www.e l sev i e r. com/ loca te / eu roneu ro

European Neuropsychopharmacology (2008) 18, 323–332

Nitric oxide involvement in the antidepressant-likeeffects of acute lithium administration in the mouseforced swimming testMehdi Ghasemia, Hamed Sadeghipour a, Afsaneh Moslehb,Hamid Reza Sadeghipour b,⁎, Ali Reza Mani c, Ahmad Reza Dehpour a

a Department of Pharmacology, School of Medicine, Medical Sciences / University of Tehran, Tehran, Iranb Department of Physiology, School of Medicine, Medical Sciences / University of Tehran, Tehran, Iranc The UCL Institute of Hepatology, Department of Medicine, Royal Free and University College Medical School, UCL,London, UK

Received 17 February 2007; received in revised form 26 July 2007; accepted 31 July 2007

⁎ Corresponding author. Tel.: +9866419484.

E-mail address: hamedsadeghipour(H.R. Sadeghipour).

0924-977X/$ - see front matter © 200doi:10.1016/j.euroneuro.2007.07.011

Abstract

In the present study we evaluated the involvement of L-arginine/nitric oxide (NO)/cGMP pathwayin the antidepressant-like effects of acute lithium administration in the mouse forced swimmingtest (FST). Lithium, at 30 and 100 mg/kg, significantly reduced the immobility times of mice inthe FST, whereas at lower doses (0.5, 5 and 10 mg/kg) had no effect on the immobility time. TheNO synthase (NOS) inhibitor NG-nitro-L-arginine methyl ester (L-NAME), at 10 and 30 mg/kg, andthe selective neuronal NOS inhibitor Nω-propyl-L-arginine (L-NPA), at 5 and 15 mg/kg, had nosignificant effects on the FST, whereas they significantly decreased the immobility time at 100and 30 mg/kg, respectively. Combination of non-effective dose of lithium (10 mg/kg) with lowdoses of L-NAME (30 mg/kg) or L-NPA (15 mg/kg) significantly reduced the immobility times in theFST. Moreover, the guanylyl cyclase inhibitor ODQ at 50 mg/kg significantly decreased theimmobility time of mice, whereas it had not significant effects on the FSTat 2, 10 and 20 mg/kg.Combination of lithium (10 mg/kg) with 20 mg/kg ODQ significantly decreased the immobilitytimes in the FST. Non-effective doses of L-arginine (750 mg/kg) or sildenafil (5 mg/kg)significantly reversed the antidepressant-like effect of 30 mg/kg lithium in the FST. Neither ofthe drugs had effect on the locomotor activity. These data indicate the involvement of L-arginine/NO/cGMP pathway in the antidepressant-like effect of lithium in the mouse FSTand alsomight suggest the concurrent administration of NOS inhibitors and lithium as an appropriatestrategy for treatment of depression.© 2007 Elsevier B.V. and ECNP. All rights reserved.

KEYWORDSLithium;Depression;Nitric oxide (NO);Nitric oxide synthase(NOS) inhibitor;L-arginine;Forced swimming test

21 66112345; fax: +98 21

@yahoo.com.sg

7 Elsevier B.V. and ECNP. All right

1. Introduction

Lithium has largely met its initial promise as the first drug tobe discovered in the modern era of psychopharmacology

s reserved.

324 M. Ghasemi et al.

(Price and Heninger, 1994). Although over 50 years haveelapsed since its effects on mania were first described (Cade,1949), lithium is still a mainstay in the treatment of mooddisorders (Souza and Goodwin, 1991; Drummond, 1987;Kleindienst and Greil, 2003). Originally lithium was assumedto have a poor antidepressant effect, but its antidepressantefficacy has been supported by several controlled studies(Worrall et al., 1979; Katona, 1988). Although antidepressanteffects of lithium have been more clearly demonstrated indepressed bipolars than in unipolars (Bourin and Prica, inpress; Baron et al., 1975), initial open trials (Johnson, 1974;Watanabe et al., 1974; Worrall et al., 1979) and double-blindcontrolled studies found lithium effective for unipolardepression (Freeman and Freeman, 2006; Soares andGershon, 1998; Mendels et al., 1972; Souza and Goodwin,1991). Even in some studies, it was found to be as effective asantidepressants (Soares and Gershon, 1998; Greil andKleindienst, 2003; Mendels et al., 1972; Watanabe et al.,1975; Worrall et al., 1979; Linder et al., 1989). However, yetno definitive mechanism for the effects of lithium has been

Figure 1 (A) Effect of acute administration of lithium chloride (0.5chloride was administered 30 min before the test. ⁎Pb0.05 and ⁎⁎Pb0administration of 30 mg/kg lithium (Li) on the mouse FST at differ⁎⁎Pb0.01 compared with the corresponding saline-treated controlchloride (0.5–100mg/kg, i.p.) on locomotor activity ofmice in the optest. (D) Effect of acute administration of 30 mg/kg Li on the open fieValues are expressed as mean±S.E.M. (n=10) and were analyzed usi

established. In this regard, there is some evidence thatlithium exerts its therapeutic effects by interfering withsignal transduction through G-protein-coupled pathways(Jope and Williams, 1994) or direct inhibition of specifictargets in signaling systems, which include inositol monopho-sphatase (Berridge and Irvine, 1989; Nahorski et al., 1991)and glycogen synthase kinase-3 (Jope, 2003). Inhibition ofserotonin receptors and up-regulation of glutamate reuptakeare other hypothesis about lithium mechanisms of action(Shaldubina et al., 2001). In addition, some authors havesuggested that nitric oxide (NO) may mediate some oflithium-induced responses in the brain (Harvey et al., 1994;Paoletti et al., 1998; Weerasinghe et al., 2004; Anai et al.,2001) or other tissues (Dehpour et al., 2000a; Sadeghipour etal., 2006, 2007). For instance, using an in vivo brainmicrodialysis method, it has been recently demonstratedthat intraperitoneally lithium administration (50 mg/kg)can significantly reduce NO3 and NOx levels in rat amygdalaafter 60 to 90 min (Maruta et al., 2005). Moreover, in ourprevious study, we showed that acute lithium administra-

–100 mg/kg, i.p.) on forced swimming test (FST) in mice. Lithium.01 compared with the saline-treated control. (B) Effect of acuteent time points after administration (15–60 min). ⁎Pb0.05 andat the same time. (C) Effect of acute administration of lithiumen field test. Lithium chloridewas administered 30min before theld test at different time points after administration (15–60 min).ng a one-way ANOVA followed by Newman–Keuls test.

