Epilepsy & Behavior 36 (2014) 159–164

Contents lists available at ScienceDirect

Epilepsy & Behavior

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

Temporally unstructured electrical stimulation to the amygdalasuppresses behavioral chronic seizures of the pilocarpine animal model

Jasiara Carla de Oliveira a, Daniel de Castro Medeiros b, Gustavo Henrique de Souza e Rezende b,Márcio Flávio Dutra Moraes b, Vinícius Rosa Cota a,⁎a Laboratório Interdiciplinar de Neuroengenharia e Neurociências, Departamento de Engenharia de Biossistemas (DEPEB), Universidade Federal de São João Del-Rei, Pça. Dom Helvécio,74, 36301-160 São João Del-Rei, MG, Brazilb Núcleo de Neurociências, Instituto de Ciências Biológicas (ICB), Universidade Federal de Minas Gerais, Av. Antônio Carlos 6627, CEP 31270-901 Belo Horizonte, MG, Brazil

⁎ Corresponding author at: DEPEB, Pça. Dom Helvécio,João Del-Rei, MG, Brazil. Tel.: +55 32 3379 2541 (office),

E-mail addresses: vrcota@ufsj.edu.br, vrcota@pq.cnpq(V.R. Cota).

http://dx.doi.org/10.1016/j.yebeh.2014.05.0051525-5050/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 17 January 2014Revised 30 April 2014Accepted 6 May 2014Available online xxxx

Keywords:Temporal lobe epilepsyPilocarpineElectrical stimulationTemporal codingBasolateral amygdalaDesynchronizationChronic seizures

Electrical stimulation applied to the basolateral amygdala in the pentylenetetrazole animal model of seizuresmay result in either a proconvulsant or an anticonvulsant effect depending on the interpulse intervals used: pe-riodic or nonperiodic, respectively. We tested the effect of this electrical stimulation temporal coding on thespontaneous and recurrent behavioral seizures produced in the chronic phase of the pilocarpine animal modelof temporal lobe epilepsy, an experimental protocol that better mimics the human condition. After 45 days ofthe pilocarpine-induced status epilepticus, male Wistar rats were submitted to a surgical procedure for the im-plantation of a bipolar electrical stimulation electrode in the right basolateral amygdala and were allowed to re-cover for seven days. The animals were then placed in a glass box, and their behaviors were recorded daily onDVD for 6 h for 4 consecutive days (control period). Spontaneous recurrent behavioral seizures when showedin animals were further recorded for an extra 4-day period (treatment period), under periodic or nonperiodicelectrical stimulation. The number, duration, and severity of seizures (according to the modified Racine's scale)during treatment were compared with those during the control period. The nonperiodically stimulated groupdisplayed a significantly reduced total number and duration of seizures. Therewas no difference between controland treatment periods for the periodically stimulated group. Results corroborate previous findings from ourgroup showing that nonperiodic electrical stimulation has a robust anticonvulsant property. In addition, resultsfrom the pilocarpine animal model further strengthen nonperiodic electrical stimulation as a valid therapeuticapproach in currentmedical practice. Our working hypothesis is that temporally unstructured electrical stimula-tion may wield its effect by desynchronizing neural networks involved in the ictogenic process.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Epilepsy is a chronic neurological disorder characterized by recur-rent and spontaneous seizures caused by hypersynchronous and exces-sive neural activity [1]. It has high prevalence, affecting about fiftymillion people worldwide [2]. Temporal lobe epilepsy (TLE) is themost common type of partial epilepsy [3], which accounts for about60% of all patients [4]. It has a focal onset, and it is the most commontype of drug-resistant epilepsy [5].

Intracranial electrical stimulation has long been considered a poten-tially viable therapy for patients with drug-resistant epilepsy who arenot eligible for ablative surgery [6]. Currently, electrical stimulation(i.e., current or voltage pulses)may be applied to the peripheral nervoussystem, in structures such as the vagus nerve (vagus nerve stimulation)

74, B. Fábricas, 36301-160 São+55 32 8861 8074 (mobile)..br, vrcota@gmail.com

[7] and the trigeminal nerve (trigeminal nerve stimulation) [8], or di-rectly to the central nervous system, in substrates such as the anteriorthalamic nucleus [9], subthalamic nuclei [10], and epileptogenic focusitself [11].

