Proc. Nadl. Acad. Sci. USAVol. 89, pp. 7189-7193, August 1992Neurobiology

Cloning, expression, and localization of a rat brain high-affinityglycine transporterJOHN GUASTELLA*, NICHOLAS BRECHAt, CHRISTINE WEIGMANNt, HENRY A. LESTER*,AND NORMAN DAVIDSON**Division of Biology, 156-29, California Institute of Technology, Pasadena, CA 91125; and tDepartments of Anatomy and Cell Biology and Medicine, BrainResearch Institute and Center for Ulcer Research, University of California, Los Angeles School of Medicine and Veterans Administration MedicalCenter-Wadsworth, Los Angeles, CA 90073

Contributed by Norman Davidson, May 6, 1992

ABSTRACT A cDNA clone encoding a glycine transporterhas been isolated from rat brain by a combined PCR andplaque-hybridization strategy. mRNA synthesized from thisclone (designated GLYT1) directs the expression of sodium-and chloride-dependent, high-affinity uptake of [3H]glycine byXenopus oocytes. [3H]Glycine transport mediated by cloneGLYT1 is blocked by sarcosine but is not blocked by methyl-aminoisobutyric acid or L-alanine, a substrate specificity sim-ilar to that described for a previously identified glycine-uptakesystem called system Gly. In situ hybridization reveals thatGLYT1 is prominently expressed in the cervical spinal cord andbrainstem, two regions of the central nervous system whereglycine is a putative neurotransmitter. GLYT1 is also stronglyexpressed in the cerebellum and olfactory bulb and is expressedat lower levels in other brain regions. The open reading frameof the GLYT1 cDNA predicts a protein containing 633 aminoacids with a molecular mass of =70 kDa. The primary struc-ture and hydropathicity profile of GLYT1 protein reveal thatthis protein is a member of the sodium- and chloride-dependentsuperfamily of transporters that utilize neurotransmitters andrelated substances as substrates.

Termination of synaptic activity is thought to occur throughremoval of neurotransmitter from the synaptic cleft by ion-coupled, high-affinity neurotransmitter transport proteins(neurotransmitter transporters) located in neuronal and glialplasma membranes. Drugs that block the action of thesetransporters can modulate neural function, probably by in-creasing the duration of neurotransmitter action. Some neu-rotransmitter transporters, such as those for the monoam-ines, are the sites of action for clinically important drugs (forexample, antidepressants) (1), as well as drugs of abuse (forexample, cocaine) (2). Other neurotransmitter transporters,such as those for the inhibitory amino acid y-aminobutyricacid (GABA), are potential targets for drugs with anticon-vulsant properties (3). Therefore, an understanding of thestructure and function of neitrotransmitter transporters maylead to a better understanding of synaptic regulation andcould also provide a rational basis for the development ofadditional, more-specific anti-transporter drugs.The recent isolation ofcDNA clones encoding transporters

for GABA (4), the monoamines norepinephrine, dopamine,and serotonin (5-10), and a nonneurotransmitter, betaine(11), has revealed that these carriers comprise a superfamilyof homologous proteins. Predictions of membrane topologysuggest that these transporters all contain =12 membrane-spanning domains, a structural motif shared by a number ofcarriers that, however, share little mutual sequence homol-ogy. The amino acid-sequence identity between the knownmembers of the neurotransmitter transporter superfamily

ranges from "-44% for the GABA transporter (GAT1) versusthe norepinephrine transporter (NET1) to =64% for NET1versus the dopamine transporter (DAT1), indicating thatsubfamilies exist within the superfamily. In addition to anoverall homology, members of this superfamily contain sev-eral regions of sequence identity distributed throughout thelength of the protein. These regions can be used to designPCR primers which can be employed to clone additionalmembers of the superfamily from brain and other tissues. Inthis paper we report the results of such a cloning approach,which has yielded a rat brain cDNAt encoding a sodium- andchloride-dependent high-affinity glycine transporter.

