Epilepsy Research (2009) 86, 113—123

journa l homepage: www.e lsev ier .com/ locate /ep i lepsyres

Effects of �-aminobutyric acid (GABA) agonists and aGABA uptake inhibitor on pharmacoresistant seizurelike events in organotypic hippocampal slice cultures

Abdul Wahab, Uwe Heinemann ∗, Klaus Albus

Institute of Neurophysiology, Charité Universitätsmedizin Berlin, Tucholskystrasse 2, D-10117 Berlin, Germany

Received 21 January 2009; received in revised form 9 May 2009; accepted 11 May 2009Available online 16 June 2009

KEYWORDSMuscimol;Alfaxalone;Nipecotic acid;In vitro seizuremodel;Epilepsy;Tonic—clonic seizurelike events

SummaryPurpose: Seizure like events (SLEs) induced by low magnesium or 4-aminopyridine in organotypichippocampal slice cultures (OHSCs) are resistant to standard antiepileptic drugs including phe-nobarbital, and 1,4-benzodiazepines [Albus, K., Wahab, A., Heinemann, U., 2008. Standardantiepileptic drugs fail to block epileptiform activity in rat organotypic hippocampal slice cul-tures. Br. J. Pharmacol. 154, 709—724]. The present study was undertaken in order to test theeffects of other compounds on SLEs in OHSCs that enhance GABA-mediated actions.Methods: Six to 12 days old Wistar rats were used to cultivate OHSCs according to the interfacemethod [Stoppini, L., Buchs, P.A., Muller, D., 1991. A simple method for organotypic cultures ofnervous tissue. J. Neurosci. Methods 37, 173—182]. Neuronal activity and extracellular potas-sium concentration were recorded under submerged conditions. SLEs were induced by loweringthe magnesium concentration. The effects of GABAA agonists muscimol and isoguvacine, theGABAB agonist baclofen, the GABA uptake blocker nipecotic acid and the neurosteroid alfax-alone on induction and ongoing SLEs were analyzed.Results: Low magnesium induced SLEs were dose dependently suppressed by the GABAA receptoragonists muscimol, isoguvacine and alfaxalone and by the GABA uptake inhibitor nipecotic acidwhereas the GABAB receptor agonist baclofen attenuated but did not suppress SLE.Discussion: Our findings demonstrate that in OHSCs GABA has an inhibitory effect on SLEs.Proconvulsant effects of GABA agonists on spontaneous neuronal activity and seizure like activity

were never observed. Our findings exclude a possible contribution of impaired/altered GABA-ergic mechanisms based on immaturity of receptors and/or low receptor density to seizuresusceptibility and pharmacoresistance in OHSCs.© 2009 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +49 30 450 528091; fax: +49 30 450 528962.E-mail address: uwe.heinemann@charite.de (U. Heinemann).

0920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.eplepsyres.2009.05.008

1

I

ApawG2icrrgisdesnBewahdtbm(acm

amrc2saPiirs(sapa

twnOmbldssh

tt

(

M

T

OtmrNthc0p2ibwA1ii

M

I(Ot1oaSmIivcwm1dtawpr(pptwv1m>w

14

ntroduction

bout 70% of adult patients diagnosed with mesial tem-oral lobe epilepsy have drug refractory seizures (Schmidtnd Loscher, 2005). In childhood 20—30% of patientsith partial epilepsies and more than 50% with Lennox-astaut syndrome are classified as intractable (Arts et al.,004). Proposed hypotheses on pathomechanisms involvedn pharmacoresistance may be classified into three generalategories: disease related mechanisms, genetics and drug-elated mechanisms (Schmidt and Loscher, 2005). Diseaseelated mechanisms may involve alterations in drug tar-ets. Genetic alterations due for example to polymorphismsn drug transporters may also contribute to pharmacore-istance (Remy and Beck, 2006). Finally tolerance as arug-related mechanism may be involved in reduced drugfficacy (Loscher and Schmidt, 2006). These hypotheses aretill under intensive investigation and the detailed mecha-isms leading to pharmacoresistance are largely unknown.esides clinical research (Summers et al., 2004; Iannettit al., 2005) and studies on brain tissue from patientsith drug-resistant seizures (Jandova et al., 2006; Oby etl., 2006) experimental models of drug-resistant epilepsyave been developed. Spontaneous seizures in epilepticogs and in amygdala kindled rats, and post status epilep-icus seizures after prolonged electrical stimulation of theasolateral amygdala all have been shown to be phar-acoresistant in subgroups of the experimental animals

Loscher, 1997; Brandt et al., 2004). All these models areposteriori models in that responders and non-responders

an be sorted out only after completion of the experi-ent.In contrast to animal in vivo models, seizure like

ctivity is a priori pharmacoresistant in some in vitroodels. Recurrent discharges induced in acute slices of

at temporal cortex by combined application of bicu-ulline and 4-aminopyridine (Bruckner and Heinemann,000) or developing during later stages of low magne-ium induced seizure like activity (Heinemann et al., 1994)re not blocked by clinically used antiepileptic drugs.harmacoresistant tonic—clonic seizure like events (SLEs)nduced by low magnesium have been reported in an intactmmature corticohippocampal preparation of 1-week oldats (Quilichini et al., 2003). We have recently demon-trated that in organotypic hippocampal slice culturesOHSCs) tonic—clonic SLEs either induced by low magne-ium or by 4-aminopyridine, are refractory to standardntiepileptic drugs such as valproic acid, carbamazepine,henytoin, phenobarbital and 1,4-benzodiazepines (Albus etl., 2008).

Drug refractory SLEs could be a property of immatureemporal cortex and as such transferred to slice cultures,hich are prepared from immature rats. The failure of phe-obarbital and 1,4-benzodiazepines to block the SLEs inHSCs prepared from immature rats (Albus et al., 2008)ay be due to the impaired/altered GABA-ergic mechanismsased on immaturity of receptors, low receptor density,ow concentration of synaptic availability of GABA and/or

epolarizing actions of GABA. To test the hypothesis of a pos-ible contribution of altered GABA-mediated mechanisms toeizure susceptibility and pharmacoresistance in OHSCs, weave investigated the effects of compounds, which produce

E

Rr

A. Wahab et al.

heir actions on GABA system with mechanisms differenthan those of phenobarbital and 1,4-benzodiazepines.