Although the literature reports an overwhelming amount of datashowing the effect of electrical stimulation on seizure suppression (fora review, see [6]), its mechanisms of action on neural network modula-tion need further investigation. While debatable, the most widelyaccepted framework for its therapeutic effectiveness posits that electri-cal stimulationwould recruit substrates and/or neural networks capableof modulating seizure-like activity in areas involved in ictogenesis orrather by impairing the coupling of neural oscillators necessary to prop-agate and sustain aberrant activity [12,13]. Among others, in silico stud-ies have shown that neural circuits that fire synchronously are coupledin a positive feedback fashion [14], that synaptic weights increase inproportion to the coincidence in neuronal firing [15], corroboratingHebbian postulates [16], and that this coincidence may be increased ordecreased by a synchronizing or desynchronizing electrical stimulation,respectively [15]. Finally, Medeiros et al. have shown that electrical

160 J.C. de Oliveira et al. / Epilepsy & Behavior 36 (2014) 159–164

stimulation may drive the temporal occurrence of preictal oscillatoryneural networks; that is, cortical preictal discharges gradually synchro-nize with electrical stimulation minutes before seizure onset [17].

In recent investigations, we approached this issue by testing the hy-pothesis that a fixed four-stimuli-per-second electrical stimulation inthe amygdaloid complex would modulate convulsive behavior of ratswith acute pentylenetetrazole (PTZ)-induced seizures according to thetemporal pattern used: structured (constant interpulse intervals orperiodic — PS) or nonstructured (random interpulse intervals ornonperiodic — NPS). It is important to highlight that the average fre-quency used (4 pulses per second) is substantially lower than what isusually considered to be anticonvulsant [6]. It was shown that PS isproconvulsant and NPS is anticonvulsant, suggesting that a putativesynchronization/desynchronization effect of structured/nonstructuredelectrical stimulation may be an underlying mechanism of action [18].In addition, these results corroborate the notion that reverberation ofneural networks has an important role in ictogenesis.

Based on the promising results obtained with the use of a nonstruc-tured temporal pattern in the suppression of acutely induced seizuresand on the urge to develop safe and efficient therapeutic alternativesto treatment-resistant epilepsy such as TLE, in this study, we set out toinvestigate the hypothesis that NPS may also have a positive effect onanimal models that best mimic this clinical scenario. For this, we testedNPS applied to the amygdaloid complex of thepilocarpine animalmodelduring chronic seizures. Pilocarpine is a cholinergic agonist that binds tomuscarinic receptors to increase cholinergic excitatory neurotransmis-sion. When administered systemically in high doses, it induces limbicseizures that become generalized and are associated with status epilep-ticus (SE) in rodents [19–22]. Such a state of enduring seizures inducesepileptogenesis due to large cellular reorganization that culminates inpermanent late neural tissue hyperexcitability [23–35]. Due not onlyto the similarities between pathophysiological mechanisms but also tothe behavioral manifestation of its spontaneous and recurrent chronicseizures, the pilocarpine animal model is considered an experimentalprotocol that mimics human TLE. The amygdala was used as a targetfor electrical stimulation because it has an important role in the cou-pling of neural oscillators within the limbic system during seizures[18,23,24,26,27] and also in the activation of modulatory circuits suchas the nucleus accumbens–temporal lobe [36,37].

2. Materials and methods

2.1. Animals and groups

All experiments were done in accordance with the Ethical Commit-tee for Animal Experimentation (Comitê de Ética em ExperimentaçãoAnimal — CETEA) of the Federal University of Minas Gerais(Universidade Federal de Minas Gerais — UFMG) and with the EthicalCommittee on Research Involving Animals (Comitê de Ética emPesquisa Envolvendo Animais — CEPEA) of the Federal University ofSão João del Rei (Universidade Federal de São João del Rei — UFSJ).The procedures for animal care were previously approved by theseorganizations under protocols 150/2006/UFMG and 01/2011/UFSJ. Atotal of 14 male Wistar rats, weighing 250–300 g, supplied by UFMG(n = 8) and UFSJ (n = 6) vivariums, were kept in a light–dark cycleof 12 h (lights on at 7 am and off at 7 pm) with free access to food andwater.

The animals were randomly divided into two groups: NPS (n = 7)received nonperiodic stimulation and PS (n = 7) received periodicstimulation. After a recovery period of 5 to 7 days, the animals wereplaced in a glass boxwith free access to food andwater, and their behav-iors were recorded on DVD for 6 h (10 am to 4 pm) daily for 4 consecu-tive days (control period — CRTL). Animals that showed spontaneousrecurrent seizures underwent electrical stimulation (pattern accordingto group) and video recording for an extra 4 days (treatment period).