MATERIALS AND METHODSRNA Isolation and PCR. Total RNA was prepared from C6

glioma cells and rat tissues by the acid guanidinium/phenolmethod (12). cDNA was synthesized directly from 5 Mg oftotal RNA or mRNA and amplified by the PCR and standardprotocols. The primers chosen for amplification consisted ofa 128-fold degenerate sense primer corresponding to bases375-394 ofGAT1 with the sequence (5'-- 3') GAG CTC GTCGAC AAG AAC GG(TCAG) GG(TCAG) GG(TCAG)GC(TC) TT and a 48-fold degenerate antisense primer cor-responding to bases 1108-1030 ofGAT1 with the sequence (5'-* 3') GGG CCC TCT AGA AA (GA)AT CTG (GA)GT(GAT)GC (GA)GC (AG)TC. The upstream primer corre-sponds in the GAT1 protein to amino acids 76-82 with thesequence KNGGGAF, and the downstream primer corre-sponds to amino acids 287-293 with the sequence DAATQIF.With these primers the PCR will generate a 677-base-pair (bp)product from GAT1 mRNA and a 710-bp product from NET1mRNA. The amplification protocol consisted of a 1-mindenaturation at 94°C, a 4-min annealing at 55°C, and a 2-minextension at 72°C for 25-30 cycles on a Perkin-Elmer DNAthermal cycler. A sample of the amplification reaction wasanalyzed by agarose gel electrophoresis, and the appropriateband was gel-purified and subcloned into pBluescript SK(-).RNA Blot. Two micrograms of mRNA was subjected to

agarose gel/formaldehyde electrophoresis and transferred tofilter paper. The blot was hybridized overnight at 42°C underhigh-stringency conditions [50%o formamide/6x standard sa-line phosphate/EDTA (SSPE; lx SSPE = 0.15 M NaCI/10mM phosphate, pH 7.4/1 mM EDTA)/5x Denhardt's solu-tion/salmon sperm DNA at 100 tg/ml]. The final stringencywash was done in 0.2x standard saline citrate at 65°C for 15min. The filter was exposed to film for 2-4 days with twointensifying screens at -70°C.cDNA Isolation and Sequencing. A AZAP rat brain cDNA

library (13) containing inserts in the 2-4 kilobase-pair (kbp)

Abbreviation: GABA, y--aminobutyric acid.tThe sequence reported in this paper has been deposited in theGenBank data base (accession no. M95413).

7189

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

May

16,

202

1

7190 Neurobiology: Guastella et al.

A B1 2 3 4 5

M_40..

FIG. 1. (A) Ethidium bromide-stained agarose gel of the DNAproducts generated with described primers and C6 glioma cell cDNAas template. Arrow denotes a major product of -700 bp. Lanes: 1 and2, products with C6 glioma cell mRNA as starting material; 3 and 4,products with C6 glioma cell total RNA; 2 and 4, products with a2-fold dilution of the respective cDNA mixtures. (B) RNA blot ofmRNA from C6 glioma cells (lane 1), rat liver and kidney (lanes 2 and3), and rat forebrain and cerebellum (lanes 4 and 5) probed with the32P-labeled 700-bp PCR product from C6 glioma cells. Arrow denotesa single transcript of -3.3 kb. A faint band, visible in lane 2 (livermRNA) of the original autoradiogram, is difficult to see in thisphotograph.

size range was screened with the same random primer-labeled PCR product used to probe the RNA blot and underthe same hybridization conditions. Six positive clones wereidentified. The phagemids were rescued from each clone andused for further analysis. Dideoxynucleotide DNA sequenc-ing was done by a combination of nested deletions and primerwalks.Xenopus Oocyte Expression and Transport Assays. Xenopus

oocytes were defolliculated and maintained in ND96 medi-um/5% horse serum (14). mRNA was synthesized from 5 ugof plasmid template and gel-purified on a Sephadex G-50 spincolumn. Ten nanograms of mRNA was microinjected intoeach oocyte, and the samples were allowed to translate for4-7 days. Controls consisted of oocytes injected with 50 nl ofwater. [3H]Glycine and drugs were added to a total volume

