1843

Regulation of the Na�-Dependent Glutamate/AspartateTransporter in Rodent Cerebellar Astrocytes

Alfonso Bernabé,1 J. Alfredo Méndez,1 L. Clara R. Hernández-Kelly,1 and Arturo Ortega1

(Accepted June 2, 2003)

The regulation of the Na�-dependent glutamate/aspartate transporter system GLAST expressedin rat and mouse cerebellar and cortical astrocytic cultures was examined. Pretreatment of thecerebellar cells with L-glutamate and 12-O-tetradecanoyl-phorbol-13-acetate (TPA), a knownCa2�/ diacylglicerol-dependent protein kinase (PKC) activator, produced a decrease in [3H]-D-aspartate uptake. This reduction was dose- and time-dependent and sensitive to PKC inhibitors.Furthermore, the L-glutamate–dependent [3H]-D-aspartate uptake decrease is a non-receptor depend-ent process, because neither of the agonists or antagonists were effective in mimicking or revert-ing the effect. Interestingly, transportable substrates could reproduce the L-glutamate effect. Insharp contrast, in cortical astrocytes, both L-glutamate and TPA pre-exposure result in an aug-mentation of the [3H]-D-aspartate uptake. These findings suggest that the Na�-dependent gluta-mate uptake GLAST undergoes a region-specific regulation.

KEY WORDS: GLAST; cerebellum; astrocytes; glutamate uptake; signal transduction.

INTRODUCTION

L-Glutamate (Glu) is the major excitatory neurotrans-mitter in the mammalian central nervous system (CNS)(1,2). Extracellular Glu concentrations are tightly regu-lated through its uptake in a sodium-dependent process(3). Thus far, five members of the sodium-dependent Glutransporter family have been identified: GLAST, GLT-1,EAAC1, EAAT4, and EAAT5 (4). In the CNS, expres-sion of GLAST and GLT-1, is generally restricted to glialcells and they are the major contributors to the Glu uptakefrom the synaptic cleft. EAAC1 and EAAT4 are locatedin neurons, whereas EAAT5 is present in both neuronsand glial cells in retina. These plasma membrane proteins

may function not merely by clearing the synaptic cleft butalso modulating excitation by rapidly buffering theneurotransmitter concentration (5). Several factors, suchas phorbol esters, arachidonic acid, nitric oxide, and freeoxygen radicals are known to regulate Glu uptake process(4). It has been reported that Glu and kainate upregulatethe expression and uptake capacity of Glu transporters incortical astrocytes cultures (6,7). In contrast, heterologousexpression of GLAST cDNA in Xenopus oocytes andhuman embryonic kidney cells has shown that phorbolesters downregulate GLAST activity (8). Interestingly,removal of all putative PKC phosphorylation sites of wild-type GLAST by site-directed mutagenesis did not abolishthe phorbol ester effect (8). More recently, the turnoverof transporters from the intracellular compartments to theplasma membrane has proven to be the mechanism bywhich the density of transporters is regulated (4).

Primary astrocyte-enriched cerebellar culturesexpress high levels of GLAST and low levels of theother transporters. This is not surprising, because GLASTis the major Glu transporter within the cerebellum (9). In

0364-3190/03/1200–1843/0 © 2003 Plenum Publishing Corporation

Neurochemical Research, Vol. 28, No. 12, December 2003 (© 2003), pp. 1843–1849

1 Departamento de Genética y Biología Molecular, Cinvestav-IPN,Apartado Postal 14-740 México, D.F., 07000, México.

2 Address reprint requests to: Arturo Ortega, Departamento Genética yBiología Molecular, Cinvestav-IPN, Apartado Postal 14–740, México,D.F., 07000, México; Fax: 525–747–7100; E-mail: arortega@mail.cin-vestav.mx

473428.qxd 9/24/03 8:59 PM Page 1843

1844 Bernabé, Méndez, Hernández-Kelly, and Ortega

fact, GLAST mRNA and protein are consistentlydetected in the cerebellum at much higher levels than inany other brain region, making cerebellar astrocytic cul-tures an ideal model to study GLAST regulation.Using a primary culture of chick cerebellar Bergmannglial cells, we have shown that phorbol esters reduceGlu transport, as well as GLAST protein and mRNAlevels (10,11). These studies suggest a plausible roleof GLAST as modulator of the cerebellar synapticactivity.