2.2. Status epilepticus induction

Status epilepticus was induced in the pilocarpine animal model byfirst applying methylscopolamine (1 mg/kg; Sigma Aldrich, St. Louis,MO, USA) by means of an intraperitoneal (i.p.) injection, followed30 min later by pilocarpine hydrochloride (320 mg/kg; Sigma Aldrich,St. Louis, MO, USA) i.p. injection. Methylscopolamine is a cholinergicantagonist that does not cross the blood–brain barrier and was used toavoid undesirable systemic effects of the cholinergic stimulation in-duced by pilocarpine (i.e., excessive defecation, urination, sweating,and bronchial secretion). Animals that did not develop SE during thefirst 30 min after pilocarpine injection received an overdose of 40% ofthe initial dose. Animals that did not develop SE evenwith the overdosewere excluded from the study. At the 90-minute mark, SE wasinterrupted by i.p. injection of diazepam (20 mg/kg; Laboratório TeutoBrasileiro, Anápolis, GO, Brazil). The animals received intensive care,including rehydration with dextrose (2 mL) i.p., during 48 h after theSE induction to ensure survival.

2.3. Stereotaxic procedures

Between 45 and 50 days after the SE induction, the animalsunderwent a surgical procedure for implantation of a bipolar stimula-tion electrode in the right basolateral amygdala. Electrodes were madeof a twisted pair of stainless-steel teflon-coated wires (model 791400,A-M Systems Inc., California, USA) and were surgically implanted atcoordinates derived from the Paxinos and Watson's atlas for rats [38]:AP = 2.8 mm and ML = 5.0 mm referenced from the bregma sutureand DV= 7.2 mm from dura mater.

Briefly, the animals were anesthetized by means of an i.p. injectioncontaining the mixture of ketamine (100 mg/kg — König do Brasil,Santana do Paraíba, SP, Brazil), xylazine (5 mg/kg — Syntec do Brasil,Cotia, SP, Brazil), and fentanyl (0.025 mg/kg — Union Chemical doBrasil, Londrina, PR, Brazil). After hair shaving and proper asepsis proce-dures, the animals were positioned in a stereotaxic frame (InsightEquipamentos Ltda, Ribeirão Preto, SP, Brazil). The electrode was fixedto the skull with zinc cement and soldered to a telephone jack (modelRJ-11), whichwas fixed onto the skull with dental acrylic. After surgery,the animals received a prophylactic pentabiotic (2.5 mg/kg) treatmentand were allowed to recover for 5–7 days before the experimentalprocedure.

2.4. Electrical stimulation

For stimulation delivery, we designed and built an electrical stimula-tor composed of a constant-voltage isolation unit driven by the outputof an MP3 player (model NWZ-B152 26B, Sony). Control signals forboth periodic electrical stimulation and nonperiodic electrical stimula-tion were digitally designed using Adobe Audition 1.0 and transformedinto a 44.1 KHz, 16-bit, mono-waveform, MP3 format compatible withthe D/A hardware output. Pulses were always square, biphasic wavesof 100 μs duration. Current amplitudes, measured by a built-in shuntresistor, varied from 400 to 600 μA due to differences in the impedanceof electrode–brain sets among the animals. Yet, there was no differencein average currents between periodic and nonperiodic groups. Pulseswere biphasic to prevent electrodeposition along the extended periodof stimulation and subsequent tissue damage, as reported in the litera-ture [39,40].

The two electrical stimulation patterns differed on theway theyweretemporally coded: 1) PS had constant interpulse intervals of 250 mswhile 2) NPS had variable interpulse intervals randomized on runtimeby an algorithm described elsewhere [18] (Fig. 1). Yet, mean frequencywas constrained at a low fixed total value of 4-stimuli-per-secondcount in both patterns.

Fig. 1. Rats were stimulatedwith two different temporal patterns. Left column: interpulse interval (IPI) histograms for: (A) periodic (PS) and (C) nonperiodic (NPS) electrical stimulation.Right column: temporal distribution of pulses every second (vertical ticks) for PS (B) and NPS (D). Electrical stimulation for both patterns was bipolar pulses of 400 to 600 μA amplitude,100 μs duration, and four-stimuli-per-second pulse count (see Cota et al. [18] for details).

161J.C. de Oliveira et al. / Epilepsy & Behavior 36 (2014) 159–164

2.5. Behavioral analysis

The number, duration, and severity – according to the modifiedRacine's scale [41] – of seizures, during control and treatment periodsof both groups, were assessed by two separate and experiencedresearchers. Their independent results were compared, and reproduc-ibility of data was assessed in order to guarantee minimum subjectivityor bias. Between researchers, there was absolutely no difference inmeasures of the severity and number of seizures, while the durationof seizure measurements differed no more than 2 s. In this last case,we used the average between measures. Only animals displaying sei-zures with Racine's index greater than or equal to 3 during the controlperiod were used to compute final results.

2.6. Histology

After the end of stimulation, the animals were deeply anesthetizedwith ketamine (100 mg/kg) and xylazine (5.0 mg/kg) and weretranscardially perfused with formaldehyde (4%) before brain removal.Coronal sections of 50 μm thickness were cut on a vibratome (Leica),mounted on glass slides, and stainedwith cresyl violet. Animals with in-correct positioning of electrodeswere not included in our analysis. Fig. 2shows a representative location of the electrode tip.

