Protective effects of lithium chloride on seizure susceptibility: Involvement of α2-Adrenoceptor

Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

Contents lists available at ScienceDirect

Pharmacology, Biochemistry and Behavior

j ourna l homepage: www.e lsev ie r .com/ locate /pharmbiochembeh

Protective effects of lithium chloride on seizure susceptibility:Involvement of α2-adrenoceptor

Borna Payandemehr a, Arash Bahremand b, Ali Ebrahimi a, Sara Ebrahimi Nasrabady c, Reza Rahimian a,Taraneh Bahremand a, Mohammad Sharifzadeh d, Ahmad Reza Dehpour a,e,⁎a Department of Pharmacology, School of Medicine, Tehran University of Medical Sciences, Tehran, Iranb Institut universitaire en santé mentale de Québec, Québec City, Québec, Canadac Motor Neuron Center, College of Physicians and Surgeons, Columbia University Medical Center, NY, USAd Department of Pharmacology and Toxicology, Pharmaceutical Sciences Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Irane Experimental Medicine Research Center, Tehran University of Medical Sciences, Tehran, Iran" to the Ahmad reza Dehpour

⁎ Corresponding author at: Department of PharmacoloUniversity of Medical Sciences, Tehran, P.O. Box: 131453652; fax: +98 21 6640 2569.

E-mail address: [email protected] (A.R. Dehpour).

http://dx.doi.org/10.1016/j.pbb.2015.03.0160091-3057/© 2015 Elsevier Inc. All rights reserved.

a b s t r a c t

Article history:Received 2 November 2014Received in revised form 19 March 2015Accepted 21 March 2015Available online 27 March 2015

Keywords:Lithium chlorideEpilepsyα2-AdrenoceptorClonidineClonic seizure thresholdMice

For more than 60 years, lithium has been themainstay in the treatment of mental disorders as a mood stabilizer.In addition to the antimanic and antidepressant responses, lithium also shows some anticonvulsant properties. Inspite of the ascertained neuroprotective effects of this alkali metal, the underlying mechanisms through whichlithium regulates behavior are still poorly understood. Among different targets, some authors suggestneuromodulatory effects of lithium are the consequences of interaction of this agent with the brain neurotrans-mitters including adrenergic system. In order to study the involvement of α2-adrenergic system in anticonvul-sant effect of lithium, we used a model of clonic seizure induced by pentylenetetrazole (PTZ) in male NMRImice. Injection of a single effective dose of lithium chloride (30 mg/kg, i.p.) significantly increased the seizurethreshold (p b 0.01). The anticonvulsant effect of an effective dose of lithium was prevented by pre-treatmentwith low and per se non-effective dose of clonidine [α2-adrenoceptor agonist] (0.05, 0.1 and 0.25 mg/kg). Onthe other hand, yohimbine [α2-adrenoceptor antagonist] augmented the anticonvulsant effect of sub-effectivedose of lithium (10 mg/kg i.p.) at relatively low doses (0.1, 0.5, 1 and 2.5 mg/kg). Moreover, UK14304 [a potentand selective α2-adrenoceptor agonist] (0.05 and 0.1 mg/kg) and RX821008 [a potent and selective α2D-adrenoceptor antagonist] (0.05, 0.1 and 0.25mg/kg) repeated the same results confirming that thesemodulatoryeffects are conducted specifically through the α2D-adrenoceptors. In summary, our findings demonstrated thatα2-adrenoceptor pathway could be involved in the anticonvulsant properties of lithium chloride in the modelof chemically induced clonic seizure.

© 2015 Elsevier Inc. All rights reserved.

1. Introduction

More than 60 years ago lithium was approved as a suggestedtreatment for mood disorders (Cade, 1949; Marmol, 2008). Since thenlithiumhas been the drug of choice for the treatment of different CNS dis-orders due to its antidepressant and anti-manic properties (Chiu andChuang, 2010; Price and Heninger, 1994). In addition, lithium alsoshows neuroprotective character in different in vivo and in vitrosettings (Cimarosti et al., 2001; Manji et al., 1999) including antiepilepticproperties (Ghasemi and Dehpour, 2011; Ghasemi et al., 2010; Minabeet al., 1988). Regarding to these protective effects, evaluation of lithiumadministration on different brain disorders like seizure, have been thesubject ofmany studies (Bahremand et al., 2010a,b; Ghasemi et al., 2010).

gy, School of Medicine, Tehran-784, Iran. Tel.: +98 21 8897

Fascinatingly, despite its widespread use as an efficacious drug in af-fective disorders and these diverse behavioral responses, the exactmechanisms of actions of lithium are not yet known. Lithium can be ef-fective in brain disturbances by regulation of variousmolecular and bio-chemical mechanisms such as gene expression, hormonal and circadianadjustment, signal transduction, ion transport and also contributing as aneurotransmitter⁄receptor-mediator (Machado‐Vieira et al., 2009). In-terfering with intracellular G-protein-coupled receptors pathways orglycogen synthase kinase-3 (Bahremand et al., 2010a; Chi-Tso andChuang, 2011) are some of the mechanisms described as the biologicbasis for the clinical efficacy of lithium. Beyond this, lithium interactswith many neurotransmitters in the body and it is believed that theseinteractions are responsible for the various effects of lithium (Chiuand Chuang, 2010; Malhi et al., 2013). We previously reported the con-tribution of nitric oxide/cGMP system in the anticonvulsant propertiesof lithium (Bahremand et al., 2010a). Also the synergistic protective ef-fects of lithium and agmatine has been illustrated and our earlier dataconfirmed the involvement of both nitric oxide system (Bahremand

38 B. Payandemehr et al. / Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

et al., 2010b) and α2-adrenoceptor (Bahremand et al., 2011) in theseantiepileptic effects. These studies along with others raise the questionof whetherα2-adrenoceptor can mediate the anticonvulsant propertiesof acute lithium administration.

