Neurochemical Research, Vol. 28, No. 2, February 2003 (© 2003), pp.

281–291

Searching for Mechanisms of N-Methyl-D-Aspartate-Induced

Glutathione Efflux in Organotypic Hippocampal Cultures*

Camilla Wallin,1 Abdul-Karim Abbas,1 Mattias Tranberg,1 Stephen G. Weber,2

Holger Wigström,1 and Mats Sandberg1,3

(Accepted August 21, 2002)

N-Methyl-D-aspartate (NMDA)-receptor stimulation evoked a selective and partly delayed ele-

vated efflux of glutathione, phosphoethanolamine, and taurine from organotypic rat hippocam-
pus slice cultures. The protein kinase inhibitors H9 and staurosporine had no effect on the
efflux. The phospholipase A2 inhibitors quinacrine and 4-bromophenacyl bromide, as well as
arachidonic acid, a product of phospholipase A2 activity, did not affect the stimulated efflux.
Polymyxin B, an antimicrobal agent that inhibits protein kinase C, and quinacrine in high con-
centration (500 ␮M), blocked efflux completely. The stimulated efflux after but not during
NMDA incubation was attenuated by a calmodulin antagonist (W7) and an anion transport
inhibitor (DNDS). Omission of calcium increased the spontaneous efflux with no or small ad-
ditional effects by NMDA. In conclusion, NMDA receptor stimulation cause an increased se-
lective efflux of glutathione, phosphoethanolamine and taurine in organotypic cultures of rat
hippocampus. The efflux may partly be regulated by calmodulin and DNDS sensitive channels.

KEY WORDS: NMDA; glutathione; phosphoethanolamine; taurine; efflux.

INTRODUCTION also essential for maintenance of the thiols of proteins

and of other antioxidants, for example, ascorbate and
Acute cerebral insults such as ischemia and brain ␣-tocopherol (18,19). A decreased glutathione level,
trauma are associated with excessive release and ex- through inhibition of its synthesis, is accompanied by
tracellular accumulation of the neurotransmitter gluta- increased excitotoxic response to NMDA, degenera-
mate (1), leading to persistent activation of glutamate tion of mitochondria, and larger infarct areas in stroke
receptors and neurodegeneration (2–5). The glutamate models (20–22). Extracellularly, glutathione has been
receptor mediated increase of reactive oxygen species suggested to have multifaceted electrophysiological ef-
(ROS) (6–8) may be a triggering event in the excito- fects by binding to its own receptors and by modulating
toxic cell death (9 –12). The reduced form of glutathione glutamatergic excitatory neurotransmission by displac-
(GSH) is a major component in cellular protection ing glutamate from its ionotropic receptors (23–25).
against ROS (for reviews see (13 –17)). Glutathione is Glutathione may also increase NMDA receptor re-
sponses by interacting with its redox sites (26–28).
* Special issue dedicated to Dr. Anders Hamberger. Furthermore, because the breakdown products of glu-
1
Department of Medical Biophysics, University of Göteborg, Med- tathione include cysteine, glycine, and cysteineglycine,
icinaregatan 11, S-405 30 Göteborg, Sweden. extracellular breakdown may supply surrounding cells
2
Department of Chemistry, University of Pittsburgh, Pittsburgh,
Pennsylvania 15260, USA.
with these glutathione precursors (29–31). Despite the
3
Address reprint requests to: Mats Sandberg, Tel: (46)-31-7733395; key function of glutathione in intracellular redox reg-
Fax: (46)-31-7733558; E-mail: [email protected] ulation and the putative extracellular effects in brain,
281
0364-3190/03/0200–0281/0 © 2003 Plenum Publishing Corporation
282 Wallin et al.

