Cloning and Functional Expression of a Human Neuropeptide Y ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 267, No. 16, Issue of June 5, pp. 10935-10938,1992 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

both phenomena being characteristic of the mammalian Y l receptor.

Cloning and Functional Expression of a Human Neuropeptide Y/PeptideYY Receptor of the Y1 Type*

Neuropeptide Y (NPY)’ is a 36-amino acid peptide with a C-terminal amide (1)which displays a remarkable degree of structural conservation in evolution (2).It belongs to a peptide Dan LarhammarS,Anders G . BlomqvistS, family that also includes peptide YY (PYY) and pancreatic Frances YeeQ, Elena JazinS, Heahyun YooQ,and polypeptide (PP). MammalianNPYand PYY show 70% Claes WahlestedtQn sequence identity, while PP is 50% homologous to NPY and From the $Department of Medical Genetics, Uppsala PYY (1, 2). NPY is widely distributed in the brain, notably Uniuersity, Uppsala, Sweden and the §Department of “limbic” regions (3), and the peripheral (4)nervous system, Neurology and Neuroscience, Cornell University Medical and is often co-localized with norepinephrine, e.g. in sympaCollege, New York, New York 10021 theticperivascularnerve fibers (3, 4). Inthebrainmany effects, including stimulation of appetite, anxiolysis/sedation, Neuropeptide Y (NPY) and peptide YY (PYY) are and modulation of pituitary hormone release, have been atstructurally related peptides that primarily function tributed to NPY/PYY (3). Among the many peripheral acas neurotransmitter and gastrointestinalhormone, re- tions of NPY, most attention has been given to its vasoconspectively. Previous functional and bindingdata have strictor effects (5). indicated theexistence of at least three distinct recepBased on functional and binding data obtained from studies tor types, Y1, Y2, and Y3, for NPY and/or PYY in of various organs andcell types, we previously suggested that mammals. We describe here a human Y 1 cDNA clone, NPY/PYY receptors are heterogeneous (5-7), and the nohY 1-6,isolated from a fetal brain library. The human menclature Y1, Y2, and Y3 receptor type was introduced (5Y 1receptor consists of 384 amino acids and hasseven 9). The Y1 receptor binds NPY PYY and with similar affinity, putative transmembrane domains like other members as well asthesyntheticanalog[Pro”]NPYand analogs of the G-protein-coupled superfamily of receptors. In thereof (lo), but C-terminal fragments of NPY andPYY were the region spanning the transmembrane domains, the Y1 receptor displays 29% sequence identity to human shown to bind poorly (6). In contrast, while the Y2 receptor tachykinin receptors, but it only shows 21% and 23% also binds NPY and PYY with high affinity, the C-terminal homology with proposed bovine (LCRl) andDrosoph- fragments, e.g. NPY13-36 (5, 6), as well as a centrally trunila (PR4) NPY receptor clones, respectively. Northern cated analog, C2NPY (11)are only slightly less potent than blot analysis of a human neuroblastoma cell line, SK- the intact peptides at this receptor type. More recently, data N-MC, previously usedby many investigators as a from several laboratories (reviewed in Ref. 9) have indicated model system for studies on the Y 1 receptor, revealed the existenceof a Y3 type of receptor, whose main charactera single 3.6-kilobase mRNA species. Reverse tran- istic is that PYY shows markedly lower affinity than NPY (7). scriptase-polymerase chain reaction analysis indicated In order to address the structural and functional relationexpression also in human cultured vascular smooth ships of the NPY/PYY receptors we pursued the isolation of muscle cells, supporting theview that theY 1 receptor is associated with NPYJPYY-evokedvasoconstriction. receptor DNA clones using several strategies. We reporthere When expressed in COSl cells, hY1-5 conferred spe- the cloning of a putative human Y1 receptor cDNA clone. cific ‘*‘I-PYY binding sites with displacement patterns This clone, hY1-5, appears to be a human homolog of a characteristic of the Y 1 receptor, i.e. PYY 2 NPY 2 previously published rat “orphan” receptor, FC5 (12). The [Le~~’,pro~~lN >> P YNPY2-36 > C2NPY > pan- latter rat clone had caught our attention because its expresby in situ hybridization (12), was creatic polypeptide > NPY 13-36 > NPY 18-36. More- sion pattern, as studied reminiscent of that of the Y1 receptor protein, as shown by over, in the Y 1 receptor-transfected COSl cells, but not in type 1 angiotensin I1 receptor-transfected con- receptor autoradiography (13).We therefore generated a potrol cells, NPY and PYY accelerated 4SCa2+ influx and lymerase chain reaction (PCR) productcorresponding to the proceeded to isolate homologous inhibitedforskolin-stimulated CAMP accumulation, ratorphanreceptorand human cDNA clones. Herein we present functional evidence * This work was supported by National Heart and Lung Institute identifying one suchclone as a human NPY/PYY receptor of Grant HL 18974 (C. W.), National Institute for Drug Abuse Grant the Y1 type. (Received for publication, February 25, 1992)

DA 06805 (C. W.), and Swedish Natural Science Research Council Grant B-BU 8524-308 (D. L.). The costsof publication of this article were defrayed in part by the payment of page charges. This article must therefore behereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact. has been submitted The nucleotide sequence(s) reported in this paper to the GenBankTM/EMBLDataBankwith accession number(s) M88461. !I To whom correspondence should be addressed: Dept.of Neurology and Neuroscience, Div. of Neurobiology, Cornell University Medical College, 411 E. 69th St., New York, NY 10021. Tel.: 212-5702900; Fax: 212-988-3672.

