Jan 25, 2016 - Julia E. Lever$. From the Department of Biochemistry and Molecular. Biology, University of Texas Medical School,. Houston, Texas 77225.
THEJOURNAL
Communication
OF
BIOLOGICAL CHEMISTRY
Vol. 268 No. 3 Issue of January 25 pp. 1509-1512 1993 for Biochemistry and Mblecular Bioloh, Inc.
0 1993 by The American So&
Printed in U.S.A.
nine kidney cell Na+/~yo-inositolcotransporter (8). These Na+-coupled transporters have several conserved amino acid residues in common with the rabbit renal Na+/phosphate cotransporter (91, as well as bacterial Na+/cotransporters for proline (IO),pantothenate (XI), and glutamate (12). Na+-coupled and Na+-independent transportersof distinct specificity mediate uptake of amino acids in mammalian cells (Received for publication, September 24,1992) (13). A Na+-independent transporterfor cationic amino acids, system y+ (14, 1.51, as well as putative regulatory subunits of Cheng-Te Kong, Shaw-Fang Yet, and a long-chain and aromatic neutral amino acid transporter, Julia E. Lever$ system L (Is),and of a transporter for cystine and dibasic From the Department of Biochemistry and Molecular and neutralamino acids, system bo,+(17,18),have beencloned Biology, University of Texas Medical School, and sequenced. These have no sequence homology with the Houston, Texas 77225 SGLT1-related family of Na+-coupled transporters. A Na+We describe the full-length sequence and functional dependent glycine transporter involved in glycinergic transexpression of a cDNA cloned from LLC-PK1 cells, mission was cloned from rat brain (19). In the present study we report the cloning and full-len~h which appears to encode a mammalian Na+-dependent neutral amino acid transporter with properties char- sequence of a Na+/amino acid cotransporter cDNA obtained acteristic of system A. This sequence, designated by low stringency screening of an LLC-PKI cDNA library for SAAT1, is 76%identical and 89%similar in amino sequences with homology to therabbit intestinalNa+/glucose acid sequence to the Na+-dependent glucose trans- cotransporter (SGLT1). SAATl exhibited 87% amino acid porter SGLTl of the same species. A leucine zipper sequence similarity to rabbit intestinal SGLT1. Expression region was detected in both SAATl and SGLTl. The of SAATl in COS-7 cells indicated that it encoded a Na+message for SAATl was a single 2.4-kilobase species dependent neutral amino acid transporter with specificity in kidney, but mRNA species of 2.4 and 3.7 kilobases characteristics of system A (13, 20). were observed in LLC-PK1cells as well asin intestine. Transcripts were alsofound in spleen, liver, and musEXPERIMENTAL PROCEDURES cle. Expression of SAATl in COS-7 cells resulted in cDNA Cloning and Sequencing-An oligo(dT)/random-primed increased levels of Na+-dependentuptake of 2-(methcDNA library in X ZAPII (Stratagene, La Jolla, CA) was prepared y1amino)isobutyric acid, a specific substrate for the from poly(A+) RNA isolated from LLC-PK, cultures and screened system A amino acid transporter. Uptake due to cDNA using the rabbitintestinalSGLT1 clone pMJC424 (2) obtained expression was inhibited by a range of amino acids that through the generosity of E. Wright (UCLA, Los Angeles, CA)under are transported by system A and exhibited a km of 0.8 low stringency conditions as described previously (21) with the omisf 0.2 mM. These results suggest that the system A sion of dextran sulfate. In order to obtain the 5' region, a second amino acid transporter is closely related to the Na+/ LLC-PK, h ZAPII library was prepared using both an internal primer based on the SAATl sequence 5'-CGCTGTTGAAGATGGAG-3' and glucose transporter SGLT1.