the mechanisms and factors determining the efflux re- Organotypic hippocampal slice cultures were prepared according to
main unknown. the method of Stoppini et al. (44) with minor modifications. Briefly,
the hippocampi of 7–9-day-old Sprague-Dawley rat pups were dis-
Glutathione efflux from brain cells is increased sected and sectioned transversely at 350–400 ␮m. Four slices were
by potassium depolarization, iron (Fe2⫹), inhibitors transferred to 30 mm diameter porous membrane inserts (Millicell-
of mitochondrial respiration (malonate, 1-methyl-4- CM 0.4 ␮m). These were placed in 6-well tissue culture plates con-
phenylpyridinium (MPP⫹), 3-nitropropionic acid and taining 1.2 ml of culture medium. The slices were maintained in a
cyanide), and after ischemia (32–38). In a previous humidified 95% air/5% CO2 atmosphere at 36°C. Cultures were al-
lowed to develop for 2 weeks before use, with media changed every
study, we showed that the NMDA-mediated gluta- 3–4 days.
thione efflux was dependent on extracellular calcium Efflux of Glutathione and Amino Acids. After removal of the
but unrelated to dantrolene sensitive intracellular cal- culture medium the cultures were washed twice with prewarmed
cium release, osmotic change and glutathione-, or ni- HBSS (⬃36°C), 1 ml was placed above and 1 ml below the mem-
tric oxide-synthesis (39). In this study our main goals brane. The slices were thereafter preincubated for 60 min and
washed again in HBSS. For preincubation and for incubation during
were (i) to evaluate if organotypic cultures of rat hip- the efflux studies 1 ml HBSS was placed below and 0.6 ml above
pocampus may be used as a model system in charac- the membrane. The cultures were incubated in 95% air/5% CO2 at-
terization of NMDA-mediated efflux of glutathione mosphere at 36°C. The incubation procedure was similar to that re-
and (ii) to initiate mechanistic studies using pharmaco- ported earlier (45). In these studies, no toxicity due to the incubation
logical tools for manipulation of activities and func- was observed 24 h after the incubation. During the experiment
HBSS was renewed every 15 min, 300 ␮l of the medium above the
tions of protein kinases, phospholipase A2, calmodulin, membrane was collected for analysis. ␤-Mercaptoethanol (␤-ME),
anion channels, glutathione synthesis, and intracellular Na2EDTA, and NaN3 were added to the samples to final concentra-
calcium. One of the potentials using cultured slices tions of 20, 1, and 5 mM, respectively. Stimulation was performed
compared to acutely isolated slices (which we will with NMDA (10 ␮M)/glycine (10 ␮M) for 15 min. When renewing
make use of in coming studies) is the possibility to cor- the medium after NMDA exposure the cultures were washed with
1 ml HBSS above the membrane and 1 ml below. When included,
relate release of glutathione and similar compounds acivicin (0.2 or 2.0 mM) and BSO (5 mM) were present during the
after an acute insult to delayed excitotoxicity one or preincubation and throughout the experiment. Arachidonic acid
more days after the acute insult. (30 ␮M), BPB (100 ␮M), dantrolene (40 ␮M), PMA (1 ␮M), PMXB
In parallel to glutathione we analysed the efflux of (10 ␮M), quinacrine (50 and 500 ␮M), staurosporine (1 ␮M), DNDS
the neurotransmitter glutamate, phosphoethanolamine (1mM), H9 (100 ␮M), W7 (100 ␮M), or calcium-free medium were
introduced 30 min before the introduction of NMDA. Experiments
(PEA, also named phosphorylethanolamine), the polar were also performed in which 500 ␮M quinacrine was introduced si-
soluble component of phosphatidylethanolamine, and multaneously with NMDA/glycine. When the drugs prepared in
taurine. The latter two were included because previous DMSO were used, the same concentrations of DMSO (0.05 or 0.1%)
studies by others and us have shown that these amines were present in the control experiments. In calcium-free medium
are released after NMDA-receptor stimulation and NaCl substituted CaCl2.
Glutathione, Amino Acid and Protein Measurements. Gluta-
other insults (4,39–43). thione and amino acids were determined using reversed phase HPLC
employing automated precolumn fluorogenic labeling with OPA/
␤-mercaptoethanol as described earlier (42). The column (300 ⫻
EXPERIMENTAL PROCEDURE 4.6 mm) was packed with Nucleosil 100-5 C18 (Macherey-Nagel, Ger-
many). The derivatives were eluted with a gradient from 0 to 90%
Drugs and Solutions. HEPES-buffered salt solution (HBSS) methanol (containing 1.25% tetrahydrofuran [v/v]) in a Na-phosphate
contained (in mM): 143.4 NaCl, 5 HEPES, 5.4 KCl, 1.2 MgSO 4 , buffer (50 mM set at pH 5.40 with 1M NaOH and containing 2.5%
1.2 NaH2PO4, 2.0 CaCl2, and 10 glucose. Culture medium consisted of tetrahydrofuran [v/v]). The autoinjector (Waters Wisp 717 plus) was
50% Eagle’s Basal Medium (BME), 23% horse serum, 25% Earl’s programmed to add 50 ␮l OPA/␤-ME to 100 ␮l sample. The method-
balanced salt solution, 7.5 mg/ml D-glucose, 25 U/ml Pest, and ology does not discriminate between thiols and disulfides; the term
1 mM glutamine. Acivicin, arachidonic acid (AA), 4-bromophenacyl glutathione therefore refers to the total concentration of reduced glu-
bromide (BPB), DL-buthionine-[S,R]-sulfoximine (BSO), dantro- tathione (GSH), glutathione dimers (GSSG), and mixed glutathione
lene, phorbol 12-myristate 13-acetate (PMA), polymyxin B-sulfate disulfides (GSX).
(PMXB), quinacrine, and staurosporine were obtained from Sigma The hippocampal slices were dissolved in 1 M NaOH, and the
(St. Louis, MO, USA). 4,4⬘-dinitrostilbene-2,2⬘disulfonic acid di- protein content was determined using BSA protein assay reagent
sodium salt (DNDS), N-(2-aminoethyl)-5-isoquinoline-sulfoamide (Pierce, Rockford, IL, USA) with bovine serum albumin as standard.
dihydrochloride (H9) and N-(6-aminohexyl)-5-chloro-1-naphtha- Absorbance was measured at 570 nm in a microplate reader (Emax,
lene sulfonamide hydrochloride (W7) were from Molecular Probes Molecular devices, Sunnyvale, CA, USA).
(Eugene, OR, USA). Arachidonic acid, 4-bromophenacyl bromide, Statistical Analysis. Data are expressed as mean ⫾ SEM and
dantrolene, PMA and staurosporine were prepared as stock solu- were analyzed for significance using the Wilcoxon matched pairs
tions in dimethyl sulfoxide (DMSO) and stored at ⫺20°C. rank sum test or the Mann-Whitney U-test for unpaired sample
Preparation of Organotypic Cultures. All animal procedures groups. Data presented on efflux rates are from n wells. A P value
were approved by the local ethics committee in Göteborg, Sweden. of ⬍.05 was considered statistically significant.
NMDA-Induced Glutathione Efflux 283

RESULTS before (0 – 60 min), during (61–75 min) and after (76 –

120 min) incubation with 10 ␮M NMDA /10 ␮M glycine
The efflux rates (see experimental procedures for (Fig. 1). Maximal efflux rates of glutathione and PEA
details) of glutathione, taurine, PEA, and glutamate from were observed after removal of NMDA (75–90 min,
organotypic hippocampal slice cultures were analysed 390.4 ⫾ 48.9% and 809.6 ⫾ 107.4%, respectively),

Fig. 1. Effect of 15-min incubation of organotypic hippocampal slice cultures with 10 ␮M NMDA/10 ␮M glycine on the efflux rates of
glutathione, PEA, taurine, and glutamate (n ⫽ 19). *P ⬍ .05 comparing NMDA-mediated efflux rates with the efflux rates before NMDA
introduction (45 – 60 min).
284 Wallin et al.