2.7. Statistical analysis

The Kolmogorov–Smirnov normality test and Student t test wereused to evaluate parametric data (number and duration of seizures)

Fig. 2. Location of the electrode tip in th

andWilcoxon test to evaluate nonparametric data (severity of seizures).Statistical significance was set at p b 0.05. Values in the text aredisplayed as means ± standard error.

3. Results

All animals had histological confirmation of the positioning of elec-trode in the BLA. Only one animal was excluded from the study becauseit had no spontaneous and recurrent behavioral seizures during thecontrol period (first 4 days of experimental protocol).

Seizures were observed during control and stimulation periods. Theanimals developed the typical behaviors of the pilocarpine animalmodel: myoclonic jerks, forelimbs and head clonus, rearing, and gener-alized tonic–clonic seizures (GTCS).

During treatment when compared to the control period, NPS wasable to significantly reduce the total number (CTRL: 3.8 ± 1.3; NPS:1.3 ± 0.6; p b 0.05) (Fig. 3A) and duration (CTRL: 26.6 ± 2.1; NPS:8.3 ± 3.9; p b 0.05) (Fig. 3B) of seizures. Seizure severity in the treat-ment period did not show significant difference (CTRL: 4.4 ± 0.3;NPS: 2.5 ± 0.9; p = 0.1077) (Fig. 3C).

In the animals of the periodically stimulated (PS) group, there wereno statistically significant differences between control and treatmentperiods regarding any of the analyzed parameters. Averages and theirrespective standard errors were the following: total seizure number(CRTL: 1.7 ± 1.7; PS: 5.4 ± 6.1; p = 0.1359) (Fig. 4A), seizure duration(CTRL: 20.2± 21.4; PS: 33.4 ± 27.1; p= 0.2675) (Fig. 4B), and seizureseverity (CTRL: 3.1 ± 2.2; PS: 3.5 ± 1.7; p = 0.7984) (Fig. 4C).

e amygdala of a representative rat.

Fig. 3. Number (A), duration (B), and severity (C) of seizures, according to the modifiedRacine's scale, in the nonperiodically stimulated group during treatment period comparedto its control period (CTRL). The number and duration of seizures was significantlydecreased by NPS.

Fig. 4. Number (A), duration (B), and severity (C) of seizures, according to the modifiedRacine's scale, in the periodically stimulated (PS) group during treatment period com-pared to its control period (CTRL). There were no significant differences.

162 J.C. de Oliveira et al. / Epilepsy & Behavior 36 (2014) 159–164

4. Discussion

Our present results show that lowpulse per second count, temporal-ly unstructured, electrical stimulation (nonperiodic or NPS) applied tothe basolateral amygdala has a positive therapeutic effect on behavioralseizures of the pilocarpine animal model of temporal lobe epilepsy(TLE), being able to reduce their number and duration during the chron-ic phase. These results corroborate previous findings from our groupusing the acute pentylenetetrazole (PTZ) animal model [18] and extendthem to another important animal model used in epileptology (i.e.,pilocarpine).

Previous studies reported the effect of fixed frequency electricalstimulation on a pilocarpine or kainate animal model, but they werenot accomplished in the chronic phase or did not show positive thera-peutic effects. Hamani et al. showed that bilateral stimulation(100 Hz) of the anterior thalamic nuclei was protective against SE in-duced by pilocarpine [42]. Jou et al. also found that anterior thalamicstimulation with high-frequency (200 Hz) and low-intensity currentsmay reduce the occurrence of seizure and SE [43]. Finally, Lado showedthat high-frequency electrical stimulation (100 Hz) led to an increase inthe frequency of seizures in animals with chronic epilepsy that receivedkainic acid [44]. To the best of our knowledge, there are no studies ofelectrical stimulation during the chronic phase of the pilocarpine animal

model targeting the amygdala or, most importantly, using unstructuredtemporal patterns with a low count of pulses per second.

The pathophysiology of the pilocarpine animal model is a complex,multivariate phenomenon, yet sharing many common mechanismswith human TLE [26]. Ultimately, molecular and cellular modificationslead to a state of highly excitable neural tissue, susceptible to developingspontaneous and recurrent seizures, that is similar to what is observedin humans affected with TLE [45]. Structures within the limbic system,prominently the hippocampus and the entorhinal cortex, are amongthe most affected neural substrates [23,46,47]. Aberrant epileptiformactivity originating in a group of hyperexcitable neurons (epileptogenicfocus) can propagate through aberrant pathological or, potentially, evenapparently normal pathways in the central nervous system [48], cou-pling different brain areas for the full epileptic event by means of syn-chronization of activity [14,49,50]. In vitro studies in healthy neuraltissue demonstrated that incoming activity from the entorhinal cortexenters the hippocampus through the perforant path passing the dentategyrus to reach Ammon's horn via mossy fibers and Schaffer collaterals[51]. Such flow of information seems to be powerfully modulated byamygdala outputs to deep layers of the entorhinal cortex [26,27]. In pi-locarpine-treated animals, epileptiform activity may also propagatefrom the entorhinal cortex to the hippocampus laterally through thetemporoammonic path to reach CA1 [52,53]. Moreover, hippocampaloutputs back to the entorhinal cortex form a reentrant circuit crucialfor sustaining seizures in a reverberatory process [24,49,54–57].