Parts of the findings have been published in abstract formWahab et al., 2007).

aterial and methods

est systems

rganotypic hippocampal slice cultures were prepared according tohe interface culture method (Stoppini et al., 1991) using cultureedia optimal for preparing and culturing neurons from postnatal

odent hippocampus (Stoppini et al., 1991; Brewer et al., 1993;oraberg et al., 1999). 6—12 day old Wistar rats were decapi-ated after receiving 4 mg kg−1 xylazin (Rompun©) and transverseippocampal slices (400 �m) were prepared with a McIlwain tissuehopper and placed on Millicell culture plate inserts (Millipore, PICM3050, 0.4 �m, 30 mm diameter) which were transferred to culturelates with 6 wells. Each well contained 1.1 ml medium composed of5 ml Hank’s balanced salt solution, 50 ml Opti-MEM and 25 ml heatnactivated horse serum. The culture plates were placed in an incu-ator at 36.5 ◦C with 5% CO2. After 3 days the medium was replacedith 1.1 ml of a chemically defined serum-free medium (Neurobasal) supplemented with 1 mM L-glutamine and B27 (Brewer et al.,993). This medium was changed every second day. Most slices werencubated for 5—35 days before experiment. However, in few slicesncubation time was up to 64 days (see Table 1).

easurements

n acute experiments slice cultures were continuously superfused4 ml min−1) with prewarmed (35.5 ± 0.2 ◦C) and oxygenated (95%2/5% CO2) minimal essential medium (MEM) containing in additiono amino acids and vitamins (in mM): NaCl 105, KCl 3, NaH2PO4.25, MgSO4 1.8, CaCl2 1.6, glucose 10, NaHCO3 26.3 mM (pH 7.40,smolality 265 mOsm kg−1). Concentrations of NaCl, and NaHCO3,nd the osmolality match the respective values in Neurobasal A.LEs were induced by superfusing the slice with MEM containing noagnesium ions and 5 mM KCl, referred to as low magnesium MEM.

ncreasing the potassium concentration from 3 to 5 mM resultedn a faster onset of SLEs particular during the first 2 weeks initro and was therefore used for all OHSCs. The extracellular con-entration of potassium [K+]o in the pyramidal cell layer of CA3as measured with a double barreled K+-selective/reference glassicroelectrode containing Fluka ionophore 60031 backfilled with

00 mM KCl/150 mM NaCl, respectively, prepared and tested asescribed (Lux and Neher, 1973). The reference electrode recordedhe DC-potential and population spikes in the pyramidal cell layer inrea CA3 in DC mode. In addition single unit and multiunit activitiesere recorded (AC-coupled) with tungsten-in-glass microelectrodesositioned in the pyramidal cell layers of area CA3 and CA1. Neu-onal activities and [K+]o were continuously recorded using Spike 2V 5.03, CED, Cambridge, UK) with sample rates of 20 kHz for fieldotentials and multi unit activity (filter cut-off 3 kHz) and 1 kHz forotassium signals (filter cut-off 0.3 kHz). Bipolar stimulation elec-rodes (tungsten wires with sharp tips, tip separation 80—200 �m)ere positioned at the hilus/CA3 border in order to assess sliceiability over time. For that population spikes were evoked every5—20 min by electrical stimulation (0.1 ms, 3—15 V) at half maxi-al and maximal stimulus strength. Slices displaying a decrease of

20% in population spike amplitude in the course of an experimentere discarded from analysis.

xperimental design

ecordings were commenced 1 h after transferring OHSCs to theecording chamber. Two experimental protocols were used. Protocol

Effects of GABA agonists and a GABA uptake inhibitor on pharmacoresistant seizure 115

Table 1 Overview of experimental data.

Drug Drug concentration n Postnatal days (PN)/days in vitro (DIV) Types of SLEs blocked (n/n)

Muscimol 1 �M 5 PN: 7, 7, 8, 8, 9 Tonic clonic 0/1DIV: 5, 26, 26, 32, 64 RSDs 0/3

Mixed 0/1

Muscimol 5 �M 5 PN: 7, 7, 7, 8, 9 Tonic clonic 0/1DIV: 10, 32, 35, 35, 64 RSDs 1/2

Mixed 1/2

Muscimol 10 �M 5 PN: 7, 8, 9, 9, 9 Tonic clonic 1/1DIV: 21, 21, 26, 32, 64 RSDs 2/2

Mixed 2/2

Isoguvacine 10 �M 6 PN: 8, 8, 8, 8, 9, 9 Tonic clonic 0/3DIV: 10, 20, 21, 28, 33, 33 RSDs 0/2

Mixed 0/1

Isoguvacine 50 �M 6 PN: 8, 8, 9, 9, 9, 9 Tonic clonic 1/3DIV: 12, 12, 20, 28, 28, 33 RSDs 1/2

Mixed 0/1

Isoguvacine 100 �M 5 PN: 8, 8, 8, 8, 9, Tonic clonic 3/3DIV: 21, 21, 21, 22, 22 RSDs 1/1

Mixed 1/1

Alfaxalone 20 �M 5 DIV: 17, 17, 19, 24, 24 Tonic clonic 0/1PN: 6, 6, 10, 10, 10 RSDs 0/3

Mixed 0/1

Alfaxalone 50 �M 5 DIV: 19, 20, 20, 24, 24 Tonic clonic 2/2PN: 10, 10, 10, 10, 10 RSDs 1/1

Mixed 1/2

(+)-Baclofen 25 �M 5 PN: 9, 9, 9, 9, 12, Tonic clonic 0/4DIV: 8, 14, 14, 33, 40 RSDs 0/1

(+)-Baclofen 200 �M 6 PN: 9, 9, 9, 9, 9, 9 Tonic clonic 0/5DIV: 14, 22, 33, 33, 40, 40 RSDs 0/1

Nipecotic acid 200 �M 9 DIV: 5, 7, 16, 20, 20, 20, 20, 22, 35 Tonic clonic 0/6PN: 6, 6, 6, 6, 7, 8, 8, 8, 12 RSDs 0/2

Mixed 0/1

Nipecotic acid 500 �M 5 PN: 8, 8, 8, 8, 12 Tonic clonic 4/4DIV: 7, 16, 16, 22, 24, RSDs 1/1

Nipecotic acid 1 mM 7 PN: 8, 8, 8, 8, 8, 8, 8 Tonic clonic 4/4DIV: 5, 13, 13, 13, 16, 24, 24 RSDs 1/1

Mixed 2/2

Taurine 2 mM 4 PN 8, 8, 8, 8 Tonic clonic 0/1DIV 13, 13, 29, 29 RSDs 0/1

Saol

D

1 was designed to assess the ability of drugs to block the induc-tion and maintenance of SLEs. It consisted of three periods withSLEs induced by low magnesium MEM each lasting 30—60 min, sepa-rated from each other by recovery periods with normal MEM lastingbetween 90 and 120 min. During the first and third SLE period nodrug were applied (control, and washout respectively) whereas the

second SLE period was run with a given drug (drug period) whichhad been preincubated for 30—60 min in normal MEM before addinginto low Mg MEM.