As a matter of fact, among different suggested mechanisms forlithium activity, it seems that adrenergic pathway and especiallyα2-adrenoceptors play a functional role in its central and peripheraleffects (Cuffí et al., 2010; Devaki et al., 2006; Marmol et al., 1992b).Even a potential role of the adrenergic nervous system has beenproposed to conduct the antidepressant and anxiolytic effects oflithium (El Khoury et al., 2001; Gould et al., 2008; Kasture et al.,2014). In another study, an increase in the inhibitory effect ofdifferent concentrations of lithium on cAMP production in the ratcerebral cortex after the blockade of α2-adrenoceptors has beenrecorded (Cuffi et al., 2003; Marmol et al., 1992a). It is evident nowthat lithium and α2-adrenoceptor agonists like clonidine showfunctional interactions in different paradigms. Lithium is able to pre-vent the development of physical dependence to clonidine (Dehpouret al., 2002) and tends to down-regulate α2-adrenoceptor functionsin some reports (Price and Heninger, 1994).

It has been known thatα2-adrenoceptors may modulate seizuresusceptibility in different seizure paradigms. Generally, α2 recep-tor agonists like clonidine suppress developing and severity ofpentylenetetrazole-induced seizures or amygdala-kindled animals(Shafaroodi et al., 2013; Shouse et al., 2007). However, there are somereports about the proconvulsive effects of clonidine in decreasing thethreshold of electroconvulsion (Loscher and Czuczwar, 1987). Similar toα2 agonists, yohimbine the α2 receptor antagonist differentially affectsseizure paradigms. In PTZ-induced seizure model yohimbine showssome proconvulsive properties at relatively high doses (Fletcher andForster, 1988; Lazarova and Samanin, 1983) but both anticonvulsant(Ludvig et al., 1986) and proconvulsant (Jackson et al., 1991) effectshave been reported for yohimbine in electrical seizures. Interestinglyrecent clinical andmolecular investigations shed light on the involvementof α2-adrenergic system in human epilepsy (Fusco et al., 2014).

Regarding the functional interaction between lithium andα2-adrenoceptors in different domains, we hypothesized that α2 re-ceptor agonists and antagonists at relatively low and non-effectivedoses may modulate the anticonvulsant properties of lithium andalter the seizure threshold. Therefore, we used α2-adrenergic receptoragonist, clonidine, and antagonist, yohimbine, in a model of PTZ-induced clonic seizure, to examine our theory. We further investigat-ed whether this modulatory effect is conducted specifically throughα2D-adrenoceptor by using UK14304 and RX821008, potent andselective α2D receptor agonist and antagonist, respectively.

2. Material and methods

2.1. Animals

MaleNMRImice, weighing 22–30 g fromour center breeding facilitieswere used in this study. The animals were placed in a temperature-controlled (22 ± 3 °C) colony room on a 12-h light/12-h dark cyclewith adequate and free access to food andwater. Eachmouse underwenttreatment once and each treatment group was composed of 8–10animals. The Institutional Review Board of Tehran University of MedicalSciences approved the project protocol and entire procedure wasaccording to the Declaration of Helsinki for Care and Use of LaboratoryAnimals (Publication No. 85-23, revised 1985).

2.2. Drugs

The drugs used were as follows: pentylenetetrazole (PTZ), clonidinehydrochloride, yohimbine hydrochloride, lithium chloride (Sigma,USA), RX821002, 2-(2,3-Dihydro-2-methoxy-1,4-benzodioxin-2-yl)-4,5-dihy dro-1H-imidazole hydrochloride, (displays selectivity for the

α2D over the α2A subtypes) and UK 14304, 5-Bromo-6-(2-imidazolin-2-ylamino) quinoxaline (Tocris, England). The UK14304 was initiallydissolved in dimethylsulfoxide (DMSO) and further diluted in salineuntil the adequate mixture (0.5% DMSO-saline, v/v) was reached. Allother drugs were dissolved in sterile physiological saline solution withappropriate concentrations that were administered in the volume of10ml/kg of mice bodyweight. In all experiments PTZwas administeredintravenously (i.v.) and all other drugs were administered intraperito-neally (i.p.).

2.3. Determination of seizure threshold

The infusion pump was adjusted to deliver PTZ (0.5%) at a constantrate (1 ml/min) in all the experiments (NE 1000, New Era Pump System,Inc). Threshold of PTZ-induced seizurewas determined by inserting a 30-gauge butterfly needle into the tail vein of unrestrained freely movinganimals. Infusionwashaltedwhen forelimb clonus followedby full clonusof the body was observed. The minimal dose of PTZ (mg/kg of mousebody weight) needed to induce clonic seizure was considered as anindex of seizure threshold. As such, the seizure threshold is dependenton the mice weight and time (Payandemehr et al., 2012, 2015).

2.4. Data analysis

Data of seizure thresholds are expressed as mean and standard errorof the mean (S.E.M.) of clonic seizure thresholds in each experimentalgroup. One-way ANOVA followed by Tukey's post hoc multiple compari-sonswere used to analyze the datawhere appropriate. In all experiments,a P-value of 0.05 was considered as the significance level between thegroups.

2.5. Experiments

Animals in experiment 1a received acute i.p. injections of differentdoses of lithium (10, 30, 50, 75mg/kg) or saline 30min before determi-nation of PTZ seizure threshold. Based on this experiment an effectivedose of 30 mg/kg and a sub-effective dose of 10 mg/kg of lithiumwere used in subsequent acute experiments. In experiment 1b, an effec-tive dose of lithium (30 mg/kg) was administered 15, 30, 45, 60 minprior to PTZ to distinct groups of mice. Based on this experiment, apre-test injection interval of 30 min was used in subsequent acuteexperiments.

