Epilepsy Research (2014) 108, 1711—1718

jo ur nal ho me p ag e: www.elsev ier .com/ locate /ep i lepsyres

Role of CB2 receptors and cGMP pathway onthe cannabinoid-dependent antiepilepticeffects in an in vivo model of partialepilepsy

Valerio Rizzoa,b,∗,1, Fabio Carletti a,1, Giuditta Gambinoa,Girolamo Schieraa, Carla Cannizzaroc, Giuseppe Ferraroa,Pierangelo Sardoa

a Dipartimento di Biomedicina Sperimentale e Neuroscienze Cliniche (Bio.Ne.C.), Sezione di Fisiologiaumana ‘‘G. Pagano’’, Università degli Studi di Palermo, Corso Tukory, 129—90134 Palermo, Italyb Department of Neuroscience, The Scripps Research Institute, Scripps Florida 130 Scripps Way, Jupiter, FL33458c Dipartimento di Scienze per la Promozione della salute, Università degli Studi di Palermo, Via del Vespro,133, 90100 Palermo, Italy

Received 4 June 2014; received in revised form 12 September 2014; accepted 1 October 2014Available online 19 October 2014

KEYWORDSCannabinoid;Temporal lobeepilepsy;AM630;sGC;Hippocampus;Electrophysiology

Summary This study aimed at providing an insight on the possible role of cannabi-noid (CB) type 2 receptors (CB2R) and cGMP pathway in the antiepileptic activity ofWIN 55,212-2, (R)-(+)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl) pyrrolo[1,2,3-de]-1,4-benzoxazin-6-Yl]-1-naphthalenylmethanone, a non-selective CB agonist, in the maximal dentateactivation (MDA) model of partial epilepsy in adult male rats. We evaluated the activityof a CB2 antagonist/inverse agonist AM630, [6-iodo-2-methyl-1-[2-(4-morpholinyl)ethyl]-1H-indol-3-yl](4-methoxyphenyl)methanone or 6-iodopravadoline, alone or in co-administrationwith WIN 55,212-2. Also, in the MDA model it was investigated the co-treatment of WIN

55,212-2 and 1H-[1,2,4]Oxadiazole[4,3-a]quinoxalin-1-one (ODQ), a specific inhibitor of the nitric oxide (NO)-activated soluble guanylyl cyclase (sGC), the cGMP producing enzyme. TheWIN 55,212-2-dependent (21 mg/kg) antiepileptic effects were significantly increased by theco-administration with AM630 and by the co-treatment with ODQ (10 mg/kg). Whereas, theadministration of AM630 (2 mg/kg), alone exerts no effects on hippocampal hyperexcitabi-lity. Our data show that pharmacological blockade of CB2 receptors and of sGC seems to

∗ Corresponding author. Tel.: +39 091 655 58 06; fax: +39 091 655 58 16.E-mail address: valerio.rizzo@unipa.it (V. Rizzo).

1 These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.eplepsyres.2014.10.0010920-1211/© 2014 Elsevier B.V. All rights reserved.

1712 V. Rizzo et al.

cooperate with WIN in its antiepileptic action. These findings shed light on CB signaling mecha-nisms, hinting that the modulation of the effects of CB agonist in the hyperexcitability phenomenamay be exerted both by targeting CB receptors and their possible downstream effectors, such asnitrergic-dependent cGMP pathway.© 2014 Elsevier B.V. All rights res

I

Srbpm2tt2eaatgeli2Ctr(

io(siea(eepcJwnaesenaerrroes

cta

tocrowea2MutFbnsoi

M

A

Mub2(ale(drw(Aa(sm

with the current Italian rules on animal experimentation(D.L. 116/92) and European directive (2010/63/EU).

Maximal dentate gyrus activation and ictal events

ntroduction

everal evidences have outlined that in the brain neu-onal excitability and synaptic function may be modulatedy either endogenous cannabinoid (eCB) signaling or cGMPathway, for instance through the control on neurotrans-itter release and ion channels conductance (Ahern et al.,

002; Castillo et al., 2012; Robello et al., 1996). Besides this,he eCB system and cGMP signaling are reported to be func-ionally related in certain neuronal paradigms (Azad et al.,001; Ghasemi et al., 2007; Howlett et al., 2004; Stefanot al., 1998). Indeed, guanine nucleotides can inhibit CBgonist binding (Devane et al., 1988), whereas cannabinoidgonists can stimulate both the production of cGMP andhe translocation of the nitric oxide (NO)-activated solubleuanylyl cyclase (sGC), the cGMP-producing enzyme (Jonest al., 2008). In the hippocampus, the distribution and co-ocalization of sGC and CB receptors have been describedn the pre-synaptic glutamatergic afferents (Burette et al.,002). In addition, anatomical evidences have proven thatA1 inhibitory synapses equipped with cannabinoid receptorype 1 (CB1R) receptors express both postsynaptic neu-onal NO synthase (nNOS) and presynaptic NO-activated sGCMakara et al., 2007).