Protocol 2 was designed to evaluate the ability of a drug toblock ongoing seizure like activity. After 20—30 min of ongoing

SpS(

Mixed 0/2

LEs induced by low magnesium MEM (control period) a drug wasdded for 30—60 min (drug period). After the drug period SLEs werebserved for another 40—60 min during wash out of the drug withow magnesium MEM (wash out period).

ata analysis and statistical procedures

LEs were considered as blocked if (protocol 1) during the drugeriod SLEs did not appear for at least 30 min, or if (protocol 2)LEs in the control period disappeared after wash in of a drugdrug period). In both protocols SLEs were rated as blocked dur-

116 A. Wahab et al.

Figure 1 Types of SLEs in OHSCs recorded in pyramidal cell layer of area CA3. Double barreled potassium sensitive microelectrodeswere used to monitor DC field potentials (FP) and changes in extracellular potassium concentration ([K+]o). (A) Left: recurrenttonic—clonic SLEs. One SLE (a) is shown on an expanded time scale in the middle. High amplitude field potentials (1), the tonic-likeperiod (2) and the clonic-like period (3) are shown on an expanded time scale on the right (a1, a2 and a3, respectively). (B) RSD-SLE.The initial part of the RSD-SLE (b) is shown on expanded time scale; in the middle. The initial few high amplitude spikes (1) thetonic-like period (2) and the recurrent short discharges resembling clonic-like events (3) are shown on an expanded time scale onthe right (b1, b2 and b3, respectively). (C) Mixed SLE changing after four tonic—clonic events to recurrent short discharges. The lasttst

FwA(ev

onic—clonic SLE and its transition into RSDs (c) is shown on an expapikes (1) the tonic-like period (2) and the recurrent short dischargeime scale on the right (c1, c2 and c3, respectively).

igure 2 (A) (a—c) shows the failure of a low concentration of musith low magnesium MEM; (b) muscimol 1 �M and low magnesium Mt 1 �M muscimol the RSD-SLE reversibly changed to tonic—clonic SLB) (a—c) shows that 10 �M muscimol reversibly blocked SLEs inducexamined after 26 DIV. [K+]o and field potentials (FP; d.c. mode) witro; MEM, minimal essential medium; OHSC, organotypic hippocam

nded time scale in the middle. The initial few high amplitudes resembling clonic-like events (3) are shown on an expanded

cimol to block SLEs induced by low magnesium MEM. (a) ControlEM; (c) after washout of muscimol with low magnesium MEM.Es. The OHSC was explanted at P 8 and examined after 26 DIV.d by low magnesium MEM. The OHSC was explanted at P 8 andere recorded in the pyramidal cell layer of CA3. DIV, days inpal slice culture; P, postnatal day; SLE, seizure-like event.

Effects of GABA agonists and a GABA uptake inhibitor on pharmacoresistant seizure 117

Figure 3 Effects of drugs on the amplitudes and time properties of tonic—clonic SLEs in OHSCs. Changes are presented as thevalue present during treatment with the drug normalized to the control value. Each column represents the mean value and theSEM calculated on the basis of the test runs listed in Table 1 (significance levels: (*) <0.05, (**) <0.01; (***) <0.001). The number oftonic—clonic SLEs evaluated for each parameter was ≥6. SLE parameters were labeled as follows: (A) latency of onset of 1. SLE; (B)maximal extracellular potassium concentration during tonic-like period; (C) duration of SLE; (D) duration of tonic-like period; (E)

ansiee of n

et

D

HmNKt3T

fidtaodt

duration of clonic-like period; (F) frequency of field potential trevents; (H) frequency of clonic-like events; (I) maximal amplitud

ing the drug period only if they reappeared during the wash outperiod.

To quantify the effects of drugs on SLEs in cases where SLEswere not blocked we measured parameters characterizing staticand dynamic properties of tonic—clonic SLEs. SLE parameterswere labeled (see Fig. 3) as follows: (A) latency of onset offirst SLE (s); (B) maximal extracellular potassium concentrationduring tonic-like period (mM); (C) duration of SLE (s); (D) dura-tion of tonic-like period (s); (E) duration of clonic-like period(s); (F) frequency of field potential transients during tonic-likeperiod (/s); (G) average duration of clonic-like events (s); (H)frequency of clonic-like events (/s); (I) maximal amplitude of neg-ative potential shift (mV); (J) duration of negative potential shift(s).

The parameter values of the first SLE in the pre drug periodwere compared with the respective values of the first SLE duringantiepileptic drug treatment and of the first SLE after wash outof the antiepileptic drug. Comparisons were repeated for the sec-ond SLEs, then for the third SLEs, and so forth. This procedure waschosen to compensate for minor changes in parameter values over

time in the course of recurrent seizure activity which were presentin some OHSCs.

Drug effects were accepted as significant only when controlvalues changed significantly and when drug effects significantlyreversed after wash out of the drug (significance level P < 0.05). Drug

R

Ac

nts during tonic-like period; (G) average duration of clonic-likeegative potential shift; (J) duration of negative potential shift.

ffects were presented as values during drug treatment normalizedo control values (set to 100%).

rugs, chemical reagents and other materials

ank’s balanced salt solution, HEPES, NaHCO3, amino acids, vita-ins and L-glutamine from Biochrom AG, 12247 Berlin, Germany;eurobasal A, B 27, Opti-MEM from Invitrogen GmbH, 76131,arlsruhe, Germany; R(+)-baclofen, S(−)-baclofen, (±)-baclofen,aurine, nipecotic acid, muscimol and alfaxalone (5�-Pregnan-�-ol-11,20-dione) from Sigma—Aldrich Chemie, GmbH, 82824aufkirchen, Germany; isoguvacine from Tocris (Ellisville, MO, USA).

Muscimol was dissolved in 0.05 M HCl and diluted in MEM to thenal concentration of 1, 5 or 10 �M. Alfaxalone was dissolved inimethylsulfoxide (DMSO) and diluted in MEM to a final concentra-ion of DMSO of ≤0.05% in MEM. DMSO at these concentrations waslso present in the control experiments and did not have any effectn tested SLE parameters. All other compounds were dissolved inistilled water, stored as stock solutions and diluted (×200—1000)o their final concentrations in MEM.

esults

pplication of low magnesium MEM induced SLEs syn-hronous in area CA3 and CA1, which were associated with

1

rahfoipaqniuD

icFlttwcTfc(cn

M

T3SapbaOSn

Stim(

vniooctdoobr

i

aa1t

aicetrh(

A

Tira2sat

B

T1i(boaateinb

N

Tanbd5eiapcs

18

ises in [K+]o of usually more than 2 mM. SLEs were char-cterized by a negative potential shift superimposed byigh frequency field potential transients (tonic-like period)ollowed by a period in which clonic-like afterdischargesccurred. The tonic-like phase is accompanied with a peakn [K+]o, whereas clonic-like phase is accompanied with rip-les in [K+]o (Fig. 1). The recovery and undershoot of [K+]o

ssociated with the late phase of a SLE and the subse-uent suppression of neuronal activity, respectively, wereot different between pre- and post-drug controls, indicat-ng that the function of the sodium/potassium-ATPase wasnaffected by seizure activity (Heinemann and Lux, 1975;’Ambrosio et al., 2002).

Three different types of SLEs were observed. The major-ty of SLEs consisted of a tonic-like period followed by alonic-like period of limited duration (tonic—clonic SLEs;ig. 1A). In a number of slices recurrent short clonic-ike discharges (RSDs) started immediately after the firstonic-like discharge (Fig. 1B) and accordingly the SLEs wereermed RSD-SLEs. Slice cultures in which tonic—clonic SLEsere eventually replaced by recurrent short clonic-like dis-harges were characterized as mixed type SLEs (Fig. 1C).onic—clonic SLEs, RSD-SLEs and mixed type SLEs accountedor 50%, 29% and 21%, of the total population (n = 78) asompared with 52%, 23% and 25% in a previous investigationAlbus et al., 2008). Tonic—clonic SLEs started 299.5 s [95%onfidence interval 223;376] after starting lowering of mag-esium concentration and lasted on average 121 s [98;143].

uscimol and isoguvacine

he effects of muscimol and isoguvacine were analyzed in2 test runs. Muscimol applied with 1 �M (n = 5) did not blockLEs in any of the 5 OHCS tested (Fig. 2A; Table 1). Muscimolt 10 �M (n = 5) (Fig. 2B) and isoguvacine at 100 �M (n = 5)revented the induction of all types of SLEs or reversiblylocked ongoing SLEs in all OHSCs tested (Table 1). Muscimolt a concentration of 5 �M blocked SLEs in two out of fiveHSC. Isoguvacine at a concentration of 50 �M blocked theLEs in 2 out of 6 OHSCs and at a concentration of 10 �M inone of the six OHSCs tested (Table 1).

