S-Methylcysteine may be a Causal Factor inMonohalomethane Neurotoxicity

Floria Pancetti1,2, Marcelo Oyarce1, Marcela Aranda2, Jorge Parodi3, Luis G. Aguayo3,Bernardo Morales2, Gotz Westphal4, Michael Muller4, Ernst Hallier4, Marc L. Zeise1,*

1Department of Management in Agriculture, Technological Faculty, University of Santiago de Chile, Santiago, Chile2Department of Biological Sciences, Faculty of Chemistry and Biology, University of Santiago de Chile, Santiago, Chile3Laboratory of Neurophysiology, Department of Physiology, University of Concepcion, Concepcion, Chile4Department of Occupational and Social Medicine, Georg-August-University, Gottingen, Germany

Received 3 July 2003; accepted 10 January 2004

Available online 12 March 2004

Abstract

S-Methylcysteine (SMC) is formed after exposure to monohalomethanes in rodents as well as in humans. The present

study was performed to study whether SMC, directly or indirectly, contributes to the well-known neurotoxicity of

monohalomethanes. We have investigated the effects of acute exposure to SMC by means of electrophysiolocal

measurements in freshly prepared hippocampal slices and dissociated hippocampal neurons in culture. For longer-

term exposures (24 h) we have used organotypic cultures (2 weeks in culture), taking electrophysiologic recordings and

assessing membrane integrity with propidium iodide (PI) fluorescence. We found that only high concentrations of SMC

(10�2 M; exposure time 30 min) in freshly isolated slices of adult rats reduce synaptically evoked population spikes in the

CA1 region. This effect was at least partially reversible. In organotypic cultures, at 5 � 10�5 M after 24 h of exposure,

SMC compromises membrane integrity as revealed by PI fluorescence, only in the dentate gyrus, spreading to pyramidal

cell layers at 5 � 10�4 M. At 5 � 10�6 and 2 � 10�5 M, under the same experimental conditions, no changes were seen

with the PI method, but we recorded increased population spike amplitudes, repetitive discharges and frequency

potentiation (at a stimulus repetition rate of 0.05 Hz). Using whole-cell patch clamp in hippocampal dissociated neurons

we have found that SMC (applied for approximately 1 s) reduces GABA-induced currents (IC50 ¼ 4:4 � 10�4 M) without

having an effect of its own, acting like a competitive antagonist at GABAA receptors. Our findings are in line with the view

that the ability of monohalomethanes to induce the formation of SMC is an important factor for their neurotoxicity,

provided that SMC is allowed to act at least for several hours. The effects exerted by SMC seem to be due, at least in part,

to its interaction with GABA receptors.

# 2004 Elsevier Inc. All rights reserved.

Keywords: S-Methylcysteine; Monohalomethanes; GABA; Hippocampus; Organotypic cultures

INTRODUCTION

Monohalomethanes are important methylatingagents that represent considerable hazards to humanhealth and the earth’s ozone layer (McCauley et al.,1999). Their use is industrial and, in the case of methylbromide, also, to a great extent, agricultural. There isconsensus that neurotoxicity of these compounds is a

major concern (Alexeef and Kilgore, 1983; Bonnefoiet al., 1991). However, the mechanisms underlyingneurotoxic action of methyl bromide as well as methyliodide (Chamberlain et al., 1999) are still unclear.

In a previous work, we have shown that concentra-tions of up to 5 � 10�2 M of the bromide ion do notirreversibly alter neuronal synaptic responses (Zeiseet al., 1999). Thus, at least in the case of methylbromide, it is most likely that methylation rather thanthe halogen anion is the primary cause for its toxicity.

The same study demonstrated that methyl bromidedoes not exert any immediate toxic effect on synaptic

NeuroToxicology 25 (2004) 817–823

* Corresponding author. Tel.: þ56-2-682-2520;

fax: þ56-2-682-2521.

E-mail address: mzeise@usach.cl (M.L. Zeise).