In experiment 2a different doses of an α2-adrenoceptor agonist,clonidine (0.05, 0.1, 0.25 and 0.5 mg/kg, i.p.), or saline were injected45 min before PTZ induced clonic seizure threshold determination.In experiment 2b animals received acute i.p. injections of differentdoses of yohimbine, an α2-adrenoceptor antagonist (0.1, 0.5, 1, 2.5and 5 mg/kg) or saline 45 min before determination of PTZ seizurethreshold.

Experiment 3 examined the involvement of α2-adrenoceptors inlithium-induced modulation of seizure threshold. In experiment 3adifferent doses of α2-adrenoceptor agonists, clonidine (0.05, 0.1and 0.25 mg/kg, i.p.) were injected 15 min before an effective doseof lithium (30 mg/kg) or saline and 45 min before PTZ-inducedseizure determination.

Animals in experiment 3b received acute i.p. injections of differ-ent doses of yohimbine, an α2-adrenoceptor antagonist (0.1, 0.5, 1and 2.5 mg/kg) or saline 15 min before a sub-effective dose of lithi-um (10 mg/kg) and 45 min before determination of PTZ seizurethreshold. Animals in experiment 4a received acute i.p. injectionsof different doses of UK14304, a potent α2-adrenoceptor agonist(0.05 and 0.1 mg/kg) or vehicle 45 min before determination ofPTZ seizure threshold. In the next part, the same doses were alsoinjected 15 min before an effective dose of lithium (30 mg/kg) and45 min before PTZ-induced seizure determination. In experiments4b, RX821002, a potent α2D-adrenoceptor antagonist (0.05, 0.1 and

39B. Payandemehr et al. / Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

0.25 mg/kg, i.p.) or saline were injected 45 min before PTZ-inducedseizure determination. The same doses of RX821002 were injected15 min before a sub-effective dose of lithium (10 mg/kg) and45 min before determination of PTZ seizure threshold. In the lastset of experiments blood level of Li (75 mg/kg, i.p) was measuredby getting proper sample of each mice, 30 min after Li injection andusing atomic absorption spectrophotometer as described before(Sharifzadeh et al., 2010).

3. Results

3.1. The effects of acute lithium administration on PTZ-induced clonicseizure threshold

Fig. 1a shows the effect of acute administration of different doses oflithium (10, 30, 50 and 75 mg/kg, i.p.) on PTZ-induced clonic seizurethreshold. One-way ANOVA revealed a significant effect for lithiumchloride (F (4, 34) = 15.558, P b 0.001) and post hoc analysis showed

a)

b)

Fig. 1. (a): Effects of lithium chloride (10, 30, 50 and75mg/kg, i.p.) administration on PTZ-induced seizure threshold in mice. Lithium chloride was administered 30 min before de-termination of PTZ seizure threshold. Data are expressed as mean ± S.E.M of seizurethreshold in each group. Each group consisted of at least 8 mice. **P b 0.01 and***P b 0.001 compared with saline control group. (b): Time course of a potent dose of lith-ium chloride (30 mg/kg) affecting PTZ-induced clonic seizure threshold in mice. Lithiumchloride was administered 15, 30, 45 or 60 min before PTZ and its effects were comparedto saline control group (30min before test). Data are expressed asmean±S.E.M of seizurethreshold in each group. Each group consisted of at least 8 mice. *P b 0.05 and **P b 0.01compared with saline control group.

a significant anticonvulsant effect for lithium at doses of 30 mg/kg andhigher compared with saline-treated control animals.

Fig. 1b shows the time-course of the anticonvulsant effect of apotent dose of lithium (30 mg/kg, i.p.). One-way ANOVA revealed asignificant effect (F4, 38 = 5.253, P b 0.01) and further post hocanalysis showed that lithium exerted a maximum anticonvulsanteffect 30 min after administration (P b 0.01, compared with saline-treated control group) and its effect decreased thereafter. Based onthis experiment, seizure threshold determination was done 30 minafter the injection of lithium throughout the study.

3.2. The effect of clonidine and yohimbine on PTZ-induced clonic seizurethreshold

Fig. 2a shows the effect of different doses of clonidine, the α2-adrenoceptor agonist (0.05, 0.1, 0.25 and 0.5 mg/kg, i.p.), on the thresh-old of PTZ-induced clonic seizure. Clonidine was injected 45min beforePTZ induced clonic seizure threshold determination. Comparison of theeffect of different doses of clonidine with saline-treated controls usingone-way ANOVA showed a mild significant anticonvulsant effect onlyat the highest level (0.5 mg/kg) (F (4, 35) = 3.954, post hoc P b 0.01),while the lower doses had no effects on seizure threshold.

Fig. 2b shows the effect of acute administration of different doses ofyohimbine, anα2-adrenoceptor antagonist (0.1, 0.5, 1, 2.5 and 5mg/kg)on PTZ-induced clonic seizure threshold. One-wayANOVA revealed thatyohimbine at low doses could not change the seizure threshold per sewhile at its highest dose (5 mg/kg) was able to induce a significantproconvulsant effect (F (5, 41) = 4.351, P b 0.01).