whereas the maximal efflux rate of taurine was ob- during NMDA /glycine incubation, but decreased the
served during the 15 min incubation period with NMDA / efflux rates of glutathione and PEA after NMDA incu-
glycine (60–75 min, 905.9 ⫾ 102.4%). Aspartate, as- bation (Fig. 3). Basal glutathione, PEA, and taurine ef-
paragine, glutamate, glutamine, ethanolamine, valine, or flux rates were not altered by 10 ␮M PMXB or by 50
isoleucine efflux rates were not affected by NMDA or 500 ␮M quinacrine (data not shown). Incubation of
incubation (data shown for glutamate, Fig. 1). the cultures with the phospholipase A2 inhibitor BPB
The efflux rate of glutathione increased during the did not alter the efflux rates of PEA or taurine (n ⫽ 4,
incubation period with HBSS prior to NMDA/glycine data not shown). It was not possible to evaluate the
introduction (Fig. 2). This effect was blocked by BSO, effect of BPB on glutathione efflux rate as BPB inter-
the glutathione synthesis inhibitor (Fig. 2). The stimulus fered with the analysis of glutathione.
evoked efflux rate of glutathione in cultures incubated in The calmodulin antagonist W7 and the anion trans-
BSO increased by 41.5 ⫾ 3.4 pmol/min/mg protein, port inhibitor DNDS had no effect on the basal or
which was significantly lower than in control experi- stimulated efflux rates of glutathione, PEA, and taurine
ments (77.5 ⫾ 11.3 pmol/min/mg, Fig. 2). The corre- during the incubation with NMDA, but attenuated the
sponding percentage increase (compared to basal efflux) efflux rates after NMDA removal (Fig. 3).
with and without BSO were during NMDA/glycine in- Arachidonic acid, dantrolene, PMA, staurosporine,
cubation 604.7 ⫾ 70.1% and 356.8 ⫾ 40.1%, respec- or H9 produced no effects on basal (data not shown) or
tively, and after removal of NMDA 217.0 ⫾ 19.6% and NMDA-mediated efflux rates of glutathione, PEA, or
390.4 ⫾ 48.9%, respectively. Efflux rates of amino acids taurine (Fig. 3).
were not affected by BSO incubation (data not shown). Omitting calcium from the medium increased the
The protein kinase C inhibitor PMXB (10 ␮M) efflux rates of glutathione, PEA, and taurine (Fig. 4).
and the phospholipase A2 inhibitor quinacrine (500 ␮M) The effect was most prominent for glutathione with
completely abolished the NMDA-mediated increases in an increase similar to that induced by NMDA/glycine
glutathione, PEA, and taurine efflux rates (Fig. 3). No in calcium-containing medium. The efflux rate of glu-
effect was observed when 50 ␮M quinacrine was used tathione was not further increased when NMDA was in-
(Fig. 3). Introducing 500 ␮M quinacrine simultane- cluded in calcium-free medium. Calcium-free medium
ously with NMDA/glycine did not alter the efflux rates abolished the NMDA-mediated increased efflux rates of
PEA and taurine.
Preincubation of the organotypic hippocampal
slice cultures with 0.2 or 2.0 mM acivicin, an inhibitor
of ␥-glutamyl transpeptidase, did not significantly alter
the glutathione efflux rate (data not shown).

DISCUSSION

The basal efflux rates and the quantitative as well

as the qualitative effects of NMDA on efflux of glu-
tathione and amino acids from organotypic hippocampal
slices were similar to those reported earlier for efflux
from acutely isolated hippocampus slices (39). In ac-
cordance with the previous results (39), maximal glu-
tathione and PEA efflux rates were reached after NMDA
withdrawal, whereas maximal taurine efflux rate was
observed during NMDA incubation. This implies that
organotypic cultures is a relevant model to search for the
mechanisms of NMDA-mediated efflux. One advantage
Fig. 2. Glutathione efflux rate from organotypic hippocampal slice
cultures in the presence (o) or absence (䉱) of 5 mM BSO (n ⫽ 5). with the organotypic culture is that long-term toxicity
BSO was present during preincubation and throughout the experiment. studies can be performed. If the mechanisms of acute
The black bar indicates the incubation period with NMDA/glycine. NMDA-stimulated glutathione efflux are elucidated,
*P ⬍ .05 comparing efflux rate with and without BSO. # P ⬍ .05
comparing efflux rate with that during the first incubation period blockage of such release will be employed to evaluate
(0–15 min). its influence on the delayed toxicity following NMDA
NMDA-Induced Glutathione Efflux 285

Fig. 3. Effect of indicated drugs on NMDA-mediated efflux rates of glutathione, PEA, taurine, and glutamate. Values are expressed as percent
of basal efflux rate, comparing the efflux rates before introduction of NMDA (45–60 min) with the efflux rates during NMDA/glycine incubation
(gray bars, 60–75 min), and with the efflux rates after NMDA incubation (black bars, 75–90 min). All drugs were introduced 30 min before
NMDA introduction and were present throughout the experiment except in experiments denoted 500 Quin*, in which quinacrine was introduced
simultaneously with NMDA. Used concentrations: PMXB, 10 ␮M; 500 Quin, 500 ␮M; staurosporine, 1 ␮M; W7, 100 ␮M; H9, 100 ␮M; DNDS,
1 mM; AA, 30 ␮M; 50 Quin, 50 ␮M; dantrolene, 40 ␮M; PMA, 1 ␮M. *P ⬍ .05 comparing efflux rates with and without the indicated
treatments (n ⱖ 4).
286 Wallin et al.

Fig. 4. Glutathione, PEA, taurine, and glutamate efflux rates in the presence (䉱) or absence (o) of calcium in the medium (n ⫽ 5). Calcium-
free medium was introduced 30 min before NMDA/glycine introduction. The black bar indicates the incubation period with NMDA/glycine.
*P ⬍ .05 comparing efflux rates with and without calcium.

receptor overactivation. The present results indicate that Some differences in the efflux comparing the two
calmodulin and DNDS-sensitive channels, but not PKC models are notable. The basal efflux and the NMDA-
or PLA2 activities, may partly be involved in the regula- stimulated efflux of glutathione were affected by the
tion of stimulated glutathione, taurine, and PEA efflux glutathione synthesis blocker BSO in organotypic slices
after but not during NMDA receptor stimulation. but not in acute slices (39). We have no obvious
NMDA-Induced Glutathione Efflux 287