After starting the application of GABAA agonists ongoingLEs became blocked after 700 and 1080 s, respectively, inhe two cases with 5 �M muscimol and after 540 and 780 sn the two cases with 50 �M isoguvacine. With 10 �M musci-ol blockade of ongoing SLEs became effective after 96.7

75;120) s and with 100 �M isoguvacine after 90 (70;111) s.Concentrations of muscimol (Fig. 3A and B) and isogu-

acine (Fig. 3 D and E) ineffective in blocking SLEsevertheless induced significant dose dependent changesn SLE parameters. Muscimol 1 �M increased the latency ofccurrence of the first SLE and strongly reduced the durationf SLEs and the duration of clonic-like events. At the higheroncentration in addition the increase in [K+]o the dura-ion of the tonic-like period, the frequency of FP transientsuring the tonic-like period and amplitude and durationf negative potential shifts were reduced. Similar changes

ccurred under isoguvacine. The shortening of SLEs causedy muscimol and isoguvacine was mainly due to a strongeduction of the duration of the clonic-like period.

RSD-SLEs reversibly changed into tonic—clonic SLEs atntermediate and low concentrations of muscimol (Fig. 2A)

inca(

A. Wahab et al.

nd isoguvacine. A similar effect has been reported incute hippocampal—entorhinal cortex slices (Pfeiffer et al.,996) where late recurrent discharges were converted toonic—clonic SLEs under 5—25 �M muscimol.

The GABAA antagonist bicuculline (10 �M) induced highmplitude population spikes associated with transientncreases of [K+]o resembling interictal activity. When bicu-ulline was applied during ongoing seizure like activitylicited by low magnesium MEM (n = 2; data not shown)he pattern of tonic—clonic SLEs changed to a patternesembling late recurrent discharges as seen in acuteippocampal—entorhinal slices under the same treatmentPfeiffer et al., 1996).

lfaxalone

he effects of the neurosteroid alfaxalone were analyzedn 10 test runs. Alfaxalone at a concentration of 50 �Meversibly blocked ongoing SLEs regardless of seizure typefter 1087 (921;1225) s (Table 1). At a concentration of0 �M it did not block ongoing SLEs but significantly reducedtrength and duration of SLEs (Fig. 3C). Similar to muscimolnd isoguvacine alfaxalone reversibly converted RSD-SLEs toonic—clonic SLEs (Fig. 4A).

aclofen

he GABAB receptor agonist (+)-baclofen was analyzed in1 test runs. It neither prevented the induction of SLEs nornhibited ongoing SLEs at concentrations of 25 and 200 �MFig. 4B; Table 1). However, in both concentrations (+)-aclofen shortened the duration of SLE and the durationsf both clonic and tonic-like periods (Fig. 3 G and H). Atconcentration of 200 �M in addition the increase of [K+]o

nd the duration of the negative potential shift during theonic-like phase of an SLE were significantly reduced. Othernantiomers, (−)-baclofen (n = 4) and (±)-baclofen (n = 4),n concentrations of 25 and 200 �M were also tested (dataot shown) and these were less effective on SLEs than (+)-aclofen.

ipecotic acid, taurine

he effects of nipecotic acid, a GABA uptake blocker, werenalyzed in 21 test runs. At a concentration of 200 �Mipecotic acid neither prevented the generation of SLEs norlocked ongoing SLEs but strongly modified strength anduration of SLEs (Fig. 3F; Table 1). At concentrations of00 �M and 1 mM nipecotic acid blocked SLEs (Fig. 5). How-ver, in 4 of the 5 OHSCs tested with 500 �M (Fig. 5A) andn four of the seven OHSCs tested with 1 mM single largemplitude population events associated with sharp transienteaks in [K+]o occurred under nipecotic acid. Reversiblehanges of RSD-SLEs or mixed SLEs in tonic—clonic SLEs aseen at intermediate and low concentrations of muscimol,

soguvacine and alfaxalone were also observed at 200 �Mipecotic acid in four of the six cases. Taurine at the con-entration of 2 mM (4 test runs) did not block ongoing seizurectivity and induced only minor changes in seizure patterndata not shown).

Effects of GABA agonists and a GABA uptake inhibitor on pharmacoresistant seizure 119

Figure 4 (A) (a—c) shows the failure of a low concentration of alfaxalone to block SLEs induced by low magnesium MEM. (a) Controlwith low magnesium MEM; (b) alfaxalone 20 �M and low magnesium MEM; (c) after washout of alfaxalone with low magnesium MEM.Application of 20 �M alfaxalone reversibly changed RSD-SLE into tonic—clonic SLEs. The OHSC was explanted at P 10 and examinedafter 24 DIV. (B) (a—c) shows the failure of 200 �M (+)-baclofen to block SLEs induced by low magnesium MEM. (a) Control withlow magnesium MEM, (b) (+)-baclofen and low magnesium MEM; (c) after washout of (+)-baclofen with low magnesium MEM. Under(+)-baclofen the duration of SLEs, in particular the duration of the clonic-like periods was reduced. The OHSC was explanted at P 9

uplem; O

and examined after 33 DIV. [K+]o and field potentials (FP; a.c. colayer of CA3. DIV, days in vitro; MEM, minimal essential mediuSLE, seizure-like event.

Discussion

We tested in the present study the effects of compoundsthat enhance GABA-mediated actions on SLEs in OHSCs.

Wdci

Figure 5 (A) (a—c) shows the blockade of low magnesium MEM indof 500 �M nipecotic acid. (a) Control with low magnesium MEM; (b)nipecotic acid with low magnesium MEM. The OHSC was explantednipecotic acid reversibly blocks a RSD-SLE induced by low magnesiacid and low magnesium MEM; (c) after washout of nipecotic acid wexamined after 5 DIV. [K+]o and field potentials (FP; a.c. coupled) wvitro; MEM, minimal essential medium; OHSC, organotypic hippocam

d in (A); D.C. mode in (B)) were recorded in the pyramidal cellHSC, organotypic hippocampal slice culture; P, postnatal day;

e found that low magnesium induced SLEs were doseependently suppressed by GABAA receptor agonists mus-imol, isoguvacine and alfaxalone and by the GABA uptakenhibitor nipecotic acid, whereas the GABAB receptor ago-

uced SLEs and appearance of interictal spikes after applicationnipecotic acid and low magnesium MEM; (c) after washout ofat P 8 and examined after 16 DIV. (B) (a—c) shows that 1 mMum MEM. (a) Control with low magnesium MEM; (b) nipecoticith low magnesium MEM. The OHSC was explanted at P 8 andere recorded in the pyramidal cell layer of CA3. DIV, days in

pal slice culture; P, postnatal day; SLE, seizure-like event.

