, J., Rezende, E., Soares, M., Fre-goneze, J., De Castro e Silva, E., 2004. Hyperglycemia inducedby acute central fluoxetine administration: role of the centralCRH system and 5-HT3 receptors. Neuropeptides 38, 98—105.

hameau, P., van Hooft, J.A., 2006. Serotonin 5-HT(3) receptors inthe central nervous system. Cell Tissue Res. 326, 573—581.

hoi, J.S., Choi, B.H., Ahn, H.S., Kim, M.J., Rhie, D.J., Yoon,S.H., Min, D.S., Jo, Y.H., Kim, M.S., Sung, K.W., Hahn, S.J.,2003. Mechanism of block by fluoxetine of 5-hydroxytryptamine3

(5-HT3)-mediated currents in NCB-20 neuroblastoma cells.Biochem. Pharmacol. 66, 2125—2132.

linckers, R., Smolders, I., Meurs, A., Ebinger, G., Michotte, Y.,2004a. Anticonvulsant action of GBR-12909 and citalopram

J

39.65 ± 2.38 8

against acute experimentally induced limbic seizures. Neu-ropharmacology 47, 1053—1061.

linckers, R., Smolders, I., Meurs, A., Ebinger, G., Michotte, Y.,2004b. Anticonvulsant action of hippocampal dopamine andserotonin is independently mediated by D2 and 5-HT1A recep-tors. J. Neurochem. 89, 834—843.

respi, F., 2010. The selective serotonin reuptake inhibitor fluoxe-tine reduces striatal in vivo levels of voltammetric nitric oxide(NO): a feature of its antidepressant activity? Neurosci. Lett.470, 95—99.

e Sarro, G.B., Donato Di Paola, E., De Sarro, A., Vidal, M.J., 1991.Role of nitric oxide in the genesis of excitatory amino acid-induced seizures from the deep prepiriform cortex. Fundam.Clin. Pharmacol. 5, 503—511.

erkach, V., Surprenant, A., North, R.A., 1989. 5-HT3 receptors aremembrane ion channels. Nature 339, 706—709.

ison, A.S., Mullins, U.L., 1995. Regulation of central 5-HT2A recep-tors: a review of in vivo studies. Behav. Brain Res. 73, 177—181.

an, P., 1994. Inhibition of a 5-HT3 receptor-mediated current bythe selective serotonin uptake inhibitor, fluoxetine. Neurosci.Lett. 173, 210—212.

avale, E., Audenino, D., Cocito, L., Albano, C., 2003. The anti-convulsant effect of citalopram as an indirect evidence ofserotonergic impairment in human epileptogenesis. Seizure 12,316—318.

isher, A., Davis, M., Croft-Baker, J., Purcell, P., McLean, A.,2002. Citalopram-induced severe hyponatraemia with coma andseizure. Case report with literature and spontaneous reportsreview. Adverse Drug React. Toxicol. Rev. 21, 179—187.

holipour, T., Ghasemi, M., Riazi, K., Ghaffarpour, M., Deh-pour, A.R., 2010. Seizure susceptibility alteration through5-HT3 receptor: modulation by nitric oxide. Seizure 19,17—22.

rant, K.A., Hellevuo, K., Tabakoff, B., 1994. The 5-HT3 antagonistMDL-72222 exacerbates ethanol withdrawal seizures in mice.Alcohol. Clin. Exp. Res. 18, 410—414.

arkin, A., Connor, T.J., Burns, M.P., Kelly, J.P., 2004. Nitric oxidesynthase inhibitors augment the effects of serotonin re-uptakeinhibitors in the forced swimming test. Eur. Neuropsychophar-macol. 14, 274—281.

ovorka, J., Herman, E., Nemcova, I.I., 2000. Treatment of interic-tal depression with citalopram in patients with epilepsy. EpilepsyBehav. 1, 444—447.

sbister, G.K., Bowe, S.J., Dawson, A., Whyte, I.M., 2004. Rela-tive toxicity of selective serotonin reuptake inhibitors (SSRIs)in overdose. J. Toxicol. Clin. Toxicol. 42, 277—285.

akus, R., Graf, M., Juhasz, G., Gerber, K., Levay, G., Halasz, P.,Bagdy, G., 2003. 5-HT2C receptors inhibit and 5-HT1A receptorsactivate the generation of spike-wave discharges in a genetic ratmodel of absence epilepsy. Exp. Neurol. 184, 964—972.

f cit

P

P

P

R

R

R

R

S

S

S

S

S

S

S

S

S

S

S

T

U

macol. Biochem. Behav. 90, 563—568.

