Journalof NeurochemistryLippincott—RavenPublishers,Philadelphia© 1998 InternationalSociety for Neurochemistry

Ethanol and Regulation of the NMDA Receptor Subunits inFetal Cortical Neurons

MeenaKumari and Maharaj K. Ticku

Departmentof Pharmacology, Universityof TexasHealth ScienceCenter, SanAntonio, Texas, U.S.A.

Abstract: Previous studies have shown that chronic etha-nol treatment up-regulates the expression of the N-methyl-o-aspartate (NMDA) receptor number and func-tion both in vitro and in vivo. In vitro chronic ethanoltreatment specifically augments mANA levels of the R2Bsubunit without altering Ri subunit mRNA levels, al-though similar treatment results in increased levels ofboth Ri and R2B polypeptides. To further delineate themolecular mechanisms involved in differential regulationof NMDA receptor subunits by chronic ethanol treatment(50 mM, 5 days), we have determined the mANA stabilityof the NMDA Ri and R2B subunits and the transcriptionrate of the respective genes using mouse fetal corticalneurons. Our observations demonstrated that ethanolstabilized the NMDA Ri mRNA over the time period ex-amined (24 h) without altering the stability of the R2BmRNA. Chronic exposure of fetal cortical neurons to etha-nol enhanced the rate of R2B gene transcription approxi-mately twofold. Taken together, these results suggestthat up-regulation of the NMDA receptor number seen incultured cortical neurons after chronic ethanol treatmentis due to the regulation of the NMDA R2B receptor sub-unit at the transcriptional level and that of the NMDARi subunit mainly at the posttranscriptional level. KeyWords: N MDA receptor—Cortical neuron—Gene ex-pression —Transcription — Posttranscriptional control.J. Neurochem. 70, 1467—1473 (1998).

In theCNS,fastexcitatoryneurotransmissionis me-diated by the neurotransmitterglutamate. The N-methyl-D-aspartate(NMDA) receptor,a memberof aclassof ligand-gatedion channels,is one subtypeofglutamate-responsivereceptors. Activation of theNMDA receptorsby glutamateresults in a rapid in-creasein intracellularcalcium(~Ca2~I

1)within neuronsexpressingthesereceptors.Theobserved[Ca

2*]jinfluxis known to be a key componentof a numberof im-portant physiologicalevents that influence neuronalactivity includingneuronaldevelopment,maintenanceof neuronalexcitability, and long-term potentiation.Overstimulationof theNMDA receptorleadsto neuro-degenerationandcell death,which is believedto resultfrom excessive[Ca2~]

1influx through the NMDA re-ceptorion channel(Simonet al., 1984;Choi, 1988).

The NMDA receptorsubunits,identifiedby molecu-lar cloning, areclassifiedinto two groups.The ubiqui-tously expressedNRI family, whichappearsto be thekey componentrequired for functional activity, haseight splice variants(Zukin and Bennett, 1995). TheNR2 family is comprisedof four members(NR2A—D) that are expressedin a region-specificmannerinthe CNS (Kutsuwadaet al., 1992). Studiesusing invitro expressionsystemsindicate that native NMDAreceptorsare composedof combinationsof the NRIsubunitandmembersof the NR2 family, eachcombi-nation exhibiting distinct pharmacologicalproperties(MirshahiandWoodward,1995;Priestleyet al., 1995).

Ethanol,which is one of the most widely abuseddrugs,is known to produceits diverseeffectsby modu-lating neurotransmissionin the brain. Although theGABAergic systemhas classically been associatedwith ethanol‘seffects in the CNS (Mehta andTicku,1995), recentstudieshave implicated the NMDA re-ceptorsystemin the developmentof ethanol-inducedtolerance,physical dependence,and withdrawal syn-drome (Lovingeret al., 1989; Hoffman, 1995). Etha-nol‘ seffecton theNMDA receptoris treatmentdepen-dent. In particular,acuteexposureto ethanol inhibitsNMDA receptor function (Lovinger et al., 1989),whereas chronic ethanol treatment increases theNMDA receptornumberin ratcerebralcortexandhip-pocampusandtheirin vivo function(Grantet al., 1990;Sannaet al., 1992). Comparableresults are obtainedwhenfetal corticalneuronsareexposedto chroniceth-anol treatmentin vitro. An increasein ~

3H]MK-80l(+)-543H]methyl-10,11-dihydro-5H-dibenzo~a,dl-

cyclohepten-5,I 0-imine maleate} binding sitesand anNMDA-mediated increase in ~Ca2~]~concentrationhavebeenreportedafter chronicexposureof fetal cor-

ResubmittedmanuscriptreceivedOctober30, 1997; acceptedNo-vember21,1997.