3.3. The effect of pre-treatment with clonidine and yohimbine on theanticonvulsant property of lithium

Fig. 3a illustrates the effect of pre-treatment with different doses ofclonidine, (0.05, 0.1 and 0.25 mg/kg, i.p.), which administered 15 minbefore an effective dose of lithium (30 mg/kg, i.p.). As seen in Fig. 3a,clonidine administration per se, at relatively low doses, which did not

a) b)

Fig. 2. Effects of α2-adrenoceptor agonist and antagonist on PTZ-induced clonic seizurethreshold inmice. Treatment of mice 45min before the test with low doses of a clonidine,an α2-adrenoceptor agonist, had no significant effect at lower doses (0.05, 0.1 and0.25 mg/kg) but increased the seizure threshold at relatively higher dose (0.5 mg/kg,i.p.) in comparison with the saline control group (a). Administration of different doses ofyohimbine, an α2-adrenoceptor antagonist, decreased the seizure threshold in compari-son to the saline control group at relatively higher dose (5 mg/kg, i.p.) but could notalter the seizure susceptibility at lower doses (0.1,0.5, 1 and 2.5mg/kg, i.p.) in comparisonto the saline control group (b). Each group consisted of at least 8mice. Data are expressedas mean ± S.E.M. of seizure threshold in each group. *P b 0.05 and **P b 0.01 comparedwith saline/saline control group.

a) b)

Fig. 3.Modulatory effects of α2-adrenoceptors on lithium anticonvulsant properties. Pre-treatment of mice with low doses of a clonidine, aα2-adrenoceptor agonist, (0.05, 0.1 and0.25 mg/kg, i.p.) inhibits the anticonvulsant effects of effective dose of lithium chloride(30 mg/kg) in PTZ-induced seizure model (a). Administration of low doses of yohimbine,a α2-adrenoceptor antagonist, (0.1, 0.5, 1 and 2.5 mg/kg, i.p.) potentiates the anticonvul-sant effect of sub-effective dose of lithium (10 mg/kg) significantly (b). Clonidine or yo-himbine was administered 15 min before i.p. injection of lithium chloride and 45 minbefore determination of PTZ seizure threshold. Each group consisted of at least 8 mice.Data are expressed as mean ± S.E.M. of seizure threshold in each group. *P b 0.05,**P b 0.01, and ***P b 0.001 compared with saline/saline control group; ##P b 0.01 and###P b 0.001 compared with corresponding lithium chloride/saline control group.

b)a)

Fig. 4. Changes of protective effects of lithiumwith administration of potent and selectiveα2-adrenoceptor agonist and antagonist. Low and per se non-effective doses of UK14304,a specific and potentα2-adrenoceptor agonist (0.05 and 0.1 mg/kg, i.p.) inhibits the anti-convulsant effect of an effective dose of lithium (30 mg/kg, i.p.) in a PTZ-induced clonicseizure (a). RX821008, a selective and potent α2D-adrenoceptor antagonist, at per selow and non-effective doses (0.05, 0.1 and 0. 25 mg/kg, i.p.) unmasks the potent anticon-vulsant effects of non-effective low dose of lithium chloride (10 mg/kg) on PTZ-inducedseizure threshold (b). RX821008 andUK14304were administered 15min before i.p. injec-tion of lithium chloride and 45 min before determination of PTZ-induced seizure thresh-old. Each group consisted of at least 8mice. Data are expressed asmean±S.E.M. of seizurethreshold in each group. **P b 0.01 and ***P b 0.001 compared with vehicle/saline controlgroup; ##P b 0.01 compared with corresponding lithium chloride/vehicle control group.

40 B. Payandemehr et al. / Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

change the seizure threshold, blocked the anticonvulsive effect oflithium chloride (F (4, 36) = 4.815, P b 0.01).

As shown in the Fig. 3b, low and per se non-effective doses ofyohimbine (0.1, 0.5, 1 and 2.5 mg/kg) administered 15 min beforethe sub-effective dose of lithium (10 mg/kg, i.p.), unmasked a potentanticonvulsant effect and increased the seizure threshold signifi-cantly and dose dependently (F (4, 35) = 16.129, P b 0.001).

3.4. The effect of pre-treatment with UK 14304 and RX821002 on theanticonvulsant property of lithium

Fig. 4a illustrates the effect of low doses of UK 14304, a potent α2-adrenoceptor agonist (0.05, 0.1mg/kg, i.p.) alone or 15min before aneffective dose of lithium (30 mg/kg, i.p.) administered to mice. Asseen in Fig. 4a, UK 14304 at doses used did not change the seizurethreshold alone (F (2, 22) = 0.313, P N 0.05). While the same dosesof UK 14304 (0.05 and 0.1 mg/kg, i.p.) administered 15 min beforean effective dose of lithium (30 mg/kg, i.p.) decreased the seizurethreshold significantly in a dose dependent manner (F (3, 32) =7.674, P b 0.01); hence it inhibited the anticonvulsant effect oflithium.

Fig. 4b shows the effect of different doses of RX821002 a potentα2D-adrenoceptor antagonist (0.05, 0.1 and 0.25 mg/kg i.p.) on thethreshold of PTZ-induced clonic seizure. RX821002 was injected45 min before PTZ-induced clonic seizure threshold determination.Comparison of the effect of different doses of RX821002 with saline-treated controls using one-way ANOVA failed to show an alterationin seizure threshold for this agent alone at these doses (F (3, 29) =0.873, post hoc P N 0.05). However, pre-treatment with these lowand per se non-effective doses of RX821002 (0.01, 0.5 and0.25 mg/kg, i.p.), 15 min before administration of a sub-effectivedose of lithium (10 mg/kg, i.p.) significantly potentiated the anti-convulsant effect of lithium chloride in a dose dependent manner(F (4, 35) = 25.112, post hoc P b 0.001) in comparison with lithiumchloride/saline control group.

