Geoffrey H. Murdoch, R. Alan North, and. Susan G. Amara. From the Vollum Institute for Advanced Biomedical. Research, Oregon Health Sciences University,.
THE JOURNAI. OF BIOI.OGICAI. CHEMISTRY Val. 268, No. 21, Issue of July 25, pp. 15329-15332. 1993 0 1993 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
Communication Cloning and Expressionof a Human Neutralhino Acid Transporter with Structural Similarity to the Glutamate Transporter Gene Family* (Received for publication, May 3, 1993, and in revised form, May 19, 1993)
Jeffrey L. ArrizaS, Michael P. Kavanaugh, Wendy A. Fairman, Yan-NaWu, Geoffrey H. Murdoch, R. Alan North, and Susan G. Amara
a m i n o a c i d transport s y s t e m y+ (9, lo), and the acidic amino a c i d t r a n s p o r tsystem X- (referred to here as glutamate transporters). Glutamate transporter subtypes were recently isolated from rat brain (11, 12) and rabbit intestine (13), and d e m o n s t r a t e d t o be structurally related members of a transp o r t e r gene family that also exhibit regions of homology with glt-P (14)and dctA (15), two bacterial nutrient uptake systems. We r e p o r t the cloning, expression, and functional characterization of a human gene product that is s t r u c t u r a l l y related t o the glutamate t r a n s p o r t e r gene family and whose properties corr e s p o n d t o the neutral a m i n o a c i d u p t a k e s y s t e mASC. EXPERIMENTAL PROCEDURES
cDNA Cloning and Sequencing-The amino acid sequence of GLASTl (11)was obtained fromthe Swiss Protein Database (accession From the Vollum Institute for Advanced Biomedical no. P24942). A synthetic oligonucleotide complementary to the amino Research, Oregon Health Sciences University, acid sequence ATINMDGTALYEALAAIFIAQ (residues395-415 of Portland, Oregon 97201 GLAST1) was synthesized because the conservationof portions of this sequence in the Escherichia coli glutamate carrier protein glt-P (14) A cDNA was isolated from human brain that encodes suggested an important andconserved structural motif. Thisoligonuan amino acid sequence 34-399’0 identical to previously cleotide sequence (5‘ CTG (A/G)GC (A/G)AT GAA (A/G)AT GGC AGC published glutamate transporter sequences. Injection of CAG GGC (C/T)TC ATA CAG GGC TGT GCC (A/G)TC CAT GTT (A/ codon RNA transcribed from this cDNA into Xenopus oocytes G)AT GGT (A/G)GC 3’)is 128-fold degenerate and utilizes human resulted in expression of a transport activity with the preferences. Human motor cortex RNAisolatedby the methodof Chomproperties of the neutral aminoacid uptake systemASC. czynski and Sacchi (16) and poly(A+) enriched by oligo(dT)-cellulose Superfusion of alanine, serine, and cysteine evoked so- chromatography were usedt o prepare a cDNA library. Mixed oligo(dT) andrandom-primed cDNA wassynthesizedusing theSuperscript dium-dependentinwardcurrents in voltage-clamped Choice System (Life Technologies Inc.) and ligated into AZAPII (Stratoocytes expressing the transporter. These currents wereagene, La Jolla,CA). The oligonucleotide was 32P-labeledfor screening dose-dependent, stereospecific, and saturable, with K , this humancortex library. Conditions for hybridization and wash were: values ranging from 29 to 88 p.Northern blotanalyses 0.5 M Na2HP0,, pH 7.15, 7% SDS for hybridization a t 50 “C, and wash of this gene,termed pASCT1, obtained by in vivo at 60 “C in 2x SSPE, 1% SDS. The plasmid revealedubiquitousexpression ASCT1, consistent with the general metabolic role as- excision (Stratagene)of a hybridizing AZAPII clone, contained a2.7-kbp cDNA insert that wassequenced on both strands by the dideoxy chain cribed to system ASC. termination method using Sequenase2.0 (U. S. Biochemical Corp.) and synthetic oligonucleotide primers (Oligos Etc.). Expression of ASCTl in Oocytes-The coding sequence of pASCTl The c e l l u l a r u p t a k e of a m i n o a c i d s is mediated by multiple was isolated with unique flanking restriction sites by the polymerase i n d e p e n d e n t t r a n s p o r tsystems distinguished primarily bysub- chain reaction using