Addresscorrespondenceand reprint requeststo Dr. M. K. TickuatDepartmentofPharmacology,Universityof TexasHealthScienceCenterat San Antonio, 7703 Floyd Curl Drive, San Antonio, TX78284-7764,U.S.A. -

Abbreviationsused: [Ca2~l~,intracellularcalcium concentration;NIB, nuclearisolationbuffer; NMDA, N-methyl-D-aspartate;SDS,sodiumdodecyl sulfate; SSC,saline—sodiumcitrate.

1467

1468 M. KUMARI AND M. K. TICKU

tical neuronsto ethanol (Hu and Ticku, 1995). Thisin vitro up-regulationof NMDA receptornumber isaccompaniedby anincreasein theNMDA RI andR2Bpolypeptidelevels (FollesaandTicku, 1996).Whereaschronic ethanolexposureaugmentsthe RI and R2Bpolypeptidelevels, similar treatmentboth in vivo andin vitro, resultsin increasedlevels of mRNA encodingthe membersof the NR2 family (Follesaand Ticku,1995; Hu et al., 1996). Theseobservationsled us tothe intriguing questionas to which mechanism(s)un-derliesthedifferential regulationof the NMDA recep-tor subunitswhenprimary culturesof cortical neuronsare exposedto chronic ethanol treatment.

In eukaryotic cells, gene expressioncan be modu-latedby changesin the rateof genetranscription,post-transcriptionalevents,andlorthe rateof protein trans-lation, and eachstepof this complex processof geneexpressionis tightly regulated.Control of genetran-scription was thoughtto be the major mechanismin-fluencingregulationof geneexpression;however,re-centresearchhas provided ampleevidencethat post-transcriptionalfactors such as mRNA stability alsoplay an importantrole in regulatinggeneexpressionin a cell- andtissue-specificmanner(CarterandMalter,1991;Sachs,1993;Pain,1996).Thegoal of thepresentstudywasto delineatewhich oneof the abovemolecu-lar mechanismsunderliesthedifferential regulationofthe NMDA receptorsubunits,namely, RI and R2B,after chronic ethanol treatment.The rationalefor ex-aminingthesetwo subunits is basedon our previousstudieswherechronic ethanoltreatmentincreasedRIand R2B polypeptide levels with a concomitantin-creasein only R2B mRNA levels. Similar treatmenthad no effect on the R2A mRNA levels (Hu et al.,1996) and R2A protein levels in 7-day-old corticalneurons(Follesa and Ticku, 1996). We havedeter-mined the mRNA stability and the rate of genetran-scription for theRi andR2B receptorsubunits,usingour in vitro model systemof mousefetalcortical neu-rons. Our results demonstratefor the first time thatchronic ethanol treatmentstabilizes the NMDA R!subunit mRNA with a simultaneousincreasein therateof R2B genetranscriptionin mousefetal corticalneurons.

MATERIALS AND METHODS

DNA clones and probesEcoRI-linearizedplasmidspFPR 1, pFPR2BR, andp15G!

weretranscribedin vitro in thepresenceof [a32P]CTP(spec.act. = 3,000 Ci/mmol) (Du Pont,Boston, MA, U.S.A.) togenerateRl, R2B, and cyclophilin-radiolabeledcRNAs,re-spectively,asdescribedelsewhere(FollesaandTicku, 1995).Levels of RI and R2B mRNA in actinomycin D-treatedsampleswere quantifiedby ribonucleaseprotectionassays.CyclophilineRNA wasusedto normalizethe results.

The NMDA R! andR2B nonradiolabeledcRNAs,corre-sponding to 221—1,306 bp (accessionno. U11418) and2,746—3,184bp (Monyeret al., 1992), respectively,as wellascyclophilin eRNAandpAMPI (GibcoBRL,GrandIsland,

NY, U.S.A.) cRNAs were immobilized onto GeneScreenPlus membrane(Du Pont) for nuclearrun-onanalysis.

Cell culture and ethanol treatmentCortical neuronswereisolatedfrom mousefetusesat 14—

15 days post coitum andcultured as describedpreviously(FollesaandTicku, 1996; Hu et al., 1996). Timed-pregnantmice(strainC57BL/6)purchasedfrom Harlan(Indianapolis,IN, U.S.A.) wereusedin accordancewith institutionalguide-lines, andprocedureswere approvedby the animal welfarecommittee. In brief, cerebralhemispheresdissectedfromfetusesundersterileconditionswere mincedandcells weredissociatedby trituration using aPasteurpipette.Dissociatedcells were countedunder an inverted microscopeusing ahemocytometerand their viability wasdeterminedby thetrypanblueexclusiontest. A known numberof viable cells(2.4 X l0~cells/75-cm2flask) were plated in poly-L-lysine-coatedplastic flasks andgrown in minimum essentialme-dium containing100 ‚aM L-glutamine,10% fetal bovine se-rum, and 10% heat-inactivatedhorseserumunder95% G,and5% CO

2 at 37°C.On day 2 of culture, a mixtureof 5-fluoro-2‘-deoxyuridineanduridine at a final concentrationof 10 ‚ag/ml wasaddedto themediumto inhibit cell prolifer-ation (Godfreyet al., 1975). Such culturescontained>92%neuronalcells,and the remainderwere glial cells, whichdonot expressNMDA receptors(Zhonget al., 1994).