As already stated, conflicting results regardingGLAST regulation by phorbol esters have beenreported (6–8,10,11). Despite the fact that GLAST isby far the major excitatory amino acid transporter inthe cerebellum, several studies concerning GLASTregulation have been performed in cortical astrocytesin culture (6,7). In this context, the aim of the presentstudy was to investigate a possible differential regula-tion of the Glu uptake in rodent cerebellar and corti-cal astrocytic cultures. This is particularly importantbecause various aspects of GLAST regulation havebeen described in lower vertabrate species, making itimperative to evaluate whether these results hold truefor other cerebellar preparations and are not exclu-sively for these species (10,11). The results presentedherein indicate a region-specific regulation of the Glutransporter GLAST.

EXPERIMENTAL PROCEDURE

Chemicals. [3H]-D-Aspartate was obtained from New EnglandNuclear (Boston, MA, USA). Tissue culture reagents were from Gibco(Gaithersburg, MD, USA). Plasticware was purchased from Costar(Cambridge, MA, USA). The PKC inhibitors, 1-(5-isoquinolinesul-fonyl)-2-methyl-piperazine dihydrochloride (H7), N-(2-aminoethyl)-5-isoquinolinesulfonamide dihydrochloride (H9), and staurosporine, aswell as agonists and antagonists, were obtained from Tocris (Ballwin,MO, USA). All other chemicals were from Sigma (St. Louis, MO,USA).

Cell Culture. Primary rat and mouse cultures of cerebellar orcortical astrocytes were prepared from rat and mouse pups, 8- and4-postnatal day, respectively, as described previously (12). Cells wereplated at a density of 0.5 million cells/ml in 24-well in Eagle’s MEMcontaining 10% fetal bovine serum (FBS), 2 mM glutamine, and gen-tamicin (50 �g/ml) and maintained at 37°C in a 5% CO2 incubator.Cells were used at confluence (day 8 and 12–14 in vitro, respectively).Immunocytochemical studies of these primary cultures demonstratethat the cultured cells are at least 95% GFAP-positive astrocytes (notshown).

[3H]-D-Aspartate Transport Assay in Astrocytes. The uptake of[3H]-D-aspartate (used as a nonmetabolizable analogue of L-Glu) wasperformed as detailed elsewhere (10). Briefly, the culture medium wasexchanged with solution A (25 mM HEPES-Tris, 130 mM Na Cl,5.4 mM KCl, 1.8 mM CaCl2, 0.8 mM MgCl2, 33.3 mM glucose, and

1 mM NaHPO4, pH 7.4) and the cells preincubated for 30 min at 37°C.The monolayers were then preincubated with glutamatergic ligands orTPA for different periods of time. The cells were thoroughly washedseveral times with solution A; this medium was exchanged for solutionA containing [3H]-D-aspartate (0.4 �Ci/ml., 14 Ci/mM) and incubatedfor 10 min. Thereafter the medium was removed by rapid aspiration,and the monolayers were washed with ice-cold solution A and solubi-lized with 0.1 M NaOH. Aliquots of the suspension were used for pro-tein determination and liquid scintillation counting. All the uptake datawere analyzed using the PRISM software program (GraphPad).

RESULTS

Glu or TPA Pre-exposure Reduce [3H]-D-AspartateTransport

The sodium-dependent Glu transporter familyuses both sodium and potassium transmembrane elec-trochemical gradients as the driving force for Gluuptake. To establish if the Glu uptake system presentin our cerebellar astrocyte cellular cultures is sodium-dependent, we performed [3H]-D-aspartate uptake stud-ies in the presence or absence of sodium ions. Asexpected, both in rat and mouse cerebellar astrocyticcultures the [3H]-D-aspartate uptake process is strictlysodium dependent (Fig. 1).