0161-813X/$ – see front matter # 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.neuro.2004.01.008

transmission in rat hippocampal slices. These and otherdata, like the prolonged delay between acute exposureto monohalomethanes and the appearing of clinicalsigns, have led to the hypothesis that toxicity is causedby metabolites rather than by direct action. Conjuga-tion of the methyl group of monohalomethanes toglutathione by gluthathione S-transferases (GSTs) isthe major pathway for metabolization of this class ofcompounds (Jager et al., 1988; Kornbrust and Bus,1983). In mammals, four cytosolic isoforms for thisenzyme have been described, called alpha, mu, pi andtheta. The theta isoform has been shown to be the onethat metabolizes monohalomethanes in human blood(Guengerich et al., 1995; Sheehan et al., 2001).

S-Methylcysteine (SMC) is a quantitatively impor-tant immediate metabolite resulting from GST activity(Goergens et al., 1994). In the present work, we testedthe hypothesis that SMC could be an essential factor forthe neurotoxicity of monohalomethanes, either as animportant mediator or by being neurotoxic of its own.

We have applied SMC to freshly isolated in vitroslices of the rat hippocampus to investigate short-termeffects, and to organotypic cultures for 24 h in order tostudy longer-lasting exposure. These preparationsleave synaptic connections largely intact, unlike cellcultures, while allowing for control of concentrationsand other experimental parameters that are not so welldefined in in vivo models. It also permits the detectionof plastic changes in synaptic transmission that may berelevant in neurological changes. We further addressedthe possibility that SMC may act via GABAA receptorsby measuring the influence of SMC on GABA-inducedcurrents in dissociated hippocampal cells.

MATERIALS AND METHODS

Freshly Isolated Slices

Slices were prepared and maintained as reportedearlier (Zeise et al., 1992). Briefly, rats (Sprague–Dawley; weighing 80–100 g, about 4 weeks old) wereanesthetized with ethyl ether and decapitated. Thebrain was dissected out and transverse hippocampalslices (350–400 mm thick) were cut using a Campdenvibratome (Campden Instruments, UK). Slices werestored at room temperature in gassed (95% CO2, 5%O2) artificial cerebrospinal fluid (aCSF) and used forrecording no earlier than half an hour and no later than12 h after preparation. They were singly transferred toa recording chamber, completely submerged in andcontinuously superfused by the same gassed aCSF at a

constant rate (2.0 ml/min). The composition of theaCSF was (mM): NaCl, 130; KCl 3.5; NaH2PO4

1.25; MgSO4 1.5; CaCl2 2.0; NaHCO3 24; and glucose10; pH 7.4.

Organotypic Cultures

Hippocampal slice cultures were prepared by theinterface culture method, as described by Stoppiniet al. (1991). Seven days old male Sprague–Dawleyrats were anesthetized by hypothermia and rapidlydecapitated. Their brains were quickly removed andplaced in ice cold Hank’s Balanced Salt Solution(HBSS; GIBCO, Langley, OK) supplemented with28 mM glucose in a 60-mm culture dish under sterileconditions. Hippocampi were dissected out under astereomicroscope. Transversal hippocampal slices(400 mm thick) were obtained using a mechanical tissuechopper (Stoelting, Kiel, WI). Hipppocampal sliceswere carefully transferred to another sterile 60 mm dishcontaining HBSS supplemented with glucose and lefton ice for 1–3 h. Then, the slices were placed onMillicell-CM sterile tissue culture plate inserts (Milli-pore, Bedford, MA) previously covered with sterilecollagen solution. The inserts were put into a six-wellculture tray, each well containing 1 ml of ‘‘optimem’’culture medium (GIBCO) supplemented with 25%heat-inactivated horse serum, 25% HBSS, penicillin,G/streptomycin sulfate (5000 units/ml and 5 mg/ml,respectively) and 28 mM glucose; pH 7.3. Slices werecultured for 3–4 days at 36.5 8C, 100% humidity, in a95% air–5% CO2 atmosphere. After 3–4 days of incu-bation the culture medium was replaced with 1 ml ofchemically defined serum-free neurobasal medium(GIBCO) containing 1 mM L-glutamine, 2% B27 sup-plement (GIBCO) and 28 mM glucose. Thereafter, themedium was changed twice a week during the next 3–4weeks. Cultures were regularly checked microscopi-cally. SMC was added with the last culture mediumexchange and left for 24 h.