From day 3 on, cortical neuronalcells were grown ineither the absenceor thepresenceof 50 mM ethanolfor 5consecutivedays, with a changeof mediumevery24 h. Onday 7, cells were usedeither for the isolation of RNA afterincubationwith actinomycinD or the isolation of nuclei.

Determination of RNA stabilityRNA stability of the NMDA RI andR2B receptorsub-

units was assayedby incubationof cortical neuronalcellswith actinomycinD to inhibit transcription.After 5 daysofethanoltreatment,cortical neuronswere fed with mediumcontainingactinomycinD (preparedin dimethyl sulfoxide)to a final concentrationof 5 ‚ag/ml. At the designatedtimepoints, mediumwasremovedandcells werelyseddirectlyin guanidiniumthiocyanatehomogenizationbuffer (Chirg-win et al., 1979).Total cellular RNA wasextractedat eachof the time points by centrifugationover a cushionof 2 mlof 5.7 M CsCl and0.1 M EDTA at 35,000rpm for 18 h at18°Cin a Beckmanultracentrifugeusingan SW 55Ti rotor(Chirgwin et al., 1979).

Equal amountsof total RNA at each of the time pointswere analyzedby the ribonucleaseprotectionassayas de-scribedearlier(FollesaandTicku, 1995; Hu et al., 1996).In brief, 25 ‚ag of total RNA was suspendedin 20 ‚al ofhybridization buffer (Ambion, Austin, TX, U.S.A.) con-taining 150,000 cpm of

32P-labeledNMDA RI and R2BcRNA probes and 15,000 cpm of 32P-labeledcyclophilineRNA probe, andincubatedat 50°Cfor 16—18 h. On thefollowing day,sampleswereRNasedigested,andrecoveredRNA—RNA hybrids were separatedon a5%sequencinggelunderdenaturingconditions.Gels were dried, exposedto aPhosphorimagerscreenas well as to x-ray film, and datawere analyzedasdescribedbelow.

Isolation of nucleiNuclei from cultured cortical neuronswere isolatedas

describedelsewhere(Kumari et al., 1996). In brief, corticalneuronsgrownin thepresenceor absenceof 50 mM ethanolfor 5 consecutivedays were washedtwice with ice-cold

J. Neurochem.,Vol. 70. No. 4, /998

REGULATION OF THE NMDA GENE EXPRESSION 1469

phosphate-bufferedsalineandharvestedwith the help of acellscraper.Plasmamembranesweredisruptedby resuspen-sion in 0.5% Triton X-100preparedin the nuclearisolationbuffer (NIB) [10 mM Tris-HCI (pH 7.4), 10 mM NaCI, 5mM MgC12, and 1 mM dithiothreitol]. Cell lysateswerecentrifugedat 2,000 rpm at 0°Cfor 10 min in a GS-6RBeckmantable-topcentrifuge.Supernatantswere discardedandnuclearpelletswere washedtwice with NIB. Washednucleiwere resuspendedin glycerol storagebuffer [50 mMTris-HC1(pH8.3),40%glycerol,5 mMMgC12,5 mMdithio-threitol, and0.1 mM EDTA] (108 nuclei/ml) andstoredasaliquotsat —70°Cuntil use.

Nuclear run-on transcription assayThe procedureof Kumari et al. (1996) was used with

the following modifications:(1) thesubstitutionof [32P]UTP

insteadof [33P]UTP,(2) isolationof newlytranscribedRNAby ultracentrifugation(Chirgwin et al., 1979), and(3) theuseof eRNAsequencesasprobesto preparedotblots.Nucleiisolatedfrom cortical neuronscultured in the absenceofethanolservedas an untreatedcontrol.