1

nafifrar

raaGGcipacnsrbssSnsrta

sSsmiidoacasadeatrtsebd

tis1nrw

5ShItoipr1on

baatoto2ePp(cienrtaFctt1aaIoparIrsroGttiaoO

20

ist baclofen attenuated but did not suppress SLE. Taurine,glycine receptor agonist, had no effects on SLEs. These

ndings demonstrate that in OHSCs GABAA receptors areunctional and mediate inhibitory effects on SLEs. Theesults do not support the hypothesis that in OHSC alterednd/or impaired GABA-ergic mechanisms contribute to drug-esistant seizure like activity.

Our findings in OHSCs are similar to observations in adultats where low magnesium induced tonic—clonic SLEs indult rat acute entorhinal cortex—hippocampal slices ofdult rats were blocked by GABAA receptor agonists andABA uptake inhibitors and only partially blocked by theABAB receptor agonist baclofen (Pfeiffer et al., 1996). Drugoncentrations effective in blocking SLEs were in fact, lowern OHSCs than in acute adult slices. In acute slices a com-lete suppression of SLEs and recurrent short discharges waschieved with 25 �M muscimol, whereas the effective con-entrations in OHSCs for muscimol were 10 �M. Similarlyipecotic acid 2.5—5 mM blocked the SLEs and recurrenthort discharges in acute slices whereas only 0.5 mM wasequired in OHSCs. Baclofen at a concentration of 100 �Mlocked tonic—clonic SLEs in the entorhinal cortex in onlyome of the acute slices whereas in OHSCs baclofen at 25 �Mignificantly shortened the duration of pharmacoresistantLEs in CA3 and 200 �M baclofen further attenuated but didot block SLEs. Limited effects of baclofen on SLEs are pos-ibly caused by rapid desensitization of postsynaptic GABAB

eceptor activation (Xiong and Stringer, 2001) and presynap-ic GABAB mediated inhibition of GABA release (Tosetti etl., 2004) following a sustained application of baclofen.

Neurosteroids preferentially act on extrasynaptic deltaubunit containing GABA—receptors (Mihalek et al., 1999;pigelman et al., 2003). They increase the conductance ofingle GABA activated channels and of open probability andean open time (Gaul et al., 2007) and thus mediate tonic

nhibition and control action potential firing and synapticnhibition of dentate granule cells and hippocampal pyrami-al neurons. In that respect their effects compare to effectsf barbiturates which however do not to block seizure likectivity (Albus et al., 2008) in OHSCs at clinically used con-entrations. In OHSCs the synthetic neurosteroid alfaxalonet 20 �M attenuated and at 50 �M suppressed low magne-ium induced SLEs. In addition to blocking tonic inhibitionlfaxalone at concentrations used in our experiments alsoirectly activates high conductance chloride channels (Gault al., 2007). It is well known that barbiturates and benzodi-zepines mediate phasic inhibition but they can also mediateonic inhibition by activating delta subunit containing GABAeceptors and �5 subunit containing GABA receptors respec-ively (Feng et al., 2004; Prenosil et al., 2006). Whetherelective effects of alfaxalone on tonic inhibition or directffects on chloride channels account for the differencesetween effects of alfaxalone and barbiturates and benzo-iazepines on SLEs is presently unclear.

Nipecotic acid is a GABA uptake inhibitor that in addi-ion can induce facilitated transport of GABA from thentracellular to the extracellular environment without a

ignificant concomitant block of uptake (Solis and Nicoll,992; Honmou et al., 1995; Patrylo et al., 2001). Theseipecotic acid-evoked responses are mediated by GABAA

eceptors and are blocked by picrotoxin. Nipecotic acidas unique among the compounds tested in that in about

D

Im

A. Wahab et al.

0% of the OHSCs investigated it caused a change fromLEs to a pattern resembling interictal activity in rat acuteippocampal—entorhinal cortex slices (Nagao et al., 1996).n OHSCs this activity was characterized by single popula-ion events synchronized between CA3 and CA1 (unpublishedbservations) associated with relatively long lasting (5—10 s)ncreases in [K+]o. Similar events in rat acute hippocam-al slices have been demonstrated to depend on a GABAA

eceptor mediated bicarbonate conductance (Staley et al.,995; Lamsa and Kaila, 1997). A possibly GABA-ergic naturef these potentials seen under nipecotic acid was however,ot tested in our experiments.

The failure of benzodiazepines and phenobarbital tolock the SLEs at therapeutic doses (Albus et al., 2008)nd the effectiveness of muscimol, isoguvacine, nipecoticcid, and alfaxalone to block the SLEs may be because ofheir differences in mechanisms to modulate the responsesf GABA as they have different binding sites at GABAA recep-or complex (Sieghart, 1992). However, the ineffectivenessf benzodiazepines in blocking SLEs in OHSCs (Albus et al.,008) and in immature rats acute hippocampal slices (Isaevt al., 2005) cannot be attributed to a lack of receptors. At10, the GABAA receptor subunits �1 and �2 in the rat hip-ocampus mediating the anticonvulsive actions of diazepamKralic et al., 2002; Fradley et al., 2007) are present inoncentrations only slightly higher (�1) or lower (�2) thann adults, and interneurons in the hippocampus stronglyxpress the �1 subunit (Yu et al., 2006). Their ineffective-ess in suppressing SLEs in OHSCs is also not due to theundown of GABA-mediated functions. Studies have shownhat during status epilepticus, GABA-ergic mechanisms failnd seizures become self-sustaining and pharmacoresistant.or example, benzodiazepines are effective early in theourse of status epilepticus but significant pharmacoresis-ance develops within few minutes and benzodiazepines failo arrest status epilepticus by 45 min (Kapur and Macdonald,997; Mazarati et al., 1998; Treiman et al., 1998; Feng etl., 2008). It is feasible that after prolonged epileptiformctivity the releasable pool of GABA may be diminished.ntracellular recordings have shown that prolonged washoutf magnesium causes decreased GABA-ergic transmission,robably due to a rundown of GABAA receptor functionfter functional deterioration of phosphokinase A whichequires Mg2+-ATP as a substrate (Whittington et al., 1995).n addition, a rapid modification in the postsynaptic GABAA

eceptor population occurs during the prolonged seizures oftatus epilepticus. This rapid modification contributes to aeduction in GABA-mediated inhibition and the developmentf benzodiazepine pharmacoresistance (Naylor et al., 2005;oodkin et al., 2007). We found however that the ineffec-iveness of 1,4-benzodiazipines and the other compoundsested in this and the earlier report (Albus et al., 2008)n suppressing SLEs was independent on whether or not wepplied them before or after inducing SLEs thus rundownf GABA may not be responsible for pharmacoresistance inHSCs.

epolarizing/excitatory actions of GABA

n the peri- and early postnatal period of rat GABAA receptorediated responses in immature neurons are depolarizing

rma

teo

tctt(cou2e

A

TvvcL

ir

c

R

A

A

B

B

B

B

C

C

D

Effects of GABA agonists and a GABA uptake inhibitor on pha

(Cherubini et al., 1991; Luhmann and Prince, 1991; Ben-Ari,2002; Romo-Parra et al., 2008 but see data in mice Wonget al., 2005): probably caused by an elevated intracellu-lar chloride concentration maintained by the activity of theNa+—K+—2Cl−-cotransporter isoform 1 (NKCC1) (Rohrboughand Spitzer, 1996). However, depolarizing synaptic responsesmay still be inhibitory in immature neurons by a shuntingmechanism (Chen et al., 1996; Psarropoulou and Descombes,1999; Lamsa et al., 2000; Wells et al., 2000; Isaev et al.,2007).