of 50 pl, and uptake assays were done as described (5). Eachdata point represents the mean of three to five oocytes.In Situ Hybridization. Rat brain and cervical spinal cord

were fixed with 4% paraformaldehyde and processed asdescribed (15). Twenty-five- to 30-tum frozen sections wereprepared in both sagittal and horizontal planes and wereincubated, free-floating, with 35S-labeled antisense RNAprobes. Control sections were probed with 35S-labeled senseRNA to determine nonspecific hybridization.Sequence Analysis. DNA and amino acid sequences were

analyzed on a personal computer with the program PCGENEand on a DEC VAX computer with the Genetics ComputerGroup package. Data bases were searched on the NationalCenter for Biotechnology Information computer with theprogram BLAST (16).

RESULTS AND DISCUSSIONCloning Strategy. C6 glioma cells have been reported to

possess transporters for a variety of neurotransmitters, in-cluding GABA (17, 18), aspartate (19), glutamate (19), andglycine (20, 21). We, therefore, chose these cells as a poten-tial source of additional members of the neurotransmitter-transporter superfamily. When RNA from C6 cells is ampli-fied by the PCR with the degenerate primers described inMaterials and Methods, a major product of -700 bp is seen(Fig. 1A). The nucleotide sequence of this PCR productencodes a peptide with significant homology to both GAT1and NET1 proteins (data not shown). RNA blot analysis withthe PCR product as a probe reveals a single transcript of -3.3kilobases (kb) in C6 cell mRNA and in rat forebrain andcerebellum mRNA (Fig. 1B). There is a barely detectablesignal in rat liver mRNA and no detectable signal in kidneymRNA. This result indicated that the gene corresponding tothis PCR product is expressed in brain but is not expressed(or is expressed at low levels) in nonneural tissue. Therefore,a size-selected AZAP rat brain cDNA library was screenedwith the C6 cell PCR product, and six positive clones were

A Water-injectedM RNA-injected

_n_[3HJGABA [3HlGLY [3H]GLU [3HJ3-ALA

Water-injected RNA-injected

0-

0

I.ZI-U8a--.0:Sis

Control No Control NoSodium Chloride

FIG. 2. Expression ofGLYT1in Xenopus oocytes. Assays weredone 4 days after injection ofoocytes with 10 ng ofmRNA or 50nl ofwater. Each point representsthe mean ± SEM of [3H]glycineuptake by three to five oocytes.All assays were done at roomtemperature for 1 hr. (A) Uptakeof [3H]GABA, [3H]glycine, [3H]-glutamic acid, and ,B-[3H]alanine.Each amino acid was used at afinal concentration of 1 ,uM. (B)Uptake of [3H]glycine in normalsaline and saline in which sodiumwas replaced by lithium (No So-dium) or saline in which chloridewas replaced by acetate (NoChloride). (C) Inhibition of [3H]-glycine uptake by three amino ac-ids or amino acid analogs. Unla-beled L-alanine (L-ALA), methyl-aminobutyric acid (MAIB), orsarcosine (SARC) was included inthe incubation mixture at a finalconcentration of 500 ,M.

8.0

6.0 [

4.0 [

2.0

0

I-

a10

8.

-4

C-

.4

Proc. Natl. Acad. Sci. USA 89 (1992)

Dow

nloa

ded

by g

uest

on

May

16,

202

1

Proc. Natl. Acad. Sci. USA 89 (1992) 7191

plaque-purified. Partial nucleotide sequences from theseclones indicated that five of the six clones were clearlyrelated to each other and to the original PCR product,whereas the sixth represented an unrelated clone. One ofthese five clones (clone 9-iB) appeared to be full-length at the5' end of the open reading frame, and this clone was chosenfor further analysis and expression.