Cloning and Expressionof a Mammalian Na'/Amino Acid Cotrans~orterwith Sequence Similarity toNa'/Glucose Cotransporters"
an oligo(dT) primer, This library was screened as described above except a 528-base pair PCR' product corresponding to bases 6731201 of the SAATl sequence and labeled by random priming was used as probe. pBluescript SK(-) was excised from positive plaqueA number of different nutrient transport systems in mam- purified phage using an in vivo excision protocol (Stratagene). DNA malian cells utilize a Na+-dependent cotransport mechanism was purified by mini-prep alkaline lysis (22) with polyethylene glycol (1).Na+-coupledtransporters cloned thus far from both bac- precipitation (23). Double-stranded DNA sequencing was carried out the dideoxy chain termination method using Sequenase@version terial and mammalian sources share significant amino acid by 2.0 (United States Biochemical Corp.) and synthetic oligonucleotide sequence homology, suggesting a ~ n d a m e n t a lsimilarity in primers (Genosys). Compressions were resolved using deaza nucleothe Na+-coupling mechanism for transport of these diverse tides. The complete sequence was determined on both strands and substrates. Na+/glucose cotransporters (SGLTl) cloned from also confirmed by automated DNA sequencing. Northern Blot Analysis-Poly(A+) RNAwas isolated fromcell rabbit intestine (2) and kidney f3), human intestine (4), and LLC-PK1 cells ( 5 ) exhibit no sequence homology with Na+- cultures or tissues (24) and resolved and transferred by Northern blot (23) to a Duralon-UV" membrane (Stratagene). The filters were UVindependent facilitative glucose transporters (6) but do share cross-linked, prehybridized for 2 h, and then hybridized overnight at a striking amino acid sequence homology with a rabbit renal 55 "C in a solution containing 50% formamide, 5 X SSPE (20 X SSPE Na+/nucleoside cotransporter (7) and the ~ a ~ n - D ca~ b y= 3 M NaCl, 0.2 M NaH2P04,0.02 M EDTA-Naz, pH 7.4), 1 X P E (1 X PE = 50 mM Tris-HC1, pH 7.5, 0.1% sodium pyrophosphate, 1% * This work was supported by Public Health Science Grant DK SDS, 0.2% polyvinylpyrrolidone, 0.2% Ficoll, and 5 mM EDTA), and 274110. The costs of publication of this article were defrayed in part 150 pg/ml denatured salmon sperm DNA. Antisense RNA was tranbe hereby scribed from the SAATI-specific probe pCR-318 using T7 RNA by the payment of page charges. This article must therefore marked "aduertisement" in accordance with 18 U.S.C. Section 1734 polymerase (Stratagene),and blots were washed for 15 min in 0.1 X SSC, 0.1% SDS twice at room temperature, then once each at 45,55, solely to indicate this fact. The n ~ ~ a t isequencefs) de reported in t h ~ s p a has ~ r been s u b ~ i t t e d and 65 "C before exposure to film. PCR Analysis-Total RNA (1 pg) from either cell cultures ortissue to the G e n B a n k ~ / E M ~Data L Bank with accession number(s) samples was denatured at 95 "C for 10 min, then reverse-transcribed L02900. $ TOwhom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, University of Texas Medical School, ' The abbreviations used are: PCR, polymerase chain reaction; kb, P. 0. Box 20708, Houston, T X 77225. Tel.: 713-792-5600; Fax: 713kilobase(s); nMGP, n-methyl-o-glucopyranoside; MeAIB, 2-(methyl794-4150. amino)isobutyric acid.