explanation for this discrepancy, but it is clear that with in unknown. Although the cultures show a well-
the present protocol a greater proportion of the efflux developed maturity of several aspects of brain func-
from the organotypic cultures appears to be derived tion (54), the expression of some genes, such as that
from a glutathione pool that turns over very rapidly encoding ␥-glutamyl transpeptidase, is likely to be de-
(46). This may, hypothetically, be due to differences in pendent on factors others than those present in the adult
maturity of the organotypic cultures compared to the horse serum that is used as additive to the culture
acutely isolated slices. This is not surprising as several medium.
aspects of NMDA-mediated efflux of taurine from The mechanisms involved in glutathione efflux in
brain slices has been shown to be dependent on the age the brain are unknown. Efflux of glutathione in liver
of the animal (43). In contrast to our previous results cells is altered by PKC activity. Activation of PKC in-
(39), omission of Ca2⫹ from the medium massively in- crease, whereas inhibition of PKC decrease glutathione
creased the basal efflux rates of glutathione. Smaller efflux rate from liver cells in vivo and in vitro (55,56).
but significant increases were observed for PEA and In our experiments, the PKC inhibitor PMXB com-
taurine. The discrepancy between the two models is pletely abolished the increased NMDA-mediated efflux
probably due to different sensitivity to calcium-free rates of glutathione, PEA, and taurine. However, incu-
medium in the acute hippocampal slice preparation and bation with the protein kinase inhibitors staurosporine
cultured hippocampal slices. Incubation in calcium- or H9 had no effect on the NMDA-mediated efflux
free medium can cause depolarization and spontaneous rates, nor did incubation with the PKC activating phor-
bursts of population spikes in hippocampal CA1 neu- bol ester PMA, implying that the effect of PMXB is not
rons (47,48). The increased glutathione efflux rate in achieved through inhibition of PKC. Although PMXB
the present experiments could thus be due to depo- is often used as a PKC inhibitor it is not selective; it
larization, which can increase glutathione efflux (36). also inhibits Ca2⫹-activated K⫹-channels (57,58) and
However, taurine and PEA efflux increased to a much calmodulin (59), and thereby enzymes such as calcium/
lower extent than glutathione, indicating that another calmodulin–dependent kinase II and calcineurin. We
factor, at present unknown, is involved in mediating cannot exclude that PMBX is acting of some isoform of
glutathione efflux during calcium-free conditions. The PKC that is not activated by PMA and not inhibited by
effect of calcium ommision reported here is not an iso- H9 or staurosporin. However the most plausable expla-
lated finding as incubation of hepatocytes and tumor nation is that PMBX exerts its effect by some other
cells in calcium-free medium also induces increased mechanism than blockage of PKC, possibly by reducing
efflux of reduced glutathione (49,50). Similar to our calcium influx after NMDA receptor activation.
previous experiments no additional efflux rate of glu- Glutamate receptor activation increases PLA2 ac-
tathione, PEA or taurine was detected when NMDA tivity, resulting in increased levels of ROS and in ara-
was included in Ca2⫹ free medium showing that the chidonic acid release. The PLA2 inhibitor quinacrine
NMDA stimulation efflux also in organotypic cultures attenuates the generation of ROS, arachidonic acid re-
is calcium-dependent. lease, glutamate cytotoxicity, and stroke injury (6, 60 –
Another major difference between the models 62). In our experiments, quinacrine (500 ␮M) com-
is that in our previous work, using acute hippocam- pletely abolished the increased NMDA-mediated ef-
pal slice preparations, incubation with the ␥-glutamyl flux rates of glutathione, PEA, and taurine. BPB,
transpeptidase inhibitor acivicin increased basal and another PLA2 inhibitor, had no effect on PEA or tau-
stimulation-induced glutathione efflux (39,51,52). This rine efflux rates (glutathione was not determined due
effect is most likely due to blockage of extracellular to interference of BPB with the analysis), nor did di-
breakdown of glutathione. Unexpectedly, incubation of rect addition of the PLA2 product arachidonic acid
organotypic hippocampal cultures with acivicin was not alter glutathione, PEA, or taurine efflux rates. These
accompanied by increased glutathione efflux. The ac- results, together with the lack of effect of 50 ␮M
tivity of this enzyme in neuronal/glial cell membranes quinacrine, a concentration generally used to inhibit
starts to appear at postnatal day 7 (53). We can thus PLA2 activity, indicates an effect of quinacrine in high
expect very low levels of ␥-glutamyl transpeptidase in concentrations unrelated to PLA2 inhibition. One con-
the slices at the start of the culturing period. The devel- ceivable mechanism is by inhibition of Ca2⫹ influx
opment of the activity of ␥-glutamyl transpeptidase in and /or Ca2⫹ release from intracellular stores.
microvessels of brain is regulated by the status of Quinacrine decrease Ca influx in synaptosomes (63)
thyroid hormones (53). As far as we know the gene and blocks Ca2⫹ induced Ca2⫹ release in the sar-
regulation of the enzyme in neuronal/glial membranes coplasmic reticulum (64). When quinacrine (500 ␮M)