325NO involvement in the antidepressant-like effects of acute lithium administration in the mouse FST

tion decreased the neurogenic relaxation of rat corpuscavernosum and this effect was reversed by coadministrationof the precursor of NO L-arginine and was potentiated by theNOS inhibitor L-NAME, indicating that NO was involved in thiseffect of lithium (Sadeghipour et al., 2007). However, theexact role of L-arginine/NO pathway in the therapeuticeffects of lithium is as yet unidentified.

Nitric oxide (NO)/cGMP pathway is implicated in thecontrol of many physiological functions (Esplugues, 2002). Inrecent studies, it has been shown that NO could be as anovel target of antidepressant action (Harkin et al., 1999,2004; Inan et al., 2004; Jefferys and Funder, 1996; Yildiz etal., 2000) using the mouse forced swimming test (FST), apreclinical behavioural paradigm that is widely used to testcompounds for antidepressant activity (Borsini, 1995; Cryanet al., 2002). When animals are exposed to the FST theytypically adopt an immobile posture, which is thought toreflect a state of behavioural despair or helplessness

Figure 2 (A) Effect of acute administration of either L-NAME (10–10inhibitors were administered 45 min before the test. ⁎Pb0.05 compar(30 mg/kg) or L-NPA (15 mg/kg) 15 min before administration of lithiutime of mice in the FSTat 30 min after Li administration. ⁎⁎Pb0.01 csaline/lithium (10 mg/kg) group; ns means non-significant. (C) EffectL-NPA (5–30mg/kg, i.p.) on the open field test in mice. NOS inhibitorsL-NAME (30mg/kg) or L-NPA (15mg/kg) 15min before administration oactivity of mice in the open field test at 30min after Li administrationusing a one-way or two-way ANOVA followed by Newman–Keuls test.

(Porsolt et al., 1977). Antidepressants reduce immobilityby increasing escape-motivated behaviours in the FSTparadigm (Borsini, 1995; Lucki et al., 2001). Moreover,some studies also have shown that NOS inhibitors display anantidepressant-like behavioural profile in the mouse FST(Harkin et al., 1999, 2004). The antidepressant-like beha-vioural effects of NOS inhibitors are blocked by pretreat-ment with the NOS substrate, L-arginine, consistent with aninvolvement of NO in this behavioural response (Harkin etal., 1999). On the other hand, animal behavioural studiesusing lithium in the FST support the antidepressantproperties of lithium (Redrobe and Bourin, 1999a,b; Hascoetet al., 1994). Moreover, lithium augmentation has beenconsidered as the first-choice procedure for patients whofail to respond to antidepressant monotherapy (Bauer andDopfmer, 1999). Pretreatment with lithium can augment theeffect of various antidepressants in the FST (Nixon et al.,1994; Hascoet et al., 1994). However, the mechanism of

0mg/kg, i.p.) or L-NPA (5–30mg/kg, i.p.) on the FST inmice. NOSed with the saline-treated control. (B) Pretreatment with L-NAMEm (Li, 10 mg/kg) exerted a significant decrease in the immobilityompared with the saline/saline control; #Pb0.05 compared withof acute administration of either L-NAME (10–100 mg/kg, i.p.) orwere administered 45min before the test. (D) Pretreatment withf lithium (Li, 10mg/kg) had no significant effect on the locomotor. Values are expressed asmean±S.E.M. (n=10) and were analyzed

326 M. Ghasemi et al.

action of the antidepressant-like effect of lithium in thismodel is not fully understood. Therefore, in the presentstudy, we investigated whether L-arginine/NO/cGMP path-way is involved in the effect of acute lithium administrationin the mouse FST.

2. Materials and methods

2.1. Animals

Male NMRI mice weighting 23–30 g (Pasteur Institute) were usedthroughout the study. Animals were housed in groups of 4–5 and wereallowed free access to food and water except for the short time thatanimals were removed from their cages for testing. All behaviouralexperiments were conducted during the period between 10:00 a.m.and 13:00 p.m. with normal room light (12-hour regular light/darkcycle) and temperature (22±1 °C). All procedures were carried out inaccordance with the institutional guidelines for animal care and use.Each mouse was used only once, and each experimental groupconsisted of at least ten animals.

Figure 3 (A) Effect of acute administration of ODQ (2–50mg/kg, i.p.⁎Pb0.05 compared with the DMSO-treated control. (B) Concurrent adsignificant antiimmobility effect in the FST at 30 min after administrcomparedwith DMSO/lithium (10mg/kg) group; nsmeans non-significathe locomotor activity of mice in the open field test. ODQ was adminlithium (10 mg/kg) and ODQ (20 mg/kg) had no significant effect on thadministration. Values are expressed asmean±S.E.M. (n=10) andwereKeuls test.

2.2. Open field locomotor activity

Immediately before the FST, the ambulatory behaviour of mice wasassessed in an open field test (Kaster et al., 2005), to ensure thatalterations in the duration of immobility are not resultant from thechanges that occur in motor activity. The apparatus consisted of awooden box measuring 40 cm×60 cm×50 cm. The floor of the arenawas divided into 12 equal squares. The animals were gently placed inthe center of the field and the number of squares crossed with allpaws (crossing) was counted in a 6-min session.

2.3. Forced swimming test (FST)

The testwas conducted using themethodof Porsolt et al. (1977). Micewere individually placed in an open cylindrical container (diameter10 cm, height 25 cm), containing 19 cm of water at 23±1 °C. Micewere allowed to swim for 6 min. Behaviour was analyzed by twoexperienced raters who were blind to drug treatment. Each mousewas judged to be immobile when it ceased struggling and remainedfloating motionless in the water, making only those movementsnecessary to keep its head above water. The duration of immobilitywas recorded during the last 4 min of the test.