163J.C. de Oliveira et al. / Epilepsy & Behavior 36 (2014) 159–164

The present data do not allow elucidation of the mechanisms bywhichNPS attains its therapeutic effect. Yet, considering the importanceof synchronization to ictogenesis and known correlations between be-havior and neural activation [58,59], we believe that NPS impairs thecoupling of micro-oscillators within the entorhinal cortex to those inthe hippocampus by imposing unstructured rhythms in the deep layersof the entorhinal cortex targeted by BLA outputs (connections in thelimbic system of pilocarpine-treated animals are preserved [60]), and,thus, it suppresses seizures. Within this theoretical framework, the de-crease in the number and duration of seizures attained by NPS couldbe explained by a putative impairment of coupling and synchronizationof micro-oscillators within the limbic system, crucial for limbic seizureinitiation, propagation, and reverberation [61], by avoiding temporal co-incidence among activities of different substrates [14,49,62]. A real-timemeasurement of neural activity, such as recordings of local field poten-tial (or electroencephalogram), is necessary in order to clarify the role ofeach temporal pattern in the synchronization or desynchronization ofthesemicro-oscillators and tomore definitely state if such amechanismis really underlying the observed effects.

Periodic stimulation (PS) did not show a clear proconvulsant effect,although there seemed to be an increasing trend in the number of sei-zures. It is plausible to believe that the fixed frequency used (4 Hz) isnot the proper stimulus design for precipitating seizures in the pilocar-pine animal model opposed to what has been demonstrated in PTZ-treated animals. On the other hand, the clear lack of an anticonvulsanteffect of PS states that the therapeutic effect of NPS is not due to itslow-frequency content.

NPS has both low frequency content and high frequency content ascan be observed in its histogram in Fig. 1. We ruled out the effect ofhigh band in the anticonvulsant effect observed through experimentsin a previous original work published in 2009 [18]. There, we tested,in animals with PTZ-induced seizures, one-per-second 50-Hz bursts ofelectrical stimulation, with 4 pulses each, and they showed no anticon-vulsant effect. Finally, we also tested a variation (called LH) of thenonperiodic (temporally unstructured) pattern of electrical stimulation,obtained by a different computational algorithm, in addition to the onethat is also used in the present work (called IH). We observed that LHvariation results were no different from control, while IH had the anti-convulsant effect. It is important to stress that all these experimentsused the very same pulse parameters (duration, amplitude, count persecond, polarity, etc.) among groups and only the temporal codingwas different. Particularly, this approach also rules out lesion-inducedeffects. If thiswere the case, all groupswould display similar anticonvul-sant or even proconvulsant effects once the very same lesion-inducingcharge (zero net charge in the present study once pulses are biphasic)was delivered in each case. It is also important to highlight that NPShad an anticonvulsant effect on dysfunctional neural tissue marked byaberrant exacerbated connectivity and excitability, including poorGABAergic inhibition [63]. This suggests that NPS does not act directlyupon neurotransmitter systems; rather, it may have an effect on mech-anisms at the circuit level.

Taken together, these results led us to the conclusion that thetemporal coding, rather than any other factor, is the responsible pa-rameter for the anticonvulsant effect. Moreover, these results andtheir plausible explanation are in agreement with previous results ofour group, using noninvasive imaging, that synchronization anddesynchronization of neural substrates may be determinant factors inictogenesis and its suppression by NPS [37], respectively.

5. Conclusions

In this work, we showed that temporally unstructured (nonperiodic)electrical stimulation applied to the basolateral amygdala, even with alow count of pulses per second, is capable of decreasing the numberand duration of spontaneous recurrent behavioral seizures of the ratsduring the chronic phase of the pilocarpine animal model, while periodic

stimulation has no clear effect. We believe these results to be of majorimportance because of three main reasons: 1) they corroborate previ-ous findings of our group using the acute PTZ animal model [18], dem-onstrating that such temporal pattern has a robust effect in controllingseizures; 2) they add empirical evidence to the putative role of syn-chronization and desynchronization as underlying mechanisms ofictogenesis and its suppression, respectively; and 3) they clearly showthat temporally unstructured electrical stimulation has a therapeuticeffect even on dysfunctional hyperexcitable neural tissue susceptibleto develop and sustain aberrant epileptiform activity.