synthetic oligonucleotide primers. Plasmid pOTVstrate preference and by ionic requirements. In m a m m a l i a n ASCTl contains thisDNA sequence subcloned into the polylinkerof an oocyte transcription vector that utilizes Xenopus 6-globin 5’-and 3’cells, neutral amino acids are t r a n s p o r t e d p r e d o m i n a n t l y by t w o s o d i u m - d e p e n d e n t s y s t e m s termed ASC (for alanine, ser- untranslated regions derived from pSP64T (17) to enhance expression levels in oocytes. Synthetic RNA was transcribed in vitro (18) and inine, and cysteine preferring)and A (for alanine preferring) (1- jected into defolliculated stage V-VI Xenopus oocytes. Transport was 3). N - M e t h y l a t e da m i n oa c i d s and the related compound assayed as described previously by measuring uptake of c3H1alanine, MAIB’ are n o t substrates for system ASC, a property that r3H1serine, or [35Slcysteine (all radiolabeled compounds from Du Pontallows it t o be operationally distinguished f r o m s y s t e m A (4). New England Nuclear) and by two-electrode voltage-clamp recording ASC-mediated t r a n s p o r t is the m a j o r c o m p o n e n t of n e u t r a l (18).Briefly, oocytes were voltage-clampedat -60 mV and continuously superfused with a buffer (ND-96) consisting of 96 mM NaC1, 2 mM KC1, a m i n o a c i d uptake in a wide variety of cell types. 1.8 mM CaCl,, 1 mM MgC12, and 5 mM HEPES pH 7.5. For transport A number of the c l a s s i c a m i n o a c i d t r a n s p o r t s y s t e m s h a v e measurements, this solution was changed to the indicated concentran o w been cloned and d e m o n s t r a t e d b ystructural homologies to tion of amino acid in ND-96. Current (1) as a function of substrate be m e m b e r s of distinct gene families. These include system Gly concentration ([SI)was fitted by least squares toI = I,,;[S]/(K, + [SI), is the maximal current and K,,, is the transport constant. (5-7), the neutral a m i n o a c i d t r a n s p o r tsystem A (81, the basic where lmax Values of K, and I,,, were determined by fitting the results from individual oocytes in which five or more different concentrations were * This work was supportedby a Hitchings award from the Burroughs applied. Data were averagedfrom all oocytes tested and are expressed Wellcome Fund and by National Institutes of Health Grants DA07595 as mean * S.E. (to S. G. A,)and DA03160 (to R. A. N.). The costs of publication of this Northern Blot Analysis-The human tissue Northern blot was obarticle were defrayed in part by the payment of page charges. This tained from Clontech Laboratories, and human brain region RNAs were article must therefore be hereby marked “advertisement” in accordance prepared (16) from tissue provided by the Oregon Brain Repository. with 18 U.S.C. Section 1734 solely t o indicate this fact. The nucleotide sequence(s) reported in this paper hasbeen submitted RNAs weresize-fractionated on denaturingformaldehydegelsand transferredtoanylonmembrane.The1.7-kbp coding sequence of to the GenBankTMIEMBL Data Bank with accession number(s) L14595. $ To whom correspondence should be addressed: Oregon Health Sci- ASCTl was radiolabeled with a-[32PldCTP (Du Pont-New England Nuclear) by the random priming method (Boehringer Mannheim). Filters ences University, The Vollum Institute L-474, 3181 SW Sam Jackson Park Rd., Portland, OR 97201-3098. Tel.: 503-494-6721; Fax: 503-494- were hybridized overnight at 42 “C with this cDNA probe (lo6 c p d m l ) 6934. in 5 x SSPE (20 x SSPE = 3 M NaCl, 0.2 M NaH,PO,, 0.02 M Na,EDTA, The abbreviations used are:MAIB, a-(methy1amino)isobutyric acid; pH 7.4). 50% formamide, 7.5 x Denhardt’s solution, 2% SDS, and 100 kb, kilobase(s); kbp, kilobase pair(s). pg/ml denatured salmon spermDNA. Autoradiography was performed
Cloning of a Human Neutral Amino
Acid Transport System
after two 30-min room temperature washes in2 x SSPE, 0.1% SDS and two 20-min washes at 50 "C in 0.1 x SSPE, 0.1% SDS. Filters were subsequently reprobed with a radiolabeled human p-actin probe (Clontech Laboratories, Palo Alto).