1509
1510
Cloning of Cotransporter a Na'lAmino Acid
in a 20-pl reaction mixture containing 1 unit/pl RNasin (Promega), 1 mM dNTPs (Boehringer Mannheim), 7.5 pg/ml oligo(dT) (17-mer, Promega), 10 units/pl Moloney murine leukemia virus reverse transcriptase (GIBCO/BRL) in 1 X R T buffer. The mixture was held at room temperature for 5 min, then incubated at37 "C for 1 h. Termination was carriedout by heating a t 90 "C for 10min and then chilling on ice. Primers (20-mers) specific for porcine SAAT1, were as follows: sense, 5'-ATACTGGTCGTCCTGGCAAT-3'; antisense, 5'-GATGTTCACTATAGTCTTCC-3'.The reverse-transcribed sample was added to 80 p1 of a mixture containing 0.03 unit/pl Hot T u b polymerase (Amersham Corp.), 25 ng of each specific primer, and 1 X HotTubreaction buffer. The predominantPCRcDNA amplification product was predicted to be 318 base pairs, and the position was between amino acid residues 538 and 643. The product wasligated to the pCR-1000'" vectorusing a TA cloning kit (InVitrogen), andpositive clones were isolated and confirmed by doublestrand sequencing.pCR318, the LLC-PK1 SAAT1-specificvector, was obtained using this procedure. Expression of SAATl in COS Cells-Two partial SAATl clones spanning the entirecoding region were constructed into a full-length clone, pCTK, by ligation at their internal ApaLI site. A XhoI/XbaI restriction fragment containing either the full-length SAATl cDNA from pCTK (2 kb) or rabbit intestinal SGLTl cDNA (2.2 kb) from pMJC424 (aspositive control) was ligated into the NheIIXhoI site of the inducible expression vector pMAMneo (Clontech). Exponentially growing COS-7 cells grown in the samemedium previously described for LLC-PK, cells (25) were washed and resuspended at 7.5 x lo6 cells/ml in 0.8 ml aliquots containing 21 mM HEPES, pH 7.05, 137 mM NaC1,5 mM KC1,0.3 mM Na2HP04,6 mM glucose, and 180 pg of DNA in a electroporation cell (BTX, P/N 640, 4-mm gap), electroporated in a BTX Electrocell Manipulator 600, cooled on ice for 5 min, andimmediately plated a t a confluent densityon 35-mm dishes. Cultures were induced one day later with 2 pM dexamethasone, and transport was assayed after 24 h as described previously for uptake of a-methyl-D-glucopyranoside (26) except 100 PM [l-'4C]2-(methylamino)isobutyric acid (MeAIB) was used as substrate.
RESULTS AND DISCUSSION
Low stringency screening with the rabbit intestinal SGLTl cDNA of two different LLC-PK, cDNA librarieswas utilized t o isolate nine overlapping SAATl cDNA clones, from which a composite full-length cDNA nucleotide sequence was obtained by sequencing both strands. Thecomposite cDNA and deduced amino acid sequence of SAATl are shown in Fig. 1. T h e full sequence encodes a protein of 660 amino acids. The 3"untranslated region contains a consensus polyadenylation sequence, AAUAAA (underlined),but does not includea poly(A+) tail. Fig. 2 demonstrates the significant amino acid sequence similarity between SAATl and several SGLTl sequences. The LLC-PK, SAATl sequenceshows 74% DNA sequence homology, 89% amino acid similarity,and 76% amino acid identity over their region of overlap to the partial LLC-PK1 SGLTl sequence reported by Ohta et al. (5). Two residues, 576 and 613, were deleted in the porcine SAATl sequence compared with porcine SGLT1. LLC-PK1 SAATl shows 88% amino acid similarity and 75% identity with the human (4) intestinal SGLTl sequence and 87% similarity and 74% identity with rabbit intestine SGLTl (2). Furthermore,hydropathyplots of both SAATl and SGLTl were nearly identical (not shown). Our alignment indicates that the porcine SAATl sequence lacks 4 residues found in human intestinal SGLTl and 2 residues are missing compared with the rabbit intestinal SGLT1. We observed only one potential N-linkedglycosylation site indicated by the asterisk (*) in the SAATl sequence (Fig. 2), corresponding to By contrast, two potential N-linked and are conservedthe in glycosylation sites, at three SGLTl sequences, although it appears thatonly AsnZ4' is glycosylated in SGLTl (27). One apparent leucinezipper region, located at the loop between transmembrane domains7 and 8, was noted in allof the SGLTlsequences and in SAAT1,indicated by the vertical
FIG. 1. Composite nucleotide anddeduced amino acid sequence of S A A T l .