As for the hyperexcitability phenomena, a functionalnvolvement of the G protein-coupled CB1R has beenbserved in the cannabinoid-mediated antiepileptic effectsHofmann and Frazier, 2013; Matsuda et al., 1990). In fact,ystemic cannabinoid treatment suppresses seizures andncreases activation threshold in epileptic rats (Wallacet al., 2003), whereas CB1R antagonists cause seizure-likectivity in hippocampal culture models of acquired epilepsyDeshpande et al., 2007) and may exacerbate paroxysmalvents in patients with temporal lobe epilepsy (Braakmant al., 2009). Though, growing evidences suggest that theathways involved in the cannabinoid antiepileptic effectsould not be exclusively CB1-mediated (Hill et al., 2013;ones et al., 2012). In this regard, in a previous papere accounted for the CB1 antagonist AM251 ineffective-ess when administered alone using the maximal dentatectivation (MDA) model of hippocampal epilepsy (Rizzot al., 2009); on the other hand, we reported that AM251ignificantly but incompletely antagonized the antiepilepticffects of WIN 55,212-2 (CB non-selective agonist, hereafteramed WIN). The finding of a partial antagonism exerted by

selective CB1 antagonist on WIN-dependent antiepilepticffect likely implies the functional involvement of a furthereceptor underpinning CB antiepileptic effects. For thiseason, in this paper we firstly focused on the cannabinoid

eceptor type 2 (CB2R), a protein associated with a varietyf brain processes (Cabral et al., 2008; Fernández-Ruizt al., 2007; Morgan et al., 2009; Van Sickle et al., 2005),uch as the decrease of neuronal firing in the prefrontal

i

Ia

erved.

ortex (den Boon et al., 2012). Therefore, we explored ifhe control by CB2R may be found on paroxysmal neuronalctivity in the MDA model.

Beyond this, irrespective of the involvement of CB recep-ors, the pathway through which eCB exert their effectsn neuronal processes is not completely clear especiallyoncerning the interplay with other neuromodulators. Asegards the epileptic condition, we previously investigatedn the intervention of the cGMP pathway within the frame-ork of nitrergic modulation of hippocampal seizures (Sardot al., 2006). In particular, we provided data that the block-de of the sGC exerted antiepileptic effects (Sardo et al.,006). In this light, in the present study, we exploited theDA model of partial epilepsy to deepen knowledge on thenderlying mechanisms of the antiepileptic effects of CBransmission and on its possible interaction with NO/cGMP.or the above described purposes, we separately targetedoth CB2 receptors, via administration of the CB2 antago-ist/inverse agonist AM630, and sGC, by administering thepecific inhibitor 1H-[1,2,4]oxadiazole[4,3-a]quinoxalin-1-ne (hereafter named ODQ), to assess if significant changesn the WIN-induced antiepileptic effects occur.

aterials and methods

nimals and surgical procedures

ale Wistar rats (weight 260—300 g, 2—3 months-old) weresed in this study. Detailed surgical procedures haveeen described in our previous papers (Carletti et al.,013). Briefly: rats were anaesthetized with urethane1.0—1.2 g/kg intraperitoneally, i.p.) (Maggi and Meli, 1986)nd after craniotomy, a stimulating electrode (coaxial bipo-ar stainless steel electrode: external diameter 0.5 mm;xposed tip 25—50 �m) was placed in the angular bundleAB) on the right side according to the stereotaxic coor-inates of the Atlas of Paxinos and Watson (1986). A glassecording electrode, filled with 1% fast Green in 2 M NaCl,as stereotaxically placed in the ipsilateral dentate gyrus

DG). The animal was grounded through a subcutaneousg/AgCl wire in the scapular region. The DG bioelectricctivity was recorded through a low level DC pre-amplifierGrass 7B, West Warwick, RI, USA) and then processed by aoftware package provided by DataWave Technologies (Long-ont, CO, U.S.A.).The experiments were conducted in strict accordance

dentification

n order to obtain stable and reproducible MDA as wells to avoid progressive changes in their duration due

Role of CB2 receptors and cGMP pathway on the cannabinoid-dependent antiepileptic effects 1713

Fig. 1 (A) Representative MDA trace. Measurements of time of onset, duration of maximal dentate gyrus activation (MDA) andafterdischarge (AD) during and after a stimulus train (400 �A, 20 Hz) delivered for 10 s to the angular bundle (AB). The recordings

rse

N). (*

w(uFue

D

DLfc(wv

were amplified using a low level DC pre-amplifier. (B) Time couvalue represents the mean of D% of each group (� Control, � WI

to repetitive AB stimulations, we modified the techniqueoriginally described by Stringer and Lothman (1989) char-acterized by variable stimulation train durations strictlyrelated to the beginning of the MDA response. Our elec-trophysiological technique has been extensively describedin our previous papers (Carletti et al., 2013). Briefly: 10 sduration trains of 20 Hz stimuli were given through theAB stimulating electrode. Individual stimuli consisted of0.3 msec biphasic pulses. The stimulus intensity was ini-tially below that necessary to elicit any response and it wasincreased in 100 �A steps in the following stimulations untilMDA occurred (threshold intensity). The stimulus train wasadministered every 2 min until a MDA appeared and thenevery 10 min for up to 3 h. MDA was recorded by the elec-trode placed in the DG and it was defined by a shift of theextracellular potential in DC-coupled recordings as well asby the presence of bursts of population spikes. Once the

MDA was elicited, the following parameters were recorded:(i) Onset duration (time from the beginning of AB stimulationto the midpoint of the DC potential shift); (ii) MDA duration(from the midpoint of the DC potential shift to the point at

rtit

of MDA parameters in controls and WIN treated animals. Each) indicates a significant difference vs baseline values (P < 0.05).

hich the evoked paroxysmal EEG events abruptly ceased);iii) afterdischarge (AD) duration (from the end of AB stim-lation to the end of the epileptiform activity) (Fig. 1A).urthermore, we analyzed the % of responses to AB stim-lation to evaluate the possible suppression of paroxysmalvents.