4. Discussion

In this study, we showed that α2-adrenoceptors modulate the anti-convulsive effect of lithium chloride dose dependently and it seemsthat at least partly this protective effect is mediated by α2D subtype ofadrenoceptors. In the present investigation, we exploited the PTZseizure test, a popular acute seizure model used to discover drugswith efficacy against non-convulsive absence or myoclonic seizures(Loscher, 2002). PTZ-induced clonic seizure is associatedwith increasedactivity in major epileptogenic centers of forebrain like the amygdalaand piriform cortex (Payandemehr et al., 2013; Swinyard andKupferberg, 1985). This model allows easy screening of different drugsand combinations with potential antiepileptic benefits (Sarkisian,2001). Valproate and ethosuximide are examples of drugs discoveredby this model. Here, we employed the more sensitive I.V. route of PTZadministration to achieve more reliable results and better detection ofeven slight effects on the convulsive tendency of different drugs(Loscher et al., 1991).

In our study, acute administration of lithium, at the doses of30 mg/kg, i.p. and higher, increased the seizure threshold significantly(Fig. 1a). We also demonstrated that the maximum effect of lithiumoccurs 30min following its acute injection (Fig. 1b), which is consistentwith our previous study (Bahremand et al., 2010a) and other studieswith different settings (Ghasemi et al., 2008; Redrobe and Bourin,1999). In accordance with our data, many studies, using various animalmodels of seizure, report a comparable effect for lithium; Roy andMukherjee reported a significant decrease in electroshock-inducedseizure susceptibility related to high lithium ion concentration in therat brain tissue after the acute administration (Roy and Mukherjee,1982). Our data confirmed blood levels of lithium in mice to be in therange of 0.2–1.5 mEq/L, which approximates the therapeutic window(Sifton, 2001). The serum level analysis after i.p administration ofhighest dose of LiCl (75mg/kg)with the time interval of 30min showedthat concentration of Li is 5.7±0.25 μg/ml (asmean±S.E.M). Thismat-ter showed all other doses are in a physiologically therapeutic range and

41B. Payandemehr et al. / Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

suggesting that the results of the present investigation are not attribut-able to lithium intoxication. Furthermore, this result verifies our previ-ous studies which reported antiepileptic activity of the acute lithiumadministration (Bahremand et al., 2010a,b, 2011).

Chronic lithium administration suppressed seizure susceptibilityand elevated the amygdala seizure threshold in animal models(Minabe et al., 1988). Conversely, there are some reports which showedeven proconvulsant properties of lithium (Atigari and Healy, 2013). Inone study, lithium carbonate failed to inhibit the development ofamygdala kindling and lithium chloride did not prevent kindledseizures in rodents (Post et al., 1984). It has been reported that lithiumalone, at the dose of 3.0 mEq/kg, decreased the seizure threshold inkindled hippocampal seizure model in rats (Clifford et al., 1985). Inthis regard, the impairment of noradrenergic function by lithium hasbeen suggested as an underlying mechanism in its potentiating effecton cholinomimetic-induced seizures (Ormandy et al., 1991). In anotherstudy, the therapeutically relevant concentrations of lithium did notinfluence the release of noradrenalin. However, the enhancement inrelease of noradrenalin by larger concentrations of lithium was seenand proposed as a possible mechanism to its toxic effects (Gross andHanft, 1990). Besides that, the functional interaction between lithiumadministration and adrenergic pathway, specifically α2-adrenoceptors,extends to the many other physiologic and pathologic responses.For example, chronic lithium administration significantly attenuatedthe withdrawal signs in clonidine-treated mice and lithium is ableto prevent the development of physical dependence to clonidine(Dehpour et al., 2002). Lithium also affects locomotor stimulationinduced by dependence-producing drugs such as amphetamine,ethanol andmorphine and this suppressive effect of lithium ismediatedvia presynaptic catecholaminergic mechanisms (Berggren et al., 1981).In addition, lithium could block the changes in dopaminergic andnoradrenergic α2 receptor sensitivity in the model of social isolationafter 6 weeks of isolation (Oehler et al., 1984). Even in the clinical trialsit has been shown that lithium, after short-term administration,exerts modulatory effects on α2-adrenoceptor sensitivity using theGH-clonidine test in comparison to the controls in both patients andvolunteers (Brambilla et al., 1988). Moreover, super sensitivity to theα2-adrenoceptor after discontinuation of lithium in patients has beendocumented (Goodnick and Meltzer, 1984).

In the present study, clonidine (0.5 mg/kg, i.p.) and yohimbine(5mg/kg, i.p.) per se could respectively elevate and decrease the seizurethreshold in relatively high doses (Fig. 2). This is in line with otherstudies indicating the modulating effect of α2-adrenoceptor agonistsand antagonists in different seizure paradigms (Amabeoku et al.,1994; Shafaroodi et al., 2013). However, it is important to know thatat higher doses, these drugs are shown to act through non-specificreceptor mediated mechanisms (Ruffolo and Hieble, 1994). As a result,we used the lower doseswhich cannot change the seizure susceptibilityalone in our series of experiments (Fletcher and Forster, 1988;Homayoun et al., 2002). Moreover, we used UK14304 and RX821008,potent and selectiveα2-adrenoceptor agonist and antagonist to confirmour results.

In recent years, α2-adrenoceptors have been the focus of epilepsystudies which investigated protective or harmful effects of differentdrugs and their mechanisms on seizure susceptibility (Fusco et al.,2014; Moezi et al., 2014; Shafaroodi et al., 2013). Recent investigationsrevealed that anticonvulsant properties of a cannabinoid agonist andadenosine mediated through α2-adrenoceptors on the model of PTZ-induced clonic seizure (Moezi et al., 2014; Shafaroodi et al., 2013).However, there are some controversies regarding the effects of α2

agonist and antagonist alone. Moezi et al. reported that clonidinecould not affect seizure susceptibility by itself, similar to the results ofyohimbine administration in another study (Moezi et al., 2014;Shafaroodi et al., 2013). The observed paradoxical effects could be theconsequence of different models or doses which were used in theselatest studies. Interestingly, in accordance with our study, Shafaroodi

et al. showed although clonidine exerts anticonvulsant effects, but thisα2 agonist can inhibit the protective effects of a cannabinoid mimetic(Shafaroodi et al., 2013). These recent results are comparable with ourstudy which shows the same properties for clonidine and/or lithiuminteraction.