Nuclei were thawedon ice just beforeuseand incubatedat 30°Cfor 30 min to reinitiatetranscriptionin thepresenceof 250 ‚aCi [a-32P]UTP(spec.act. = 3,000 Ci/mmol) (DuPont) asdescribed(Kumari et al., 1996). The reactionwasstoppedby theadditionof 2.7 ml of 5 M guanidiniumthiocy-anatehomogenizationbuffer(Chirgwinet aI., 1979)andthesuspensionswere sonicatedfor 30 s. 32P-RNA waspelletedby ultracentrifugationas described(Chirgwin et al., 1979).Each32P-RNApelletwasdissolvedin 0.5 ml diethyl pyrocar-bonate-treatedwater and mixed with an equal volume of2X hybridizationbuffer [IX hybridizationbuffer = 10 mil‘!N-tris(hydroxymethyl)methyl-2-aminoethanesulfonicacid(TES) (pH 7.5), 1% sodiumdodecylsulfate (SDS), 10 mIt!EDTA, 250‚ag/ml E. coli tRNA, 0.3 M NaC1,lx Denhardt‘sbuffer (0.02% polyvinylpyrrolidone,0.02% bovine serumalbumin,and0.02% Ficoll), 10% dextransulfate, and 100‚ag/ml herring spermDNA]. Equal amountsof heat-dena-turedradioactiveRNA (6 X I o~cpmlml) wereusedto probedot blotscarryingmouseNMDA R! andNMDA R2B eRNAsequencesspotted onto Gene ScreenPlus membranesinquantities of 2, 0.2, and0.02 ‚ag. pAMP1 plasmid sequencewas used to monitor nonspecificbinding and cyclophilin(PIB15)wasusedasan internalcontrol.Hybridizationswereperformed at 65°Cfor 48—52 h in heat-sealablepouches.Membraneswere washedin 2X SSC (lx SSC = 0.15 MNaC1 and0.015 M sodiumcitrate, pH 7) and 1% SDS atroom temperature2 X 15 min and0.2x SSC,0.1% SDS at65°Cfor 60 mm, and hybridization signalswereanalyzedon a Phosphorlmager(Molecular Dynamics, Sunnyvale,CA, U.S.A.).

Quantitation of autoradiogramsTheNMDA Ri andR2B subunitmRNAs andcyclophilin

mRNA contentwere quantifiedon aPhosphorlmagerusingthe ImageQuantprogram.To control for equalsampleload-ing, the signal density of NMDA RI andR2B mRNA werenormalizedby dividing the signal density of RI - andR2B-protectedfragmentby the signal density of cyclophilin-pro-tected fragmentfrom the samesample,and the amount ofmRNA wasexpressedin arbitrary units. Resultswere ex-pressedas percentagesof control andplotted on a logarith-mie scalevs. time. Datapoints aremean ±SE valuesofthreeseparateexperiments.The half-livesof R! andR2Bwere determinedfrom the slopeof the line.

For nuclear run-on analyses,the density of the R! orR2B hybridizationsignalswasdivided by the density of thecyclophilin hybridization signal. Values obtainedfor un-treatedcontrols were comparedwith valuesof the corre-sponding subunits of ethanol-treatedsamplesto determinechangesin the rate of RI and R2B genetranscription.

RESULTS

Effect of ethanol on mRNA stabilityRegulationof steady-statelevels of mRNA mayoc-

curthroughregulationof mRNA stability (Ross,1995).Previousreportsfrom our laboratoryhaveshownthatchronic ethanoltreatmentof fetal corticalneuronsre-sultsin anincreasein NMDA receptornumber(Hu andTicku, 1995).This up-regulationof receptornumberisaccompaniedby a selectiveincreasein NMDA R2BmRNA levels (Hu et al., 1996). To explorethe mecha-nism(s)by which chronic ethanol treatmentup-regu-latestheNMDA receptornumber,wehavedeterminedthe mRNA stability of the NMDA Ri and R2B sub-units by using mousefetal cortical neuronsas an invitro model system.Neuronsculturedin the presenceor absenceof 50 mM ethanol for 5 consecutivedayswere treatedwith actinomycinD (5 ‚ag/ml) to blockfurther transcription.The cultures were subsequentlyterminatedat different time intervals.Total RNA ex-tractedfrom eachof the time points wasanalyzedforthe RI andR2B subunitmRNAsby ribonucleasepro-tection assays.Cyclophilin was used as a control tonormalizethe results(FollesaandTicku, 1995;Hu etal., 1996). The half-lives of Ri and R2B mRNAs,determinedfrom the rate of their mRNA degradationin the presenceof actinomycinD, reflectsthe stabilityof the respectivemRNAs.

Figure lA depictsthe levels of NMDA Ri andR2BmRNAs in cortical neuronsgrown in the absenceofethanol.The hybridization signals of Ri- and R2B-protectedfragmentswere quantified as describedinMaterials and Methods. The signal density of eachprotectedfragmentbandrepresentingRi andR2B sub-units was divided by the signal density of protectedfragment representingcyclophilin to normalize forequalsampleloading.Linear regressionanalysisof thehybridization signals (ratio of Ri/cyclophilin; R2B/cyclophilin) indicatedthat thehalf-life of theNMDARi subunitmRNA was 16 hand thatof theR2B sub-unit mRNA 15.7 h (Fig. lB and C). After chronicethanoltreatment(50 mM, 5 days)thehalf-life of theNMDA R2B subunit mRNA was not altered (t

112= 15.1 h) (Fig. 2A andB). In contrast,theNMDA RIsubunitmRNA did not showany sign of decayduringthetime periodexamined(Fig. 2B). For theRi subunit,incubationtime with actinomycin D was extendedto24 h. No detectabledecayof the NMDA Ri subunitmRNA was observedafter 24 h of incubation withactinomycin D (datanot shown). Theseobservationssuggestthat ethanol selectively stabilizedtheNMDARi subunitmRNA up to thetime periodexamined(24h) andhad no effecton the decayrate of theNMDA