rug treatment

MSO was purchased from Sigma Chemical Co. (Sigma, St.ouis,MO, USA), while WIN, AM630 and ODQ were purchasedrom Tocris Bioscience (Bristol, UK). The study took intoonsideration six groups of rats (n = 6 rats each). The 1stuntreated controls) and 2nd groups (vehicle treated)ere studied for a period of about 200 min in order toerify possible modifications of MDA parameters due to the

epetitive stimulations or to the vehicle administration. Inhe remaining groups the animals received WIN (21 mg/kg,.p., 3rd group), AM630 (2 mg/kg, i.p., 4th group), a co-reatment with AM630 and WIN (2 mg/kg and 21 mg/kg, i.p,

1

rOa2d(w(lgedatWObs

H

Ietsttw1issl

S

Trttoa

foeMgttMNwo

ebioec

oIpmpttscnai6tmoDw

R

C

Iiwtrto(

Mt

At

rhAcatwDirraot

tf

r

714

espectively; 5th group) and a co-treatment with WIN andDQ (21 mg/kg and 10 mg/kg, i.p, respectively; 6th group),t dosages previously described (García-Gutiérrez et al.,012; Rizzo et al., 2009; Sardo et al., 2006). All drugs wereissolved in the same final vehicle volume for each animal300 �l of DMSO). Each single pharmacological treatmentas performed after five consecutive stable MDA responses

baseline period) and the subsequent observation periodasted 120 min after the drug injection (2nd, 3rd and 4throup). In the last two groups, receiving two treatmentsach, due to different pharmacokinetic profiles of therugs administered, an interval was interposed betweendministrations so as to allow coincident actions. In fact,he 5th group was pretreated with AM630 30 min beforeIN injection, and the 6th group was administered withDQ 30 min after receiving pretreatment with WIN. Foroth co-treated groups the observational period after theecond drug administration lasted 120 min.

istology

n our histological procedure, recording and stimulatinglectrode positions were verified and marked through ion-ophoretic Fast Green injection (50 �A for 10 min) and amall electrolytic lesion (20 mA for 10 s), respectively. Athe end of each experiment, the animals were anaes-hetized by an overdose of pentobarbital i.p., then theyere whole-body perfused with normal saline followed by0% buffered formalin. The brains were removed, postfixedn the same fixative overnight and then cryoprotected in 30%ucrose/PBS. Subsequently, brains were sliced in 30—50 �merial coronal sections and stained by using Nissl-cresyl vio-et method (Sardo et al., 2008).

tatistical analysis

he chi-square (X2) test was used to compare the % ofesponses to AB electrical stimulation following each drugreatment within the same experimental group. A between-reatments X2 test was used to compare the % of responsesf 4th, 5th and 6th group (5th and 6th after second drugdministration) with WIN treated group.

In order to normalize individual data, within a group,or each studied parameter (the duration of onset, MDAr AD), data from each animal were expressed as % differ-nce (D%) versus the baseline values represented by the lastDA responses of the period preceding vehicle (in the 2ndroup) or single drugs (in 3th and 4th groups) administra-ion. As regards the 5th and 6th group, data collected afterhe second drug administration were compared to the lastDA response of either the baseline period or pretreatment.evertheless, data and related graphs of 5th and 6th groupsere referred to the comparison with the last MDA responsef baseline period.

Then, in order to study the time course of effects, inach group D% data were averaged (mean ± S.D.) on theasis of the time elapsed from the first stimulation follow-

ng a single or combined pharmacological treatment (timef stimulus: 10 min for 1st stimulus, 20 min for 2nd stimulus,tc.). The time course of response parameters in untreatedontrols and vehicle-treated animals was analyzed using a

ta1D

V. Rizzo et al.

ne-way repeated measures analysis of variance (ANOVA).n drug-treated animals, post-pharmacological treatmentarameters were statistically analyzed using an one-wayultivariate ANOVA test (MANOVA) followed by Bonferroniost-hoc test. The factor studied in this analysis was theime elapsed from drug administration, with 13 levels (theime 0 control value plus the twelve post administrationtimulations). The occurrence of null MDA response (andonsequently the lack of related response parameters) didot allow the use of a repeated measures MANOVA. The samenalysis was used for a further between-treatments compar-son to assess the effects of the treatments of 4th, 5th andth groups with respect to WIN group. The factor studied inhis analysis was the treatment, with 4 levels (each treat-ent group). In this study, the variance ratio and the degrees

f freedom (DF) are indicated by F(DF among groups, DF within groups).ifferences were considered statistically significant when Pas less than 0.05.

esults

ontrol, vehicle and WIN-treated groups

n untreated controls and vehicle-treated groups, repet-tive AB stimulations always induced a MDA responsehose parameters were not altered along the experimen-

al observation period. The WIN treatment reduced the % ofesponses and also significantly influenced the MDA parame-ers, increasing the onset time and shortening the durationf both the MDA and AD, with respect to control groupFig. 1B), as previously reported (Rizzo et al., 2009).

odulation of WIN effects by AM630 and ODQ onhe number of MDA responses

comprehensive bar graph showing the effects of eachreatment on MDA responses is reported in Fig. 2.