) on the FST inmice. ODQwas administered 30min before the test.ministration of lithium (10 mg/kg) and ODQ (20 mg/kg) exerted aation. ⁎Pb0.05 compared with the DMSO/saline control; #Pb0.05nt. (C) Effect of acute administration of ODQ (2–50mg/kg, i.p.) onistered 30 min before the test. (D) Concurrent administration ofe locomotor activity of mice in the open field test at 30 min afteranalyzed using a one-way or two-wayANOVA followedbyNewman–

Figure 4 (A) Effect of simultaneous administration of lithiumchloride (Li, 30 mg/kg) with either L-NAME (30 mg/kg) or L-NPA(15 mg/kg) or ODQ (20 mg/kg) on the FST in mice at 45 min afteradministration. ⁎Pb0.05 compared with the DMSO/saline con-trol, ⁎⁎Pb0.01 compared with the saline/saline control; #Pb0.05compared with saline/lithium (30 mg/kg)-treated group; nsmeans non-significant. (B) Effect of simultaneous administrationof lithium chloride (Li, 30 mg/kg) with either L-NAME (30 mg/kg)or L-NPA (15 mg/kg) or ODQ (20 mg/kg) on the Open field testat 45 min after administration. Values are expressed as mean±S.E.M. (n=10) and were analyzed using a two-way ANOVAfollowed by Newman–Keuls test.

327NO involvement in the antidepressant-like effects of acute lithium administration in the mouse FST

2.4. Drugs and treatment

The following drugs were used in the study: lithium chloride, L-arginine, sildenafil, NG-L-arginine methyl ester (L-NAME), [1H-[1,2,4]Oxadiazole[4,3-a]quinoxalin-1-one] (ODQ) (Sigma, St Louis, MO,USA), Nω-propyl-L-arginine (L-NPA, Tocris, Bristol, UK). All drugswere dissolved in saline except ODQ which was dissolved in 15%dimethyl sulfoxide (DMSO: in saline). All solutions were preparedimmediately before the experiments and all injections wereadministered intraperitoneally (i.p.) in a constant volume of 5 ml/kg body weight. Lithium chloride (0.5, 5, 10, 30 and 100 mg/kg) wasadministered 30 min before the FST in separate groups. In this step,both the per se non-effective and potent doses of lithium were foundfor assessment in our next experiments.

For evaluating the effects of L-NAME (10, 30, 100 mg/kg), L-NPA(5, 15, 30 mg/kg), ODQ (2, 10, 20 and 50 mg/kg), L-arginine (750 mg/kg, a dose that produces no effect in the FST; Kaster et al., 2005) andsildenafil (5 mg/kg, a dose that produces no effect in the FST; Kasteret al., 2005) in the mouse FST. L-NAME, L-NPA, L-arginine andsildenafil were administered 45 min and ODQ was administered30 min before the test in separate groups of animals. In theseexperiments, the non-effective doses of L-NAME, L-NPA andODQwerefound for our next experiments.

For evaluating the possible role of L-arginine/NO/cGMP pathwayin the effect of lithium in the FST, the non-effective (subactive) dosesof either L-NAME or L-NPA were separately administered 15 minbefore administration of non-effective (subactive) dose of lithiumchloride. Moreover, in another separate group the subactive dose ofODQ was administered immediately before administration of non-effective (subactive) dose of lithium chloride. Thirty minutes afterlithium administration, animals were assessed in the FST. In otherseparate experimental groups, either L-arginine (750 mg/kg) orsildenafil (5 mg/kg) was injected into the animals 15 min beforeadministration of a potent dose of lithium in the FST. Thirty min afteradministration of the active dose of lithium chloride, animals wereassessed in the FST.

2.5. Statistical analysis

Comparisons between experimental and control groups wereperformed by one-way or two-way ANOVA followed by Newman–Keuls test when appropriate. A value of Pb0.05 was considered to besignificant.

3. Results

The results depicted in Fig. 1A show that the administration oflithium chloride decreased the immobility time of mice in theFST. ANOVA revealed a significant effect of lithium (F5,54=4.296,Pb0.01). Lithium chloride in concentrations of 30 mg/kg(Pb0.01) and 100 mg/kg (Pb0.05) significantly decreased theimmobility time in the FST, whereas at doses of 0.5, 5 and 10 mg/kg produced no significant antiimmobility effect. Fig. 1B showsthe time-dependent effect of lithium chloride (30 mg/kg) on theimmobility time in the FST. The highest antiimmobility effect oflithium chloride was observed 30 min after lithium injection(Pb0.01). At 15 and 45 min, lithium chloride (30 mg/kg) had notany significant effect on the immobility time in the FST (Fig. 1B).As shown in Fig. 1C, different doses of lithium chloride had nosignificant effect on locomotor activity of mice in the open fieldtest (F5,54=0.388, PN0.05). Also, there was no time-dependentalteration in locomotor activity of mice after administration of30 mg/kg lithium chloride (Fig. 1D; F7,72=0.1617, PN0.05).

As shown in Fig. 2A, the administration of either thenonselective NOS inhibitor L-NAME or the selective nNOS inhibitor

L-NPA had antiimmobility effects on mice in the FST. ANOVArevealed a significant effect of L-NAME (F3,36=3.230, Pb0.05)and L-NPA (F3,36=3.433, Pb0.05). Either 100 mg/kg L-NAME or30 mg/kg L-NPA significantly (Pb0.05) decreased the immobilitytime of mice in the FST. Fig. 2B shows that combination of per senon-effective doses of L-NAME (30 mg/kg) and lithium chloride(10 mg/kg) significantly (Pb0.01) exerted an antidepressant-likeeffect in the FST (F3,36=4.762, Pb0.01). Similar result wasobtained when both per se non-effective doses of L-NPA (15 mg/kg) and lithium chloride (10 mg/kg) were administered before theFST (F3,36=8.336, Pb0.001). As shown in Fig. 2C, neither L-NAME(F3,36=0.6099, PN0.05) nor L-NPA (F3,36=0.2269, PN0.05) hadeffect on the locomotor activity of mice in the open field test. Inaddition, comparedwith saline/saline treated animals, concurrentadministration of lithium chloride (10 mg/kg) with either L-NAME(F3,36=0.1788, PN0.05) or L-NPA (F3,36=0.3905, PN0.05) did notalter the locomotor activity of mice in the open field test (Fig. 2D).