bars in Fig. 2. Aleucinezipper hasnot previously been reported for SGLTl sequences but has beenfound in the family of facilitative glucose transporters (28) and may be involved in transporter subunit oligomerization (29). Aminoacidresidues Gly380,Ala417, Gly426,and Arg427 were conserved in SAATl and all of the Na+ symporters, as well as the Escherichia coli Na+/proline (lo), glutamate (12) and pantothenate symporters(11),the rabbit Na+/phosphate (9), rabbit Na+/nucleoside (7), and Madin-Darby caninekidney cell Na+/myo-inositol transporter (8). Leu281was conserved in all except the proline cotransporter. This conserved region, designated the SOB motif (12), located next to the leucinezipper region may play an important role in Na' binding. Arg3°0 is alsoconserved in SAATl and the three SGLTl sequences and corresponds to A r P in the E. coli Na+/proline cotransporter shown by mutation to be implicatedin Na' bindingand energeticcoupling totheNa+ gradient (30).Asp2', the residue that is mutated in SGLTl of patients withglucose-galactose malabsorption(31), is preserved in SAATl and corresponds to Aspz5 in the sodium/ nucleoside cotransporter SNSTl (7). In order to preventcross-hybridization between the closely related sequences SAATl and SGLT1, an SAAT1-specific probe pCR318, which contained a 318-base pair insert corresponding to amino acidresidues 538-643, was utilized for Northern blot analysis. This region exhibited the most sequence diversity between the various Na+-coupled transport-
Cloning of a Na+/Amino Acid Cotransporter
1511
FIG.2. Alignment of deduced amino acid sequences for LLC-PK1 pig renal SAATl with SGLTl sequencesfrom LLC-PKlcells and rabbit and human intestine. The sequences are aligned against the top sequence shown, using the Genetics Computer group program GAP (38). Porcine SGLTl indicates the LLC-PK sequence (5); rabbit (2) and human (4) intestine SGLTl sequences are shown. Vertical bars indicate leucine zipper regions. The predicted SAATl N-linked glycosylation site is indicated by the asterisk (*). Gaps and identical amino acidresidues are indicated bv dots and dashes. remectively.
TABLE I Expression of SAA TI in COS- 7 cells Uptake was assayed at 24 h after addition of 2 p~ dexamethasone and values aremeans f S.E. of triplicatedeterminations. Na'stimulateduptake is shown aftersubtraction of Na+-independent uptake. Similar results were obtained in 3 independent experiments. Insert Na+-stimulated Na+-independent Vector MeAIB uptake MeAIB uptake cDNA
4.9kb 4 3.7kb
-.
2.4kb 4
nmollhlmp protein
1.4kb 4
1
2
3
4
5
6
FIG. 3. Expression of SAATl mRNA in LLC-PK cells and porcine tissues. Samples (1 pg) of poly(A+) RNA from each of the indicated sources were analyzed by Northern blot. The filter was hybridized with antisense RNA transcribed from the SAAT1-specific probe pCR-318. Lane I, LLC-PK, cells, confluent; lane 2, kidney; lane 3, intestine; lane 4, liver; lane 5,spleen; lane 6, muscle.
pCTK-MAMneo SAATl pMJC-MAMneo SGLTl None pMAMneo None" None COS-7 cells subjected to the addition of vector DNA.
4.0 f 0.02 1.8 f 0.06 2.0 f 0.3 1.9 f 0.1
0.2 f 0.02 0.2 f 0.005 0.3 f 0.1 0.1 f 0.008
electroporation procedure without
the region corresponding to amino acid residues 538-643 was identicalinsequence to that shown for porcineLLC-PKI SAATl in Fig. 1. A single 4.9-kb transcript was detected in liver, whereas skeletal muscle exhibited a predominant 1.4kb bandwith faintly visible bands of 2.4 and 4.9 kb. In spleen, predominant bandsof 2.4 and 4.9 kb were evident with afaint band of 3.7 kb (Fig. 3). The transcript sizes and tissue distriers. Control experiments established that the SAATl antibution of SAATl differ from those observed for SNSTl and sense probe transcribed from pCR318 did not hybridize with six other distinct partiallycharacterized cDNAs isolatedfrom sense RNA transcribed in vitro from an LLC-PKl SGLTl a rabbit kidney library by high stringency hybridization with template but did specifically hybridize with sense RNA tran- SGLTl (7). Southern