J. Neurochem.,Vol. 70, No. 4, 1998

1470 M. KUMARI AND M. K. TICKU

FIG. 1. Determination of the NMDA Ri and R2B receptor subunit mRNA stability in mouse cortical neurons cultured in the absenceof ethanol. Cortical neurons were cultured in the absence of ethanol for 5 days. The next day, actinomycin D (5 ‚ag!ml) was added tothe medium and cells were incubated further for the indicated times. Total RNA was isolated and the NMDA Ri and R2B mRNA5 werequantified by ribonuclease protection assay. A: A gel autoradiogram showing protected fragments identified by name on the right. Ri,NMDA Ri subunit; R2B, NMDA R2B; P1B15, cyclophilin. Lane M, end-labeled markers, 4X174 Hinfl digested (numbers indicatenucleotides); lane P, undigested probe aliquot of the hybridization solution containing the cRNA probe for Ri (189 bases), R2B (530bases), and cyclophilin (301 bases); lane i, O h; lane 2, 6 h; lane 3, 8 h; lane 4, 10 h; lane 5, 12 h; lane 6, 14 h; and lane 7, 16 h.Hybridization signals of Ri (B) and R2B (C) were quantified on a Phosphorlmager as described in Materials and Methods. Results areexpressed as percentages of control and plotted on a logarithmic scale vs. time. Data points are mean ±SE values of three separateexperiments. The half-lives of Ri and R2B mRNA were determined from the slope of the line.

R2B subunit mRNA. For mRNA stability assays,cyclophilin was used tonormalizeRi and R2Bsignals(Figs. JA and 2A);however,we noted that ethanolhad an mRNA-stabilizing effecton the cyclophilinmessage. -

Ethanol and the rate of gene transcriptionThe lack of ethanol‘seffect on the stability of the

NMDA R2B subunit mRNAin the presentstudy(Figs.1 and 2) implies thatup-regulationof the R2B subunitmRNA after chronic ethanoltreatmentin vitro (Hu etal., 1996)may be aresultof direct effectof ethanolonR2B mRNA synthesis. Toinvestigatethis possibility,nuclearrun-on analysiswasperformedon nuclei iso-lated from cortical neuronsgrown in thepresenceorabsenceof 50 mA‘! ethanol for 5 days. RadiolabelednascentRNA generatedby incubationof nuclei intran-scription buffer werehybridizedwith the NMDA RiandR2B cRNAs blotted ontoGeneScreenPlusmem-brane. Cyclophilin and vector pAMP1 cRNAs wereincluded as positive and negativecontrols, respec-tively. Results shown in Fig.3 indicatedthat ethanolproduced(1) a small(25%) but statisticallysignificantincreasein the rate of NMDA Ri genetranscriptionand (2) anapproximately twofold increasein the rateof NMDA R2B genetranscriptionrelativeto theun-treatedcontrols.

DISCUSSION

In thepresentstudy,we attemptedto reveala molec-ular explanationfor ethanol-inducedup-regulationof

NMDA receptornumberin culturedmouse fetalcorti-cal neurons(Hu and Ticku, 1995). Thesecells wereselectedfor thepresentstudybecause theyareknownto respond to chronic ethanol treatmentin the samemanneras cortical neuronsof intact animals(Follesaand Ticku, 1995; Hu et al., 1996). During week 1 inculture, cortical neuronsexpress NMDARi andR2Bsubunits andvery low levelsof the NMDA R2A sub-unit. On thecontrary, expressionof the R2C subunitbeginsonly on day2i in culture(Zhong et al., 1994).Hence, the presentinvestigation examinesthe effectof ethanol on the regulation of the NMDA Ri andR2B subunitexpression atthetranscriptional andpost-transcriptionallevels.

Resultsof thepresent study indicatedthat thehalf-life of theNMDA Ri subunit mRNAwas16h andthatof theR2B subunitmRNA 15.7 h in cortical neuronsculturedin theabsenceof ethanol.Exposureto chronicethanoltreatmentresultedin completestabilization ofthe NMDA RI mRNA up to the time period (24 h)examined.Wewereunabletodeterminethe exact half-life of theNMDA Ri mRNA in thepresenceof etha-nol, becausetreatmentof cells with actinomycinD forextendedperiods of time leadsto generalcytotoxicity(Scottand Tomkins,1975).Thehalf-life of theNMDAR2B mRNA afterchronic ethanoltreatment(t

112 = 15.1h) wasnot significantly differentfrom thatof untreatedcontrols (t112 = 15.7 h), suggestingthat the effect ofchronicethanol treatmenton themRNA stability wasRi subanitspecific.