The results depicted in Fig. 3A show the effect of differentdoses of the soluble gyanylyl cyclase inhibitor ODQ on the

328 M. Ghasemi et al.

immobility time of mice in the FST. ANOVA revealed a significanteffect of ODQ (F4,45=3.081, Pb0.05). 50 mg/kg ODQ significantly(Pb0.05) decreased the immobility time in the FST, whereas atdoses of 2, 10 and 20 mg/kg it had not any significantantiimmobility effect in the FST. As shown in Fig. 3B, concurrentadministration of low doses of lithium chloride (10 mg/kg) andODQ (20 mg/kg) exerted a significant (Pb0.05) antiimmobilityeffect in the FST (F3,36=3.932; Pb0.05). Fig. 3C shows that ODQdid not have any significant effect on the locomotor activity ofmice in the open field test (F4,45=0.4318, PN0.05). Also,compared with DMSO/saline animals, concurrent administrationof ODQ (20 mg/kg) and lithium chloride (10 mg/kg) did notalter the locomotor activity of mice in the open field test(F3,36=0.4275, PN0.05; Fig. 3D).

As mentioned above, lithium chloride (30 mg/kg) at 45 minafter injection had no significant effect on the immobility time ofmice in the FST (Fig. 1B). We further evaluated whethercoadministration of lithium chloride and non-effective doses ofL-NAME, L-NPA and ODQ had any effect on the immobility time at45 min. The results depicted in Fig. 4A show that when 30 mg/kg

Figure 5 (A) Effect of pretreatment with either L-arginine(750 mg/kg, i.p.) or sildenafil (5 mg/kg, i.p.) on the lithiumchloride (30 mg/kg)-induced reduction in immobility time ofmice in the FST. ⁎⁎Pb0.01 compared with the saline/salinecontrol; #Pb0.05 compared with the saline/lithium group. (B)Pretreatment with L-arginine (750 mg/kg) or sildenafil (5 mg/kg,i.p.) 15 min before administration of lithium (Li, 30 mg/kg) hadno significant effect on the locomotor activity ofmice in the openfield test at 30 min after Li administration. Values are expressedas mean±S.E.M. (n=10) and were analyzed using a two-wayANOVA followed by Newman–Keuls test.

lithium chloride was administered concomitantly with subactivedose of L-NAME (30mg/kg) the immobility time of mice in the FSTwas significantly (Pb0.01) decreased (F3,36=7.083, Pb0.001)45 min after their injection compared with saline/saline treatedanimals. Similarly, simultaneous administration of lithium at thesame dose and non-effective dose of L-NPA (15 mg/kg) exerted asignificant (Pb0.01) decrease in the immobility time of mice at45 min (F3,36=5.584, Pb0.01) compared with saline/salinetreated animals. Our data also showed a significant (Pb0.05)decrease in the immobility time of mice when lithium chloride atthe same dose was concomitantly administered with 20 mg/kgODQ (F3,36=3.093, Pb0.05). Also, concurrent administration oflithium chloride with L-NAME, L-NPA and ODQ had not any effecton the locomotor activity of mice in the open field test (Fig. 4B).

Fig. 5A shows that although either L-arginine (750 mg/kg) orsildenafil (5 mg/kg) individually did not alter the immobilitytime of mice in the FST, they significantly (Pb0.05) reversed theantidepressant-like effect of 30 mg/kg lithium chloride in theFST (F3,36=9.029, Pb0.001 and F3,36=6.442, Pb0.01, respec-tively). Neither L-arginine (F3,36=0.3258, PN0.05) nor sildenafil(F3,36=0.4499, PN0.05), concurrent with vehicle or lithiumchloride, alter the locomotor activity of mice in the open fieldtest (Fig. 5B).

4. Discussion

The FST is the most widely employed screening test forantidepressants in rodents (Cryan et al., 2002; Porsolt et al.,1977). Rodents when forced to swim in a cylinder from whichthey cannot escape will, after an initial period of vigorousactivity, display a characteristic immobile posture which canbe readily identified and is said to reflect a state of‘behavioural despair’. A variety of antidepressant drugsincrease escape-oriented behaviour in the test (Borsini andMeli, 1988). In the present study, we showed that acutelithium administration caused a decrease in the immobilitytime of mice in the FST. This effect of lithium, which wassignificant at doses of 30 and 100 mg/kg, was significantlyobserved 30 or 60 min after its (30 mg/kg) injection, whereaseither 15 or 45 min after lithium (30 mg/kg) injection theantidepressant-like effect of lithiumwas not observed. Theseresults are consistent with previous studies which haveindicated that lithium decreases time immobile in the FSTin a dose-dependent manner (O'Brien et al., 2004). Somestudies also have reported that acute lithium administrationreduces time immobile in mice if animals are tested 30 minafter injection but not at 45 min (Redrobe and Bourin, 1999a;Nixon et al., 1994). In the latter study, however, pretreat-ment with lithium was able to augment the effect of variousantidepressants when given at a single bolus 45 min beforethe test. Acute lithium administration also reportedlyenhances the effect of antidepressants on the FST (Redrobeand Bourin, 1999b). Moreover, O'Brien et al. (2004) haveshown that lithium treatment of mice for a period of 10 dayshad a robust effect on time immobile of mice in the FST. Theyalso showed that this effect was in a dose-dependent mannersince mice which had been treated with 0.2% lithium (for5 days) followed by 0.4% lithium (for 5 days) had less timeimmobile compared with mice which had been treated with0.2% lithium for 10 days. The effect of lithium in the FSTcouldrepresent a parallel to the mood stabilizing action of lithiumon the depressive phase of bipolar disorder (Bourin and Prica,

329NO involvement in the antidepressant-like effects of acute lithium administration in the mouse FST

in press; O'Brien et al., 2004). Interestingly, lithium treat-ment also parallels the effects of antidepressants on learnedhelplessness (Sherman and Petty, 1980) and immobilizationstress in rats (Hata et al., 1995). However, there are somediscrepancies regarding the antidepressant-like effect oflithium in the FST. In one study, Kitamura et al. (2002)demonstrated that there was no significant alteration in theimmobility time of rats in the FST 8 h after oral administrationof lithium (10–100 mg/kg). Wegener et al. (2003) also failedto demonstrate any effect of subchronic lithium treatment(60 mM/kg, orally) on the immobility of rats in the FST. EvenTomasiewicz et al. (2006) reported that subacute treatmentof rats with lithium (100 mg/kg) caused a prodepressant-likeeffect in the FSTand significantly increased the occurrence ofimmobility in lithium-treated animals in this test. Thesedivergent effects of lithium may be influenced by theduration of lithium treatment. In addition, it is not clearfrom previous studies whether the doses used were within arange that is free of non-specific behavioural effects that cancomplicate data interpretation. On the other hand, thedifferences in animal species or strainmay play a role in thesediscrepancies. For instance, strain differences in mice havebeen recorded using the FST (Bai et al., 2001). It has beensimilarly reported that the FSTmodel efficacy varies betweendifferent strains of rats, in that the immobility responserecorded in Sprague–Dawley was significantly smaller thanthe immobility recorded in the Wistar–Kyoto rat strainfollowing different antidepressant therapy (Lopez-Rubalcavaand Lucki, 2000).