5-HT3 and nitric oxide modulate the anticonvulsant effect o

Kabuto, H., Yokoi, I., Endo, A., Takei, M., Kurimoto, T., Mori, A.,1994. Chronic administration of citalopram inhibited El mouseconvulsions and decreased monoamine oxidase-A activity. ActaMed. Okayama 48, 311—316.

Katsurabayashi, S., Kubota, H., Tokutomi, N., Akaike, N., 2003.A distinct distribution of functional presynaptic 5-HT recep-tor subtypes on GABAergic nerve terminals projecting to singlehippocampal CA1 pyramidal neurons. Neuropharmacology 44,1022—1030.

Kelly, C.A., Dhaun, N., Laing, W.J., Strachan, F.E., Good, A.M., Bate-man, D.N., 2004. Comparative toxicity of citalopram and thenewer antidepressants after overdose. J. Toxicol. Clin. Toxicol.42, 67—71.

Keltner, N.L., McAfee, K.M., Taylor, C.L., 2002. Mechanisms andtreatments of SSRI-induced sexual dysfunction. Perspect. Psy-chiatr. Care 38, 111—116.

Kirkby, R.D., Carroll, D.M., Grossman, A.B., Subramaniam, S., 1996.Factors determining proconvulsant and anticonvulsant effects ofinhibitors of nitric oxide synthase in rodents. Epilepsy Res. 24,91—100.

Kumar, A., Garg, R., Gaur, V., Kumar, P., 2010. Possible role of NOmodulators in protective effect of trazodone and citalopram(antidepressants) in acute immobilization stress in mice. IndianJ. Exp. Biol. 48, 1131—1135.

Limebeer, C.L., Litt, D.E., Parker, L.A., 2009. Effect of 5-HT3antagonists and a 5-HT(1A) agonist on fluoxetine-induced condi-tioned gaping reactions in rats. Psychopharmacology (Berl) 203,763—770.

Loscher, W., 2002. Animal models of epilepsy for the development ofantiepileptogenic and disease-modifying drugs. A comparison ofthe pharmacology of kindling and post-status epilepticus modelsof temporal lobe epilepsy. Epilepsy Res. 50, 105—123.

Loscher, W., 2009. Preclinical assessment of proconvulsant drugactivity and its relevance for predicting adverse events inhumans. Eur. J. Pharmacol. 610, 1—11.

Loscher, W., Honack, D., Fassbender, C.P., Nolting, B., 1991. Therole of technical, biological and pharmacological factors in thelaboratory evaluation of anticonvulsant drugs III. Pentylenete-trazole seizure models. Epilepsy Res. 8, 171—189.

Löscher, W., Czuczwar, S.J., 1985. Evaluation of the5-hydroxytryptamine receptor agonist 8-hydroxy-2-(di-n-propylamino) tetralin in different rodent models of epilepsy.Neurosci. Lett. 60, 201—206.

Luo, L., Tan, R.X., 2001. Fluoxetine inhibits dendrite atrophy of hip-pocampal neurons by decreasing nitric oxide synthase expressionin rat depression model. Acta Pharmacol. Sin. 22, 865—870.

Mainardi, P., Leonardi, A., Albano, C., 2008. Potentiation of brainserotonin activity may inhibit seizures, especially in drug-resistant epilepsy. Med. Hypotheses 70, 876—879.

Morales, M., Bloom, F.E., 1997. The 5-HT3 receptor is present indifferent subpopulations of GABAergic neurons in the rat telen-cephalon. J. Neurosci. 17, 3157—3167.

Nelson, E.B., Keck Jr., P.E., McElroy, S.L., 1997. Resolution offluoxetine-induced sexual dysfunction with the 5-HT3 antagonistgranisetron. J. Clin. Psychiatry 58, 496—497.

Oke, A., Adhiyaman, V., Aziz, K., Ross, A., 2001. Dose-dependentseizure activity associated with fluoxetine therapy. QJM 94, 113.

Osonoe, K., Mori, N., Suzuki, K., Osonoe, M., 1994. Antiepilep-tic effects of inhibitors of nitric oxide synthase examinedin pentylenetetrazol-induced seizures in rats. Brain Res. 663,338—340.

Papp, M., Nalepa, I., Antkiewicz-Michaluk, L., Sánchez, C., 2002.Behavioural and biochemical studies of citalopram and WAY100635 in rat chronic mild stress model. Pharmacol. Biochem.

Behav. 72, 465—474.

Pericic, D., Lazic, J., Jazvinscak Jembrek, M., Svob Strac, D., 2005.Stimulation of 5-HT1A receptors increases the seizure thresholdfor picrotoxin in mice. Eur. J. Pharmacol. 527, 105—110.