Functional Expression. Oocytes injected with 9-lB mRNAwere tested for their ability to take up radiolabeled neuro-transmitters or related compounds. Fig. 2A shows thatoocytes expressing clone 9-lB do not significantly accumu-late [3H]GABA, [3H]glutamate, or /[3H]alanine above back-ground levels. In contrast, the accumulation of [3H]glycine isenhanced markedly in oocytes injected with 9-lB RNA,reaching levels of -10-20 pmol after 1 hr of incubation in 2.5tkM [3H]glycine. This result indicated that clone 9-lB en-codes a glycine transporter, and it was designated GLYT1.Uptake by oocytes injected with GLYT1 mRNA depends

on both extracellular sodium and extracellular chloride,indicating that the transporter encoded by this clone mediatesion-coupled active transport of [3H]glycine (Fig. 2B). Inter-estingly, the moderate level of [3H]glycine uptake by controloocytes is also sodium- and chloride-dependent, indicatingthat Xenopus oocytes possess an endogenous transport sys-tem for glycine. This fact is not surprising because uptakesystems for metabolically important amino acids are found inboth neural and nonneural tissues (22). Oocytes injected withGLYT1 mRNA take up [3H]glycine by a high-affinity mech-anism (Fig. 3 A and B). Kinetic analysis revealed a Km of 33± 8 uM (mean ± SEM, n = 3), which is similar to the Kmvalues reported previously for high-affinity glycine uptake inthe central nervous system (23-25). This Km value is almost3-fold lower than that reported for [3H]glycine uptake by C6glioma cells (Km = 95 ,IM) (21). The Vm,, value for the clonedtransporter varied, depending on the batch of oocytes, from120 to 390 pmol per oocyte per hr (mean = 246 ± 78). Theoocyte glycine transporter also appears to be a high-affinitytransporter with a Km of =20 ,uM and a Vma. of =20 pmol peroocyte per hr (data not shown).At least three uptake systems able to transport glycine,

termed system Gly, system ASC, and system A, have beendemonstrated in a variety of tissues and cell types by usingamino acids and amino acid analogs as substrates and com-petitive inhibitors (26). Therefore, the substrate specificity ofGLYT1 was examined by using several of the subtype-specific compounds (at a final concentration of 500 ,uM) ascompetitive inhibitors of [3H]glycine uptake by mRNA-injected oocytes (Fig. 2C). Sarcosine (N-methylglycine), asystem Gly substrate, inhibits [3H]glycine uptake by oocytesinjected with GLYT1 mRNA by >85%. By contrast, L-ala-nine, a system ASC substrate, and methylaminoisobutyricacid, a system A substrate, are almost completely ineffectiveas inhibitors. This inhibition profile suggests that GLYT1represents a system Gly-like transporter, in agreement withthe substrate specificity defined for the native glycine trans-porter expressed by C6 cells (21). However, the substratespecificity of GLYT1 differs from that described in earlierwork on the rat spinal cord (23), where sarcosine wasreported to have no significant inhibitory effect at concen-trations as high as 1 mM. Thus, GLYT1 may represent oneof several glycine-transporter subtypes expressed in brain.The oocyte endogenous glycine transporter shows an in-

hibition profile different from GLYT1. The former is inhib-ited almost 90%o by L-alanine (Fig. 2C), L-serine, and L-Cys-teine (data not shown). This transporter is not blocked bymethylaminoisobutyric acid and is only partially inhibited bysarcosine (Fig. 2C). Therefore, the endogenous transporterappears similar to system ASC.

Sequence Analysis. Nucleotide sequencing of the GLYT1cDNA revealed an open reading frame encoding a protein of

633 amino acids with a molecular mass of 70,570 Da (Fig. 4).The protein contains four putative N-glycosylation siteslocated in a region, which, for GATi, is proposed to form alarge extracellular loop (4). There are five putative proteinkinase C phosphorylation sites. A Kyte-Doolittle analysis ofGLYT1 yielded a hydrophobicity profile almost identical tothat seen for GATi (4) and other members of this transportersuperfamily, suggesting that the membrane topology ofGLYT1 is similar to proposed models (4, 5, 8). Furtheranalysis with CLUSTAL and PALIGN (PCGENE) showed thatGLYT1 is homologous to all known members of the neuro-transmitter-transporter superfamily, with identities rangingfrom 37% to 41% and overall homologies (identity plussimilarity) of 53% to 61%. A search of the GenBank andProtein Identification Resource data bases with the programBLAST did not reveal significant homologies between GLYT1and any cloned genes other than the known members of thissuperfamily.