blot analysis of LLC-PK, cell DNA scribed from an SAATl template under the same stringency using a SAAT1-specific probe revealed a different pattern of conditions used in the Northern blot. The antisense RNA hybridization from that obtained using a probe specific for probe transcribedfrom pCR318 detected both 2.4- and 3.7-kb SGLTl (not shown). bands by Northern blot analysis of LLC-PKI poly(A+)RNA In order to determine the function of SAAT1, the full(Fig. 3). Thiscell population was recloned twice from a single length cDNA including the start codon was subcloned into cell in order to avoid complications arising from the consid- the dexamethasone-inducible mammalian expression vector erable cellular heterogeneity of the parental LLC-PKI cell pMAMneo, which provided the missing poly(A+)tail. The line (32, 33). By contrast, an SGLTl probe detected tran- resulting construct, pCTK-MAMneowas used to transiently scripts of 2.2 and 3.9 kb in LLC-PKl cells (5).A single 2.4-kb transfect COS-7 cells by electroporation. Initial studies indiSAATl transcript was observed in kidney, while transcripts cated that pCTK-MAMneo was not able to express Na+of 2.4, 3.7, and 4.9 kb were observed in poly(A+) RNA from dependent a-methyl-D-glucopyranoside (aMGP)uptake in small intestine (Fig. 3). PCR analysis of poly(A+)RNA from COS cells ( 22. Hattori, M. & Sakaki, Y.(1986)Anal. Biochem. 162,232-238 cysteine, proline (34%) > glycine (25%). By contrast leucine, 23. Maniatis T., Fritsch, E. F. & Samhrook, J. (1989)Molecular Cloni A Laboritory Manual, 2nd Ed., Cold SpringHarbor Laboratory,?old glutamic acid or histidine did not inhibit uptake. The less Spring Harbor, NY 24. Davis, L. G., Dihner, M.D. & Battey, J. F. (1986)Basic Methods in specificnonmetabolizable analog 2-aminoisobutyricacid, Molecular Biology, Elsevier Science Publishing,Co., New York which is transportedby systems A, ASC, and L (13), exhibited 25. Yoneyama, Y. & Lever, J. E.(1988)Am. J. Physrol. 266,C816-C821 26. Amsler, K. & Cook, J. S. (1982)Am. J. Physiol. 242, C94-C101 a 50% increase in transport after expression of SAATl in 27. Mendlein, J., Hediger, M., Coady, M. & Wright, E. M. (1988)FASEB J. 2, COS-7 cells compared with mock-transfected cells. ~lnm .__"" White, M. K. & Weber, M. J. (1989)Nature 340,103-104 In summary, SAAT1, which appears to encode the system 28. 29. Chakerian A. E., Tesmer, V. M., Manly, S. P., Brackett, J. K., Lynch, M. A Na+/amino acid cotransporter, isclosely related insequence J., Hoh,'J. T. & Matthews, K. S. (1991)J. Biol. Chem. 266, 1371-1374 M., Mogi, T., Yamamoto, H., Yamato, I. & Anraku, Y. (1988)J. t o other members of the SGLT1-related gene family, which 30. Ohsawa, Bacteriol. 170,5185-5191 includes transportersfor various substrates including glucose 31. Turk, E., Zabel, B., Mundlos, S., Dyer, J. & Wright, E. M. (1991)Nature 360,354-356 and nucleosides. SAATl does not share anysequence homol- 32. Wohlwend, A., Vassalli, J.-D., Belin, D. & Orci, L. (1986)Am. J. Physiol. 260, C682-C687 ogy with any of the Na+-independent aminoacid transporters R. (1989)J. Cell. Physiol. 33. Van Den Bosch. L.. De Smedt, H. & Borghgraef, -~ clonedthusfar (14-19). Thebroadtissuedistribution of 141,483-489 J. I. &Johnstone, R. M. (1988)Proc. Natl. Acad. Sci. U. s. A. SAATl transcripts is consistent with the known ubiquity of 34. McCormlck. 85,787717881 system A, although the possibility of tissue-specific isoforms 35. Doyle, F. A. & McGivan, J. D. (1992)Biochem. J. 281.95-102 C.R., Mahon, B., Levy, H. L., Reade, T. M., Kronick, J., Lemieux, remains to be established. The protein components involved 36. Scriver, B. & Laber e C (1987) Am. J. Human Genet. 40,401-412 in the systemA transport process have not yet been unequiv- 37. Guidotti, G. 8,'Borghetti, A. F. & Gazzola, G. C. (1978) Biochirn. Biophys. Acta 516,329-366 ocally identified,although a candidate polypeptide of apparent 38. Devereux. J.. Haeberli. P.. & Smithies. 0.(1984)Nucleic Acids Res. 12, molecular mass 120-130 kDa copurified with system A trans'