Cultured Hippocampal Neurons

Hippocampal neurons were dissociated and main-tained as described before (Aguayo and Pancetti,1994). Briefly, cells were taken from 18- to 19-day-timed pregnant Sprague–Dawley rats and maintainedin vitro for 12–17 days on 35 mm tissue culture dishescoated with poly(L-lysine). The neuronal feeding med-ium consisted of 90% minimal essential medium(GIBCO), 10% heat-inactivated horse serum (GIBCO)and a mixture of nutrient supplements.

818 F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823

SMC Application

SMC (Sigma, St. Louis, MO) was dissolved in aCSFand applied by adding it to the perfusion fluid of freshlyprepared slices. After 30 min of exposure, perfusionwas switched back to control solution.

In organotypic cultures, 2 weeks after dissection,SMC was added to the culture medium used for regularreplacement. Twenty-four hours later, cultures wereeither transferred to the recording chamber perfusedwith aCSF devoid of SMC, or they were kept in theirwells and PI was added for the evaluation of cellmembrane integrity (see below).

Electrophysiology

Extracellular Field RecordingsA bipolar platinum electrode was lowered until it just

touched the slice in the area of the Schaffer collaterals.Application of rectangular electrical pulses (100 ms,100–3000 mA) elicited field population spikes thatwere recorded from the CA1 pyramidal cell layer usingglass electrodes filled with 2 M NaCl. By the mean ofinput/output relationships, stimulus intensities weredetermined that yielded the maximal differencesbetween control and SMC responses (typically justbelow the one that elicited maximum response incontrol; Fig. 1A). Population spike amplitudes weremeasured as the average of 10 repetitive stimulationresponses. Only when these values did not deviatemore than 10% for three consecutive trials, measure-ments were taken as baseline set to 100% in theevaluation (see Fig. 1). Slices that displayed visibledamage, repetitive discharges in control or populationspike amplitudes of less than 1 mV were discarded.

Electrophysiologic recordings from organotypic cul-tures were performed as described above using aspecial recording chamber that allowed the placementof the plate inserts.

Whole Cell Patch Clamp RecordingsFor whole cell patch clamp recordings, the culture

dish containing dissociated hippocampal neurons wascontinuously perfused with an external solution con-taining 150 mM NaCl, 5.4 mM KCl, 2.0 mM CaCl2,1.0 mM MgCl2, 10 mM HEPES, pH 7.4, and 10 mMglucose. The internal solution contained 120 mMCsCl, 10 mM BAPTA, 10 mM HEPES, 4 mM MgCl2and 2 mM ATP-disodium, pH 7.35. The cells werestabilized at room temperature (20–22 8C) for at least30 min before starting the recordings. The whole-cellrecordings were done using an Axopatch-1D amplifier

(Axon Instruments, Union City, CA). The holdingpotential was �60 mV. Electrodes were pulled fromborosilicate capillary glass (World Precision Instru-ments, Sarasota, FL) on a vertical puller (Sutter Instru-ments, Novato, CA). The current signals were filteredat 2 kHz and stored for off-line analysis on a PCinterfaced with a TL-1 board (Axon Instruments).