terminal portion. The sequence similarity of ASCTl with known glutamate transporters suggested thatthis proteinmightfunction to transport glutamate or other amino acids. We have previously utilized expression in oocytes and voltage clamp recording to RESULTS A novel cDNA clone was isolated during thecourse of screen- characterize system y+ (101, and others have utilized similar ing a human motor cortex cDNA library for sequences related assays to characterize the activity of glutamate transporters to rat (11)and E. coli (14) glutamate transporters.Nucleotide (13).Superfusion of ASCTl RNA-injected oocytes voltagesequence analysis of this clone predicted an open reading frame clamped at -60 mV with the amino acids alanine, serine, and of 532 amino acids and corresponding t o a protein of -56 kDa, cysteine produced inward currents that were absent in unin2 A ) . These uptake currents were doseflanked by -180 base pairs of 5'-untranslated sequence and jectedcontrols(Fig. -0.9 kbp of 3"untranslated sequence. The start of the coding dependent and saturable (Fig. 2, B and C, and Table I). Threalso elicited currents,but with lower sequence was defined by the first ATG downstream of an in- onine and valine frame stop codon, and the surroundingsequences ( G C C E G ) affinities. Lysine or glutamate did not elicit detectable reconform to a consensus translation initiation site(19). A frag- sponses (Fig. 2 A ) , and the system A-specific substrate MAIB ment of this cDNA sequence (-nucleotides 1156-1492) was neither evoked a detectable current nor inhibited the current among the randomly selected clones sequenced by Adams et al. elicited by alanine. N-Methylalanine at 1 mM did not produce a (20)and referred to as EST02446. We termthis sequence detectable current nor did proline, glycine, glutamine, asparaASCTl because of the functionalproperties of the encoded gine, methionine, arginine, or leucine (data not shown). The rate of radiolabeled amino acid uptake (in pmol/min.oocyte, transport activity. The ASCTl amino acid sequence shown in Fig. 1 exhibits determined at 100 p ~in)ASCT1-expressing oocytes was 18.0 f 2.0 for alanine, 20.1 2 5.1 for serine, and19.2 2 5.9 for cysteine, remarkable similarity to those reported for glutamate transporter subtypes. Amino acid identities among these sequences while the comparable values in uninjected controls were 0.6 2 have been highlighted to illustrate the conservation of gluta- 0.1, 0.4 f 0.1, and 1.0 * 0.3, respectively ( n = 5). Uptakeof 100 mate transporter structural motifs within theASCTl sequence. PM L3H1alanine was unaffected by the addition of 10 mM MAIB ASCTl has 39,34, and 39% sequenceidentity with rat GLASTl (107% of control; n = 2). ASCT1-mediated transport wasalsoexamined for ster(ll),rat GLTl (12), and rabbitEAACl (13), respectively. This sequence identity iscomparable withthe identity seen between eospecificity and ion sensitivity. The current induced by o-alathe subtypesthemselves. All sequences exhibit unrelated NH2 nine a t 1 mM was only 12 f 3% of the current obtained with 1 and COOH termini, and diversity in a putative extracellular mM L-alanine ( n = 3). The inward current elicited by 1 mM transporter domain containing conserved potential N-linked L-alanine superfusion was not seen when Tris was substituted glycosylation sites (Fig. 1).Another notable difference resides for Na+ (Fig. 2 A ) . The currentinduced by 1mM alanine was also in a region highly conserved in the glutamate transporters (seereduced 61 4% ( n = 4) when extracellular