According to our study the net effect of GABA onSLEs in OHSCs is inhibitory and therefore, a depolariz-ing/excitatory action of GABA is unlikely to contribute tothe susceptibility of OHSCs to seizure like activity andto the pharmacoresistance in our model. These conclu-sions are supported by further observations (unpublished).GABAA agonists reversibly suppressed spontaneous neuronalactivity at concentrations which usually were ineffec-tive in suppressing SLEs. The GABAA antagonist bicucullineincreased spontaneous neuronal activity and induced burstlike firing resembling recurrent short discharges and interic-tal activity. When combining bicuculline and low magnesiumSLEs changed their tonic—clonic pattern to a patternresembling late recurrent discharges as seen in acutehippocampal—entorhinal slices under the same treatment(Pfeiffer et al., 1996).

Our findings are in agreement with previous findingsof other groups in acute immature slices. For example,in an intact 1-week old acute corticohippocampal forma-tion in vitro low magnesium induced tonic—clonic SLEs areresistant to a number of antiepileptic drugs (Quilichini etal., 2003) but these SLEs can be blocked with isoguvacine(Quilichini et al., 2002). In different in vivo experiments andacute hippocampal and neocortical preparations from imma-ture animals, the application of GABAA receptor antagonistsinduce epileptiform activities and exacerbate seizure-likeactivity when added with other epileptogenic agents andconditions (Swann and Brady, 1984; Daval and Sarfati, 1987;Gomez-Di Cesare et al., 1997; Khalilov et al., 1997; Wellset al., 2000; Isaev et al., 2005). It has also been shown thatisoguvacine blocked the SLEs in high potassium low magne-sium in vivo model in immature rat (Isaev et al., 2005). Incontrast, in different in vitro seizure models, GABAA recep-tor antagonists significantly or even completely suppressesseizure-like events, whereas GABAA agonists increases thefrequency and duration of the seizure-like events (Dzhalaand Staley, 2003; Khalilov et al., 2003; Khazipov et al.,2004). It has been suggested that Cl− accumulation inimmature neurons is due to the overexpression of NKCC1transporter which is responsible for depolarizing actions ofGABA (Rohrbough and Spitzer, 1996; Dzhala et al., 2005).Bumetanide, a NKCC1 transporter inhibitor, has been shownto inhibit the SLEs in acute high-potassium model in hip-pocampal slices (Dzhala et al., 2005), whereas it has beenshown to increase the frequency of ILE induced by 1 �Mkainate (Kilb et al., 2007). In OHSCs explanted from 2 to7 days old rats and incubated for 6—10 days (n = 8), we

have found that 10—20 �M bumetanide did not affect eitherlow magnesium induced SLEs or the phenytoin resistanceof low magnesium induced SLEs but reversibly increasedspontaneous activity when infused with normal MEM (unpub-lished observations). Thus, the role of GABA in controlling

D

coresistant seizure 121

he epileptiform activity in the immature brain seems to bextremely complex with a net effect depending on varietyf factors.

Our study does not exclude the possibility that suscep-ibility to seizure like activity and resistance of SLEs tolinically employed antiepileptic drugs is caused by imma-urity of receptors and channels other than subservinghe GABA system. As discussed in more detail elsewhereAlbus et al., 2008) other candidate mechanisms of pharma-oresistance are aberrant recurrent excitatory connectivity,verexpression of drug transporters and changes in molec-lar drug targets of antiepileptic drugs (Remy and Beck,006). These hypothesized mechanisms are presently underxperimental investigation.

cknowledgements

his work was supported by set (Stiftung zur Erforschungon Ersatz- und Ergänzungsmethoden zur Einschränkungon Tierversuchen) and by DFG grant SFB-TR3 and Epi-ure grants. We are grateful for technical assistance to Ch.ehmann.

We confirm that we have read the Journal’s position onssues involved in ethical publication and affirm that thiseport is consistent with those guidelines.

None of the authors has any conflict of interest to dis-lose.

eferences

lbus, K., Wahab, A., Heinemann, U., 2008. Standard antiepilep-tic drugs fail to block epileptiform activity in rat organotypichippocampal slice cultures. Br. J. Pharmacol. 154, 709—724.

rts, W.F., Brouwer, O.F., Peters, A.C., Stroink, H., Peeters, E.A.,Schmitz, P.I., van Donselaar, C.A., Geerts, A.T., 2004. Course andprognosis of childhood epilepsy: 5-year follow-up of the Dutchstudy of epilepsy in childhood. Brain 127, 1774—1784.

en-Ari, Y., 2002. Excitatory actions of GABA during development:the nature of the nurture. Nat. Rev. Neurosci. 3, 728—739.

randt, C., Volk, H.A., Loscher, W., 2004. Striking differencesin individual anticonvulsant response to phenobarbital in ratswith spontaneous seizures after status epilepticus. Epilepsia 45,1488—1497.

rewer, G.J., Torricelli, J.R., Evege, E.K., Price, P.J., 1993. Opti-mized survival of hippocampal neurons in B27-supplementedNeurobasal, a new serum-free medium combination. J. Neu-rosci. Res. 35, 567—576.

ruckner, C., Heinemann, U., 2000. Effects of standard anti-convulsant drugs on different patterns of epileptiform dis-charges induced by 4-aminopyridine in combined entorhinalcortex—hippocampal slices. Brain Res. 859, 15—20.

hen, G., Trombley, P.Q., van den Pol, A.N., 1996. Excitatory actionsof GABA in developing rat hypothalamic neurons. J. Physiol. 494(Pt 2), 451—464.

herubini, E., Gaiarsa, J.L., Ben-Ari, Y., 1991. GABA: an excita-tory transmitter in early postnatal life. Trends Neurosci. 14,515—519.

’Ambrosio, R., Gordon, D.S., Winn, H.R., 2002. Differential role of

KIR channel and Na+/K+-pump in the regulation of extracellularK+ in rat hippocampus. J. Neurophysiol. 87, 87—102.

aval, J.L., Sarfati, A., 1987. Effects of bicuculline-induced seizureson benzodiazepine and adenosine receptors in developing ratbrain. Life Sci. 41, 1685—1693.