Treatment of the cultures for 30 min with increas-ing concentrations of the protein kinase C activatorTPA (0.1 nM–1 �M) or with Glu (0.01–1 mM) beforethe uptake assay, resulted in a progressive reductionof [3H]-D-aspartate transport (Fig. 2). Next, we exam-ined the time dependency of this pre-exposure to TPAor Glu to elicit a significant decrease in a 10 min[3H]-D-aspartate uptake assay. As shown in Fig. 3,30 min of pre-exposure is sufficient to reduce thetransport of the labeled amino acid. Long-term pre-treatment of cell cultures with TPA or Glu leads toa sustained decrease of the transporter activity, sug-gesting that besides a short-term effect that couldreflect a phosphorylation-regulated translocation ofthe transporter molecules to the plasma membrane, along-term effect that may involve a transcriptionallevel of regulation could be triggered by Glu and TPA(Fig. 3). This result is opposite to what has beenreported in other culture systems, so we decided toexplore whether in our assay conditions, we couldreproduce the well-documented TPA and Glu upregu-lation of [3H]-D-aspartate uptake in mouse corticalastrocytes (6,7). The results are shown in panel C ofFig. 3, in line with the published work, an increase inthe uptake activity was obtained. Therefore one can beconfident that in cerebellar astrocytes, Glu and TPAreduce GLAST uptake activity.

473428.qxd 9/24/03 8:59 PM Page 1844

GLAST Regulation in Cerebellar Astrocytes 1845

Fig. 1. [3H]-D-aspartate uptake in rodent cerebellar astrocytes issodium-dependent. Panel A, Rat cerebellar astrocytes. Panel B, Mousecerebellar astrocytes. In both cases the cell monolayers were incu-bated with 0.4 �Ci/ml of [3H]-D-aspartate for the indicated times inthe presence of sodium (open bars) or choline (filled bars). The mono-layers were extensively washed and solubilized with 0.1 N NaOH.Values represent means � SD of at least three independent experi-ments performed in triplicate. *P � .05 (Student’s t test), as comparedto control values.

Fig. 2. Increasing concentrations of TPA or Glu in the preincubationmedia reduce the Na�-dependent aspartate uptake in rodent cerebel-lar astrocytes in culture. Cells were preincubated with the indicatedconcentrations of TPA (panel A) or Glu (panel B) for 60 min at roomtemperature. After extensive washing, the media was replaced withsolution A, containing 0.4 �Ci/ml [3H]-D-aspartate, and the uptakeactivity was measured for 10 min. Values represent means � SD ofat least three independent experiments performed in triplicate. *P �.05 (Student’s t test), as compared to control values

[3H]-D-Aspartate Uptake Reduction Is PKCMediated and Non–Receptor Dependent

At this stage, we decided to investigate the role ofPKC both in TPA- and Glu-induced decrease of the[3H]-D-aspartate uptake. To this end, we pre-exposed themonolayers to several PKC inhibitors: H-7, H-9, or stau-rosporine at the appropriate concentrations at which theyare specific for PKC inhibition (13,14). The results arepresented in Fig. 4; the C kinase blockers inhibit boththe TPA and the Glu effect. These results strongly sug-gest the involvement of PKC in the Glu-dependent reg-ulation of the transporter.

Next, we decided to use specific glutamatergic ago-nists and antagonists to characterize pharmacologicallythe Glu response. Cells were exposed for 1 h to a fixedconcentration of 1 mM Glu, kainate (KA), DL-2-amino-4-phosphonobuttyric acid (DL-AP4), N-methyl-D-aspar-tate (NMDA), or 0.5 mM quisqualate (Quis). As depictedin the Fig. 5, none of the Glu agonists tested was ableto mimic the Glu-induced effect. Similarly, when thecells were pre-exposed for 45 min to the Glu antagonistsdizocilpine (MK-801, 10 �M) and 6,7-dinitroquinoxa-line-2,3-dione (DNQX, 25 �M) before 1 mM Glu, noneof these antagonists reverted the Glu effect. A similar

473428.qxd 9/24/03 8:59 PM Page 1845

1846 Bernabé, Méndez, Hernández-Kelly, and Ortega

1Fig. 3. Kinetics of the Glu and TPA effect on [3H]-D-aspartateuptake in rodent cultured astrocytes. Monolayers were exposed forthe indicated time periods to 100 nM TPA or 1 mM Glu andremoved before the uptake assay. Panel A, Rat cerebellar astrocytes.Panel B, Mouse cerebellar astrocytes. Panel C, Mouse cortical astro-cytes. Values represent means � SD of at least three independentexperiments performed in triplicate. *P � .05 (Student’s t test), ascompared to control values.