K+ was increased the ASCTl sequence IFQC, residues 384-387) and may be re- from 2 to 20 mM. The expression of ASCTl mRNA was analyzed by Northern lated to the determination of substrate specificity. The absence of a signal peptide sequence suggests the amino terminus is blotting of RNAs from various human tissues. Two prominent a smaller hybridizing intracellular, but other aspects of the transmembranetopology RNA species 5.1 and 4.0 kb in length, and of this transporter family are currently controversial because band corresponding to 2.4 kb, were detectablein all lanes (Fig. many of the hydrophobic regions predictedby sequence hydrop- 3A). The nature of these multiple RNA species is unclear but athy analyses are more consistentwithP-sheets than with may reflect differential RNA processing. ASCTl RNA appears a-helices. Six membrane-spanning regions are predicted in the to be most abundantly expressed in brain, muscle, and panNH2-terminal portion of ASCT1, consistent with domains pro- creas but was detectable in all tissues examined. The expresposed for the glutamate transporters (11-13), and as many as sion of ASCTl in various regions of human brain was also four transmembrane regions may be present in the COOH- examined and found to be uniformly distributed (Fig. 3B 1. ASCT? GUST1 GLTI
M E K S N E T N G Y L D S A O A G Q A A G P G A P G ~ A A G R A R R C A R F L R R O A L V ~ - - ~ S G ~ L A ~ A G L ~ A A ~ . G L M T K S N G E E P R M G S R M T R F O O G V R K R T L L A K K K V O N I l K E D V K S Y L F R N A F V M A S T E G A N N M P K O V E V R M ~ D S ~ ~ S S E E P K ~ R N ~ G M R M C O K L G K N L
S L M R T NToL
FIG.1. ASCTl is structurallyrelated tomammalian glutamate transporters. Thededucedamino acid sequence of human ASCTl is shown aligned with the sequences reported for the glutamate transporter subtypes GLASTl (111, GLTl (121, and EAACl (13).Gaps (indicated by dashes) wereintroduced intotheproteinsequencestofacilitate their alignment. Positions where at least three of the four protein sequences have identicalamino acid residuesareindicated by white on black lettering. Potential N-linkedglycosylation sites (N-X-Sfl') are outlined within boxes.
487 O E L A E V K V E A I P N C K S E E E T S P L V T H O N P A G P V A S A P ~ L E S K E S V L 507
5 3 2
V E M G N S V I E E N E M K K P V O L I A O D N E P E K P V A D S E l K M 5 4 3 S ~ ~ R M ~ E D I E M I K T ~ S ~ Y D ~ T K N H R E ~ N S N Q ~ Y I * * H N S Y V I D E C K V T L A ~ N G K ~ A D ~ ~ ~ E E E P ~ K ~5E7 K3 V S S E V N I V N P F A L E S A T L D N E D S D T K K S Y l N G G F A ~ D K S D l I S F T Q T S O F5 1 1
Cloning ofHuman a Neutral Amino Acid Pansport System
conserved structuralfeatures of the GLAST1,GLT1, and EAACl transporter subtypes, asdepicted in Fig. 1. Moreover, ASCTl also contains the heptapeptidesequence AA(W)FIAQ that is found in the bacterial genes for a glutamatelaspartate carrier protein (14) anda dicarboxylate carrier protein ( E ) , a s well as in the mammalian sequences. These structural parallels with acidic amino acid carrier proteins are strikingin view of the marked difference ASCTl has in substrate specificity. These findings demonstrate that the functional diversityof this 10 30 IW M .a Inno gene family is broader than previously suspected. "G+"C/-i/ In agreement with previous studies of system ASC transport, 1m Mn Mo Roo lm ASCTl demonstrated a broad specificity for neutral amino aclsmlm addl gLt ids. Alanine, serine, cysteine, and threonine were the most FIG.2. Inward currents evoked by amino acids in voltageeffective substrates, and cysteine had the highest measured clamped ASCT1-expressing oocytes. A, amino acids (1 mM) were ASCTl exhibited sodium desuperfused for the duration indicated by the bar. Sodium-dependent transport affinity (K, = 29 p ~ ) . currents were induced by L-alanine, 1.