Fig. 1. Effect of SMC on population spike amplitudes in freshly

isolated brain slices recording from the CA1 soma region. For each

concentration of SMC and the control, data were collected from six

experiments corresponding to six different slices stemming from at

least three different animals (error bars: S.E.M.). SMC was added

to the perfusion fluid (arrow) for 30 min immediately after the third

baseline measurement. (A) Dose–response curves in control and

10�3 M SMC; measured 30 min after wash in of SMC, 40 min after

the beginning of recording; control: I/O data were measured

correspondingly 40 min after the beginning of recording. No

significant differences were reached at any stimulus intensity

(control: n ¼ 7; SMC: n ¼ 9; error bars: S.E.M.) (B) Application

of SMC at various concentrations for 30 min: Only the reduction of

responses at 10�2 M, measured at the end of the application period,

is significantly different from control (ANOVA, Tukey’s test;

P < 0:05). At this concentration, recovery values just failed to be

significantly different from the baseline (Student’s paired t-test;

P ¼ 0:056).

F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823 819

Propidium Iodide (PI) Uptake

Propidium iodide is a polar compound that enterscells only when membranes are severely damaged andbecomes brightly red fluorescent after binding tonucleic acids (Macklis and Madison, 1990). In orga-notypic cultures, after 24 h of exposure to SMC orunder control conditions, 5 mg/ml PI was added to eachwell together with control culture medium. Three hourslater, pictures of the cultures were taken using afluorescence microscope (Olympus BX-50, NIKON,Melbeach, NY). Determinations were repeated threetimes and performed in duplicate (using two wellsunder the same conditions).

Statistical Analysis

For the experiments with freshly prepared hippo-campal slices, the recordings obtained for each group(control, 1, 5 and 10 � 10�3 M SMC) were collected

using slices from at least three different animals. Thedata were analyzed using the program Instat (GraphpadSoftware Inc.). The percentage change in averagedpopulation spike amplitude � S:E:M: between groupswas analyzed using one-way ANOVA. Significancewas determined using Tukey’s test with a level ofsignificance set at P < 0:05. For the data obtainedwith 10 � 10�3 M SMC, recovery data were comparedwith the same data set at baseline and application timeusing Student’s paired t-test. Electrophysiologic dataobtained from organotypic cultures were evaluatedusing the one-way ANOVA test.

RESULTS

Acute effects of SMC were examined by exposingfreshly prepared hippocampal slices for half an hour tolow millimolar concentrations. In Fig. 1A, an input/output relationship for control and 10�3 M SMC isdisplayed. Even though the difference between both

Fig. 2. Fluorescence caused by PI uptake into hippocampal cells. Organotypic cultures from 1-week-old rat pups, kept in culture for 2 weeks

and exposed to SMC during the last 24 h (DG: dentate gyrus). Two more assays, performed in duplicate, yielded qualitatively similar results.

820 F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823

curves did not reach significance at any point, themaximal difference was consistently detected at stimu-lus intensities that just induced maximal controlresponses. The absolute maximum was not altered bySMC (Fig. 1A). As shown in Fig. 1B, while SMC at1 mM caused no detectable change, at higher concen-trations, it reduced the synaptic population spike inCA1 pyramidal cells (24:6 � 4:3 % at 10�2 M 30 minafter start of application). We observed no sign of atoxic effect in these experiments, including no repetitivedischarges or changes in the time course of the fieldresponse. There was a highly significant recovery, but itmay not have been complete (the wash out data failed tobe significantly different from the control). There wasno difference in response in control and SMC-treatedslices when stimulus repetition frequency was increasedfrom 0.2 to 0.5 Hz (not shown).

In organotypic cultures, that had been maintained inculture for 2 weeks and exposed to SMC during 24 h, at5 � 10�6 and 2 � 10�5 M, there was no specific signalof fluorescence employing PI. However, at 5 � 10�5 MSMC, we detected damage to the integrity of the cellmembranes as revealed by the PI uptake. The signal at5 � 10�5 M was confined to the dentate gyrus (identi-fied by comparison with Nissl stainings; not shown),being more intense and widespread, involving also theCA1 region at 5 � 10�4 M (Fig. 2; n ¼ 3, determina-tions taken in duplicate).