The major challenge in the treatment of bipolar depres-sion is the tendency for antidepressant medications toprecipitate episodes of mania, or to increase cycle frequencyor symptom intensity (Boerelin et al., 1998; Peet, 1994).Thus, exploring the utility of mood stabilizers such as lithiumas monotherapy for bipolar depression or their combinationswith agents that have no such side effects is important. Afterit was introduced into clinical use, lithium has been shown tohad a major impact on the treatment of bipolar disorder(Solomon et al., 1995). On the other hand, it has beenreported that depressed patients with a previous history ofmania or hypomania are more likely to respond to lithiumthan patients without such previous episodes (Soares andGershon, 1998). Depression associated with bipolar disorderhas been regarded as amore severe depression, whichmay bemore difficult to treat, and lithium treatment is beneficial(Freeman and Freeman, 2006; Baron et al., 1975; Mendelset al., 1972; Bourin and Prica, in press). In addition, somestudies found lithium effective for unipolar depression(Soares and Gershon, 1998; Greil and Kleindienst, 2003;Mendels et al., 1972; Khan et al., 1987). Overall, the responserates reported for lithium in the acute treatment of unipolardepression have been about 30 to 40%,which is lower than therates for bipolar depression or the rates with antidepressants(Goodwin and Jamison, 1990). Moreover, some studies haveshown that when lithium is added the ineffective primaryantidepressant drug, about 50 percent of patients withrefractory depression have a response, usually within one tothree weeks (Heninger et al., 1983). Improvement may occurwithin 48 h (Coryell et al., 1992), but this is uncommon.According to the present data, we showed that acute lithiumadministration had antidepressant-like effects in the Porsoltmouse FST. However, since clinical use for improving the

unipolar depression or the depressive facet of bipolardisorders requires nearly prolonged treatment, more studiesexamining the effects of chronic lithium administration in themouse FSTare warranted.

Although the FST has predictive validity as a model ofantidepressant activity (Cryan et al., 2002), it also has somedrawbacks represented by the possibility of obtaining somefalse positives or negatives (Borsini and Meli, 1988). Drugsenhancing motor activity may give a “false” positive effect inthe forced swimming test. Therefore, this test would not beas a good test for antidepressants such as bupropion, nomi-fensine, and amineptine, since these agents increase motoractivity (Borsini and Meli, 1988). In the present study, activedoses of lithium in the FST had no effect on the locomotoractivity of mice in the open field test. Therefore it excludesthe possibility that the antidepressant-like effects of lithiumin the FSTmight be due to its effect on the locomotor activityin mice. This result is in agreement with previous studies(Nixon et al., 1994; Wegener et al., 2003).

Some studies have demonstrated that NOS inhibitionproduces anxiolytic and antidepressant-like behaviouraleffects in a variety of animal paradigms including the originalrat and mouse FSTs (Harkin et al., 1999, 2004; Inan et al.,2004; Jefferys and Funder, 1996; Yildiz et al., 2000). Some ofthese effects are overcome by administration of the NOprecursor L-arginine, consistent with an involvement of NO inthis behavioural response (Harkin et al., 1999; Jefferys andFunder, 1996; Yildiz et al., 2000). Moreover, Harvey (1996) hasalso proposed a potential role for NO in affective disorders(Harvey, 1996). In agreement with previous studies, thepresent data showed that either the competitive nonspecificNOS inhibitor L-NAME, at 100 mg/kg, or the selective nNOSinhibitor L-NPA, at 30mg/kg, decreased the immobility time ofmice in the FST, indicating that these drugs in theseconcentrations have antidepressant-like effects in the FST.Also, our data indicated that these agents at active orsubactive doses had no effect on the locomotor activity inthe open field test.

In the present study, we examined the effect ofsimultaneous administration of lithium with L-NAME, L-NPA,ODQ, L-arginine and sildenafil in the FST. Our data show thatneither low concentrations of NOS inhibitors (L-NAME, L-NPA)nor low concentration of the guanylyl cyclase inhibitor ODQnor low concentrations of lithium, when administeredindependently, significantly affected the immobility timeof mice in the FST. However, when these agents and lithiumwere combined at the same low concentrations, theyexerted a significant antidepressant-like effect in the FST.This result may reflect the involvement of neuronal NO/cGMP pathway in the effects of lithium in the FST. Further,we demonstrated that the antidepressant-like effect of anactive dose of lithium in the FSTwas significantly preventedby coadministration of a dose of L-arginine or sildenafil whichper se did not alter the immobility time of mice in the FST. Inagreement with our data, it was previously reported that theadministration of a low and per se non-effective dose of L-NAME potentiated the inhibitory effects of low doses oflithium on morphine's biphasic modulation of susceptibilityto pentylenetetrazole-induced clonic seizure in mice (Honaret al., 2004). In the same study, L-arginine at a per se non-effective dose reversed the inhibitory effects of lithium(Honar et al., 2004). Furthermore, according to another

330 M. Ghasemi et al.

reported observation, while coadministration of lithium withL-NAME exerts a synergistic and potent inhibition ofmorphine withdrawal syndrome in mice, cotreatment withL-arginine decreases the inhibitory effect of lithium in thesame model (Dehpour et al., 2000b). Recently, we also haveshown that acute lithium administration decreased theneurogenic relaxation of rat corpus cavernosum and thisinhibitory effect of lithium was reversed by coadministrationof a non-effective dose of L-arginine and was potentiated byconcurrent administration of a low and non-effective dose ofL-NAME (Sadeghipour et al., 2007). Taken together andaccording to our study, the present findings may suggest arole for L-arginine/NO/cGMP pathway in the antidepressant-like effects of lithium in the mouse FST.