1

D

D

F

F

F

G

G

G

H

H

H

I

I

I

J

K

K

K

K

K

K

L

L

L

L

L

L

M

M

N

N

N

O

P

P

P

22

zhala, V.I., Staley, K.J., 2003. Excitatory actions of endogenouslyreleased GABA contribute to initiation of ictal epileptiformactivity in the developing hippocampus. J. Neurosci. 23,1840—1846.

zhala, V.I., Talos, D.M., Sdrulla, D.A., Brumback, A.C., Mathews,G.C., Benke, T.A., Delpire, E., Jensen, F.E., Staley, K.J., 2005.NKCC1 transporter facilitates seizures in the developing brain.Nat. Med. 11, 1205—1213.

eng, H.J., Bianchi, M.T., Macdonald, R.L., 2004. Pento-barbital differentially modulates alpha1beta3delta andalpha1beta3gamma2L GABAA receptor currents. Mol.Pharmacol. 66, 988—1003.

eng, H.J., Mathews, G.C., Kao, C., Macdonald, R.L., 2008. Alter-ations of GABAA-receptor function and allosteric modulationduring development of status epilepticus. J. Neurophysiol. 99,1285—1293.

radley, R.L., Guscott, M.R., Bull, S., Hallett, D.J., Goodacre, S.C.,Wafford, K.A., Garrett, E.M., Newman, R.J., O’Meara, G.F.,Whiting, P.J., Rosahl, T.W., Dawson, G.R., Reynolds, D.S., Atack,J.R., 2007. Differential contribution of GABAA receptor subtypesto the anticonvulsant efficacy of benzodiazepine site ligands. J.Psychopharmacol. 21, 384—391.

aul, S., Ozsarac, N., Liu, L., Fink, R.H., Gage, P.W., 2007. Theneuroactive steroids alphaxalone and pregnanolone increase theconductance of single GABAA channels in newborn rat hippocam-pal neurons. J. Steroid Biochem. Mol. Biol. 104, 35—44.

omez-Di Cesare, C.M., Smith, K.L., Rice, F.L., Swann, J.W., 1997.Axonal remodeling during postnatal maturation of CA3 hip-pocampal pyramidal neurons. J. Comp. Neurol. 384, 165—180.

oodkin, H.P., Sun, C., Yeh, J.L., Mangan, P.S., Kapur, J., 2007.GABAA receptor internalization during seizures. Epilepsia 48(Suppl 5), 109—113.

einemann, U., Dreier, J., Leschinger, A., Stabel, J., Draguhn, A.,Zhang, C., 1994. Effects of anticonvulsant drugs on hippocampalneurons. Hippocampus 4, 291—296.

einemann, U., Lux, H.D., 1975. Undershoots following stimulusinduced rises of extracellular potassium concentration in cere-bral cortex of cat. Brain Res. 93, 63—76.

onmou, O., Kocsis, J.D., Richerson, G.B., 1995. Gabapentin poten-tiates the conductance increase induced by nipecotic acid in CA1pyramidal neurons in vitro. Epilepsy Res. 20, 193—202.

annetti, P., Spalice, A., Parisi, P., 2005. Calcium-channel blockerverapamil administration in prolonged and refractory statusepilepticus. Epilepsia 46, 967—969.

saev, D., Isaeva, E., Khazipov, R., Holmes, G.L., 2005. Anticonvul-sant action of GABA in the high potassium-low magnesium modelof ictogenesis in the neonatal rat hippocampus in vivo and invitro. J. Neurophysiol. 94, 2987—2992.

saev, D., Isaeva, E., Khazipov, R., Holmes, G.L., 2007. Shuntingand hyperpolarizing GABAergic inhibition in the high-potassiummodel of ictogenesis in the developing rat hippocampus. Hip-pocampus 17, 210—219.

andova, K., Pasler, D., Antonio, L.L., Raue, C., Ji, S., Njunt-ing, M., Kann, O., Kovacs, R., Meencke, H.J., Cavalheiro,E.A., Heinemann, U., Gabriel, S., Lehmann, T.N., 2006.Carbamazepine-resistance in the epileptic dentate gyrus ofhuman hippocampal slices. Brain 129, 3290—3306.

apur, J., Macdonald, R.L., 1997. Rapid seizure-induced reductionof benzodiazepine and Zn2+ sensitivity of hippocampal dentategranule cell GABAA receptors. J. Neurosci. 17, 7532—7540.

halilov, I., Khazipov, R., Esclapez, M., Ben-Ari, Y., 1997. Bicucullineinduces ictal seizures in the intact hippocampus recorded invitro. Eur. J. Pharmacol. 319 (2—3), R5—6.

halilov, I., Holmes, G.L., Ben Ari, Y., 2003. In vitro formation of asecondary epileptogenic mirror focus by interhippocampal prop-agation of seizures. Nat. Neurosci. 6, 1079—1085.

hazipov, R., Khalilov, I., Tyzio, R., Morozova, E., Ben Ari, Y.,Holmes, G.L., 2004. Developmental changes in GABAergic

P

A. Wahab et al.

actions and seizure susceptibility in the rat hippocampus. Eur.J. Neurosci. 19, 590—600.

ilb, W., Sinning, A., Luhmann, H.J., 2007. Model-specific effectsof bumetanide on epileptiform activity in the in-vitro intacthippocampus of the newborn mouse. Neuropharmacology 53,524—533.

ralic, J.E., O’Buckley, T.K., Khisti, R.T., Hodge, C.W., Homan-ics, G.E., Morrow, A.L., 2002. GABAA receptor alpha-1 subunitdeletion alters receptor subtype assembly, pharmacological andbehavioral responses to benzodiazepines and zolpidem. Neu-ropharmacology 43, 685—694.

amsa, K., Kaila, K., 1997. Ionic mechanisms of spontaneousGABAergic events in rat hippocampal slices exposed to 4-aminopyridine. J. Neurophysiol. 78, 2582—2591.

amsa, K., Palva, J.M., Ruusuvuori, E., Kaila, K., Taira, T.,2000. Synaptic GABAA activation inhibits AMPA-kainate receptor-mediated bursting in the newborn (P0-P2) rat hippocampus. J.Neurophysiol. 83, 359—366.

oscher, W., 1997. Animal models of intractable epilepsy. Prog. Neu-robiol. 53, 239—258.

oscher, W., Schmidt, D., 2006. Experimental and clinical evidencefor loss of effect (tolerance) during prolonged treatment withantiepileptic drugs. Epilepsia 47, 1253—1284.

uhmann, H.J., Prince, D.A., 1991. Postnatal maturation of theGABAergic system in rat neocortex. J. Neurophysiol. 65,247—263.

ux, H.D., Neher, E., 1973. The equilibration time course of (K+) 0in cat cortex. Exp. Brain Res. 17, 190—205.

azarati, A.M., Baldwin, R.A., Sankar, R., Wasterlain, C.G., 1998.Time-dependent decrease in the effectiveness of antiepilep-tic drugs during the course of self-sustaining status epilepticus.Brain Res. 814, 179—185.

ihalek, R.M., Banerjee, P.K., Korpi, E.R., Quinlan, J.J., Fire-stone, L.L., Mi, Z.P., Lagenaur, C., Tretter, V., Sieghart, W.,Anagnostaras, S.G., Sage, J.R., Fanselow, M.S., Guidotti, A.,Spigelman, I., Li, Z., DeLorey, T.M., Olsen, R.W., Homanics,G.E., 1999. Attenuated sensitivity to neuroactive steroids in �-aminobutyrate type A receptor delta subunit knockout mice.Proc. Natl. Acad. Sci. USA 96, 12905—12910.