Fig. 4. Effect of staurosporine (500 nM), H-7 (100 �M), or H-9 (100�M) on Na�-dependent [3H]-D-aspartate uptake in rat (panel A) ormouse (panel B) cerebellar astrocytes in culture. The uptake activitywas measured for 10 min. Values represent means � SD of at leastthree independent experiments performed in triplicate. *P � .05 (Stu-dent’s t test), as compared to control values.

scenario, meaning, a non–receptor-mediated regulationof Glu transport, has been reported in cultured chickBergmann glial cells and in cultures of cortical astro-cytes, albeit in the latter case, Glu treatment upregulatesGlu transport and GLAST surface expression (7,15).Thus, GLAST regulation by Glu is most likely anon–receptor-mediated and region-specific event.

[3H]-D-Aspartate Uptake Reduction Is Dependenton Substrate Transport

The observation that the reduction in [3H]-D-aspartateuptake elicited by Glu pre-exposure is a non–receptor-mediated effect led us to explore whether the decrease in

473428.qxd 9/24/03 8:59 PM Page 1846

GLAST Regulation in Cerebellar Astrocytes 1847

Fig. 5. Effect of glutamatergic ligands on the Na�-dependent [3H]-D-aspartate uptake in rat (panel A) or mouse (panel B) cerebellar astro-cytes in culture. Glutamatergic agents were used as follows: Glu, KA,AP-4, NMDA (1 mM), Quis (0.5 mM). The antagonists DNQX(25 �M) and MK-801 (10 �M) were added 45 min before Glu. A con-centration of 100 �M of glycine was present in the experiments usingNMDA. Values represent means � SD of at least three independentexperiments performed in triplicate. *P � .05 (Student’s t test), as com-pared to control values.

Fig. 6. Effect of the application of transportable substrates on [3H]-D-aspartate uptake in rat and mouse cerebellar astrocytes in culture. Mono-layers were incubated for 60 min with the indicated substrates, all at 1mM concentration. The cultured cells were then washed, and the [3H]-D-aspartate uptake was measured for 10 min. Values represent means� SD of at least three independent experiments performed in triplicate.*P � .05 (Student’s t test), as compared to control values.

uptake activity was dependent on the substrate translo-cation through the Glu carrier. For this purpose, cultureswere exposed to transportable as well as nontrans-portable GLAST substrates. Exposure of the cells toD-aspartate (Asp) (1 mM), DL-threo-�-hydroxyaspartate(THA) (1 mM), aspartate-�-hydroxamate (BHA) (1 mM),or L-�-aminoadipate (AAA) (1 mM) for 1 h, resulted ina differential effect on [3H]-D-aspartate uptake. The resultsshown in Fig. 6 demonstrate that the transportable ana-logues ASP and THA inhibit the transport in a similar

fashion as Glu, strongly suggesting that the transport processitself leads to a conformational change in the transportermolecule that eventually signals, via PKC, to modifythe translocation of the protein in and out of the plasmamembrane (4).

DISCUSSION

Within the cerebellum, most, if not all, of the [3H]-D-aspartate uptake activity is carried through GLAST (9).In this context, one can be confident that cerebellar astro-cytic cultures are an excellent model to study GLASTregulation. In fact, using chick cerebellar Bergmann gliacells in culture, we have outlined important features ofGLAST regulation, such as its regulation by phorbolesters, Glu, receptor tyrosine kinases, and, more recently,its transcriptional regulation (10,11,15,16). Nevertheless,experiments carried out with cortical rat and mouse astro-cytes in culture report different regulation processes, forexample, after Glu or TPA pre-exposure. Whereas inBergmann glial cells, Glu elicits a receptor-independentinhibition of the transporter activity, in rat cortical astro-cytes, Glu treatment results in an enhancement of thetransporter activity and expression (6,7). To investigatewhether this discrepancy could reflect a tissue differen-tial regulation of GLAST, rather than a species differ-ence, we decided to study GLAST modulation in rodentcerebellar astrocytes in culture.

473428.qxd 9/24/03 8:59 PM Page 1847

1848 Bernabé, Méndez, Hernández-Kelly, and Ortega

The activity and expression of Glu transporters ishighly regulated. Different levels of Glu transporter reg-ulation have been examined in diverse systems such assynaptosomal preparations, cell lines, and Xenopus lae-vis oocytes and in primary cultures. Molecular cloninghas revealed that all cloned Glu transporters exhibit puta-tive phosphorylation sites for protein kinase A (PKA)and protein kinase C. Therefore, second messengers andprotein kinases apparently regulate both the activity andexpression of these transporters. Major mechanisms bywhich Glu transporters are regulated include phosphory-lation, translocation from/to plasma membrane, and asso-ciation with proteins (4). Rapid changes in activity arethe result of phosphorylation events and translocationto/from the cell surface. Using rat and mouse cerebellarcultured astrocytes, we detected a significant decrease intransporter activity upon Glu or TPA treatment. Thisdownregulation occurred rapidly, after 5 min from theapplication of stimuli. These results are in line with thefact that phorbol esters decrease the functional activityof the transporter EAAC1 within minutes and induce itsinternalization from the plasma membrane (17).