288 Wallin et al.

was introduced simultaneously with NMDA, there was Ycf1p (77), the yeast orthologue of the multidrug re-
no effect on glutathione, PEA, or taurine efflux rates, sistance associated proteins (MRP). DIDS also inhibits
indicating that quinacrine per se did not inhibit glutathione transport in retinal Müller cells (78). The
NMDA-receptor activation. During the incubation pe- similarity in effect of DNDS and W7 on glutathione,
riod after NMDA removal, glutathione and PEA efflux PEA, and taurine efflux rates in our study may indicate
rate were reduced compared to control experiments, a relationship between calmodulin and an anion trans-
possibly because quinacrine must reach a certain in- port system. CFTR as well as other anion channels
tracellular concentration to exert its action. have phosphorylation sites that can alter their activi-
Part of the cytosolic Ca2⫹ increase after NMDA ties (79–81). A speculative working hypothesis is thus
receptor activation results from intracellular stores (65). that a least two different processes are involved in
Dantrolene, a blocker of ryanodine receptor–dependent efflux of glutathione, PEA, and taurine efflux after
Ca2⫹ release, attenuates the NMDA-mediated rise in NMDA-stimulation. One such process appears to occur
intracellular Ca2⫹ (66). As in our previous experi- via a DNDS-sensitive channel that is regulated by a
ments with acute hippocampal slices, dantrolene did not Ca2⫹-calmodulin dependent activity. Further exper-
change glutathione, PEA, or taurine efflux rates, sug- iments using specific inhibitors against calcineurin,
gesting that the efflux rate is not regulated by ryanodine CaMKII, NO synthase, and adenylate cyclase as well
receptor–dependent Ca2⫹ release. This is at variance as other chloride-channel blockers (but see below)
with studies of the CA1 area of the rat hippocampus may help to clarify the mechanisms in NMDA-
(67). In this report, NMDA-mediated taurine efflux was stimulated efflux of glutathione, PEA, and taurine.
partly, but not completely, reduced by dantrolene. The The basal efflux rates of glutathione, PEA, and
discrepancy may be related to the differences in prepa- taurine were not affected by PMXB, quinacrine, W7,
rations, sampling area, and collection technique. or DNDS, and the efflux during NMDA-stimulation
The calmodulin inhibitor W7 attenuated the ef- was unchanged by W7 and DNDS, suggesting dif-
flux rate of glutathione, PEA, and taurine after NMDA ferently regulated effluxes of glutathione, PEA, and
removal. Increased intracellular calcium concentra- taurine before, during, and after NMDA receptor stim-
tions leads to activation of calmodulin, which in turn ulation. Alternatively, the efflux have different cellu-
stimulates several enzymes such as Ca2⫹/calmodulin- lar origin. Glutathione is present in both neurons and
dependent kinase II (CaMKII), neuronal NO synthe- glia (29,30,82–86), but as no functional NMDA recep-
tase, adenylate cyclase, and calcineurin (32,68–70). tors appear to be located on hippocampal glia cells
Several responses to glutamate-receptor activation are (87), the most straightforward explanation for the
altered by the calmodulin inhibitor W7. Glutamate- increase in glutathione efflux is neuronal release.
mediated elevations of cAMP and Ca2⫹ are attenuated However, it cannot be excluded that NMDA-receptor
by W7, presumably via inhibition of calmodulin- stimulation induces a release of a substance from neu-
dependent activation of adenylate cyclase and CaMKII, rons that evokes glutathione release from the glial
respectively (71). Increased Na⫹,K⫹-ATPase after glu- cells. One such substance may be K⫹, which is re-
tamate application to cultured cerebellar neurones leased following NMDA receptor overactivation (88).
was reduced by W7 and by the calcineurin inhibitor Because high K⫹-evoked efflux of taurine from glial
cyclosporin, suggesting an involvement of calmodulin- cells appear to be dependent on calmodulin (89,90),
calcineurin–dependent dephosphorylation (72). Inter- we can hypothesize the efflux after NMDA may, at
estingly, W7 prevents glutamate-induced depletion of least partly, be due to glial uptake of K⫹, which leads
ATP and neuronal death in cultured cerebellar neu- to passive osmotic uptake of water and opening of
rones (73). Also the anion transport inhibitor DNDS DNDS-sensitive volume activated anion channels via
attenuated the efflux rate of glutathione, PEA, and calmodulin (89,90).
taurine after NMDA removal. DNDS decrease the per- The likely unselective effects of the drugs quin-
meability of glutathione through the cystic fibro- acrine and PMXB on NMDA-mediated calcium influx
sis transmembrane conductance regulator (CFTR) is not unique. Several of the drugs used in studying
(74,75), a phosphorylation and ATP-dependent mem- the mechanisms of organic anion transport, such as
brane transport protein, permeable both to chloride furosemide, piretamide, and bumetanide, as well as ni-
and larger organic anions (74,75). Interestingly, the flumic and flufenamic acids, appear to block NMDA-
CFTR mRNA and protein is expressed in the hip- induced currents by binding to the NMDA-receptor
pocampus (76). Another disulphonic stilbene deriva- (91), a finding that often is neglected. Similarly, it was
tive, DIDS, inhibits glutathione transport in yeast by recently reported that combinations of antioxidants such
NMDA-Induced Glutathione Efflux 289