The co-treatment AM630-WIN enhanced WIN effect ineducing the % of responses to AB stimulation, as describedenceforth. In particular, a within-treatment analysis onM630-WIN group displayed significant changes on the per-entage of responses to AB stimulation. Indeed, data showed

marked decrease from 30th to 60th minutes and from 100tho 110th minutes, with a maximal effect at 30th minutehen no animals exhibited any MDA response (�2 = 12.000,F = 1, P = 0.0005). A further analysis on the % of responses

n AM630-WIN group showed significant differences withespect to WIN alone: in fact, the co-treatment continuouslyeduced the % of responses from 30th to 70th minutes, with

significant decrease at 3rd stimulus (maximal reductionbserved: 50%, �2 = 4.000, DF = 1, P = 0.0455), with respecto WIN alone.

The assessment of AM630 alone proved to be ineffec-ive on MDA responses since AB stimulations were alwaysollowed by DG activation in the observation period.

The co-treatment ODQ-WIN enhanced WIN effect ineducing the % of responses to AB stimulation. The within-

reatment �2 test revealed that in ODQ-WIN group there was

clear reduction of the % of responses from 10th minutes to20th minutes, with a maximal effect of −83.3% (�2 = 8.571,F = 1, P < 0.01) at all the stimulations where the number of

Role of CB2 receptors and cGMP pathway on the cannabinoid-dep

Fig. 2 The bar graph shows the percentage of responses tothe AB stimulation recorded in the experimental groups, asindicated in the legend: WIN, AM630-WIN, AM630 and ODQ-WIN, during the twelve progressive stimuli. Within-treatmentstatistically significant differences were indicated for P < 0.05(*), P < 0.01 (**) or P < 0.001 (***). (◦) Indicates a significant dif-ference between the effects of AM630-WIN treatment versus

pcop

COptfMwPPPiAotwaftp

eppetefWow−TsArtoPo1oPctsrt

D

A

WIN (P < 0.05). () Indicates a significant difference between theeffects of ODQ-WIN treatment versus WIN (P < 0.05).

responses was equal to 1. A further comparison on the %of responses in ODQ-WIN group with respect to WIN aloneshowed significant reductions of % of responses at 10th,70th and 80th minutes (in all the stimulations the reductionobserved was: 66.6%; �2 = 6.000, DF = 1, P = 0.0143 at 10thminute; �2 = 5.33, DF = 1, P = 0.02 at 70th and 80th minutes).

Modulation of WIN effects by AM630 and ODQ onthe MDA parameters

Comparisons within each treatment (Fig. 3)In the group receiving pre-treatment with AM630 followedby WIN administration, one-way MANOVA revealed a signifi-cant multivariate main effect for stimulation time (Wilks’� = 0.275, F(33, 98) = 1.632, P = 0.034). Moreover, significantunivariate main effects for stimulation time were obtainedfor all parameters (Onset, F(11, 35) = 2.335, P = 0.028; MDAduration, F(11, 35) = 3.300, P = 0.00348; AD duration, F(11,35) = 3.272, P = 0.0036). Post hoc analysis revealed thatAM630-WIN induced an increase in the onset parameterfrom 60th to 120th minutes, with a maximum effect of+86.05 ± 10.22% at 70th minute (P = 0.0003). Moreover, areduction of MDA duration was evident from 60th to 120thminutes, with a maximum effect of −85.49 ± 3.92% at 70thminute (P = 0.0004). Similarly, the co-treatment induced areduction in the AD duration from 60th to 120th minutes,with a maximum effect of −87.04 ± 3.20% at 70th minute(P = 0.0011). In contrast, one-way MANOVA for stimulationtime did not reveal a significant multivariate main effectfor stimulation time for the group treated with AM630 alone(Wilks’ � = 0.496, F(36, 187) = 1.391, P = 0.0830), when com-

pared to baseline period. Moreover, significant univariatemain effects for stimulation time were not obtained forany parameter in AM630 group. As for the comparison ofthe effects of ODQ-WIN versus baseline period on the MDA

tcei

endent antiepileptic effects 1715

arameters, the data did not reach an adequate statisti-al outcome due to the massive ODQ-WIN-induced reductionf the % of responses (n = 1 or 2 at most stimulation timeoints).

Fig. 3

omparisons between treatments (Fig. 3)ne-way (treatment, four levels) MANOVA performed byooling the data of all time stimulation points for eachreatment revealed a significant multivariate main effector treatment (Wilks’ � = 0.399, F(9, 424) = 21.591, P < 0.0001).oreover, significant univariate main effects for timeere obtained for all parameters (Onset, F(3, 176) = 20.139,

< 0.0001, power = 100; MDA duration, F(3, 176) = 22.929, < 0.0001, power = 100; AD duration, F(3, 176) = 16.737, < 0.0001, power = 100). Post hoc analysis showed signif-cant differences between the treatments: in particular,M630-WIN co-treatment significantly increased the meannset time versus WIN treatment (from 13.49 ± 2.9%o 56.61 ± 4.63% for AM630 + WIN, +319.64%, P < 0.0001),hereas MDA and AD duration were not significantlyffected. Remarkably, significant differences were observedor these two parameters between AM630 and all the otherreatments, the latter inducing marked reduction in thearameters duration.