It has been established for some years now thatlong-term treatment with phorbol esters downregulatesPKC activity and a dramatic reduction in its immuno-reactivity levels is present (18). With this in mind, wedecided to incubate the cultured cells for 24 h with eitherTPA or Glu. Under such circumstances, the [3H]-D-aspartate uptake activity did not return to control levels,suggesting that, besides a PKC-dependent short-termreduction in transporter activity, this protein also mightbe regulated at a different level, such as its gene expres-sion. This level of regulation is present, as already stated,in chick cultured Bergmann glial cells, in which long-term TPA treatment reduces GLAST mRNA steady-statelevels, without a significant change in GLAST mRNAhalf-life (11). In fact, as stated above, one of the mainreasons to undertake the present study, was to gaininsight into the discrepancy with the referred results inchick cerebellar Bergmann glial cells, with the reportedincrease in GLAST uptake and protein levels as a resultof treatment of rat and mouse cortical astrocytes withphorbol esters, Glu, or even dibutyril cAMP (6,7). Theresults presented in this communication argue for aregion-specific regulation, rather than a species-specificdifferential mode of regulation.

Of particular interest are the results obtained whenthe cells were exposed to Glu agonists and antagonists.None of the agonists tested—KA, Quis, AP-4, orNMDA—reproduced the Glu effect; moreover, neitherof the antagonists used (DNQX or MK-801) revertedthe Glu effect. One could argue that the pharmacolog-

ical agents tested cannot rule out that a Glu receptor isinvolved in the described effects. This suggestion isunlikely because the agonists were tested at concentra-tions high enough to activate both ionotropic ormetabotropic receptors. Another simple interpretation isthat Glu competes with the [3H]-D-aspartate used tomeasure the uptake leading to an isotopic dilution andthus explaining why the receptor blockers had no effect.This is an oversimplified view because the Glu treat-ment precedes the exposure of the cultured cells to theradioactive tracer, and in fact the cells are washed sev-eral times before the uptake process. The amount ofGlu that could remain in the medium would be mini-mal, and thus no isotopic dilution would be present.Another possibility is that Glu exposure would lead toa sustained depolarization that could result in anincreased intracellular calcium concentration, throughthe activation of AMPA receptors expressed in the cere-bellar astrocytes. This would lead to sodium entrythrough the Na�/Ca� exchanger, inhibiting GLAST.This mechanism is ruled out by the fact that none ofthe Glu receptor agonists could mimic the effect(Fig. 5). In any event, when the possibility that the Glueffect could be mediated through the transporter itselfwas explored, we could establish that only transportableanalogues of Glu, such as Asp and THA, reproduce thedecrease in [3H]-D-aspartate uptake, reinforcing thehypothesis that the transport of the substrate leads to aconformational change of the transporter molecule thateventually results in a decrease of the uptake activity(Fig. 6). Interestingly, this effect is sensitive to PKCblockers (Fig. 4), suggesting that the transporter mightfunction as a signal transducer.

The GLAST-mouse gene promoter region has beencharacterized. A putative AP-1 binding site has beenreported, predicting the interaction of inducible tran-scription factors (19). If one takes into consideration thatGlu induces AP-1 DNA binding in practically all thesystems tested, a plausible explanation for the sustainedGlu effect on GLAST activity could be a transcriptionaldownregulation. Recent experiments of our groupdemonstrate that GLAST undergoes a transcriptionalcontrol upon Glu exposure (16).

SUMMARY

The GLAST-mediated Glu uptake differencesreported between both cerebral cortex and cerebellarastrocytes cultures are likely to be due to a region-related differential regulation.

473428.qxd 9/24/03 8:59 PM Page 1848

GLAST Regulation in Cerebellar Astrocytes 1849

ACKNOWLEDGMENTS

This work was supported by a Conayct-Mexico grant(33058-N) to A. O., A. B., and J. A. M. are supported by a Conacyt-Mexico scholarship.