Considering the common features of the pilocarpine animalmodel ofepilepsy andhuman temporal lobe epilepsy (spontaneous recurrent sei-zures and limbic pathophysiology), these present results suggest thattemporally unstructured electrical stimulation of the basolateral amyg-dala should be considered as a therapeutic approach in clinical trials.

Conflict of interest

The authors state that no other people or organization have inappro-priately influenced this work. Therefore, there is no pertinent claim of aconflict of interest.

Acknowledgments

We are grateful to Brazilian agencies CNPq (grants #484588/2011-7and #484704/2012-5) and FAPEMIG (grant #APQ 01818-12) for finan-cial support. Márcio Flávio Dutra Moraes is a recipient of CNPq researchfellowship.

References

[1] Fisher RS, van Emde Boas W, Blume W, Elger C, Genton P, Lee P, et al. Epileptic sei-zures and epilepsy: definitions proposed by the International League Against Epilep-sy (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia 2005;46:470–2.

[2] World Health Organization. Epilepsy in the WHO Africa region, bridging the gap:the global campaign against epilepsy “out of the shadows”. Geneva: World healthOrganization; 2004.

[3] Téllez-Zenteno JF, Hernández-Ronquillo L. A review of the epidemiology of temporallobe epilepsy. Epilepsy Res Treat 2012;2011:1–5.

[4] Wiebe S. Epidemiology of temporal lobe epilepsy. Can J Neurol Sci 2000;27:S20–1.[5] Engel Jr J. Clinical neurophysiology, neuroimaging, and the surgical treatment of

epilepsy. Curr Opin Neurol Neurosurg 1993;6:240–9.[6] Theodore WH, Fisher RS. Brain stimulation for epilepsy. Lancet Neurol 2004;3:

111–8.[7] E. Ben-Menachem. Vagus-nerve stimulation for the treatment of epilepsy. Lancet

Neurol 2002;1:477–82.[8] DeGiorgio CM, Shewmon DA,Whitehurst T. Trigeminal nerve stimulation for epilepsy.

Neurology 2003;61:421–2.[9] Mirski MA, Rossell LA, Terry JB, Fisher RS. Anticonvulsant effect of anterior thalamic

high frequency electrical stimulation in the rat. Epilepsy Res 1997;28:89–100.[10] Chabardès S, Kahane P, Minotti L, Koudsie A, Hirsch E, Benabid A. Deep brain stimu-

lation in epilepsy with particular reference to the subthalamic nucleus. EpilepticDisord 2002;4:83–93.

[11] Vonck K, Boon P, Achten E, De Reuck J, Caemaert J. Long-term amygdalohippocampalstimulation for refractory temporal lobe epilepsy. Ann Neurol 2002;52:556–65.

[12] McIntyre CC, Savasta M, Kerkerian-Le Goff L, Vitek JL. Uncovering the mechanism(s)of action of deep brain stimulation: activation, inhibition, or both. Clin Neurophysiol2004;115:1239–48.

[13] McIntyre CC, Grill WM, Sherman DL, Thakor NV. Cellular effects of deep brain stim-ulation: model-based analysis of activation and inhibition. J Neurophysiol 2004;91:1457–69.

[14] Abeles M, Hayon G, Lehmann D. Modeling compositionality by dynamic binding ofsynfire chains. J Comput Neurosci 2004;17:179–201.

[15] Tass PA, Hauptmann C. Therapeutic modulation of synaptic connectivity withdesynchronizing brain stimulation. Int J Psychophysiol 2007;64:53–61.

[16] Hebb DO. The organization of behavior. New York: Wiley; 1949.[17] Medeiros DC, Oliveira LB, Mourão FAG, Bastos CP, Cairasco NG, Pereira GS, et al. Tem-

poral rearrangement of pre-ictal PTZ induced spike discharges by low frequencyelectrical stimulation to the amygdaloid complex. Brain Stimul 2013;7:170–8.

[18] Cota VR, DdC Medeiros, MRSdP Vilela, Doretto MC, Moraes MFD. Distinct patternsof electrical stimulation of the basolateral amygdala influence pentylenetetrazoleseizure outcome. Epilepsy Behav 2009;14:26–31.

[19] Clifford DB, Olney JW, Maniotis A, Collins RC, Zorumski CF. The functional anatomyand pathology of lithium–pilocarpine and high-dose pilocarpine seizures. Neurosci-ence 1987;23:953–68.

[20] Persinger MA, Makarec K, Bradley JC. Characteristics of limbic seizures evoked byperipheral injections of lithium and pilocarpine. Physiol Behav 1988;44:27–37.