Severalreportshaveshownthat a diversegroupof

J. Neurochem.,Vol. 70, No. 4, 1998

REGULATIONOF THE NMDA GENE EXPRESSION 1471

and Deeley,1988; Carterand Malter, 1991;Busabmet al., 1993; Crowe, 1993;Lopez-Barahonaet al.,1993). For instance,estrogenstabilizes vitellogeninmRNA in primary culturesof Xenopushepatocytes.Inthese cells, thehalf-life of vitellogeninmRNA is seento extendfrom i6 h in untreatedcontrols to‘=3 weeksafter estrogentreatment(Brock and Shapiro, 1983).Szabo et al. (1995)haveshown that acute ethanoltreat-mentof monocytesin vitro resultedin destabilizationof interleukin-IbmRNA. To ourknowledge,the pres-ent studydemonstratesfor the first time that chronicethanol treatmenthad aselectivestabilizingeffect onthe NMDA R! receptor subunitmRNA, therebyinflu-encinggene expressionat theposttranscriptionallevel.

Earlier reportsfrom our laboratory have demon-strated anincreasein the R2B subunitmRNA and

FIG. 2. Determination of the NMDA Ri and R2B receptor sub-unit mRNA stability in ethanol-treated mouse cortical neurons.Cortical neurons were cultured in the presence of 50 mMethanolfor 5 days. The next day, actinomycin D (5 jig/mI) was added tothe medium. Cells were incubated further for the indicated times.Total RNA was isolated and the NMDA Ri and R2B subunitmRNA levels were quantified by ribonuclease protection assay.A: A gel autoradiogram showing protected fragments identifiedby name on the right. Ri, NMDA Ri subunit; R2B, NMDA R2B;PiBi5, cyclophilin. Lane M, end-labeled markers, 4iX174 Hbf Idigested (numbers indicate nucleotides); lane 1, 0 h; lane 2, 4h; lane 3, 6 h; lane 4, 8 h; lane 5, 10 h; and lane 6, i2 h. B:Hybridization signals were quantified as described in Materialsand Methods, and results expressed as percentages of controlare plotted on a logarithmic scale vs. time. Data points are mean±SE values of three separate experiments. The half-life of R2BmRNA was determined from the slope of the line. Ri subunit,O; R2B subunit, S.

factors such asserum,hormones, cytokines,phorbolester, and ionsalter the mRNA stability in a cell- andtissue-specificmanner(Wiskocil et al., 1980; Brockand Shapiro, 1983; Paekand Axel, 1987; Cochrane

FIG. 3. Nuclear run-on transcription assay to determine the tran-scription rates of NMDA Ri and R2B genes in mouse fetal corti-cal neurons. A: Nuclei were isolated from neurons with or withoutchronic ethanol treatment and run-on transcripts were labeledwith [32P]UTPas described in Materials and Methods. Radiola-beled nuclear transcripts were hybridized to dot blots carrying2, 0.2, and 0.02 jig quantities of unlabeled cRNA probes asindicated on the left. P1B15, cyclophilin; Ri, NMDA Ri subunit;R2B, NMDAR2B subunit; and pAMP1,vector sequence includedto monitor nonspecific binding. B: Using the ImageQuant pro-gram (Molecular Dynamics), the signal density of NMDA Ri (Ri)and NMDA R2B (R2B) bands were quantitated and normalizedto that of cyclophilin to correct for differences in hybridizationconditions. The ratio in the ethanol-treated lane was divided bythe corresponding value in the vehicle-treated lane to calculatethe percentage change in the transcriptional rate of the NMDARi and R2B genes in response to chronic ethanol treatment.The data are expressed as percentages of control (mean ±SEof three separate experiments). *p < 0.05, compared with controlgroup (analysis of variance and Scheffé‘s test).

J. Neurochem..Vol. 70. No. 4. /998

1472 M. KUMARI AND M. K TICKU

polypeptide levels in cultured fetal cortical neuronsafter chronic ethanol treatment(Follesa and Ticku,1996;Hu et al., i996). The lack of ethanol‘seffect ontheR2B mRNA half-life in thepresentstudysuggestedthatposttranscriptionalmechanismsdid notplay a rolein the up-regulationof the R2B mRNA. It is likelythatmechanismsoperatingat the nuclearlevel maybemore importantto explaina significant increasein theNMDA R2B subunit mRNA levels. To test this hy-pothesis,we performednuclearrun-onanalysison nu-clei isolatedfrom cortical neuronsculturedin thepres-enceof ethanol.Nuclei isolated from neuronsgrownin the absenceof ethanolservedas untreatedcontrol.Ourobservationsindicatedthat chronic ethanoltreat-ment produceda twofold increasein the rate of theR2B genetranscriptioncomparedwith the untreatedcontrol. Theseresults suggestthat the NMDA R2BsubunitmRNA genetranscriptionand not mRNA sta-bility mediated up-regulation of the R2B subunitmRNA levels in cortical neuronsafterchronicethanoltreatment.Although ethanol (50 mM, 5 days) showedatendencyto increasetherateof Ri genetranscription,the majoreffect of ethanolon the NMDA RI subunitappearsto be the stabilizationof NMDA Rl message.