In order to further explore the time course of theffects, a one-way (treatment, four levels) MANOVA waserformed by analyzing data for each stimulation timeoint. This analysis revealed a significant multivariate mainffect for treatment for each stimulation with the excep-ion of 3rd, 5th, 8th and 12th. Significant univariate mainffects were observed at different time points, and wereurther highlighted by post hoc test: in particular, AM630-IN co-treatment showed a significant increase of the time

f onset versus WIN alone, from 60th to 110th minutes,ith the maximal effect at 110th minute of +80.56% (from8.76% ± 15.2 to +71.8 ± 13.9%; P = 0.0003; F(1,6) = 59.236).he same analysis for MDA and AD durations did not revealignificant differences. Furthermore, the comparison ofM630- versus WIN-induced effects showed a significanteduction of the mean MDA duration in WIN group from 40tho 100th minutes, with a maximum effect at 60th minutef −54.79% (from −15.6% ± 23.82 to −70.39% ± 13.70;

= 0.0114; F(1,9) = 10.033). Lastly, a significant decreasef AD duration was shown in WIN group from 40th to00th minutes, with a maximum effect at 90th minutef −85.13% (from +17.65 ± 42.51% to −67.48 ± 10.86%;

= 0.007; F(16,51) = 13.662), compared to AM630. As for theomparison of the effects of ODQ-WIN versus WIN alone onhe MDA parameters, the data did not reach an adequatetatistical outcome due to the massive ODQ-WIN-inducededuction of the % of responses (n = 1 or 2 at most stimulationime points).

iscussion

functional involvement of CB transmission in the modula-

ion of paroxysmal events has been widely investigated, butlear and definitive conclusions are not still available. Sev-ral studies targeted the CB1 receptors in an attempt to clar-fy the mechanism underlying the possible CB modulation of

1716 V. Rizzo et al.

Fig. 3 Effects of AM630 and AM630-WIN on the time course of MDA parameters during the 12 progressive stimuli. Each valuerepresents the mean of D% of each treatment (� AM630 or � AM630-WIN) per stimulus versus baseline values. The symbol NRindicates no response to the stimulation (3rd stimulus of AM630-WIN group). Within-treatment statistically significant D% is indicatedf * nt di ◦

W grou

httgtosrk(2htetmnTeeAmsdCbaiWCoap2

iWesrrstmerbiwcueFWaf

tNAdagpI

or P < 0.05 ( ) vs baseline values. Between-treatment significaIN group. Between-treatment significant differences in AM630

yperexcitability processes. As a matter of fact, CB1 recep-ors are linked to an inhibitory G-protein modulating, amonghe others, A-type K+ channels and N and P/Q type voltage-ated Ca2+ currents so as to stabilize the membrane poten-ials (Deadwyler et al., 1995; Pan et al., 1996). This actionn membrane conductance may determine the CB-mediateduppression of presynaptic neurotransmitter release, eitheregarding Glutamate or GABA release, a neuronal processnown as depolarization-induced suppression of excitationDSE) or inhibition (DSI), respectively (Kreitzer and Regehr,001; Wilson and Nicoll, 2001). In detail, CB1-mediated DSEas been hypothesized to be involved in the reduction ofhe seizure discharge in hippocampal cultures (Deshpandet al., 2007). Thus, it is conceivable that the antiepilep-ic properties of CB agonists may consists of a modulationainly directed towards the inhibition of the glutamatergic

eurotransmission than GABA release (Monory et al., 2006).hough, in the MDA experimental rat model of human partialpilepsy, the lack of a complete blockade of WIN-inducedffect after pre-treatment with a selective CB1 antagonist,M251, (Rizzo et al., 2009) raised the possibility of a furtherechanism for CB-mediated modulation of hippocampal

eizures. In this regard, multiple evidences report a wideistribution of CB2 in neuronal and glial cells in severalNS areas such as cerebral cortex, hippocampus, thalamus,rain stem and cerebellum (Gong et al., 2006), suggesting

potential implication of CB2 receptors, besides CB1R role,n mediating CB signaling (García-Gutiérrez et al., 2012).ith the aim of exploring the possible modulatory role ofB2 on the antiepileptic action exerted by WIN 55,212-2

n the MDA model, we administered the antagonist/inversegonist, AM630 (Bolognini et al., 2012), known for its highotency and affinity for rat CB2 receptors (Mukherjee et al.,004). Our results showed that, when compared to the effect

OWri

fferences in AM630-WIN group are indicated for P < 0.05 ( ) vsp are indicated for P < 0.05 (#) vs WIN group.

nduced by WIN alone, the co-treatment with AM630 andIN significantly reduced the severity of ictal events and,

ven more, the percentage of responses to the stimulation,uggesting that AM630 improves WIN efficacy. A possibleeduced proneness to the epileptogenic phenomena, asevealed by the increase of onset time, is not associated toignificant differences in MDA and AD parameters betweenhe AM630-WIN and WIN groups. Taken together, these dataight suggest that the efficacy of the co-treatment is

xerted mainly by augmenting seizure threshold in the DG,ather than on the epileptic discharge, once elicited. On theasis of the enhancement of WIN-induced effects follow-ng AM630 pretreatment, one group of animals was treatedith the CB2 antagonist/inverse agonist to assess its effi-acy when administered alone, but this treatment resultednexpectedly fruitless, suggesting that CB2 exerts no directffects on hippocampal hyperexcitability in MDA model.rom a pharmacological point of view, we hypothesize thatIN may have greater occupancy at CB1Rs when CB2Rs are

ntagonized by AM630, hence eliciting a better response byacilitating WIN selectivity on CB1-mediated pathway.