164 J.C. de Oliveira et al. / Epilepsy & Behavior 36 (2014) 159–164

[21] Turski WA, Cavalheiro EA, Schwartz M, Czuczwar SJ, Kleinrok Z, Turski L. Limbic sei-zures produced by pilocarpine in rats: behavioural, electroencephalographic andneuropathological study. Behav Brain Res 1983:315–35.

[22] Turski WA, Cavalheiro EA, Bortolotto ZA, Mello LM, Schwarz M, Turski L. Seizuresproduced by pilocarpine in mice: a behavioral, electroencephalographic and mor-phological analysis. Brain Res 1984;321:237–53.

[23] Avoli M, Nagao T, Köhling R, Lücke A, Mattia D. Synchronization of rat hippocampalneurons in the absence of excitatory amino acid-mediated transmission. Brain Res1996;735:188–96.

[24] Bertram EH, Zhang DX, Mangan P, Fountain N, Rempe D. Functional anatomy oflimbic epilepsy: a proposal for central synchronization of a diffusely hyperexcitablenetwork. Epilepsy Res 1998;32:194–205.

[25] Cavalheiro EA, Leite JP, Bortolotto ZA, Turski WA, Ikonomidou C, Turski L. Long termeffects of pilocarpine in rats: structural damage of the brain triggers kindling andspontaneous recurrent seizures. Epilepsia 1991:778–82.

[26] Kajiwara R, Takashima I, Mimura Y, Witter MP, Iijima T. Amygdala input pro-motes spread of excitatory neural activity from perirhinal cortex to the entorhinal–hippocampal circuit. J Neurophysiol 2003;89:2176–84.

[27] Koganezawa N, Taguchi A, Tominaga T, Ohara S, Tsutsui KI, Witter MP, et al. Signif-icance of the deep layers of entorhinal cortex for transfer of both perirhinal andamygdala inputs to the hippocampus. Neurosci Res 2008;61:172–81.

[28] Lothman EW, Rempe DA, Mangan PS. Changes in excitatory neurotransmissionin the CA1 region and dentate gyrus in a chronic model of temporal lobe epilepsy.J Neurophysiol 1995;74:841–8.

[29] Mangan PS, Rempe DA, Lothman EW. Changes in inhibitory neurotransmissionin the CA1 region and dentate gyrus in a chronic model of temporal lobe epilepsy.J Neurophysiol 1995;74:829–40.

[30] Mello LE, Cavalheiro EA, Tan AM, Pretorius JK, Babb TL, Finch DM. Granule cell dis-persion in relation to mossy fiber sprouting, hippocampal cell loss, silent periodand seizure frequency in the pilocarpine model of epilepsy. Epilepsy Res 1992:51–9.

[31] Mello LE, Cavalheiro EA, Tan AM. Circuit mechanisms of seizures in the pilocarpinemodel of chronic epilepsy: cell loss andmossy fiber sprouting. Epilepsia 1993:985–95.

[32] Nadler JV, Perry BW, Cotman CW. Selective reinnervation of hippocampal area CA1and the fascia dentata after destruction of CA3–CA4 afferents with kainic acid. BrainRes 1980;182:1–9.

[33] Parent JM, Yu TW, Leibowitz RT, Geschwind DH, Sloviter RS, Lowenstein DH. Dentategranule cell neurogenesis is increased by seizures and contributes to aberrantnetwork reorganization in the adult rat hippocampus. J Neurosci 1997:3727–38.

[34] Rempe DA, Mangan PS, Lothman EW. Regional heterogeneity of pathophysiologicalalterations in CA1 and dentate gyrus in a chronic model of temporal lobe epilepsy.J Neurophysiol 1995;74:816–28.

[35] Sloviter RS. Feedfoward and feedback inhibition of hippocampal principal cell activ-ity evoked by perforant path stimulation: GABA-mediatedmechanisms that regulateexcitability in vivo. Hippocampus 1991;1:31–40.

[36] McDonald AJ, Mascagni F, Guo L. Projections of the medial and lateral prefrontal cor-tices to the amygdala: a Phaseolus vulgaris leucoagglutinin study in the rat. Neurosci-ence 1996;71:55–75.

[37] Mesquita MBS, Castro Medeiros D, Cota VR, Richardson MP, Williams S, MoraesMFD. Distinct temporal patterns of electrical stimulation influence neural recruit-ment during PTZ infusion: an fMRI study. Prog Biophys Mol Biol 2011;105:109–18.

[38] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. San Diego: AcademicPress; 2007.

[39] Gerken GM, Judy MM. Electrode polarization and the detection of electrical stimula-tion of the brain. Physiol Behav 1977;18:825–32.

[40] Tehovnik EJ. Electrical stimulation of neural tissue to evoke behavioral responses.J Neurosci Methods 1996;65:1–17.