as TEMPO, catalase, trolox, and ascorbate had pro- necrosis by attenuating extracellular glutamate concentrations.
found blocking effects on NMDA currents (92). In our J. Neurochem. 69:412–417.
6. Lafon-Cazal, M., Pietri, S., Culcasi, M., and Bockaert, J. 1993.
experiments, we have used concentrations of drugs NMDA-dependent superoxide production and neurotoxicity.
that have been shown effective against their supposed Nature 364:535–537.
single targets in other studies. However, we realize 7. Bindokas, V. P., Jordan, J., Lee, C. C., and Miller, R. J. 1996.
Superoxide production in rat hippocampal neurons: selective
that findings of inhibited efflux by any drug needs ad- imaging with hydroethidine. J. Neurosci. 16:1324 –1336.
ditional information to rule out unselective effects on 8. Dugan, L. L., Sensi, S. L., Canzoniero, L. M., Handran, S. D.,
NMDA receptors and/or calcium influx. One way to Rothman, S. M., Lin, T. S., Goldberg, M. P., Choi, D. W., Dugan,
L. L., Sensi, S. L., Canzoniero, L. M., Handran, S. D., Rothman,
at least partly get around this problem is to use dif- S. M., Lin, T. S., Goldberg, M. P., and Choi, D. W. 1995. Mito-
ferent types of drugs, which we have done in the case chondrial production of reactive oxygen species in cortical neu-
of protein kinase C and phospholipase A2. In the case rons following exposure to N-methyl-D-aspartate. J. Neurosci.
15:6377–6388.
of DNDS and W7, we do not suspect unselective ef- 9. Siesjö, B. K. 1981. Cell damage in the brain: a speculative syn-
fects because the efflux during NMDA application thesis. J. Cereb. Blood Flow Metab. 1:155–185.
was unchanged. 10. Coyle, J. T. and Puttfarcken, P. 1993. Oxidative stress, gluta-
mate, and neurodegenerative disorders. Science 262:689–695.
In conclusion, NMDA-receptor stimulation cause 11. Love, S. 1999. Oxidative stress in brain ischemia. Brain Pathol.
an increased efflux of glutathione, PEA, and taurine 9:119–131.
in organotypic cultures of rat hippocampus. The pres- 12. Nicholls, D. G. and Budd, S. L. 1998. Neuronal excitotoxicity:
the role of mitochondria. Biofactors 8:287–299.
ent experiments indicate that efflux during and after 13. Brains, J. S. and Shaw, C. A. 1997. Neurodegenerative disor-
NMDA incubation may be regulated by different mech- ders in humans: the role of glutathione in oxidative stress-
anisms. The organotypic cultures appear to be a suit- mediated neuronal death. Brain Res. Brain Res. Rev. 25:335–358.
14. Dringen, R. 2000. Metabolism and functions of glutathione in
able model for further studies on the mechanisms of brain. Prog. Neurobiol. 62:649– 671.
NMDA-stimulated efflux, particularly if coupled to par- 15. Hayes, J. D. and McLellan, L. I. 1999. Glutathione and gluta-
allel measurements of toxicity. thione-dependent enzymes represent a co-ordinately regulated
defence against oxidative stress. Free Radic. Res. 31:273–300.
16. Meister, A. 1994. Glutathione-ascorbic acid antioxidant system
in animals. J. Biol. Chem. 269:9397–9400.
17. Shan, X. Q., Aw, T. Y., and Jones, D. P. 1990. Glutathione-
ACKNOWLEDGMENT dependent protection against oxidative injury. Pharmacol. Ther.
47:61–71.
Professor Anders Hamberger is gratefully appreciated for his 18. Meister, A. and Anderson, M. E. 1983. Glutathione. Annu. Rev.
never-failing inspiring enthusiasm in neurochemical research and for Biochem. 52:711–760.
19. Cooper, A. J. and Kristal, B. S. 1997. Multiple roles of glu-
his robust support during the last 25 years. We are grateful for the ex-
tathione in the central nervous system. Biol. Chem. 378:793–802.
pert technical assistance by Barbro Jilderos. The work was supported 20. Jain, A., Mårtensson, J., Stole, E., Auld, P. A., and Meister, A.
by the Swedish Natural Science Research Council (B 5101), Gunvor 1991. Glutathione deficiency leads to mitochondrial damage in
och Josef Anérs stiftelse, Åhlén-stiftelsen och Alzheimerfonden. MS brain. Proc. Natl. Acad. Sci. USA 88:1913–1917.
and SGW are supported by the National Institute of Health (GM 21. Mizui, T., Kinouchi, H., and Chan, P. H. 1992. Depletion of
44842). HW is supported by the Swedish Medical Research Counsil brain glutathione by buthionine sulfoximine enhances cerebral
(05954). ischemic injury in rats. Am. J. Physiol. 262:H313–H317.
22. Bridges, R. J., Koh, J. Y., Hatalski, C. G., and Cotman, C. W.
1991. Increased excitotoxic vulnerability of cortical cultures with
reduced levels of glutathione. Eur. J. Pharmacol. 192:199–200.
23. Ogita, K., Enomoto, R., Nakahara, F., Ishitsubo, N., and
Yoneda, Y. 1995. A possible role of glutathione as an endoge-
REFERENCES nous agonist at the N-methyl-D-aspartate recognition domain in
rat brain. J. Neurochem. 64:1088–1096.
1. Fonnum, F. 1984. Glutamate: A neurotransmitter in mammalian 24. Varga, V., Jenei, Z., Janaky, R., Saransaari, P., and Oja, S. S.
brain. J. Neurochem. 42:1–11. 1997. Glutathione is an endogenous ligand of rat brain N-
2. Simon, R. P., Swan, J. H., Griffiths, T., and Meldrum, B. S. methyl-D-aspartate (NMDA) and 2-amino-3-hydroxy-5-methyl-
1984. Blockade of N-methyl-D-aspartate receptors may protect 4-isoxazolepropionate (AMPA) receptors. Neurochem. Res. 22:
against ischemic damage in the brain. Science 226:850–852. 1165 –1171.
3. Benveniste, H., Drejer, J., Schousboe, A., and Diemer, N. H. 25. Janaky, R., Ogita, K., Pasqualotto, B. A., Bains, J. S., Oja, S. S.,
1984. Elevation of the extracellular concentrations of glutamate Yoneda, Y., and Shaw, C. A. 1999. Glutathione and signal trans-
and aspartate in rat hippocampus during transient cerebral is- duction in the mammalian CNS. J. Neurochem. 73:889–902.
chemia monitored by intracerebral microdialysis. J. Neurochem. 26. Gilbert, K. R., Aizenman, E., and Reynolds, I. J. 1991. Oxidized
43:1369–1374. glutathione modulates N-methyl-D-aspartate- and depolarization-
4. Hagberg, H., Lehmann, A., Sandberg, M., Nyström, B., Jacob- induced increases in intracellular Ca2⫹ in cultured rat forebrain
son, I., and Hamberger, A. 1985. Ischemia-induced shift of in- neurons. Neurosci. Lett. 133:11–14.
hibitory and excitatory amino acids from intra- to extracellular 27. Kohr, G., Eckardt, S., Luddens, H., Monyer, H., and Seeburg,
compartments. J. Cereb. Blood Flow Metab. 5:413–419. P. H. 1994. NMDA receptor channels: subunit-specific potenti-
5. Kawaguchi, K., Huerbin, M., and Simon, R. P. 1997. Lesioning ation by reducing agents. Neuron 12:1031–1040.
of deep prepiriform cortex protects against ischemic neuronal 28. Sucher, N. J. and Lipton, S. A. 1991. Redox modulatory site of
290 Wallin et al.