Cellular Localization. In situ hybridization was used tostudy distribution of GLYT1 mRNA in the rat central ner-vous system (Fig. 5 A-D). There is prominent labeling in alllamina of the cervical spinal cord, the brainstem nuclei

350

0

4U,

V)

C-

300

250

200

150

100

50

00 20 40 60 80 100 120

[Glycine] (JLM)

V/s

FIG. 3. Kinetics of [3H]glycine uptake. Michaelis-Menten (A)and Eadie-Hofstee (B) plots of [3H]glycine uptake by oocytesinjected with GLYT1. V, velocity; V/S, velocity/substrate. Thesevalues represent the difference between [3H]glycine uptake bymRNA-injected oocytes and water-injected control oocytes.

Neurobiology: Guastefla et al.

Dow

nloa

ded

by g

uest

on

May

16,

202

1

7192 Neurobiology: Guastella et al.

GLYTI M V -

GATI MI- -

NETl !M L L

GLYT1GATINE I

GLYrlGATINET1

GLYFIGATINETl

GLYTlGATINEll

G D LG D A

Proc. Natl. Acad. Sci. USA 89 (1992)

v

- - G K G A K G M L N C. A V P S E A - T K 7[D Q N TIA-I'DNSKVADGQI STEVS IEAPVASDKIPKTIVKrKT1VV i,,yKsA R M N P Q V Q P E N N G A D I C P EQ F-L R A R K T A E Lj 1 tu K F F V i, F;i.!

p

QM T]V FTI LI FF L I I

TRG..NWGNQIETVLTSVGYAVG-LGNRFDRIDTIWIKGRFDIFLMSCIVGIYIAIIGLGNVWRFP'~ CG-'KNmA.(';.,FLIIF'I FL

P R G T W G KQ I D F L L S V G FA Y L AN V W R F P I Cl N K N GI G i;. A F L I _ F I

FFM-'j~LSF [QFASQ[C C [G VWR S [T]MFK GVGN(,~ G NIlMVVST YITGI

A G V P l. F L L F C S L G Q Y T S G G L G V W K L A P|M| F K i, V i A A V L S W L N I

AG M .PLF.YM EL A L LiG Q Y N R E [G A A T [VW K I C F F K C t( A L. A Y VG F

.. -smu- u - -

lNV V _ C I A FYYF FY Y I V I I S W A I y Y L YY Y N J I I A W S L Y Y. L F

T A I. S G N L S H 1 F N Y' M T S - - - -V L G N 1-i T K Y 1)

GLYTI C G V S W vGATI T L|A IAWVNET! CWLM Y V V I

GLYTI [G IIM YYL TGATl G IlL F YiI TNET! L N AW L H

S sM7TiH V L P W A Y 7 N N P WN T P D .

1 11-1T PX KQ|CI|DNDNP I N TI'R(ISL FW 1 N L DT[D C H TWNS PNNC

L.. Q R T S

K N K F T

P S F W R 1. V K L S D

A Vr V|IE F W N9 M H1 Q M1 D

P A AL F Y E R G V L. L E

V V F L C L t R C V K S S G K V VL V Y F C I W K G V G W T G K V VV L Y F S L W K G V K T S G K V V

P K W D K I L E A K V W G A A SP N F R K L S D S E V W L F) A A TI D F Y R L K E A T W IID A A|T

Y F TIA TI FY F SIA IYLW I TL L

L

L

U- U- EKE- U U U-

Q I FQ I FIFQI

G C A W

GI IL G L

JA G F

G

D Ar N=F ZSr

'r FS

1.

2V N T

D D F E R I jP 1 IC1 L PK PIC;|Q R VI P AC. D G I r Q V, QI 1

F V R G V T I1.FTG,T\T TF F R ('\V I fl,A K EL \,' H V 'I L. P G AP N

ASlil!