In order to analyze the downstream effectors of CB recep-ors activation in the phenomena examined, we focused onO/cGMP pathway by administering the sGC inhibitor, ODQ.s previously reported, ODQ alone induced a significantecrease of the severity of ictal events, but did not induceny change in the number of responses, in any case sug-esting a functional involvement of the NO/sGC metabolicathway in the DG paroxysmal activity (Sardo et al., 2006).n the present study, animals were co-treated with WIN and

DQ so as to assess if the inhibition of sGC may impact onIN antiepileptic properties. Our data showed a significant

eduction in the percentage of responses to AB stimulationnduced by WIN and ODQ co-treatment. Moreover, this fall

-dep

R

A

A

A

B

B

B

B

C

C

C

C

D

d

D

D

F

Role of CB2 receptors and cGMP pathway on the cannabinoid

in the percentage of responses was significantly enhanced,when compared to WIN alone. This impressive decrease ofMDA responses suggests an interplay between cGMP and CB1

controlling the excitability of DG neurons in experimentally-induced seizures. Providing evidences to the hypothesis ofthis co-role, physiological and anatomical studies stronglybridge NO and eCB inasmuch as regulating neuronal hyper-excitability in pathophysiological states such as epilepsy(Bahremand et al., 2009; Jones et al., 2008; Makara et al.,2007; Stringer and Erden, 1995). The underlying mecha-nism may regard the involvement of NO/cGMP pathwayin CB-mediated DSE and DSI, in which a retrograde con-trol of glutamate and GABA release is exerted in in CB1

expressing axon terminals (Chevaleyre et al., 2006; Llanoet al., 1991; Makara et al., 2007). Indeed, the inhibitionof NO/cGMP signaling at various levels diminished DSI in aCB1R-dependent manner (Makara et al., 2007), hence mod-ulating interneuron GABA-inhibition. In agreement with theaforementioned studies, our data convey the idea that, inthe MDA model, NO/cGMP pathway could be targeted as aneuromodulator after previous CB1 activation (e.g. underWIN pre-treatment). On this basis, providing that nNOS isa Ca-dependent enzyme (Mergia et al., 2009; Neitz et al.,2011), it is conceivable that CB1 antiepileptic activity maybe exerted by reducing NMDA-dependent calcium signalingin glutamate synapses, thus lowering post-synaptic NO pro-duction. If so, according to previously reported evidences(Arancio et al., 2001), the blockade of presynaptic sGC inexcitatory synapses enhances the WIN antiepileptic effectslikely by further weakening glutamate transmission. As far asthese signaling systems might act on common synaptic func-tions such as ion channels permeability, neurotransmitterrelease and synaptic plasticity, any change in cGMP synthe-sis within CB1-expressing axon terminals, by modulating thesGC activity, could have consequences on the CB1-inducedeffects (Ahern et al., 2002; Castillo et al., 2012; Deadwyleret al., 1995; Feil and Kleppisch, 2008).

In conclusion, our study showed that the manipulationof CB2 receptor is able to potentiate the antiepilepticaction of the CB agonist, WIN, in a model of hippocam-pal seizures; a similar effect is achieved by blocking theactivity of sGC, a putative downstream effector of CBsignaling. Taken together, these results shed light on themechanisms underpinning CB antiepileptic effects and hint apossible novel pharmacological approach for the treatmentof excitotoxicity-derived diseases.

Acknowledgement

This work was supported by grants of Italian Ministry ofthe University and the Scientific Research (M.I.U.R.), MIUR-UNIPA ORPA07BLYM Rome, Italy.

Appendix A. Supplementary data

Supplementary data associated with this arti-cle can be found, in the online version, athttp://dx.doi.org/10.1016/j.eplepsyres.2014.10.001.

F

endent antiepileptic effects 1717

eferences

hern, G.P., Klyachko, V.A., Jackson, M.B., 2002. cGMP and S-nitrosylation: two routes for modulation of neuronal excitabilityby NO. Trends Neurosci. 25, 510—517.

rancio, O., Antonova, I., Gambaryan, S., Lohmann, S.M., Wood,J.S., Lawrence, D.S., Hawkins, R.D., 2001. Presynaptic role ofcGMP-dependent protein kinase during long-lasting potentiat-ion. J. Neurosci. 21 (1), 143—149 (The Official Journal of theSociety for Neuroscience).

zad, S.C., Marsicano, G., Eberlein, I., Putzke, J., Zieglgänsberger,W., Spanagel, R., Lutz, B., 2001. Differential role of the nitricoxide pathway on delta(9)-THC-induced central nervous systemeffects in the mouse. Eur. J. Neurosci. 13 (3), 561—568.

ahremand, A., Nasrabady, S.E., Shafaroodi, H., Ghasemi, M.,Dehpour, A.R., 2009. Involvement of nitrergic system in the anti-convulsant effect of the cannabinoid CB(1) agonist ACEA in thepentylenetetrazole-induced seizure in mice. Epilepsy Res. 84(2—3), 110—119.

olognini, D., Cascio, M.G., Parolaro, D., Pertwee, R.G., 2012.AM630 behaves as a protean ligand at the human cannabinoidCB2 receptor. Br. J. Pharmacol. 165 (8), 2561—2574.

raakman, H.M., van Oostenbrugge, R.J., van Kranen-Mastenbroek,V.H., de Krom, M.C., 2009. Rimonabant induces partial seizuresin a patient with a history of generalized epilepsy. Epilepsia 50(9), 2171—2172.

urette, A., Zabel, U., Weinberg, R.J., Schmidt, H.H., Valtschanoff,J.G., 2002. Synaptic localization of nitric oxide synthase and sol-uble guanylyl cyclase in the hippocampus. J. Neurosci. 22 (20),8961—8970.

abral, G.A., Raborn, E.S., Griffin, L., Dennis, J., Marciano-Cabral,F., 2008. CB2 receptors in the brain: role in central immunefunction. Br. J. Pharmacol. 153 (2), 240—251.