Concerning the possible effects of lithium on constitutiveNOS and its downstream signaling pathway, several differ-ences in results and methodology exist. Some studies haveindicated that chronic lithium treatment caused an en-hancement of nNOS activity in the brain and increased thenNOS gene expression in the hypothalamus and astrocytes(Anai et al., 2001; Feinstein, 1998; Bagetta et al., 1993).Using NADPH-diaphorase staining, we have recently shownthat both eNOS and nNOS were increased in corpuscavernosum of chronic lithium-treated rats (Sadeghipour etal., 2006). However, it was reported that lithium inhibitedmuscarininc receptor-mediated cGMP synthesis in mouseneuroblastoma clone and this effect appeared to be at thelevel of guanylyl cyclase, but not NO synthase (Shintani etal., 1994). Harvey et al. (1994) reported that chronic lithiumtreatment of rats significantly increased the corticalaccumulation of cGMP levels which was inhibited by L-NAME. They also failed to demonstrate any effect ofchemically relevant lithium plasma levels on plasma nitrogenoxides (NOx), the end-products of the NO metabolism,following chronic oral lithium therapy (Harvey et al.,1994). In contrast, using an in vivo brain microdialysismethod, it has been recently demonstrated that intraper-itoneally lithium administration (50 mg/kg) can significantlyreduce NOB3 and NOx levels in rat amygdala after 60 to90 min (Maruta et al., 2005). Moreover, Wegener et al. (2004)found that lithium inhibited the hippocampal NOS activity inrats in a concentration-dependent manner, although only thehighest dose (above 5 mM) was significantly lower thancontrol. In another study, Hashioka et al. (2007) haverecently demonstrated that a 24 h-pretreatment withlithium chloride could inhibit NO production by interfer-on-γ-activated microglia in a dose-dependent mannerreaching significance at a dose of 0.1 mM. Taking ourfindings together with these reports into consideration, itseems that NO is likely to play a role in the antidepressant-like effects of lithium in the mouse FST. However, furtherstudies, using different animal species and differentexperimental models, are needed to verify the exactmechanism of L-arginine/NO pathway in the antidepres-sant-like effects of lithium.

In summary, in the present study we found that acutelithium administration had antidepressant-like effect in themouse FST. Moreover, our data indicated that subactive dosesof lithium show a strong antidepressant effect together withlow doses of either L-NAME or L-NPA in the FST. Further, L-arginine, in concentration by which had no effect onimmobility time of mice in the FST, significantly prevented

the antidepressant-like effect of lithium. Therefore, it seemsthat the L-arginine/NO pathway may be involved in theantidepressant-like effect of lithium in the mouse FST.

Role of the funding source

Funding for this study was provided by Departments of Physiologyand Pharmacology. They had no further role in study design; in thecollection, analysis and interpretation of data; in the writing of thereport; and in the decision to submit the paper for publication.

Contributors

M. Ghasemi designed the study, wrote the protocol, and wrote themanuscript. M. Ghasemi and H. Sadeghipour performed theexperiments and undertook the statistical analysis. M. Ghasemiand A. Mosleh performed the additional experiments for providingthe revised manuscript. H. R. Sadeghipour and A. R. Dehpourmanaged the literature searches and analysis and A. R. Mani revisedthe manuscript written by M. Ghasemi.

Conflict of interest

The authors declare that they have no conflicts of interest.

Acknowledgement

The authors wish to express their thanks to Prof Kevin Moore (UCL)for his support during the preparation of the revised manuscript.

References

Anai, H., Ueta, Y., Serino, R., Nomura, M., Nakashima, Y.,Yamashita, H., 2001. Activation of hypothalamic neuronal nitricoxide synthase in lithium-induced diabetes insipidus rats.Psychoneuroendocrinology 26, 1009–1020.

Bagetta, G., Corasaniti, M.T., Melino, G., Paoletti, A.M., Finazzi-Agro,A., Nistico, G., 1993. Lithium and tacrine increase the expressionof nitric oxide synthase mRNA in the hippocampus of rat.Biochem. Biophys. Res. Commun. 197, 1132–1139.

Bai, F., Li, X., Clay, M., Lindstrom, T., Skolnick, P., 2001. Intra- andinterstrain differences in models of “behavioural despair”.Pharmacol. Biochem. Behav. 70, 187–192.

Baron, M., Gershon, E.S., Rudy, V., Jonas, W.Z., Buchsbaum, M., 1975.Lithium carbonate response in depression. Prediction by unipolar–bipolar illness, average-evoked response, catechol-O-methyl trans-ferase, and family history. Arch. Gen. Psychiatry 32, 1107–1111.

Bauer, M., Dopfmer, S., 1999. Lithium augmentation in treatment-resistant depression: meta-analysis of placebo-controlled stud-ies. J. Clin. Psychopharmacol. 19, 427–434.

Berridge, M.J., Irvine, R.F., 1989. Inositol phosphates and cellsignalling. Nature 341, 197–205.

Boerelin, H.L., Gitlin, M.J., Zoellner, L.A., Hammen, C.I., 1998.Bipolar depression and antidepressant-induced mania: a natu-ralistic study. J. Clin. Psychiatry 59, 374–379.

Borsini, F., 1995. Role of the serotonergic system in the forcedswimming test. Neurosci. Biobehav. Rev. 19, 377–395.

Borsini, F., Meli, A., 1988. Is the forced swimming test a suitablemodel for revealing antidepressant activity? Psychopharmacology94, 147–161.

Bourin, M., Prica, C., in press. The role of mood stabilizers in thetreatment of the depressive facet of bipolar disorders. Neurosci.Biobehav. Rev. [Electronic publication ahead of print].

Cade, J.F.J., 1949. Lithium salts in the treatment of psychoticexcitement. Med. J. Aust. 2, 349–352.

331NO involvement in the antidepressant-like effects of acute lithium administration in the mouse FST

Coryell, W., Endicott, J., Keller, M., 1992. Rapidly cycling affectivedisorder: demographics, diagnosis, family history, and course.Arch. Gen. Psychiatry 49, 126–131.

Cryan, J.F., Markou, A., Lucki, I., 2002. Assessing antidepressantactivity in rodents: recent developments and future needs.Trends Pharmacol. Sci. 23, 238–245.