I

.5c

..2

.1 b21-F

h4~

IGII IIT M AIK- 'NS;K7 HN7 1L|SGT1 IA L S FHHNINIVL I A F A.q NI KSF ID

I sv cT S

IY R DISIY R DIS I I

IYR DA L L

Al Y P E Al L T Ll LA| Y P E AV T Q L

LIYP E A| I S TL

I L Q KR N R RK R H R

I T C AEC I N| S CI T Ss I 1

|V Y A|G F|VM F A1G FJV|F V S GFA

I F S II F S II F S I

LWVIG

u.n_

IP IP IS G

S P LW|S|P L|WLS T F W

K T Y V T L GH7AE L F 1 - A AIV|CK L F T - F GWV T

S L LA I LJA V V

F F FF FF FV

V A G F L G I

I V S Y L I G L

F S T FW L A L

M LM LM L

ILL ML A LG

P L T S Q AS N I T Q GF IC I T K G

M AM AM A

N H G V D V S R %AD H PG AFVH V IK R S! AD AAASSCJ P LIAIFl LH E I K V N I E D JA T F G A G V F I

LGTQFC L LFfTL-v[I DSQF( TV|FIG FI1TL D S S M C C ME A V I L

G I YG I YG I Y

W L LV F KV L T

I-LL

MT] NY AFIDI Y SLW I F A

AA

A! VA L VC L A

.'vv

E VN N F TA'E FP R 1, F.D F Q %V

41I4,ih

c F T iLVVS' M|S;LLF| NI\

ruT I L F .N V'

GLYTI I M C VGAT1 F E C VNErl M E A I

S I M YS I S WG V S W

Y G H R NIY G V N R

Y V D R

Y F Q DF Y D NF s ND

GLYTI V1QYRHIHYNHHfQYHYGWAVAIGATI A V Q MTI P|LT|MG SYIV FI P|KIWIG Q G V

NETI I I N F KWP LW Y D D WI FW PHA N W V

QlRL[K N A T KH S

QIR LQ V M I Q P -

E| R LA Y G I TLWE

I QIQ|M|ml

L

V

m

G F PG S R P

G F RLP

P L F F Q IC I WW K LG L Y W R L

C'W!c t;c w

_m_umu___

|G F L M A L S 1 V I C IW L MIAL SSI V iW E L. S1M V V

R D W G P A L L E H R T G R Y A P T I1 1 PS PS E - - D I V R P E N G P E Q r Q A G z < A SNE - - -HLVAQRD I RQFQLI Q - A

rPP~

KR WN.;sIS I. 1:K F V

S

TS

A- 1 FN1 A YV Y

_ T I I F F I 1 I F TPI V A V FLV SPlAF~I. i.:VVVVSQi L K1) PT T i\1F i LLK,j1.

F P. ( ..K F A-N

I V G S N G S S R L Q D s R_ &13599

-_ 617

FIG. 4. Amino acid sequence of GLYT1 aligned with the amino acid sequences of GAT1 (4) and NET1 (5). Residues identical among allthree proteins are boxed; residues representing conservative substitutions are shaded. Putative transmembrane regions are denoted by bars,

potential N-glycosylation sites are marked with asterisks, and potential protein kinase C phosphorylation sites are marked by arrowheads.Alignment was done with the program CLUSTAL in PCGENE.

(including the spinal trigeminal nucleus, Fig. 5A), the cere-

bellum (Fig. 5B), and the olfactory bulb (Fig. SC). Lighterlabeling is present in other medullary, pontine, midbrain,thalamic, and hypothalamic nuclei (data not shown). Label-ing above background could not be detected in the cerebralcortex, caudate-putamen and globus pallidus, hippocampalformation, or septal nuclei. Furthermore, there is no specificlabeling in white-matter fiber tracts. No specific labeling wasobserved in control sections incubated with sense RNAprobes (Fig. SD). Thus, GLYT1 is not expressed in all regionsof the central nervous system. Instead, its expression islimited to those areas where glycine is thought to function as