We could assume that there might be a common cellular pathwaysuch as cAMP for the interactions between lithium and α2-adrenergicreceptor modulators. There are some studies reporting the inhibitoryeffect of both systems which is mediated through the reduction ofcAMP release. Similarly, an increase in the inhibitory effect of differentconcentrations of lithium on cAMP production in the rat cerebral cortexafter the blockade of α2-adrenoceptors has been recorded (Cuffi et al.,2003; Marmol et al., 1992a). It has been known that accumulation ofcAMP leads to neuronal hyperexcitability and could be a prominentmodulator of seizure susceptibility in different seizure paradigms(Bahremand et al., 2011). In addition, lithium treatment may affectdifferent subtypes of adrenoceptors, interact with the adrenergic nerveterminal vesicles, and change the turnover and concentration ofnorepinephrine and other neurotransmitters in the brain (Berggren,1988; Devaki et al., 2006; Farah et al., 2013). It has been proposed thatchronic lithium administration alters the α2-adrenoceptor turnover;furthermore, inhibitory effect of lithium on cAMP levels is conditionedby α2D-adrenoceptors in rat brain (Carbonell et al., 2004; Cuffi et al.,2003). Considering the frequent clinical use of lithium and based on theunderlying mechanism of mentioned interaction between lithium andα2-adrenergic pathway, more studies with cellular and molecularapproaches are needed to ascertain the exact role of each of these biologicprocesses. This approach can develop modalities to enhance the efficacyof lithium and minimize its side effects.

We previously showed the additive antiepileptic effects of agmatineand lithium could be mediated through α2-adrenoceptors (Bahremandet al., 2011), in this study we tried to find the exact role of these recep-tors in lithium properties on seizure susceptibility. In our study, first weshowed that lithium chloride exerts some protective effects againstPTZ-induced seizures in mice. Also our data confirmed that selecteddoses of LiCl lead to the therapeutic range of lithium in blood which isup to 1meq Li/L (Sifton, 2001). Thenwe reported the efficacy of lithiumwas augmented when combined with certain doses of α2 receptor an-tagonists. These findings first may shed more light on the mechanismsof action of lithium that are not fully understood yet; and second,might offer the α2-adrenergic pathway as a good candidate to improvethe effects or to curtail the side effects of lithium in patients by its mod-ulatory role. As a matter of fact, in one study α2-adrenoceptor function,was attenuated by repeated lithium administration and it is proposedthat these processes may explain the emergence of lithium as an ad-junct to the treatment of refractory depressive illness (Goodwin et al.,1986). On the other hand, more recent study reported that chronic lith-ium treatment can regulate theα2-adrenoceptor gene expression in ratbrain (Cuffí et al., 2010). In the same way, our results may warrantfurther investigation of lithium alone and in combination with othercompounds like α2 receptor antagonists as a potential treatment innon-convulsive myoclonic seizures or refractory seizures by usingvalid seizure models and relevant clinical trials in the future.

In summary, we showed that acute administration of lithium chlo-ride dose dependently increases the PTZ-induced clonic seizure thresh-old in mice. Then we demonstrated that anticonvulsant property oflithium chloride at least partly, is mediated through the α2-adrenergicpathway. Finally, using potent and specificα2 receptor agonists and an-tagonists we confirmed this interaction is possibly mediated throughthe α2D subtype of adrenergic receptors.

References

Amabeoku G, Chikuni O, Bwakura E. Gamma aminobutyric acid mediation of the anticon-vulsant effect of clonidine on pentylenetetrazol-induced seizures in mice. PharmacolRes 1994;29:273–80.

42 B. Payandemehr et al. / Pharmacology, Biochemistry and Behavior 133 (2015) 37–42

Atigari O, Healy D. Pro-convulsant effects: a neglected dimension of psychotropic activity.Aust N Z J Psychiatry 2013;47:998–1001.

Bahremand A, Nasrabady SE, Ziai P, Rahimian R, Hedayat T, Payandemehr B, et al. Involve-ment of nitric oxide–cGMP pathway in the anticonvulsant effects of lithium chlorideon PTZ-induced seizure in mice. Epilepsy Res 2010a;89:295–302.

Bahremand A, Ziai P, Khodadad TK, Payandemehr B, Rahimian R, Ghasemi A, et al.Agmatine enhances the anticonvulsant effect of lithium chloride onpentylenetetrazole-induced seizures in mice: involvement of L-arginine/nitric oxidepathway. Epilepsy Behav 2010b;18:186–92.

Bahremand A, Ziai P, Payandemehr B, Rahimian R, Amouzegar A, Khezrian M, et al. Addi-tive anticonvulsant effects of agmatine and lithium chloride on pentylenetetrazole-induced clonic seizure in mice: involvement of [alpha] 2-adrenoceptor. Eur JPharmacol 2011;666(1):93–9.

Berggren U. The effect of short-term lithium administration on noradrenaline metabolismin the limbic forebrain and hemispheres of rats. J Neural Transm 1988;73:71–5.

Berggren U, Engel J, Liljequist S. The effect of lithium on the locomotor stimulationinduced by dependence-producing drugs. J Neural Transm 1981;50:157–64.

Brambilla F, Catalano M, Lucca A, Smeraldi E. Effect of lithium treatment on the GH-clonidine test in affective disorders. Eur J Clin Pharmacol 1988;35:601–5.