[41] Racine RJ. Modification of seizure activity by electrical stimulation: I. After-dischargethreshold. Electroencephalogr Clin Neurophysiol 1972;32:269–79.

[42] Hamani C, Ewerton FIS, Bonilha SM, Ballester G, Mello LEAM, Lozano AM. Bilateralanterior thalamic nucleus lesions and high-frequency stimulation are protectiveagainst pilocarpine-induced seizures and status epilepticus. Neurosurgery 2004:54.

[43] Jou SB, Kao IF, Yi PL, Chang FC. Electrical stimulation of left anterior thalamic nucleuswith high-frequency and low-intensity currents reduces the rate of pilocarpine-induced epilepsy in rats. Seizure 2013;22:221–9.

[44] Lado FA. Chronic bilateral stimulation of the anterior thalamus of kainate-treatedrats increases seizure frequency. Epilepsia 2006;47:27–32.

[45] Curia G, Longo D, Biagini G, Jones RSG, Avoli M. The pilocarpine model of temporallobe epilepsy. J Neurosci Methods 2008;172:143–57.

[46] Bernardo LS, Prince DA. Acetylcholine induced modulation of hippocampal pyrami-dal neurons. Brain Res 1981:227–34.

[47] Nagao T, Alonso A, Avoli M. Epileptiform activity induced by pilocarpine in the rathippocampal–entorhinal slice preparation. Neuroscience 1996;72:399–408.

[48] Thom M, Bertram EH. Chapter 14 — temporal lobe epilepsy. In: Stefan H,Theodore WH, editors. Handbook of clinical neurology epilepsy. Elsevier;2012. p. 225–40.

[49] Avoli M, DAntuono M, Louvel J, Köhling R, Biagini G, Pumain R, et al. Network andpharmacological mechanisms leading to epileptiform synchronization in the limbicsystem in vitro. Prog Neurobiol 2002;68:167–207.

[50] Uhlhaas PJ, Singer W. Neural synchrony in brain disorders: relevance for cognitivedysfunctions and pathophysiology. Neuron 2006;52:155–68.

[51] Traub RD, Miles R. Neuronal networks of the hippocampus. Cambridge: CambridgeUniversity Press; 1991.

[52] Barbarosie M, Louvel J, Kurcewicz I, Avoli M. CA3-released entorhinal seizuresdisclose dentate gyrus epileptogenicity and unmask a temporoammonic pathway.J Neurophysiol 2000;83:1115–24.

[53] Soltesz I, Jones RSG. Hippocampus forum: the direct perforant path input to CA1.Hippocampus 1995;5:101–46.

[54] Colder BW, Wilson CL, Frysinger RC, Chao LC, Harper RM, Engel J. Neuronal synchro-ny in relation to burst discharge in epileptic human temporal lobes. J Neurophysiol1996;75:2496–508.

[55] de Guzman P, D'antuono M, Avoli M. Initiation of electrographic seizures by neuro-nal networks in entorhinal and perirhinal cortices in vitro. Neuroscience 2004;123:875–86.

[56] Dutra Moraes MF, Galvis-Alonso OY, Garcia-Cairasco N. Audiogenic kindling in theWistar rat: a potential model for recruitment of limbic structures. Epilepsy Res2000;39:251–9.

[57] Si Imamura, Tanaka S, Akaike K, Tojo H, Takigawa M, Ji Kuratsu. Hippocampal tran-section attenuates kainic acid-induced amygdalar seizures in rats. Brain Res2001;897:93–103.

[58] Eells JB, Clough RW, Browning RA, Jobe PC. Comparative fos immunoreactivity in thebrain after forebrain, brainstem, or combined seizures induced by electroshock, pen-tylenetetrazol, focally induced and audiogenic seizures in rats. Neuroscience2004;123:279–92.

[59] Garcia-Cairasco N, Wakamatsu H, Oliveira JAC, Gomes ELT, Del Bel EA, Mello LEAM.Neuroethological and morphological (Neo–Timm staining) correlates of limbic re-cruitment during the development of audiogenic kindling in seizure susceptibleWistar rats. Epilepsy Res 1996;26:177–92.

[60] Kemppainen S, Pitkänen A. Damage to the amygdalo-hippocampal projection intemporal lobe epilepsy: a tract-tracing study in chronic epileptic rats. Neuroscience2004;126:485–501.

[61] Bertram EH. Neuronal circuits in epilepsy: do they matter? Exp Neurol 2013;244:67–74.

[62] Womelsdorf T, Schoffelen JM, Oostenveld R, Singer W, Desimone R, Engel AK, et al.Modulation of neuronal interactions through neuronal synchronization. Science2007;316:1609–12.

[63] Olsen RW, Avoli M. GABA and epileptogenesis. Epilepsia 1997;38:399–407.