the NMDA receptor-channel complex: regulation by oxidized 47. Valiante, T. A., Perez Velazquez, J. L., Jahromi, S. S., and Carlen,
glutathione. J. Neurosci. Res. 30:582–591. P. L. 1995. Coupling potentials in CA1 neurons during calcium-
29. Kranich, O., Hamprecht, B., and Dringen, R. 1996. Different free-induced field burst activity. J. Neurosci. 15:6946 – 6956.
preferences in the utilization of amino acids for glutathione syn- 48. Haas, H. L. and Jefferys, J. G. 1984. Low-calcium field burst
thesis in cultured neurons and astroglial cells derived from rat discharges of CA1 pyramidal neurones in rat hippocampal slices.
brain. Neurosci. Lett. 219:211–214. J. Physiol. (Lond.) 354:185–201.
30. Sagara, J. I., Miura, K., and Bannai, S. 1993. Maintenance of neu- 49. Fariss, M. W., Olafsdottir, K., and Reed, D. J. 1984. Extra-
ronal glutathione by glial cells. J. Neurochem. 61:1672–1676. cellular calcium protects isolated rat hepatocytes from injury.
31. Dringen, R., Kranich, O., Löschmann, P. A., and Hamprecht, B. Biochem. Biophys. Res. Commun. 121:102–110.
1997. Use of dipeptides for the synthesis of glutathione by 50. Brodie, A. E. and Reed, D. J. 1991. Calcium chelation induced
astroglia-rich primary cultures. J. Neurochem. 69:868 – 874. glutathione efflux from tumor cells and prevention by ruthe-
32. Eliot, L. S., Dudai, Y., Kandel, E. R., and Abrams, T. W. 1989. nium red or neomycin. Biochem. Biophys. Res. Commun. 176:
Ca2⫹/calmodulin sensitivity may be common to all forms of neu- 276–279.
ral adenylate cyclase. Proc. Natl. Acad. Sci. USA 86:9564–9568. 51. Li, X., Orwar, O., Revesjö, C., and Sandberg, M. 1996. Gamma-
33. Zeevalk, G. D., Bernard, L. P., Sinha, C., Ehrhart, J., and Nick- glutamyl peptides and related amino acids in rat hippocampus
las, W. J. 1998. Excitotoxicity and oxidative stress during inhibi- in vitro: effect of depolarization and gamma-glutamyl transpep-
tion of energy metabolism. Dev. Neurosci. 20:444 – 453. tidase inhibition. Neurochem. Int. 29:121–128.
34. Han, J., Cheng, F. C., Yang, Z., and Dryhurst, G. 1999. In- 52. Li, X., Wallin, C., Weber, S. G., and Sandberg, M. 1999. Net ef-
hibitors of mitochondrial respiration, iron (II), and hydroxyl flux of cysteine, glutathione and related metabolites from rat hip-
radical evoke release and extracellular hydrolysis of glutathione pocampal slices during oxygen/glucose deprivation: dependence
in rat striatum and substantia nigra: potential implications to on gamma-glutamyl transpeptidase. Brain Res. 815:81–88.
Parkinson’s disease. J. Neurochem. 73:1683–1695. 53. Hemmings, S. J. and Storey, K. B. 1999. Brain gamma-glu-
35. Landolt, H., Lutz, T. W., Langemann, H., Stauble, D., Mende- tamyltranspeptidase: characteristics, development and thyroid
lowitsch, A., Gratz,l O., and Honegger, C. G. 1992. Extracellular hormone dependency of the enzyme in isolated microvessels
antioxidants and amino acids in the cortex of the rat: monitoring and neuronal/glial cell plasma membranes. Mol. Cell Biochem.
by microdialysis of early ischemic changes. J. Cereb. Blood Flow 202:119–130.
Metab. 12:96–102. 54. Bahr, B. A., Kessler, M., Rivera, S., Vanderklish, P. W., Hall,
36. Zängerle, L., Cuenod, M., Winterhalter, K. H., and Do, K. Q. R. A., Mutneja, M. S., Gall, C., and Hoffman, K. B. 1995. Stable
1992. Screening of thiol compounds: Depolarization-induced maintenance of glutamate receptors and other synaptic compo-
release of glutathione and cysteine from rat brain slices. J. Neu- nents in long-term hippocampal slices. Hippocampus 5:425 – 439.
rochem. 59:181–189. 55. Raiford, D. S., Sciuto, A. M., and Mitchell, M. C. 1991. Effects
37. Andiné, P., Orwar, O., Jacobson, I., Sandberg, M., and Hagberg, of vasopressor hormones and modulators of protein kinase C
H. 1991. Extracellular acidic sulfur-containing amino acids and on glutathione efflux from perfused rat liver. Am. J. Physiol.
gamma-glutamyl peptides in global ischemia: postischemic re- 261:G578–G584.
covery of neuronal activity is paralleled by a tetrodotoxin- 56. Sato, C., Liu, J. H., Tang, L., Sakai, Y., Yauchi, T., Izumi, N.,
sensitive increase in cysteine sulfinate in the CA1 of the rat Liu, J., Takano, T., and Marumo, F. 1992. Possible involvement
hippocampus. J. Neurochem. 57:230–236. of protein kinase C and calcium in GSH efflux from Hep G2
38. Orwar, O., Li, X., Andine, P., Bergstrom, C. M., Hagberg, H., cells. Life Sci. 51:2057–2063.
Folestad, S., and Sandberg, M. 1994. Increased intra- and ex- 57. Varecka, L., Peterajova, E., and Pogady, J. 1987. Polymyxin B,
tracellular concentrations of gamma-glutamylglutamate and re- a novel inhibitor of red cell Ca2⫹-activated K⫹ channel. FEBS
lated dipeptides in the ischemic rat striatum: involvement of Lett. 225:173–177.
glutamyl transpeptidase. J. Neurochem. 63:1371–1376. 58. Weik, R. and Lonnendonker, U. 1990. Polymyxin B as a highly ef-
39. Wallin, C., Weber, S. G., and Sandberg, M. 1999. Glutathione fective gating modifier of high-conductance Ca2(⫹)-activated K⫹
efflux induced by NMDA and kainate: implications in neuro- channels in mouse skeletal muscle. Pflugers Arch. 415:671– 677.
toxicity? J. Neurochem. 73:1566 –1572. 59. Hegemann, L., van Rooijen, L. A., Traber, J., and Schmidt,
40. Lehmann, A., Lazarewicz, J. W., and Zeise, M. 1985. N-Methyl- B. H. 1991. Polymyxin B is a selective and potent antagonist of
aspartate-evoked liberation of taurine and phosphoethanolamine calmodulin. Eur. J. Pharmacol. 207:17–22.
in vivo: Site of release. J. Neurochem. 45:1172–1177. 60. Lazarewicz, J. W., Salinska, E., and Wroblewski, J. T. 1992.
41. Menendez, N., Herreras, O., Solis, J. M., Herranz, A. S., and NMDA receptor-mediated arachidonic acid release in neurons:
Martin del Rio, R. 1989. Extracellular taurine increase in rat role in signal transduction and pathological aspects. Adv. Exp.
hippocampus evoked by specific glutamate receptor activation Med. Biol. 318:73–89.
is related to the excitatory potency of glutamate agonists. Neu- 61. Gunasekar, P. G., Kanthasamy, A. G., Borowitz, J. L., and
rosci. Lett. 102:64–69. Isom, G. E. 1995. NMDA receptor activation produces concur-
42. Sandberg, M., Butcher, S. P., and Hagberg, H. 1986. Extracel- rent generation of nitric oxide and reactive oxygen species: im-
lular overflow of neuroactive amino acids during severe insulin- plication for cell death. J. Neurochem. 65:2016–2021.
induced hypoglycemia: in vivo dialysis of the rat hippocampus. 62. Phillis, J. W. 1996. Cerebroprotective action of the phospholi-
J. Neurochem. 47:178–184. pase inhibitor quinacrine in the ischemia/reperfused gerbil hip-
43. Oja, S. S. and Saransaari, P. 2000. Modulation of taurine release pocampus. Life Sci. 58:L97–L101.
by glutamate receptors and nitric oxide. Prog. Neurobiol. 62: 63. Baba, A., Ohta, A., and Iwata, H. 1983. Inhibition by quinacrine
407– 425. of depolarization-induced acetylcholine release and calcium influx
44. Stoppini, L., Buchs, P. A., and Muller, D. 1991. A simple method in rat brain cortical synaptosomes. J. Neurochem. 40:1758–1761.
for organotypic cultures of nervous tissue. J. Neurosci. Meth. 64. Fernandez-Belda, F., Soler, F., and Gomez-Fernandez, J. C.
37:173–182. 1989. Quinacrine inhibits the calcium-induced calcium release in
45. Vornov, J. J., Park, J., and Thomas, A. G. 1998. Regional vul- heavy sarcoplasmic reticulum vesicles. Biochim. Biophys. Acta
nerability to endogenous and exogenous oxidative stress in 985:279–285.
organotypic hippocampal culture. Exp. Neurol. 149:109–122. 65. Mody, I. and MacDonald, J. F. 1995. NMDA receptor-dependent
46. Steinherz, R., Mårtensson, J., Wellner, D., and Meister, A. 1990. excitotoxicity: the role of intracellular Ca2⫹ release. Trends
Transport into brain of buthionine sulfoximine, an inhibitor of Pharmacol. Sci. 16:356–359.
glutathione synthesis, is facilitated by esterification and admin- 66. Hayashi, T., Kagaya, A., Takebayashi, M., Oyamada, T., Ina-
istration of dimethylsulfoxide. Brain Res. 518:115–119. gaki, M., Tawara, Y., Yokota, N., Horiguchi, J., Su, T. P., and
NMDA-Induced Glutathione Efflux 291