Cade JF. Lithium salts in the treatment of psychotic excitement. Med J Aust 1949;2:349–52.

Carbonell L, Cuffı ML, Forn J. Effect of chronic lithium treatment on the turnover of α2-adrenoceptors after chemical inactivation in rats. Eur Neuropsychopharmacol 2004;14:497–502.

Chi-Tso C, Chuang D-M. Neuroprotective action of lithium in disorders of the centralnervous system. Zhong Nan Da Xue Xue Bao Yi Xue Ban 2011;36:461.

Chiu C-T, Chuang D-M. Molecular actions and therapeutic potential of lithium inpreclinical and clinical studies of CNS disorders. Pharmacol Ther 2010;128:281–304.

Cimarosti H, Rodnight R, Tavares A, Paiva R, Valentim L, Rocha E, et al. An investigation ofthe neuroprotective effect of lithium in organotypic slice cultures of rat hippocampusexposed to oxygen and glucose deprivation. Neurosci Lett 2001;315:33–6.

Clifford DB, Podolsky A, Zorumski CF. Acute effects of lithium on hippocampal kindledseizures. Epilepsia 1985;26:689–92.

Cuffi ML, Carbonell L, Marmol F, Gimenez J, Forn J. Effect of alpha 2-adrenoceptorblockade on lithium action in the rat brain. Brain Res 2003;994:73–80.

Cuffí ML, Artells R, Navarro A, Ciruela F, Carbonell L. Regulation of α2-adrenoceptor geneexpression by chronic lithium treatment in rat brain. Methods Find Exp ClinPharmacol 2010;32:721–5.

Dehpour AR, Sadr SS, Azizi MR, Namiranian K, Farahani M, Javidan AN. Lithium inhibitsthe development of physical dependence to clonidine in mice. Pharmacol Toxicol2002;90:89–93.

Devaki R, Shankar Rao S, Nadgir SM. The effect of lithium on the adrenoceptor-mediatedsecond messenger system in the rat brain. J Psychiatry Neurosci 2006;31:246–52.

El Khoury A, Johnson L, Åberg-Wistedt A, Stain-Malmgren R. Effects of long-term lithiumtreatment on monoaminergic functions in major depression. Psychiatry Res 2001;105:33–44.

Farah R, Khamisy-Farah R, Amit T, Youdim MBH, Arraf Z. Lithium's gene expressionprofile, relevance to neuroprotection A cDNA microarray study. Cell Mol Neurobiol2013;33:411–20.

Fletcher A, Forster EA. A proconvulsant action of selective alpha 2-adrenoceptor antago-nists. Eur J Pharmacol 1988;151:27–34.

Fusco M, Vago R, Striano P, Bonaventura C, Zara F, Mei D, et al. The α2B-adrenergicreceptor is mutant in cortical myoclonus and epilepsy. Ann Neurol 2014;75:77–87.

Ghasemi M, Dehpour AR. The NMDA receptor/nitric oxide pathway: a target for thetherapeutic and toxic effects of lithium. Trends Pharmacol Sci 2011;32:420–34.

Ghasemi M, Sadeghipour H, Mosleh A, Sadeghipour HR, Mani AR, Dehpour AR. Nitricoxide involvement in the antidepressant-like effects of acute lithium administrationin the mouse forced swimming test. Eur Neuropsychopharmacol 2008;18:323–32.

Ghasemi M, Shafaroodi H, Nazarbeiki S, Meskar H, Heydarpour P, Ghasemi A, et al.Voltage-dependent calcium channel and NMDA receptor antagonists augmentanticonvulsant effects of lithium chloride on pentylenetetrazole-induced clonicseizures in mice. Epilepsy Behav 2010;18:171–8.

Goodnick PJ, Meltzer HY. Neurochemical changes during discontinuation of lithiumprophylaxis. I. Increases in clonidine-induced hypotension. Biol Psychiatry 1984;19:883–9.

Goodwin GM, DeSouza RJ, Wood AJ, Green AR. Lithium decreases 5-HT1A and 5-HT2receptor and alpha 2-adrenoceptor mediated function in mice. Psychopharmacology(Berl) 1986;90:482–7.

Gould TD, O'Donnell KC, Dow ER, Du J, Chen G, Manji HK. Involvement of AMPA receptorsin the antidepressant-like effects of lithium in the mouse tail suspension test andforced swim test. Neuropharmacology 2008;54:577–87.

Gross G, Hanft G. Does lithium in vitro and ex vivo alter the release of [3H]noradrenalinefrom brain tissue and the sensitivity of presynaptic autoreceptors? Neuropharmacol-ogy 1990;29:831–5.

Homayoun H, Khavandgar S, Dehpour AR. The role of alpha2-adrenoceptors in the mod-ulatory effects of morphine on seizure susceptibility in mice. Epilepsia 2002;43:797–804.

Jackson HC, Dickinson SL, Nutt DJ. Exploring the pharmacology of the pro-convulsanteffects of alpha 2-adrenoceptor antagonists in mice. Psychopharmacology (Berl)1991;105:558–62.

Kasture SB, Mane-Deshmukh RV, Arote SR. Non-linear dose effect relationship in anxio-lytic and nootropic activity of lithium carbonate and Nardostachys jatamansi in rats.Orient Pharm Exp Med 2014:1–6.

Lazarova M, Samanin R. Potentiation by yohimbine of pentylenetetrazol-induced seizuresin rats: role of alpha 2 adrenergic receptors. Pharmacol Res Commun 1983;15:419–25.

Loscher W. Animal models of epilepsy for the development of antiepileptogenic anddisease-modifying drugs. A comparison of the pharmacology of kindling and post-status epilepticus models of temporal lobe epilepsy. Epilepsy Res 2002;50:105–23.