arletti, F., Ferraro, G., Rizzo, V., Cannizzaro, C., Sardo, P., 2013.Antiepileptic effect of dimethyl sulfoxide in a rat model of tem-poral lobe epilepsy. Neurosci. Lett. 546, 31—35.

astillo, P.E., Younts, T.J., Chávez, A.E., Hashimotodani, Y., 2012.Endocannabinoid signaling and synaptic function. Neuron 76 (1),70—81.

hevaleyre, V., Takahashi, K.A., Castillo, P.E., 2006.Endocannabinoid-mediated synaptic plasticity in the CNS.Annu. Rev. Neurosci. 29, 37—76.

eadwyler, S.A., Hampson, R.E., Mu, J., Whyte, A., Childers, S.,1995. Cannabinoids modulate voltage sensitive potassium A-current in hippocampal neurons via a cAMP-dependent process.J. Pharmacol. Exp. Ther. 273 (2), 734—743.

en Boon, F.S., Chameau, P., Schaafsma-Zhao, Q., van Aken, W.,Bari, M., Oddi, S., Kruse, C.G., Maccarrone, M., Wadman, W.J.,Werkman, T.R., 2012. Excitability of prefrontal cortical pyrami-dal neurons is modulated by activation of intracellular type-2cannabinoid receptors. Proc. Natl. Acad. Sci. U.S.A. 109 (9),3534—3539.

eshpande, L.S., Sombati, S., Blair, R.E., Carter, D.S., Martin, B.R.,DeLorenzo, R.J., 2007. Cannabinoid CB1 receptor antagonistscause status epilepticus-like activity in the hippocampal neu-ronal culture model of acquired epilepsy. Neurosci. Lett. 411(1), 11—16.

evane, W.A., Dysarz, F.A., Johnson, M.R., Melvin, L.S., Howlett,A.C., 1988. Determination and characterization of a cannabinoidreceptor in rat brain. Mol. Pharmacol. 34 (5), 605—613.

eil, R., Kleppisch, T., 2008. NO/cGMP-dependent modula-tion of synaptic transmission. Handb. Exp. Pharmacol. 184,529—560.

ernández-Ruiz, J., Romero, J., Velasco, G., Tolón, R.M., Ramos,J.A., Guzmán, M., 2007. Cannabinoid CB2 receptor: a new targetfor controlling neural cell survival? Trends Pharmacol. Sci. 28 (1),39—45.

1

G

G

G

H

H

H

J

J

K

L

M

M

M

M

M

M

M

N

P

P

R

R

S

S

S

S

S

V

W

epilepsy. J. Pharmacol. Exp. Ther. 307 (1), 129—137.

718

arcía-Gutiérrez, M.S., García-Bueno, B., Zoppi, S., Leza, J.C.,Manzanares, J., 2012. Chronic blockade of cannabinoid CB2receptors induces anxiolytic-like actions associated with alter-ations in GABA(A) receptors. Br. J. Pharmacol. 165 (4), 951—964.

hasemi, M., Sadeghipour, H., Shafaroodi, H., Nezami, B.G.,Gholipour, T., Hajrasouliha, A.R., Tavakoli, S., Nobakht, M.,Moore, K.P., Mani, A.R., Dehpour, A.R., 2007. Role of the nitricoxide pathway and the endocannabinoid system in neurogenicrelaxation of corpus cavernosum from biliary cirrhotic rats. Br.J. Pharmacol. 151 (5), 591—601.

ong, J.P., Onaivi, E.S., Ishiguro, H., Liu, Q.R., Tagliaferro,P.A., Brusco, A., Uhl, G.R., 2006. Cannabinoid CB2 receptors:immunohistochemical localization in rat brain. Brain Res. 1071(1), 10—23.

ill, T.D., Cascio, M.G., Romano, B., Duncan, M., Pertwee, R.G.,Williams, C.M., Whalley, B.J., Hil, A.J., 2013. Cannabidivarin-rich cannabis extracts are anticonvulsant in mouse and rat viaa CB1 receptor-independent mechanism. Br. J. Pharmacol. 170(3), 679—692.

ofmann, M.E., Frazier, C.J., 2013. Marijuana, endocannabinoids,and epilepsy: potential and challenges for improved therapeuticintervention. Exp. Neurol. 244, 43—50.

owlett, A.C., Breivogel, C.S., Childers, S.R., Deadwyler, S.A.,Hampson, R.E., Porrino, L.J., 2004. Cannabinoid physiology andpharmacology: 30 years of progress. Neuropharmacology 47(Suppl. 1), 345—358.

ones, J.D., Carney, S.T., Vrana, K.E., Norford, D.C., Howlett,A.C., 2008. Cannabinoid receptor-mediated translocation ofNO-sensitive guanylyl cyclase and production of cyclic GMP inneuronal cells. Neuropharmacology 54 (1), 23—30.

ones, N.A., Glyn, S.E., Akiyama, S., Hill, T.D.M., Hill, A.J., Weston,S.E., Burnett, M.D.A., Yamasaki, Y., Stephens, G.J., Whalley,B.J., Williams, C.M., 2012. Cannabidiol exerts anti-convulsanteffects in animal models of temporal lobe and partial seizures.Seizure 21 (5), 344—352 (The Journal of the Epilepsy Associa-tion).

reitzer, A.C., Regehr, W.G., 2001. Retrograde inhibition of presyn-aptic calcium influx by endogenous cannabinoids at excitatorysynapses onto Purkinje cells. Neuron 29, 717—727.