REFERENCES

1. Fonnum, F. 1984. Glutamate: A neurotransmitter in mammalianbrain. J. Neurochem. 42:1–11.

2. Nicholls, G. 1993. The glutamatergic nerve terminal. Eur. J.Biochem. 212:613–631.

3. Robinson, M. B. and Dowd, L. A. 1997. Heterogeneity and func-tional properties of subtypes of sodium-dependent glutamatetransporters in the mammalian central nervous system. Adv. Phar-macol. 37:69–115.

4. Robinson, M. B. 2002. Regulated trafficking of neurotransmittertransporters: Common notes but different melodies. J. Neurochem.80:1–11.

5. Barbour, B., Keller, B. U., Llano, I., and Marty, A. 1994. Pro-longed presence of glutamate during excitatory synaptic trans-mission to cerebellar Purkinje cells. Neuron 12:1331–1343.

6. Gegelashvili, G., Civenni, G., Racagni, G., Danbolt, N. C.,Schousboe, I., and Schousboe, A. 1996. Glutamate receptor ago-nists up-regulate glutamate transporter GLAST in astrocytes. Neu-roreport 8:261–265.

7. Duan, S., Anderson, C. M., Stein, B. A., and Swanson R. A.1999. Glutamate induces rapid upregulation of astrocyte gluta-mate transport and cell-surface expression of GLAST. J. Neu-rosci. 23:10193–10200.

8. Conradt, M. and Stoffel, W. 1997. Inhibition of the high-affinitybrain glutamate transporter GLAST-1 via direct phosphorylation.J. Neurochem. 68:1244–1251.

9. Ullensvang, K., Lehre, K. P., Storm-Mathiesen, J., and Danbolt,N. C. 1997. Differential developmental expression of two rat brainglutamate transporter proteins GLAST and GLT. Eur. J. Neurosci.9:1646–1655.

10. González, M. I. and Ortega, A. 1997. Regulation of the Na�-dependent high-affinity glutamate/aspartate transporter in culturedBergmann glia by phorbol esters. J. Neurosci. Res. 50:585–590.

11. Espinoza-Rojo, M., López-Bayghen, E., and Ortega, A. 2000.GLAST: Gene expression regulation by phorbol esters. Neuro-report 12:2827–2832.

12. McCarthy, K. D. and de Vellis, J., 1980. Preparation of separateastroglial and oligodendroglial cell cultures from rat cerebral tis-sue. J Cell Biol. 85:890–902.

13. Kawamoto, S. and Hidaka, H. 1984. 1-(5-Isoquinolinesulfonyl)-2-methylpiperazine (H-7) is a selective inhibitor of protein kinase Cin rabbit platelets. Biochem. Biophys. Res. Commun. 125:258–264.

14. Tamaoki, T., Nomoto, H., Takahashi, I., Kato, Y., Morimoto, M.,and Tomita, F. 1986. Staurosporine: A potent inhibitor of phos-pholipids/Ca��-dependent protein kinase. Biochem. Biophys.Res. Commun. 135:397–402.

15. González, M. I. and Ortega, A. 2000. Regulation of high-affinityglutamate uptake activity in Bergmann glia cells by glutamate.Brain Res. 866:73–81.

16. López-Bayghen, E., Espinoza-Rojo, M., and Ortega A. 2003. Glu-tamate down-regulates GLAST expression through AMPA recep-tors in Bergmann glia cells. Molec. Brain Res. 115:1–9.

17. Troti, D., Peng. J., Dunlop. J., and Hediger, M. A. 2001. Inhib-ition of the glutamate transporter EAAC1 expressed in Xenopusoocytes by phorbol esters. Brain Res. 914:196–203

18. Chen, C. C., Cheng, C. S., Chang, J., and Huang, H. C. 1995.Differential correlation between translocation and down-regula-tion of conventional and new protein kinase C isoenzymes in C6

glioma cells. J. Neurochem. 64:818–824.19. Hagiwara, K., Tanaka, S., Takai, Y., Maeno-Hikichi, Y., Mukainaka,

Y., and Wada, K. 1996. Genomic organization, promoter analysis,and chromosomal localization of the gene for the mouse glial high-affinity glutamate transporter S1ca3. Genomics 33:508–515.

473428.qxd 9/24/03 8:59 PM Page 1849