a neurotransmitter and to other specific regions where gly-

cine might play a neurotransmitter role.Labeling in the brain stem (Fig. SA) and cervical spinal cord

is clearly associated with neuronal cell bodies, although hy-

bridization to glial cells cannot be ruled out. Cerebellar label-

ingis confined to the Purkinje cell layer, where the silvergrains

form a continuous band (Fig. 5B). The Purkinje cells them-

selves are not labeled, and labeling does not appear to be

associated with basket, Golgi, or Lugaro cells, which may be

located near the Purkinje cell layer. Instead, the labeling

pattern indicates that GLYT1 is expressed in Bergmann glia,

which are located in this layer. The absence ofGLYT1 mRNAin the cerebellar Golgi neurons contrasts with previous reports

that, in the rat, these neurons take up [3H]glycine (27) and

contain glycine immunoreactivity (28, 29); thus, this finding

provides further evidence for the possibility of multiple gly-

cine-transporter subtypes. In the olfactory bulb, there is

prominent labeling of many cells located in the glomerular

layer (Fig. 5C). Labeled cells also occur in the externalplexiform layer, the mitral cell body layer, and the granule cell

GLYT1GATINETrGLYrlGATINET!

GLYrlGATINETI

GLYTIGATINET1

GLYTIGATINErl

(,

[

Dow

nloa

ded

by g

uest

on

May

16,

202

1

Proc. Natl. Acad. Sci. USA 89 (1992) 7193

FIG. 5. Localization of GLYT1 expression in the rat centralnervous system. (A) Sagittal section through the medulla; numerousheavily labeled cells appear in the spinal trigeminal nucleus. (B)Horizontal section through cerebellum showing a band of labeling inthe Purkinje cell layer. ML, molecular cell layer; PCL, Purkinje celllayer; GCL, granule cell layer; WM, white matter. (C) Horizontalsection through the main olfactory bulb. (D) Control section. Sameas C, except section was incubated with sense mRNA probe. GL,glomerular layer; EPL, external plexiform layer; MCL, mitral celllayer; GRL, granule cell layer. (Bar = 200 pm forA and 100 ,m forB-D.)

layer (Fig. 5C). Although localization of [3H]glycine uptake inthe olfactory bulb is equivocal (30), recent immunocytochem-ical studies suggest that the periglomerular neurons and somecells in the granule cell layer contain glycine (29). As in theother labeled regions, we cannot rule out the possibility thatsome glia cells in the olfactory bulb express GLYT1.

Availability of the GLYT1 cDNA should facilitate ourunderstanding of glycinergic metabolism in the central ner-

vous system. In addition, further in situ hybridization studiesof GLYT1 expression will undoubtedly aid in identifyingputative glycinergic neurons in the spinal cord, the brain-stem, and in forebrain structures not conventionally associ-ated with glycinergic neurotransmission.

Note Added in Proof. After this paper was submitted, Smith et al. (31)reported the isolation of a rat brain glycine transporter cDNA clonewhose sequence and distribution is similar, but not identical, to thatof GLYT1.

We thank M. Quick for oocyte preparation, L. Czyzyk for assis-tance with nucleotide sequencing, and A. Figl for assistance witholigonucleotide synthesis. This work was supported by NationalInstitutes of Health Grant EY04067, Neuroendocrine Anatomy CoreGrant DK4130, and Veterans Administration Medical ResearchFunds to N.B.; National Institutes of Health Grant NS-11756 toH.A.L. and N.D.; and a National Institutes of Health PostdoctoralFellowship to J.G.

1. Glowinski, J. & Axelrod, J. (1964) Nature (London) 204,1318-1319.

2. Horn, A. S. (1990) Prog. Neurobiol. 34, 387-400.3. Krogsgaard-Larsen, P., Labouta, I. M., Meldrum, B.,

Croucher, M. & Schousboe, A. (1981) in Neurotransmitter,Seizures and Epilepsy, eds. Morselli, P. L., Loscher, W.,Lloyd, K. G., Meldrum, B. & Reynolds, E. H. (Raven, NewYork), pp. 23-33.