agao, T., Alonso, A., Avoli, M., 1996. Epileptiform activity inducedby pilocarpine in the rat hippocampal-entorhinal slice prepara-tion. Neuroscience 72, 399—408.

aylor, D.E., Liu, H., Wasterlain, C.G., 2005. Trafficking of GABAAreceptors, loss of inhibition, and a mechanism for pharmacore-sistance in status epilepticus. J. Neurosci. 25, 7724—7733.

oraberg, J., Kristensen, B.W., Zimmer, J., 1999. Markers for neu-ronal degeneration in organotypic slice cultures. Brain Res. BrainRes. Protoc. 3, 278—290.

by, E., Caccia, S., Vezzani, A., Moeddel, G., Hallene, K., Guiso, G.,Said, T., Bingaman, W., Marchi, N., Baumgartner, C., Pirker, S.,Czech, T., Lo, R.G., Janigro, D., 2006. In vitro responsivenessof human-drug-resistant tissue to antiepileptic drugs: insightsinto the mechanisms of pharmacoresistance. Brain Res. 1086,201—213.

atrylo, P.R., Spencer, D.D., Williamson, A., 2001. GABA uptake andheterotransport are impaired in the dentate gyrus of epilepticrats and humans with temporal lobe sclerosis. J. Neurophysiol.85, 1533—1542.

feiffer, M., Draguhn, A., Meierkord, H., Heinemann, U., 1996.Effects of gama-aminobutyric (GABA) agonists and GABA uptakeinhibitors on pharmacosensitive and pharmacoresistant epilep-tiform activity in vitro. Br. J. Pharmacol. 119, 569—577.

renosil, G.A., Schneider Gasser, E.M., Rudolph, U., Keist, R.,

Fritschy, J.M., Vogt, K.E., 2006. Specific subtypes of GABAAreceptors mediate phasic and tonic forms of inhibition in hip-pocampal pyramidal neurons. J. Neurophysiol. 96, 846—857.

sarropoulou, C., Descombes, S., 1999. Differential bicuculline-induced epileptogenesis in rat neonatal, juvenile and adult

rma

S

S

T

T

W

W

W

W

X

Effects of GABA agonists and a GABA uptake inhibitor on pha

CA3 pyramidal neurons in vitro. Dev Brain Res. 117, 117—120.

Quilichini, P.P., Diabira, D., Chiron, C., Ben-Ari, Y., Gozlan, H., 2002.Persistent epileptiform activity induced by low Mg2+ in intactimmature brain structures. Eur. J. Neurosci. 16, 850—860.

Quilichini, P.P., Diabira, D., Chiron, C., Milh, M., Ben-Ari, Y., Gozlan,H., 2003. Effects of antiepileptic drugs on refractory seizuresin the intact immature corticohippocampal formation in vitro.Epilepsia 44, 1365—1374.

Remy, S., Beck, H., 2006. Molecular and cellular mechanisms ofpharmacoresistance in epilepsy. Brain 129, 18—35.

Rohrbough, J., Spitzer, N.C., 1996. Regulation of intracellularCl− levels by Na(+)-dependent Cl− cotransport distinguishesdepolarizing from hyperpolarizing GABAA receptor-mediatedresponses in spinal neurons. J. Neurosci. 16, 82—91.

Romo-Parra, H., Trevino, M., Heinemann, U., Gutierrez, R., 2008.GABA actions in hippocampal area CA3 during postnatal devel-opment: differential shift from depolarizing to hyperpolarizingin somatic and dendritic compartments. J. Neurophysiol. 99,1523—1534.

Schmidt, D., Loscher, W., 2005. Drug resistance in epilepsy: putativeneurobiologic and clinical mechanisms. Epilepsia 46, 858—877.

Sieghart, W., 1992. GABAA receptors: ligand-gated Cl− ion channelsmodulated by multiple drug-binding sites. Trends Pharmacol.Sci. 13, 446—450.

Solis, J.M., Nicoll, R.A., 1992. Postsynaptic action of endogenousGABA released by nipecotic acid in the hippocampus. Neurosci.Lett. 147, 16—20.

Spigelman, I., Li, Z., Liang, J., Cagetti, E., Samzadeh, S., Mihalek,R.M., Homanics, G.E., Olsen, R.W., 2003. Reduced inhibition andsensitivity to neurosteroids in hippocampus of mice lacking theGABAA receptor delta subunit. J. Neurophysiol. 90, 903—910.

Staley, K.J., Soldo, B.L., Proctor, W.R., 1995. Ionic mechanisms of

neuronal excitation by inhibitory GABAA receptors. Science 269,977—981.

Stoppini, L., Buchs, P.A., Muller, D., 1991. A simple method fororganotypic cultures of nervous tissue. J. Neurosci. Methods 37,173—182.

Y

coresistant seizure 123

ummers, M.A., Moore, J.L., McAuley, J.W., 2004. Use of verapamilas a potential P-glycoprotein inhibitor in a patient with refrac-tory epilepsy. Ann. Pharmacother. 38, 1631—1634.

wann, J.W., Brady, R.J., 1984. Penicillin-induced epileptogenesisin immature rat CA3 hippocampal pyramidal cells. Brain Res.314, 243—254.

osetti, P., Bakels, R., Colin-Le Brun, I., Ferrand, N., Gaiarsa,J.L., Caillard, O., 2004. Acute desensitization of presynapticGABAB-mediated inhibition and induction of epileptiform dis-charges in the neonatal rat hippocampus. Eur. J. Neurosci. 19,3227—3234.

reiman, D.M., Meyers, P.D., Walton, N.Y., Collins, J.F., Colling, C.,Rowan, A.J., Handforth, A., Faught, E., Calabrese, V.P., Uthman,B.M., Ramsay, R.E., Mamdani, M.B., 1998. A comparison of fourtreatments for generalized convulsive status epilepticus Veter-ans Affairs status epilepticus cooperative study group. N. Engl.J. Med. 339, 792—798.

ahab, A., Albus, K., Heinemann, U., 2007. Effects of GABA agonistsand a GABA uptake inhibitor on pharmacoresistant epileptiformactivity in hippocampal slice cultures. Abstr. Soc. Neursci. 161,1.

ells, J.E., Porter, J.T., Agmon, A., 2000. GABAergic inhibitionsuppresses paroxysmal network activity in the neonatal rodenthippocampus and neocortex J. Neurosci. 20, 8822—8830.

hittington, M.A., Traub, R.D., Jefferys, J.G.R., 1995. Erosionof inhibition contributes to the progression of low magne-sium bursts in rat hippocampal slices. J. Physiol. London 468,723—734.

ong, T., Zhang, X.L., Asl, M.N., Wu, C.P., Carlen, P.L., Zhang, L.,2005. Postnatal development of intrinsic GABAergic rhythms inmouse hippocampus. Neuroscience 134, 107—120.

iong, Z.Q., Stringer, J.L., 2001. Effects of postsynaptic GABABreceptor activation on epileptiform activity in hippocampal

slices. Neuropharmacology 40, 131—138.

u, Z.Y., Wang, W., Fritschy, J.M., Witte, O.W., Redecker, C., 2006.Changes in neocortical and hippocampal GABAA receptor subunitdistribution during brain maturation and aging. Brain Res. 1099,73—81.