Dehpour, A.R., Aghadadashi, H., Ghafourifar, P., Roushanzamir, F.,Ghahremani, M.H., Meysamee, F., Rassaee, N., Koucharian, A.,2000a. Effect of chronic lithium administration on endothelium-dependent relaxation in rat aorta. Clin. Exp. Pharmacol. Physiol.27, 55–59.

Dehpour, A.R., Sadr, S.S., Nouroddini, M., Shadan, F., Nourozi, A.,Farahani, M., Sahebgharani, M., 2000b. Comparison of simulta-neous administration of lithium with L-NAME or L-arginine onmorphine withdrawal syndrome in mice. Hum. Psychopharmacol.Clin. Exp. 15, 87–93.

Drummond, A.H., 1987. Lithium and inositol lipid-linked signalingmechanisms. Trends Pharmacol. Sci. 8, 129–133.

Esplugues, J.V., 2002. NO as a signalling molecule in the nervoussystem. Br. J. Pharmacol. 135, 1079–1095.

Feinstein, D.L., 1998. Potentiation of astroglial nitric oxide synthasetype-2 expression by lithiumchloride. J. Neurochem. 71, 883–886.

Freeman, M.P., Freeman, S.A., 2006. Lithium: clinical considerationsin internal medicine. Am. J. Med. 119, 478–481.

Goodwin, F.K., Jamison, K.R., 1990. Manic-Depressive Illness. OxfordUniversity Press, New York.

Greil, W., Kleindienst, N., 2003. Lithium in the long-term treatmentof bipolar disorders. Eur. Arch. Psychiatry Clin. Neurosci. 253,120–125.

Harkin, A., Bruce, K.H., Craft, B., Paul, I.A., 1999. Nitric oxidesynthase inhibitors have antidepressant-like properties in mice:1. Acute treatments are active in the forced swim test. Eur.J. Pharmacol. 372, 207–213.

Harkin, 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.

Harvey, B.E., 1996. Affective disorders and nitric oxide: a role inpathways to relapse and refractoriness? Hum. Psychopharmacol.11, 309–319.

Harvey, B.H., Carstens, M.E., Taljaard, J.J., 1994. Evidence thatlithium induces a glutamatergic: nitric oxide-mediated responsein rat brain. Neurochem. Res. 19, 69–474.

Hascoet, M., Bourin, M., Khimake, S., 1994. Additive effect oflithium and clonidine with 5-HT1A agonists in the forcedswimming test. Prog. Neuro-Psychopharmacol. Biol. Psychiatry18, 381–396.

Hashioka, S., Klegeris, A., Monji, A., Kato, T., Sawada, M., McGeer,P.L., Kanba, S., 2007. Antidepressants inhibit interferon-γ-induced microglial production of IL-6 and nitric oxide. Exp.Neurol. 206, 33–42.

Hata, T., Itoh, E., Nishikawa, H., 1995. Behavioural characteristics ofSART stressed mice in the forced swim test and drug action.Pharmacol. Biochem. Behav. 51, 849–853.

Heninger, G.R., Charney, D.S., Sternberg, D.E., 1983. Lithiumcarbonate augmentation of antidepressant treatment: an effec-tive prescription for treatment-refractory depression. Arch. Gen.Psychiatry 40, 1335–1342.

Honar, H., Riazi, K., Homayoun, H., Demehri, S., Dehghani, M.,Vafaie, K., Ebrahimkhani, M.R., Rashidi, N., Gaskari, S.A.,Dehpour, A.R., 2004. Lithium inhibits the modulatory effects ofmorphine on susceptibility to pentylentetrazole-induced clonicseizure in mice: involvement of a nitric oxide pathway. Brain Res.1029, 48–55.

Inan, S.Y., Yalcin, I., Aksu, F., 2004. Dual effects of nitric oxide in themouse forced swimming test: possible contribution of nitric oxidemediated serotonin release and potassium channel modulation.Pharmacol. Biochem. Behav. 77, 457–464.

Jefferys, D., Funder, J., 1996. Nitric oxide modulates retention ofimmobility in the forced swimming test in rats. Eur. J. Pharmacol.295, 131–135.

Johnson, G., 1974. Antidepressant effect of lithium. Compr.Psychiatry 15, 43–47.

Jope, R.S., 2003. Lithium and GSK-3: one inhibitor, two inhibitoryactions, multiple outcomes. Trends Pharmacol. Sci. 24, 441–443.

Jope, R.S., Williams, M.B., 1994. Lithium and brain signal transduc-tion systems. Biochem. Pharmacol. 47, 429–441.

Katona, C.L., 1988. Lithium augmentation in refractory depression.Psychiatr. Dev. 6, 153–171.

Kaster, M.P., Ferreira, P.K., Santos, A.R.S., Rodrigues, A.L.S.,2005. Effects of potassium channel inhibitors in the forcedswimming test: possible involvement of L-arginine-nitric oxide-soluble guanylate cyclase pathway. Behav. Brain Res. 165,204–209.

Khan, M.C., Wickham, E.A., Reed, J.V., 1987. Lithium versus placeboin acute depression: a clinical trial. Int. Clin. Psychopharmacol.2, 47–54.

Kitamura, Y., Araki, H., Gomita, Y., 2002. Influence of ACTH on theeffects of imipramine, desipramine and lithium on duration ofimmobility of rats in the forced swim test. Pharmacol. Biochem.Behav. 71, 63–69.

Kleindienst, N., Greil, W., 2003. Lithium in the long-term treatmentof bipolar disorders. Eur. Arch. Psychiatry Clin. Neurosci. 253,120–125.

Linder, J., Fyro, B., Pettersson, U., Werner, S., 1989. Acuteantidepressant effect of lithium is associated with fluctuationof calcium and magnesium in plasma. A doubleblind study on theantidepressant effect of lithium and clomipramine. Acta Psy-chiatr. Scand. 80, 27–36.

Lopez-Rubalcava, C., Lucki, I., 2000. Strain differences in thebehavioural effects of antidepressant drugs in the rat forcedswimming test. Neuropsychopharmacol. 22, 191–199.

Lucki, I., Dalvi, A., Mayorga, A.J., 2001. Sensitivity to the effects ofpharmacologically selective antidepressants in different strainsof mice. Psychopharmacology 155, 315–322.

Maruta, S., Suzuki, E., Yokoyama, M., Sato, T., Inada, K., Watanabe,S., Miyaoka, H., 2005. Effects of intraperitoneally injectedlithium, imipramine and diazepam on nitrate levels in ratamygdale. Psychiatry Clin. Neurosci. 59, 358–361.