-serine, and L-cysteine in ASCT1- pendence and stereospecificity for the L-enantiomer of alanine expressing oocytes but not in uninjected controls. B , currents elicited and did not transport N-methylalanine. Thesystem A-specific were dose-dependent and saturable. C , concentration response data substrate MAIB was not a substrate for ASCT1, and itdid not were fitted to the Michaelis-Menten equation to obtain the kinetic painhibit alanine uptake. These are defining characteristics of rameters listed in Table I. system ASC (2, 3). Despite the structural homology of ASCTl TARLF. I with glutamate transporters but consistentwith system ASC, Kinetic parameters of amino acid substrates glutamate and aspartate were not effective substrates. As there have been reports that system ASC does transport glutamate Amino acid K", I,.." with low affinity when the pH is reduced (21,22), further studIta1 ies will be required to analyze this phenomenon. L-Alanine 71 14 (1.0) Although the substrates of ASCTl and the glutamate transL-Serine 88 2 11 1.2 f 0.08 29 2 6 1.0 = 0.04 porters differ, the structural similaritiesof these proteins sugL-Cysteine !.-Threonine 137 = 19 1.4 0.03 gest that they may function by similar mechanisms. Uptake of L-Valine 390 f 8 0.6 f 0.11 glutamate is anelectrogenic process that appears to be coupled " Maximum currents obtained with each substrate were normalized to the cotransportof Na' and the countertransportof intracelto themaximal current induced by alanine in the same oocyte ( n = 3-5). lular K' and of an additional pH-changing anion(23-25). Consistent with the notion of a conserved ionic mechanism, ASCTlmediated transport was found to be electrogenic and sensitive to elevated extracellular K+ as well as reduced extracellular Na'. System ASC has been suggested to be electroneutral (26), and K' sensitivity has not been reported. Previous studies of 2.4system ASC have necessarily been conducted in cells with mul1.4tiple partially overlapping transport activities. However, the overexpression of the cloned ASCTl gene product in Xenopus oocytes and theutilization of voltage-clamp control allow direct demonstration of these properties. The substrate specificity of ASCT1-mediated transport argues that this cDNA clone enB P I z 3 4 5 9 codes the biochemically defined cellular uptake system ASC. 7 ASCTl mRNA was found to be expressed in all human tis4. sues examined, with the highest levels detected in brain, skeletal muscle, and pancreas. The abundantexpression in brain is 2. particularly intriguing because cysteine, the ASCTl substrate with the highest affinity, has been reported to have neurotoxic I. effects in the central nervous system(27).Thus, ASCTl and the glutamate transporters may have similar roles in preventing the accumulation of endogenous neurotoxic amino acids. The FIG.3. Expression of ASCTl mRNA in human tissues. A, North- molecular cloning of ASCTl will provide an avenue toexplore e m blot analysis using theradiolabeled coding sequence of ASCTl as a further therole of this neutral aminoacid transport system in hybridization probe. Two pg of'polylA') RNA was loaded per lane. Lane neuronal andcellular physiology. Moreover, the structuralsim1, heart; lane 