As shown in Fig. 3, at 5 � 10�6 and 2 � 10�5 Msynaptic responses were significantly altered. (Noelectrophysiologic recording could be performed incultures exposed to SMC 5 � 10�5 M or more, pre-sumably due to the damage exerted by SMC as indi-cated by the PI uptake). Firstly, the field populationspike amplitude was enhanced and repetitive dis-charges were observed. Moreover, when stimulation

was repeated continuously, using a stimulus repetitionfrequency of 0.05 Hz during 10 min, responses ofslices that had been exposed to SMC (5 � 10�6 and2 � 10�5 M) increased in amplitude while the responseamplitude in control cultures was not changed signifi-cantly (Fig. 3). Qualitatively identical results wereobtained in two more controls and two more applica-tions of SMC at either concentration (population spikeamplitudes were always larger with SMC than incontrol and no statistically significant change couldbe detected with repetitive stimulation in control, but atleast 50% enhancement with SMC).

In dissociated hippocampal cells, SMC, up to5 � 10�3 M, applied alone did not evoke any current.However, when co-applied with GABA, SMC reducedGABA-induced currents with an IC50 of almost5 � 10�4 M (Fig. 4).

Fig. 3. Modulation of responses to synaptic stimulation in

organotypic cultures (every trace represents the average of three

measurements). Left column: responses at the beginning of

stimulation; right column: responses after 10 min of repetitive

stimulation at 0.05 Hz.

Fig. 4. Effect of SMC on GABA-induced currents in dissociated hippocampal cells. SMC, when applied together with GABA, reduces the

GABA-induced current in a concentration-dependent manner (IC50 ¼ 4:41 � 10�4 M þ 70 S.E.M; n ¼ 7). Applications of SMC alone were

without effect (not shown).

F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823 821

DISCUSSION

In a previous study (see Introduction), we could notfind serious impairment of electrical properties inhippocampal neurons exposed to methyl bromide upto millimolar concentrations (Zeise et al., 1999). Thus,in the present investigation we examined the effects ofSMC, that is one of the first metabolites appearing inthe brain after monohalomethane exposure. SMC isgenerated mainly by mediation of GSTs inside oroutside of the brain. At the moment, data are lackingas to whether SMC present in the brain is formed thereor transported into it after having been synthesized inthe blood.

Our results show that, at an exposure time of 30 min,SMC lowers excitability to synaptic stimulation. Thiseffect attains statistical significance only at a concen-tration of 10�2 M. It is not impossible that this valuemay be reached in an accident involving heavy expo-sure and may cause sedation in the person intoxicated.However, concentrations measured from blood serumin a serious accident (Garnier et al., 1996) were con-siderably lower. Thus, we may assume that the ‘‘acuteeffect’’ of SMC, lowering neuronal excitability, thatwas reversible, at least partially, cannot be consideredas being of much practical importance related totoxicity. The mechanism of this reduction of fieldresponse amplitude is unclear but probably not dueto increased GABAergic inhibition because in disso-ciated hippocampal cells SMC acted like a competitiveantagonist (see below).

A very different picture appeared when we exposedorganotypic cultures to SMC at medium micromolarconcentrations for 24 h. This resulted in membranedamage as revealed by the intake of PI (5 � 10�5 Mand above; Fig. 2). Interestingly, at 5 � 10�5 M the PIsignal was strictly limited to the dentate gyrus. Pyr-amidal cells were affected only at 5 � 10�4 M. Thehigher vulnerability of the dentate gyrus may have todo with neuronal maturation. It has been shown that inrodents this structure matures only in the postnatalperiod (Bayer and Altman, 1974), leaving it moresusceptible to the actions of neurotoxic substances.More specifically, some GST isoforms are expressedonly in mature tissues (Eaton and Bammler, 1999).