Mendels, J., Secunda, S.K., Dyson, W.L., 1972. A controlled study ofthe antidepressant effect of lithium carbonate. Arch. Gen.Psychiatry 26, 154–157.

Nahorski, S.R., Ragan, C.I., Challiss, R.A., 1991. Lithium and thephosphoinositide cycle: an example of uncompetitive inhibitionand its pharmacological consequences. Trends Pharmacol. Sci.12, 297–303.

Nixon, M.K., Hascoet, M., Bourin, M., Colombel, M.C., 1994.Additive effects of lithium and antidepressants in the forcedswimming test: further evidence for involvement of theserotoninergic system. Psychopharmacology (Berl) 115, 59–64.

O'Brien, W.T., Harper, A.D., Jové, F., Woodget, J.R., Maretto, S.,Piccolo, S., 2004. Glycogen synthase kinase-3β haploinsuffi-ciency mimics the behavioural and molecular effects of lithium.J. Neurosci. 6791–6798.

Paoletti, A.M., Piccirilli, S., Costa, N., Rotiroti, D., Bagetta, G.,Nistico, G., 1998. Systemic administration of N omega-nitro-L-arginine methyl ester and indomethacin reduces the elevation ofbrain PGE2 content and prevents seizures and hippocampaldamage evoked by LiCl and tacrine in rat. Exp. Neurol. 149,349–355.

Peet, M., 1994. Induction of mania with selective serotonin reuptakeinhibitors and tricyclic antidepressants. Br. J. Psychiatry 164,549–550.

Porsolt, R.D., Bertin, A., Jalfre, M., 1977. Behavioural despair inmice: a primary screening test for antidepressants. Arch. Int.Pharmacodyn. Ther. 229, 327–336.

332 M. Ghasemi et al.

Price, L.H., Heninger, G.R., 1994. Lithium in the treatment of mooddisorders. N. Engl. J. Med. 331, 591–598.

Redrobe, J.P., Bourin, M., 1999a. The effect of lithium administra-tion in animal models of depression: a short review. Fundam.Clin. Pharmacol. 13, 293–299.

Redrobe, J.P., Bourin, M., 1999b. Evidence of the activity of lithiumon 5-HT1B receptors in the mouse forced swimming test:comparison with carbamazepine and sodium valproate. Psycho-pharmacology 141, 370–377.

Sadeghipour, H., Ghasemi, M., Nobakht, M., Ebrahimi, F., Dehpour,A.R., 2006. Effect of chronic lithium administration on endothe-lium-dependent relaxation of rat corpus cavernosum: the role ofnitric oxide and cyclooxygenase pathways. BJU Int. 99, 177–182.

Sadeghipour, H., Ghasemi, M., Ebrahimi, F., Dehpour, A.R., 2007.Effect of lithium on endothelium-dependent and neurogenicrelaxation of rat corpus cavernosum: role of nitric oxide pathway.Nitric Oxide 16, 54–63.

Shaldubina, A., Agam, G., Belmaker, R.H., 2001. The mechanism oflithium action: state of the art, ten years later. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 25, 855–866.

Sherman, A.D., Petty, F., 1980. Neurochemical basis of the action ofantidepressants on learned helplessness. Behav. Neural Biol. 30,119–134.

Shintani, F., Kana, S., Nakai, T., Nakamura, R., Sato, K., Yagi, G.,Richelson, E., Kato, R., Asai, M., 1994. Inhibition by lithium ofcyclic GMP formation without inhibition of nitric oxide genera-tion in the mouse neuroblastoma cells (N1E-115). Neuropsycho-pharmacol. 11, 119–124.

Soares, J.C., Gershon, A., 1998. The lithium ion: a foundation forpsychopharmacological specificity. Neuropsychopharmacol. 19,167–182.

Solomon, D.A., Keitner, G.I., Miller, I.W., Shea, M.T., Keller, M.B.,1995. Course of illness and maintenance treatments for patientswith bipolar disorder. J. Clin. Psychiatry 56, 5–13.

Souza, F.G.M., Goodwin, G.M., 1991. Lithium treatment and pro-phylaxis in unipolar depression: a meta-analysis. Br. J. Psychiatry158, 666–675.

Tomasiewicz, H.C., Mague, S.D., Cohen, B.M., Carlezon, W.A., 2006.Behavioural effects of short-term administration of lithium andvalproic acid in rats. Brain Res. 1093, 83–94.

Watanabe, S., Ishino, H., Otsuki, S., 1974. Lithium and affectivepsychoses. Folia Psychiatr. Neurol. Jpn. 28, 267–305.

Watanabe, S., Ishino, H., Otsuki, S., 1975. Double-blind comparisonof lithium carbonate and imipramine in treatment of depression.Arch. Gen. Psychiatry 32, 659–668.

Weerasinghe, G.R., Rapoport, S.I., Bosetti, F., 2004. The effect ofchronic lithium on arachidonic acid release and metabolism in ratbrain does not involve secretory phospholipase A2 or lipoxygen-ase/cytochrome P450 pathways. Brain Res. Bull. 63, 485–489.

Wegener, G., Bandpey, Z., Heiberg, I.L., Mork, A., Rosenberg, R.,2003. Increased extracellular serotonin level in rat hippocampusinduced by chronic citalopram is augmented by subchroniclithium: neurochemical and behavioural studies in the rat.Psychopharmacology 166, 188–194.

Wegener, G., Bandpey, Z., Heiberg, I.L., Volke, V., Trabace, L.,Rosenberg, R., Harvey, B.H., 2004. Combined chronic treatmentwith citalopram and lithium does not modify the regionalneurochemistry of nitric oxide in rat brain. J. Physiol. Pharmacol.55, 575–586.

Worrall, E.P., Moody, J.P., Peet, M., Dick, P., Smith, A., Chambers,C., Adams, M., Naylor, G.J., 1979. Controlled studies of the acuteantidepressant effects of lithium. Br. J. Psychiatry 135, 255–262.

Yildiz, F., Erden, B.F., Ulak, G., Utkan, T., Gacar, N., 2000.Antidepressant-like effect of 7-nitroindazole in the forcedswimming test in rats. Psychopharmacology (Berl.) 149, 41–44.