lano, I., Leresche, N., Marty, A., 1991. Calcium entry increasesthe sensitivity of cerebellar Purkinje cells to applied GABA anddecreases inhibitory synaptic currents. Neuron 6 (4), 565—574.

aggi, C.A., Meli, A., 1986. Suitability of urethane anesthesia forphysiopharmacological investigations in various systems. Part 1:General considerations. Experientia 42 (2), 109—114.

akara, J.K., Katona, I., Nyíri, G., Németh, B., Ledent, C., Watan-abe, M., de Vente, J., Freund, F.T., Hájos, N., 2007. Involvementof nitric oxide in depolarization-induced suppression of inhi-bition in hippocampal pyramidal cells during activation ofcholinergic receptors. J. Neurosci. 27 (38), 10211—10222 (TheOfficial journal of the Society for Neuroscience).

atsuda, L.A., Lolait, S.J., Brownstein, M.J., Young, A.C., Bonner,T.I., 1990. Structure of a cannabinoid receptor and functionalexpression of the cloned cDNA. Nature 346 (6284), 561—564.

ergia, E., Koesling, D., Friebe, A., 2009. Genetic mouse mod-els of the NO receptor ‘soluble’ guanylyl cyclases. Handb. Exp.

Pharmacol. 191, 33—46.

onory, K., Massa, F., Egertová, M., Eder, M., Blaudzun, H., West-enbroek, R., Kelsch, W., Jacob, W., Marsch, R., Ekker, M.,Long, J., Rubenstein, J.L., Goebbels, S., Nave, K.A., During, M.,

W

V. Rizzo et al.

Klugmann, M., Wölfel, B., Dodt, H.U., Zieglgänsberger, W., Wot-jak, C.T., Mackie, K., Elphick, M.R., Marsicano, G., Lutz, B.,2006. The endocannabinoid system controls key epileptogeniccircuits in the hippocampus. Neuron 51 (4), 455—466.

organ, N.H., Stanford, I.M., Woodhall, G.L., 2009. FunctionalCB2 type cannabinoid receptors at CNS synapses. Neuro-pharmacology 57 (4), 356—368.

ukherjee, S., Adams, M., Whiteaker, K., Daza, A., Kage, K., Cassar,S., Meyer, M., Yao, B.B., 2004. Species comparison and phar-macological characterization of rat and human CB2 cannabinoidreceptors. Eur. J. Pharmacol. 505 (1—3), 1—9.

eitz, A., Mergia, E., Eysel, U.T., Koesling, D., Mittmann, T., 2011.Presynaptic nitric oxide/cGMP facilitates glutamate release viahyperpolarization-activated cyclic nucleotide-gated channels inthe hippocampus. Eur. J. Neurosci. 33 (9), 1611—1621.

an, X., Ikeda, S.R., Lewis, D.L., 1996. Rat brain cannabinoid recep-tor modulates N-type Ca2+ channels in a neuronal expressionsystem. Mol. Pharmacol. 49 (4), 707—714.

axinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordi-nates. Academic Press, San Diego, IL, USA.

izzo, V., Ferraro, G., Carletti, F., Lonobile, G., Cannizzaro, C.,Sardo, P., 2009. Evidences of cannabinoids-induced modula-tion of paroxysmal events in an experimental model of partialepilepsy in the rat. Neurosc. Lett. 462 (2), 135—139.

obello, M., Amico, C., Bucossi, G., Cupello, A., Rapallino, M.V.,Thellung, S., 1996. Nitric oxide and GABAA receptor functionin the rat cerebral cortex and cerebellar granule cells. Neuro-science 74 (1), 99—105.

ardo, P., Carletti, F., D’Agostino, S., Rizzo, V., Ferraro, G., 2006.Involvement of nitric oxide-soluble guanylyl cyclase pathway inthe control of maximal dentate gyrus activation in the rat. J.Neural Transm. 113 (12), 1855—1861.

ardo, P., D’Agostino, S., Carletti, F., Rizzo, V., La Grutta,V., Ferraro, G., 2008. Lamotrigine differently modulates 7-nitroindazole and L-arginine influence on rat maximal dentategyrus activation. J. Neural Transm. 115 (1), 27—34.

tefano, G.B., Rialas, C.M., Deutsch, D.G., Salzet, M., 1998. Anan-damide amidase inhibition enhances anandamide-stimulatednitric oxide release in invertebrate neural tissues. Brain Res.793 (1—2), 341—345.

tringer, J.L., Erden, F., 1995. In the hippocampus in vivo, nitricoxide does not appear to function as an endogenous antiepilepticagent. Exp. Brain Res. 105, 391—401.

tringer, J.L., Lothman, E.W., 1989. Maximal dentate gyrus activa-tion: characteristics and alterations after repeated seizures. J.Neurophysiol. 62, 136—143.

an Sickle, M.D., Duncan, M., Kingsley, P.J., Mouihate, A., Urbani,P., Mackie, K., Stella, N., Makriyannis, A., Piomelli, D., Davi-son, J.S., Marnett, L.J., Di Marzo, V., Pittman, Q.J., Patel,K.D., Sharkey, K.A., 2005. Identification and functional charac-terization of brainstem cannabinoid CB2 receptors. Science 310(5746), 329—332.

allace, M.J., Blair, R.E., Falenski, K.W., Martin, B.R., DeLorenzo,R.J., 2003. The endogenous cannabinoid system regulatesseizure frequency and duration in a model of temporal lobe

ilson, R.I., Nicoll, R.A., 2001. Endogenous cannabinoids medi-ate retrograde signalling at hippocampal synapses. Nature 410,588—659.