The electrophysiologic changes recorded in organo-typic cultures at 5 � 10�6 and 2 � 10�5 M SMC (anincrease in population spike amplitude and highersusceptibility to frequency potentiation) may haveexcitotoxic consequences, particularly in the longrun. The effect could be explained by a decrease inGABergic inhibition, since in dissociated hippocampal

cells from rat embryos, SMC reduced GABA-inducedcurrents in a concentration-dependent manner beingwithout effect when applied alone. Thus, SMCappeared to act as a competitive GABAA receptorantagonist with an IC50 of 4:4 � 10�4 M (Fig. 4). Thisvalue is very compatible with our fluorescence data(extensive damage at 5 � 10�4 M), but does not seemto explain the electrophysiologic data obtained with5 � 10�6 and 2 � 10�5 M, although even minor reduc-tion of GABAergic inhibition may cause significantmodifications in the behavior of the neuronal circuitry.

There is, however, an apparent contradiction with theresults of the brain slice study where a reduction offield responses was observed at high concentrations ofSMC (10�2 M). Why did SMC not act as an inhibitor ofGABA-induced currents in brain slices, augmentingsynaptically induced field responses, while, in thewhole cell clamp measurements, this effect was quiteclear? It may be speculated that, since organotypiccultures and dissociated cells were taken from newbornor embryonic animals, while freshly prepared sliceswere from almost adult specimens, these differentresults are due to differences in maturation of post-synaptic responses as described for the CA1 region(Wang et al., 2002).

The frequency potentiation, at a repetition rate aslow as 0.05 Hz mentioned above, may also be causedby mechanisms involving presynaptic GABAB recep-tors since GABAergic inhibition of pyramidal neuronsis subject to frequency dependency causing inhibitorypostsynaptic potentials to decrease when stimulusrepetition frequencies exceed 0.1 Hz (Deisz andPrince, 1989). This phenomenon is crucial for theinduction of LTP (Mott and Lewis, 1991). SMC mightshift the frequency for the induction of potentiation byan interaction with these receptors.

As mentioned above, we believe that a concentrationof 10�2 M SMC in brain tissue will not occur fre-quently, if ever, rendering low its presumable impor-tance in monohalomethane poisoning. Concentrationsof 5 � 10�6 to 5 � 10�5 M, however, that alteredsynaptic responses in organotypic cultures after 24 hof exposure are likely to occur in exposed people. In aprevious work, we determined bromide level in theblood of workers exposed to methyl bromide to be onaverage about 4 mg/l greater than that of workers notexposed (Muller et al., 1999). This corresponds to about5 � 10�5 M of bromide. In individual cases and acci-dental poisonings, concentrations may be much higher.This may lead to concentrations of SMC in the mediummicromolar range and above. Thus, we interpretour results in such a way that short-term exposure of

822 F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823

hippocampal neurons to SMC even at very high con-centrations does not induce toxicity immediately,whereas medium micromolar concentrations of SMCapplied for 24 h alter the responses in a way that iscompatible with toxic effects.

Taken together, our results are compatible with a roleof SMC as an intermediate and/or directly actingsubstance that underlies neurotoxicity of monohalo-methanes. This also implies a role for GSTs, that arenecessary for the generation of SMC at least frommethyl bromide and methyl chloride, whose methylat-ing power is not strong enough to produce importantamounts of SMC without the interference of theseenzymes. However, even in methyl iodide metabolism,GSTs seem to play a major role (Bonnefoi et al., 1991).

The role of GSTs in the formation of SMC in thebrain should be further investigated as well as theinteraction of SMC with GABA receptors in order toclarify whether an excitotoxic action caused byreduced GABAergic inhibition is essential for theobserved neurotoxicity of SMC.

ACKNOWLEDGEMENTS

This study was supported by the Foundation Volks-wagen, project no. I/73 050 to MZ; FONDECYT,project no. 1030220 to BM and FP; and the Universityof Santiago de Chile through DICYT, project no.0075PV to FP and project no. 029971Z to MZ..

REFERENCES

Aguayo LG, Pancetti FC. Ethanol modulation of the gamma-

aminobutyric acid A- and glycine-activated Cl� current in

cultured mouse neurons. J Pharmacol Exp Ther 1994;270:61–9.

Alexeef BV, Kilgore WW. Methylbromide. Residue Rev 1983;88:

101–53.