4. Guastella, J., Nelson, N., Nelson, H., Czyzyk, L., Keynan, S.,Miedel, M. C., Davidson, N., Lester, H. A. & Kanner, B. I.(1990) Science 249, 1303-1306.

5. Pacholcyzk, T., Blakely, R. D. & Amara, S. G. (1991) Nature(London) 350, 350-353.

6. Kilty, J. E., Lorange, D. & Amara, S. G. (1991) Science 254,578-579.

7. Shimada, S., Kitayama, S., Lin, C.-L., Patel, A., Nanthaku-mar, E., Gregor, P., Kuhar, M. & Uhl, G. (1991) Science 254,576-578.

8. Usdin, T. B., Mezey, E., Chen, C., Brownstein, M. J. & Hoff-man, B. J. (1991) Proc. Natl. Acad. Sci. USA 88, 11168-11171.

9. Blakely, R. D., Berson, H. E., Fremeau, R. T., Caron, M. G.,Peek, M. M., Prince, H. K. & Bradely, C. C. (1991) Nature(London) 354, 66-70.

10. Hoffman, B. J., Mezey, E. & Brownstein, M. J. (1991) Science254, 579-580.

11. Yamauchi, A., Uchida, S., Kwon, H. M., Preston, A. S.,Robey, R. B., Garcia-Perez, A., Burg, M. B. & Handler, J. S.(1992) J. Biol. Chem. 267, 649-652.

12. Chomczynski, P. & Sacchi, N. (1987) Anal. Biochem. 162,156-159.

13. Snutch, T. P., Leonard, J. P., Gilbert, M. M., Lester, H. A. &Davidson, N. (1990) Proc. Natl. Acad. Sci. USA 87, 3391-3395.

14. Quick, M. W., Naeve, J., Davidson, N. & Lester, H. A. (1992)BioTechniques, in press.

15. Brecha, N. C., Sternini, C. & Humphrey, M. F. (1991) Cell.Mol. Neurobiol. 11, 497-509.

16. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman,D. J. (1990) J. Mol. Biol. 215, 403-410.

17. Hutchison, H. T., Werrbach, K., Vance, C. & Haber, B. (1974)Brain Res. 66, 265-274.

18. Schrier, B. K. & Thompson, E. J. (1974) J. Biol. Chem. 249,1769-1780.

19. Erecinska, M., Troeger, M. B., Wilson, D. F. & Silver, I. A.(1986) Brain Res. 369, 203-214.

20. Zafra, F. & Gimenez, C. (1986) Brain Res. 397, 108-116.21. Zafra, F. & Gimenez, C. (1989) Biochem. J. 258, 403-408.22. Schultz, S. G. & Curran, P. F. (1970) Physiol. Rev. 50, 637-710.23. Johnston, G. A. R. & Iversen, L. L. (1971) J. Neurochem. 18,

1951-1961.24. Staatz-Benson, C. & Potashner, S. J. (1987) J. Neurochem. 49,

128-137.25. Debler, E. A. & Lajtha, A. (1987) J. Neurochem. 48,1851-1856.26. Christensen, H. N. (1975) Curr. Top. Membr. Transp. 6, 227-

258.27. Wilkin, G. P., Csillag, A., Balazs, R., Kingsbury, A. E., Wil-

son, J. E. & Johnson, A. L. (1981) Brain Res. 216, 11-33.28. Ottersen, 0. P., Davanger, S. & Storm-Mathisen, J. (1987)

Exp. Brain Res. 66, 211-221.29. Pourcho, R. G., Goegel, D. J., Jojich, L. & Hazlett, J. C.

(1992) Neuroscience 46, 643-656.30. Halasz, N. & Shepherd, G. M. (1983) Neuroscience 10,579-619.31. Smith, K. E., Borden, L. A., Hartig, P. R., Branchek, T. &

Weinshank, R. L. (1992) Neuron 8, 927-935.

Neurobiology: Guastefla et al.

Dow

nloa

ded

by g

uest

on

May

16,

202

1