W

alopram 227

ericic, D., Svob Strac, D., 2007. The role of 5-HT7 receptors in thecontrol of seizures. Brain Res. 1141, 48—55.

rice, P.L., 1999. Citalopram: another SSRI antidepressant. S. D. J.Med. 52, 99—100.

ruus, K., Vaarmann, A., Rudissaar, R., Allikmets, L., Matto, V.,2002. Role of 5-HT1A receptors in the mediation of acutecitalopram effects: a 8-OH-DPAT challenge study. Prog. Neu-ropsychopharmacol. Biol. Psychiatry 26, 227—232.

edrobe, J.P., Bourin, M., 1997. Partial role of 5-HT2 and 5-HT3receptors in the activity of antidepressants in the mouse forcedswimming test. Eur. J. Pharmacol. 325, 129—135.

eiser, G., 1990. Mechanism of stimulation of cyclic-GMP level ina neuronal cell line mediated by serotonin (5-HT3) receptors.Involvement of nitric oxide, arachidonic-acid metabolism andcytosolic Ca2+. Eur. J. Biochem. 189, 547—552.

iad, M., Hamon, M., Emerit, M.B., 1994. Pharmacological evidencefor the involvement of calcium/calmodulin in serotonin 5-HT3receptor-mediated cation permeability in NG 108-15 cells. J.Neurochem. 62, 2224—2232.

opert, N., Guy, N., 1991. Serotonin facilitates GABAergic trans-mission in the CA1 region of rat hippocampus in vitro. J. Physiol.441, 121—136.

algado-Commissariat, D., Alkadhi, K.A., 1997. Serotonin inhibitsepileptiform discharge by activation of 5-HT1A receptors in CA1pyramidal neurons. Neuropharmacology 36, 1705—1712.

argent, A.I., Deppe, S.A., Chan, F.A., 1993. Seizure associated withondansetron. Clin. Pharm. 12, 613—615.

evoz-Couche, C., Maisonneuve, B., Hamon, M., Laguzzi, R., 2002.Glutamate and NO mediation of the pressor response to 5-HT3receptor stimulation in the nucleus tractus solitarii. Neuroreport13, 837—841.

harma, A., Raina, V., 2001. Generalised seizures followingondansetron. Ann. Oncol. 12, 131—132.

ingh, V.P., Jain, N.K., Kulkarni, S.K., 2003. Fluoxetine sup-presses morphine tolerance and dependence: modulation ofNO-cGMP/DA/serotoninergic pathways. Methods Find Exp. Clin.Pharmacol. 25, 273—280.

nyder, S.H., Bredt, D.S., 1991. Nitric oxide as a neuronal messen-ger. Trends Pharmacol. Sci. 12, 125—128.

pecchio, L.M., Iudice, A., Specchio, N., La Neve, A., Spinelli, A.,Galli, R., Rocchi, R., Ulivelli, M., de Tommaso, M., Pizzanelli,C., Murri, L., 2004. Citalopram as treatment of depression inpatients with epilepsy. Clin. Neuropharmacol. 27, 133—136.

tahl, S.M., 1998. Mechanism of action of serotonin selectivereuptake inhibitors: serotonin receptors and pathways medi-ate therapeutic effects and side effects. J. Affect. Disord. 51,215—235.

tahl, S.M., Grady, M.M., Moret, C., Briley, M., 2005. SNRIs: theirpharmacology, clinical efficacy, and tolerability in comparisonwith other classes of antidepressants. CNS Spectr. 10, 732—747.

tarr, M.S., Starr, B.S., 1993. Paradoxical facilitation of pilocarpine-induced seizures in the mouse by MK-801 and the nitric oxidesynthesis inhibitor L-NAME. Pharmacol. Biochem. Behav. 45,321—325.

winyard, E.A., Kupferberg, H.J., 1985. Antiepileptic drugs: detec-tion, quantification, and evaluation. Fed Proc. 44, 2629—2633.

suda, M., Suzuki, T., Misawa, M., 1997. Aggravation of DMCM-induced seizure by nitric oxide synthase inhibitors in mice. LifeSci. 60, PL339—PL343.

lak, G., Mutlu, O., Akar, F.Y., Komsuoglu, F.I., Tanyeri, P., Erden,B.F., 2008. Neuronal NOS inhibitor 1-(2-trifluoromethylphenyl)-imidazole augment the effects of antidepressants acting viaserotonergic system in the forced swimming test in rats. Phar-

aring, W.S., Gray, J.A., Graham, A., 2008. Predictive factors forgeneralized seizures after deliberate citalopram overdose. Br.J. Clin. Pharmacol. 66, 861—865.