Ethanol‘seffect on the rateof genetranscriptionisnot anisolatedobservation.Chronic ethanol treatmenthas been shown to increasethe rate of cytochromeP450CYP 2E1 genetranscriptionin rat liver in vivo(Badgeret al., 1993) and tyrosine hydroxylasegeneexpressionin NIE-I 15 neuroblastomacells in vitro(Gayeret al., 1991). However, it is noteworthy thatin the case of the NMDA receptor,chronic ethanoltreatmentmediatesits actionvia two distinctregulatorymechanisms,one,operatingat the transcriptionallevelfor the NMDA R2B subunit,anda second,operatingmainly at the posttranscriptionallevel for the NMDAR! subunit, with a net effect on the up-regulationofNMDA receptornumberand their function reportedearlier (Hu andTicku, 1995). Our resultsalso indicatethat ethanol appearsto stabilize cyclophilin mRNA.Future experimentsthat delineate the exact mecha-nisms by which ethanolbrings aboutsuch varied ef-fects on different subunits of the samereceptorwillhelp us understandthe involvementof the NMDA re-ceptor in ethanol-mediatedalcohol dependenceandwithdrawal syndrome.

Acknowledgment: We thank Antje Anji for help withdata analysis, Bradley Lega for technical assistance,andMrs. SadiePhillips for excellentsecretarialhelp.Thanksarealso due to Dr. John R. McCarrey for critical evaluationof the manuscript.This study was supportedby NationalInstitutesof Health—NIAAA grant AA10552.

REFERENCES

BadgerT.M.,HuangJ.,RonisM.,andLumpkinC.K.(l993)Induc-tion of cytochromeP450 2E1 duringchronic ethanolexposureoccursvia transcriptionof theCYP 2E1 genewhen blood alco-

hoI concentrationsare high. Biochem.Biophys.Res. Commun.190, 780—785.

BrockM. L. andShapiroD. J. (1983)EstrogenstabilizesvitellogeninmRNA againstcytoplasmicdegradation.Cell 34, 207—214.

BusabmK. J., GeiserA. G., RobertsA. B., and SpornM. B. (1993)Synergistic increaseof phorbol ester-inducedc-fos mRNA ex-pressionby retinoic acid throughstabilizationof thec-fosmes-sage. Oncogene8, 2267—2273.

CarterB. Z. and Malter J. S. (1991) Biology of disease:regulationof mRNA stability andits relevanceto disease.Lab. Invest. 65,610—621.

Chirgwin J.M., Przybyla A. E., MacDonaldR. J., and RutterW. J.(1979) Isolation of biologically active ribonucleicacid fromsourcesenrichedin ribonuclease.Biochemistry18, 5294—5299.

Choi D.W. (1988)Glutamateneurotoxicityanddiseasesof thener-vous system.Neuron 1, 623—634.

CochraneA. W. and DeeleyR. G. (1988) Estrogen-dependentacti-vation oftheavianvery low density apolipoproteinII andvitel-logenin genes.Transientalterationsin mRNA polyadenylationand stability early during induction. J. Mol. Biol. 203,555—567.

CroweD. L. (1993)Retinoicacidmediatespost-transcriptionalregu-lation of keratin 19 mRNA levels. J. Cell Sci. 106, 183—188.

FollesaP. andTicku M. K. (1995) Chronicethanoltreatmentdiffer-entially regulatesNMDA receptorsubunit mRNA expressionin rat brain. Mol. Brain Res. 29, 99—106.

Follesa P. and Ticku M. K. (1996) Chronic ethanol-mediatedup-regulation of the N-methyl-D-aspartatereceptor polypeptidesubunits in mousecortical neurons in culture. J. Biol. Chem.271, 13297—13299.

GayerG. G., GordonA., andMiles M. F. (1991)Ethanolincreasestyrosine hydroxylase gene expression in N1E-115 neuro-blastomacells.J. Biol. Chem.266,22279—22284.

GodfreyE. W., Nelson P. G., ShrierB. K., BreuerA. C., andRan-som B. R. (1975) Neuronsfrom fetal rat brain in a new cellculturesystem: a multidisciplinaryanalysis.Brain Res.90, I —

21.Grant K. A., Valverius P., Hudspith M., and Tabakoff B. (1990)

EthanolwithdrawalseizuresandtheNMDA receptorcomplex.Eur. J. Pharmacol.176, 289—296.

Hoffman P.L. (1995) Effects of alcoholon excitatoryamino acidreceptorfunction, in HandbookofExperimentalPharmacology,Vol. 114 (Kranzler H. R., cd), pp. 75—160. Springer Verlag,Berlin.

Hu X. J. andTicku M. K. (1995)Chronicethanoltreatmentupregu-lates the NMDA receptorfunction and binding in mammaliancorticalneurons.Mol. Brain Res.30, 347—356.