Yamawaki, S. 1997. Effect of dantrolene on KCI⫺ or NMDA- 79. Berger, R., Jensen, A., and Paschen, W. 1998. Metabolic dis-
induced intracellular Ca2⫹ changes and spontaneous Ca2⫹ oscil- turbances in hippocampal slices of fetal guinea pigs during and
lation in cultured rat frontal cortical neurons. J. Neural Transm. after oxygen-glucose deprivation: is nitric oxide involved? Neu-
104:811–824. rosci. Lett. 245:163–166.
67. Menendez, N., Solis, J. M., Herreras, O., Galarreta, M., Cone- 80. Bettendorff, L., Kolb, H. A., and Schoffeniels, E. 1993. Thi-
jero, C., and Martin del Rio, R. 1993. Taurine release evoked by amine triphosphate activates an anion channel of large unit
NMDA receptor activation is largely dependent on calcium conductance in neuroblastoma cells. J. Membr. Biol. 136:281–
mobilization from intracellular stores. Eur. J. Neurosci. 5:1273– 288.
1279. 81. Fuller, C. M., Ismailov, II, Keeton, D. A., and Benos, D. J. 1994.
68. Smigel, M. D. 1986. Purification of the catalyst of adenylate cy- Phosphorylation and activation of a bovine tracheal anion chan-
clase. J. Biol. Chem. 261:1976–1982. nel by Ca2⫹/calmodulin-dependent protein kinase II. J. Biol.
69. Walters, J. D. and Johnson, J. D. 1988. Inhibition of cyclic Chem. 269:26642–26650.
nucleotide phosphodiesterase and calcineurin by spermine, a 82. Philbert, M. A., Beiswanger, C. M., Waters, D. K., Reuhl, K. R.,
calcium-independent calmodulin antagonist. Biochim. Biophys. and Lowndes, H. E. 1991. Cellular and regional distribution of re-
Acta 957:138–142. duced glutathione in the nervous system of the rat: histochemical
70. Abu-Soud, H. M., Yoho, L. L., and Stuehr, D. J. 1994. Calmod- localization by mercury orange and o-phthaldialdehyde-induced
ulin controls neuronal nitric-oxide synthase by a dual mecha- histofluorescence. Toxicol. Appl. Pharmacol. 107:215–227.
nism: activation of intra- and interdomain electron transfer. J. 83. Huang, J. and Philbert, M. A. 1995. Distribution of glutathione
Biol. Chem. 269:32047–32050. and glutathione-related enzyme systems in mitochondria and
71. Tsuji, K., Nakamura, Y., Ogata, T., Shibata, T., and Kataoka, cytosol of cultured cerebellar astrocytes and granule cells. Brain
K. 1995. Transient increase of cyclic AMP induced by gluta- Res. 680:16–22.
mate in cultured neurons from rat spinal cord. J. Neurochem. 84. Langeveld, C. H., Schepens, E., Jongenelen, C. A., Stoof, J. C.,
65:1816 –1822. Hjelle, O. P., Ottersen, O. P., and Drukarch, B. 1996. Presence
72. Marcaida, G., Kosenko, E., Minana, M. D., Grisolia, S., and of glutathione immunoreactivity in cultured neurones and astro-
Felipo, V. 1996. Glutamate induces a calcineurin-mediated de- cytes. Neuroreport 7:1833–1836.
phosphorylation of Na⫹,K(⫹)-ATPase that results in its activation 85. Rice, M. E. and Russo-Menna, I. 1998. Differential compart-
in cerebellar neurons in culture. J. Neurochem. 66:99–104. mentalization of brain ascorbate and glutathione between neu-
73. Marcaida, G., Minana, M. D., Grisolia, S., and Felipo, V. 1995. rons and glia. Neuroscience 82:1213–1223.
Lack of correlation between glutamate-induced depletion of 86. Ong, W. Y., Hu, C. Y., Hjelle, O. P., Ottersen, O. P., and Halli-
ATP and neuronal death in primary cultures of cerebellum. well, B. 2000. Changes in glutathione in the hippocampus of
Brain Res. 695:146 –150. rats injected with kainate: depletion in neurons and upregulation
74. Linsdell, P. and Hanrahan, J. W. 1998. Glutathione permeabil- in glia. Exp. Brain Res. 132:510–516.
ity of CFTR. Am. J. Physiol. 275:C323– C326. 87. Seifert, G. and Steinhauser, C. 1995. Glial cells in the mouse
75. Linsdell, P. and Hanrahan, J. W. 1998. Adenosine triphosphate- hippocampus express AMPA receptors with an intermediate Ca2⫹
dependent asymmetry of anion permeation in the cystic fibrosis permeability. Eur. J. Neurosci. 7:1872–1881.
transmembrane conductance regulator chloride channel. J. Gen. 88. Yu, S. P., Yeh, C., Strasser, U., Tian, M., and Choi, D. W. 1999.
Physiol. 111:601–614. NMDA receptor-mediated K⫹ efflux and neuronal apoptosis.
76. Mulberg, A. E., Resta, L. P., Wiedner, E. B., Altschuler, S. M., Science 284:336–339.
Jefferson, D. M., and Broussard, D. L. 1995. Expression and 89. Pasantes-Morales, H., Cardin, V., and Tuz, K. 2000. Signaling
localization of the cystic fibrosis transmembrane conductance events during swelling and regulatory volume decrease. Neu-
regulator mRNA and its protein in rat brain. J. Clin. Invest. 96: rochem. Res. 25:1301–1314.
646 – 652. 90. Mongin, A. A., Cai, Z., and Kimelberg, H. K. 1999. Volume-
77. Rebbeor, J. F., Connolly, G. C., Dumont, M. E., and Ballatori, dependent taurine release from cultured astrocytes requires per-
N. 1998. ATP-dependent transport of reduced glutathione on missive [Ca(2⫹)](i) and calmodulin. Am. J. Physiol. 277:C823–
YCF1, the yeast orthologue of mammalian multidrug resistance C832.
associated proteins. J. Biol. Chem. 273:33449–33454. 91. Lerma, J. and Martin del Rio, R. 1992. Chloride transport block-
78. Kannan, R., Yi, J. R., Tang, D., Li, Y., Zlokovic, B. V., and ers prevent N-methyl-D-aspartate receptor-channel complex acti-
Kaplowitz, N. 1996. Evidence for the existence of a sodium- vation. Mol. Pharmacol. 41:217–222.
dependent glutathione (GSH) transporter: expression of bovine 92. Vergun, O., Sobolevsky, A. I., Yelshansky, M. V., Keelan, J.,
brain capillary mRNA and size fractions in Xenopus laevis Khodorov, B. I., and Duchen, M. R. 2001. Exploration of the
oocytes and dissociation from gamma-glutamyltranspeptidase role of reactive oxygen species in glutamate neurotoxicity in rat
and facilitative GSH transporters. J. Biol. Chem. 271:9754–9758. hippocampal neurones in culture. J. Physiol. 531:147–163.