2, brain; lane 3, placenta; lane 4 , lung; lane 5 , liver; lane may aid 6, skeletal muscle; lane 7, kidney; lane 8,pancreas. In the lower panel, ilarity of ASCTl with glutamate transporter subtypes thesame filter was reprobed with a human8-actin cDNA. 8-Actin in the elucidation of structure/function relationships in this mRNA is approximately 2.0 kb in length, but in heart and skeletal transporter gene family. muscle an abundant 1.7-kb actin mRNA species also hybridizes. B , 20 pg of total RNA prepared from human brain regions was analyzed as in A. Lane I , motor cortex; lane 2, frontal cortex; lane 3, hippocampus; lane 4, basal ganglia; lane 5,cerebellum. Hybridization of this filter with the p-actin probe is shown in the hottom panel. DISCUSSION
On the basis of amino acid sequence homology, ASCTl is a member of the gene family recently defined by the cloning of mammalian glutamate transporter subtypes(11-13). The predicted ASCTl aminoacid sequence contains manyof the highly
REFERENCES 1. Christensen, H. N. (1984) Riochim. Riophys. A d a 779,255-269 2. Stein, W. D.f 1986) 'IFnnsportand Diffilsion across Cell Memhranes, pp. 467474, Academic Press, San Diego. CA 3. Christensen. H.N. f 1990) Physiol. Rev. 70,43-77 4. Christensen. H. N.. Liang. M. & Archer, E. G . (1967) J . R i d . Chem. 242, 5237-5246 5. Smith, K. E., Borden. L. A,, Hartig, P. R., Branchek. T. & Weinshank. R. L. 11992) Neuron 8,927-935 6. Lui, Q.-R., Nelson, H., Mandiyan, S.. Lopez-Corcuera, B. & Nelson, N.(1992) FEUS k t ( . 305, 110-114 7. Guastella. J.. Brecha, N., Weigmann, C.. Lester. H. A. & Davidson, N. (1992)
Cloning ofHuman a Neutral Amino Acid Transport System
Proc. Natl. Acad. Sci. U. S. A. 89, 7189-7193 8. Kong, C.-T., Yet, S.-F. & Lever, J. E. (19931 J. Biol. Chem. 268, 1509-1512 9. Kim. J . W.. Closs., E. I.., Albritton. L. M. & Cunnineham. J. M. (1991) Nature Y
352, 725-728 10. Wang, H., Kavanaugh, M.P., North, R. A. & Kabat, D. (1991) Nature 352, 729-731 11. Storck, T.,Schulte, S., Hofmann, K. & Stoffel, W. (19921 Proc. Natl. Acad. Sci. U. S. A. 89, 10955-10959 12. Pines, G., Danbolt, N. C., Bjoras, M., Zhang, Y., Bendahan, A., Eide, L., Koepsell, H., Storm-Mathisen, J., Seeberg, E. & Kanner, B. I. (1992)Nature 360, 464-467 13. Kanai, Y. & Hediger, M. A. 11992) Nature 360,4 6 7 4 7 1 14. Wallace, B., Yang,Y.-J., Hong, J . & Lum, D. (1990)J . Becteriol. 172,3214-3220 15. Engelke, T., Jording, D., Kapp, D. & Puhler, A. (1989) J. Bacteriol. 171,55515560 16. Chomczynski, P. & Sacchi, N. (19871 Anal. Biochem. 162, 156-159 17. Kreig, P. A. & Melton, D. A. (1984) Nucleic Acids Res. 12,7057-7070 18. Kavanaugh, M. P., Aniza, J. L., North, R. A. & Amara, S. G. (1992) J. Biol.
Chem. 267,22007-22009 19. Kozak, M. (1987) Nucleic Acids Res. 15, 8125-8132 20. Adams, M. D., Dubnick, M., Kerlavage, A. R., Moreno, R., Kelley, J. M., Utterback, T. R., Nagle, J . W., Fields, C. & Venter, J . C. 11992) Nature 355, 632-634 21. Maenz, D. D., Chenu, C., Breton, S. & Berteloot, A. (19921J. Biol. Chem. 267, 1510-1516 22. Makowske, M. & Christensen, H. N. (1982) J. B i d . Chem. 257, 14635-14638 23. Ericinska, M., Wantorsky, D. &Wilson, D. F. (1983)J . B i d .Chem. 258,90699077 24. Nicholls, D. & Attwell, D. (19901 Pends Pharmacol. Sci. 11, 462-468 25. Bouvier, M., Szatkowski, M., Amato, A. & Attwell, D. 11992) Nature 360, 471-474 26. Bussolati, O., Laris, P. C., Rotoli, B. M., Dall'Asta, V. & Gazzola, G. C. (1992) J. B i d . Chem. 267, 8330-8335 27. Olney, J. W., Zorumski, C., Price, M. T. & Labruyere, J. (1990) Scienre 248, 596-599