Hu X. J., Follesa P., andTicku M. K. (1996)Chronic ethanoltreat-ment producesa selectiveupregulationof theNMDA receptorsubunitgene expressionin mammaliancultured corticalneu-rons. Mol. Brain Res.36, 211—218.

Kumari M., StroudJ. C., Anji A., andMcCarreyJ. R. (1996)Differ-ential appearanceof DNase I-hypersensitivesites correlateswith differentialtranscriptionofPgk genesduringspermatogen-esis in the mouse.J. Biol. Che,n.271, 14390—14397.

KutsuwadaT., KashiwabuchiN., Mon H., SakimuraK., KushiyaE., Araki K., Meguro H., Masaki H., Kumanishi T., ArakawaM., andMishina M. (1992) Moleculardiversity of the NMDAreceptorchannel.Nature 358, 36—41.

Lopez-BarahonaM., MinanoM., Mira E., IglesiasT., StunnenbergH. G., Rodriguez-PenaA., Bernai J., and Munoz A. (1993)Retinoicacid posttranscriptionallyup-regulatesproteolipidpro-tein gene expressionin C6 glioma cells. J. Biol. Chem. 268,25617—25623.

LovingerD. M., White G., andWeightF. F. (1989)Ethanol inhibitsNMDA-activatedion currentin hippocampalneurons.Science243, 1721—1724.

Mehta A. K. and Ticku M. K. (1995) Effect of alcoholon GABA-mediated neurotransmission,in Handbook of ExperimentalPharmacology, Vol. 144 (Kranzler H. R., cd), pp. 103—119.SpringerVerlag,Berlin.

MirshahiT.andWoodwardJ. J. (1995)Ethanolsensitivityofhetero-

J. Neurochem.,Vol. 70, No. 4, /998

REGULATIONOF THE NMDA GENE EXPRESSION 1473

meric NMDA receptors:effects of subunitassembly,glycineand NMDAR 1 Mg

2tinsensitivemutants.Neuropharmacology34, 347—355.

MonyerH., SprengelR., SchoepferR., HerbA., Higuchi M., LomeliH., BurnashevN., SakmannB., andSeeburgP. H. (1992)Heter-omeric NMDA receptors:molecularandfunctional distinctionof subtypes.Science256, 1217—1221.

Paek I. and Axel R. (1987) Glucocorticoids enhancestability ofhumangrowthhormonemRNA. Mol. Cell. Bio!. 7, 1496—1507.

Pain V. M. (1996)Initiation ofprotein synthesisin eukaryoticcells.Eur. J. Biochem.236,747—77 I.

PniestleyT., LaughtonP., Myers J., Le BourdellésB., KerbyJ., andWhitingP. J. (1995)Pharmacologicalpropertiesof recombinanthumanN-methyl-D-aspartatereceptorscomprisingNR1aJNR2AandNRla!NR2B subunitassembliesexpressedin permanentlytransfectedmousefibrobiast cells. Mol. Pharmacol.48, 841—

848.RossJ. (1995)mRNA stability in mammaliancells.Microbiol. Rev.

59, 423—450.SachsA. B. (1993)MessengerRNA degradationin eukaryotes.Cell

74, 413—421.SannaE., SerraM., CossuA., ColomboG., FollesaP., and Biggio

G. (1992)GABAA andNMDA receptorfunctionduringchronic

administrationof ethanol,in Advancesin BiochemicalPsycho-pharmacology,Vol. 47, pp. 3 17—324. RavenPress,NewYork.

Scott W. A. andTomkinsG. M. (1975)The useof inhibitors in thestudyof hormonemechanismsin cell culture.MethodsEnzymol.40, 273—293.

Simon R. P., SwanJ.H., Gniffiths T., and Meldrum B. 5. (1984)BlockadeofN-methyl-n-aspartatereceptorsmayprotectagainstischemicdamagein the brain.Science226, 850—852.

SzaboG., MandrekarP., andCatalanoD. (1995) Inhibition of super-antigen-inducedT cell proliferationand monocyteIL-1 beta,TNF-alpha,and IL-6 productionby acuteethanoltreatment.J.LeukocyteBio!. 58, 342—350.

Wiskocil R., BenskyP., DowerW., GoldbergerR. F., GordonJ. I.,andDeeleyR.G. (1980)Coordinateregulationof two estrogen-dependentgenesin avian liver. Proc. Nat!. Acad. Sd. USA 77,4474—4478.

ZhongJ.,RussellS. L., PnitchettD. B., Molinoff P. B., andWilliamsK. (1994) Expressionof mRNAs encodingsubunits of the N-methyl-o-aspartatereceptorin cultured cortical neurons.Mol.Pharmacol.45, 846—853.

Zukin R. S. and BennettM. V. L. (1995) Altematively spliced iso-formsof theNMDARI receptorsubunit.TrendsNeurosci. 18,306—3 13.

J. Neurochem.,Vol. 70, No. 4, 1998