Bayer SA, Altman J. Hippocampal development in the rat:

cytogenesis and morphogenesis examined with autoradiography

and low-level X-irradiation. J Comp Neurol 1974;158:55–79.

Bonnefoi MS, Davenport CJ, Morgan KT. Metabolism and

toxicity of methyl iodide in primary dissociated neural cell

cultures. Neurotoxicology 1991;12:33–46.

Chamberlain MP, Sturgess NC, Lock EA, Reed CJ. Methyl iodide

toxicity in rat cerebellar granule cells in vitro: the role of

glutathione. Toxicology 1999;139:27–37.

Deisz RA, Prince DA. Frequency-dependent depression of

inhibition in guinea-pig neocortex in vitro by GABAB receptor

feed-back on GABA release. J Physiol Lond 1989;412:513–41.

Eaton DL, Bammler TK. Concise review of the glutathione

S-transferases and their significance to toxicology. Toxicol Sci

1999;49:156–64.

Garnier R, Rambourg-Schepens MO, Muller A, Hallier E.

Glutathione transferase activity and formation of

macromolecular adducts in two cases of acute methyl bromide

poisoning. Occup Environ Med 1996;53:211–5.

Goergens HW, Hallier E, Muller A, Bolt HM. Macromolecular

adducts in the use of methyl bromide as fumigant. Toxicol Lett

1994;72:199–202.

Guengerich FP, Thier R, Persmark M, Taylor JB, Pemble SE,

Ketterer B. Conjugation of carcinogens by Q class glutathiones

S-transferases: mechanisms and relevance to variations in

human risk. Pharmacogenetics 1995;5:S103–7.

Jager R, Peter H, Sterzel W, Bolt HM. Biochemical effects of

methyl chloride in relation to its tumorigenicity. J Cancer Res

Clin Oncol 1988;114:64–70.

Kornbrust DJ, Bus JS. The role of glutathione and cytochrome P-

450 in the metabolism of methyl chloride. Toxicol Appl

Pharmacol 1983;67:246–56.

Macklis JD, Madison RD. Progressive incorporation of propidium

iodide in cultured mouse neurons correlates with declining

electrophysiological status: a fluorescence scale of membrane

integrity. J Neurosci Methods 1990;31:43–6.

McCauley SE, Goldstein AH, DePaolo DJ. An isotopic approach

for understanding the CH(3)Br budget of the atmosphere. Proc

Natl Acad Sci USA 1999;96:10006–9.

Mott DD, Lewis DV. Facilitation of the induction of long-term

potentiation by GABAB receptors. Science 1991;252:1718–20.

Muller M, Reinhold P, Lange M, Zeise M, Jurgens U, Hallier E.

Photometric determination of human serum bromide levels—a

convenient biomonitoring parameter for methyl bromide

exposure. Toxicol Lett 1999;107:155–9.

Sheehan D, Meade G, Foley VM, Dowd CA. Structure, function

and evolution of glutathione transferases: implications for

classification of non-mammalian members of an ancient

enzyme superfamily. Biochem J 2001;360:1–16.

Stoppini L, Buchs P-A, Muller D. A simple method for

organotypic cultures of nervous tissue. J Neurosci Methods

1991;37:173–82.

Wang DS, Inokuchi H, Tanaka E, Isagai T, Li JS, Higashi H.

Postnatal changes in the overall postsynaptic currents evoked in

CA1 pyramidal neurons of the rat hippocampus. Life Sci

2002;72:341–53.

Zeise ML, Teschemacher A, Arriagada J, Zieglgansberger W.

Corticosterone reduces synaptic inhibition of rat hippocampal and

neocortical neurons in vitro. J Neuroendocrinol 1992;4:107–12.

Zeise ML, Jofre D, Morales P, Espinoza J, Nalli A, Aranda M.

Methyl bromide decreases excitability without having

immediate toxic effects in rat hippocampal CA1 neurons in

vitro. Neurotoxicology 1999;20:827–32.

F. Pancetti et al. / NeuroToxicology 25 (2004) 817–823 823