Loscher W, Czuczwar SJ. Comparison of drugs with different selectivity for central alpha1- and alpha 2-adrenoceptors in animal models of epilepsy. Epilepsy Res 1987;1:165–72.

Loscher W, Honack D, Fassbender CP, Nolting B. The role of technical, biological andpharmacological factors in the laboratory evaluation of anticonvulsant drugs. III.Pentylenetetrazole seizure models. Epilepsy Res 1991;8:171–89.

Ludvig N, Gyorgy L, Vizi ES. Evidence that serotonin receptors are involved in the anticon-vulsant effect of yohimbine in mice. Psychopharmacology (Berl) 1986;88:82–5.

Machado‐Vieira R, Manji HK, Zarate Jr CA. The role of lithium in the treatment of bipolardisorder: convergent evidence for neurotrophic effects as a unifying hypothesis.Bipolar Disord 2009;11:92–109.

Malhi GS, Tanious M, Das P, Coulston CM, Berk M. Potential mechanisms of action oflithium in bipolar disorder. CNS Drugs 2013;27:135–53.

Manji HK, Moore GJ, Chen G. Lithium at 50: have the neuroprotective effects of thisunique cation been overlooked? Biol Psychiatry 1999;46:929–40.

Marmol F. Lithium: bipolar disorder and neurodegenerative diseases possible cellularmechanisms of the therapeutic effects of lithium. Prog Neuropsychopharmacol BiolPsychiatry 2008;32:1761–71.

Marmol F, Carbonell L, Cuffi ML, Forn J. Demonstration of inhibition of cyclic AMPaccumulation in brain by very low concentrations of lithium in the presence ofalpha-adrenoceptor blockade. Eur J Pharmacol 1992a;226:93–6.

Marmol F, Puig-Parellada P, Forn J. Interactions of lithium with the adrenergic system inelectrically-stimulated guinea pig myenteric plexus isolated preparations. Actions ofa phosphodiesterase inhibitor. Methods Find Exp Clin Pharmacol 1992b;14:623–8.

Minabe Y, Emori K, Kurachi M. Effects of chronic lithium treatment on limbic seizuregeneration in the cat. Psychopharmacology (Berl) 1988;96:391–4.

Moezi L, Mansoori E, Niknahad H, Shafaroodi H. The role of alpha-2 adrenoceptors in theanticonvulsant effects of adenosine on pentylenetetrazole-induced seizure thresholdin mice. Pharmacol Biochem Behav 2014;126:36–42.

Oehler J, Jahkel M, Schmidt J. Inhibition of isolation-induced changes in aminergictransmission by chronic lithium treatment. Pharmacol Biochem Behav 1984;21:181–4.

Ormandy GC, Song L, Jope RS. Analysis of the convulsant-potentiating effects of lithium inrats. Exp Neurol 1991;111:356–61.

Payandemehr B, Bahremand A, Rahimian R, Ziai P, Amouzegar A, Sharifzadeh M, et al. 5-HT3 receptor mediates the dose-dependent effects of citalopram onpentylenetetrazole-induced clonic seizure in mice: Involvement of nitric oxide.Epilepsy Res 2012;101(3):217–27.

Payandemehr B, Rahimian R, Bahremand A, Ebrahimi A, Saadat S, Moghaddas P, et al. Roleof nitric oxide in additive anticonvulsant effects of agmatine and morphine. PhysiolBehav 2013;118:52–7.

Payandemehr B, Ebrahimi A, Gholizadeh R, Rahimian R, Varastehmoradi B, Gooshe M,et al. Involvement of PPAR receptors in the anticonvulsant effects of a cannabinoidagonist, WIN 55,212-2. Prog Neuropsychopharmacol Biol Psychiatry 2015;57:140–5.

Post RM, Weiss SR, Pert A. Differential effects of carbamazepine and lithium on sensitiza-tion and kindling. Prog Neuropsychopharmacol Biol Psychiatry 1984;8:425–34.

Price LH, Heninger GR. Lithium in the treatment of mood disorders. N Engl J Med 1994;331:591–8.

Redrobe JP, Bourin M. The effect of lithium administration in animal models of depres-sion: a short review. Fundam Clin Pharmacol 1999;13:293–9.

Roy U, Mukherjee BP. Correlation of lithium effect on electroshock-induced seizure in ratswith its concentration in brain and plasma. Arch Int Pharmacodyn Ther 1982;255:81–8.

Ruffolo Jr RR, Hieble JP. Alpha-adrenoceptors. Pharmacol Ther 1994;61:1–64.Sarkisian MR. Overview of the current animal models for human seizure and epileptic

disorders. Epilepsy Behav 2001;2:201–16.Shafaroodi H, Moezi L, Bahremand A, Dehpour AR. The role of α2-adrenoceptors in the

anti-convulsant effects of cannabinoids on pentylenetetrazole-induced seizurethreshold in mice. Eur J Pharmacol 2013;714:1–6.

Sharifzadeh M, Rezaei H, Ghamsari MR. Interactive effects of acute and chronic lithiumwith dopamine receptor antagonists on naloxone-induced jumping in morphine-dependent mice. Iran J Pharm Res 2010:217–23.

Shouse MN, Scordato JC, Farber PR, de Lanerolle N. The alpha2 adrenoreceptor agonistclonidine suppresses evoked and spontaneous seizures, whereas the alpha2adrenoreceptor antagonist idazoxan promotes seizures in amygdala-kindled kittens.Brain Res 2007;1137:58–68.

Sifton DW, editor. Physicians' desk reference. 55th ed. Montvale, NJ: Medical EconomicsCompany, Inc; 2001.

Swinyard EA, Kupferberg HJ. Antiepileptic drugs: detection, quantification, and evalua-tion. Fed Proc 1985;44:2629–33.