EXPERIMENTALPROCEDURES

Screening of a Human Fetal Brain cDNA Library-The X ZAP11 cDNA library (Stratagene)was made from mRNA of a human female fetal (17-18 weeks of gestation)brain, using botholigo(dT)and The abbreviations used are: NPY, neuropeptide Y; PYY, peptide YY; PP, pancreatic polypeptide; TM, transmembrane; PCR, polymerasechainreaction; bp, basepair(s); kb,kilobase pair(s);SDS, sodium dodecyl sulfate; RT-PCR, reverse transcriptase-polymerase chain reaction.

10935

10936

Human Yl Receptor

random sequence primers. The librarywas screened witha 32P-labeled 500-bp human PCR product corresponding a to portion of the coding region of a rat orphan receptor, FC5 (12). This PCR product was obtained using the aforementioned cDNA library as template and using the following PCR conditions: 5 min at 95 "C for 1 cycle, then 1 min a t 93 'C, 1 min a t 45 "C and 2 min at 72 "C for 35 cycles. The 5' primer was a23-mer with the sequence TTCCAAAATGTATCACTTGCGGC corresponding to positions 547-569. The 3' primer was a 25-mer complementary to positions 1023-1047 and had the sequence TAGTCTCGTAGTCCGTCCGTCTCGA. Hybridization was carried out for 16 h a t 42 "C in 25% formamide, 1 M NaCl, 10% dextran sulfate, 5 X Denhardt's solution, and 1% SDS. Filters were washed twice for 5 min at room temperature in 2 X SSC, 0.2% SDS and twice for 30 min a t 42 "C in 2 X SSC, 0.5% SDS. DNA Sequencing-Phagemids were rescued from the X vector and transfectedintoXL-1Bluebacteriaas describedby Stratagene. Deletion subclones were generatedusingvariousrestriction sites. Four internal synthetic primers were used for sequencing (see below). Sequence determinations were performed with dideoxy chain termination(14) using anautomatedfluorescent dye DNAsequencer (Applied Biosystems) or [C~-~'SS]~ATP followed by autoradiography; 43% of the sequence was determined on both strands; 48% of the sequence was determined both manually and by automated fluorescent sequencing; 76% was determined either on both strands or with both sequencing procedures; 87% of the coding regions was determined either on both strands or with both sequencing procedures. Because the 3' primer initially predicted from the rat FC5 sequence (12) did not function optimally in our attempts amplify to DNA with PCR, we sequenced the rat FC5 clone (kindly provided by Dr. Rolf Sprengel, Heidelberg). The following three nucleotidedifferences from the published sequence were observed, numbered as in Ref. 12: position 1031, one G should be deleted in the published sequence; position 1034, one G should be deleted; position 1098, one G should be inserted in the published sequence. Two of these differences occur in the region corresponding to the 3' primer used for PCR. Northern Hybridization-After purification of mRNAs by standard guanidinium isothiocyanate/oligo(dT)-cellulosemethods, hybridization was as described above for the cDNA screening except that 2X SSC was used instead of 1M NaC1. Washes were done as above except that the final two washes were done at 65 "C in 0.2 X SSC, 0.1% SDS. The probewas a 1.4-kb XhoI-EcoRI fragment from hY1-5, corresponding to the second half of the coding region plus the 3'untranslated region. Southern Hybridization-Genomic DNA was purified from human leucocytes and digested with restriction enzymes. Probe, hybridization, and washes were as described for the Northern hybridization. Reverse Transcriptase-PCR (RT-PCR)-RNAs, total ormessenger, were prepared from several cell types and cDNAs were synthesized by use of R T (BRL). Using human primers, three forward (CTCTTGCTTATGGA/GGCTGTGA, TATGTAGGTATTGCTGTGATTTG, TATACCACTCTTCTC/TT/CTGGTGCTG) and one reverse primer complementary to the sequence TGGG/ATTCCTGAACAAAAACTTCCAG, corresponding toregions in the vicinity of T M 1, 4, 5, and 7, respectively, of thehumanY1receptor,the following PCR conditions were selected 5 min a t 95 "C for 1 cycle, then 1 min a t 93 "C, 2 min at 55 "C and2 min at 72 "C for 35 cycles. Functional Expression-The entire cDNA insert of clone hY1-5, excised with HindIII-NotI, was transferred to the eucaryotic expression vector pCDM8 (Invitrogen) digested with the same enzymes. The resulting pCDM8/hY1-5 construct was subsequently used for transfection into COS1 monkeykidney cells grown either in 150-mm (for preparation of cell membranes) or 35-mm(for second messenger analyses) dishes. Plasmid DNAwas transfected over 3.5 h using DEAE-dextranin Dulbecco's modifiedEagle'smedium with 10% 1 Nuserum, 5% C02. Control cells were transfected with rat type angiotensin I1 receptor, pBa23.i4, in pCDM8 (a kind gift of Dr. T. J. Murphy, Emory University, Atlanta). Within 60 h, the cellswere harvested and membraneswere prepared for 1251-PYY(Du Pont-New England Nuclear) binding, and incubated for 100 min a t 22 "C in a final volume of 0.5 ml with membranes from 2 X lo6 cells with other binding conditions asdescribed (5, 7). Alternatively, intact cells were studied with respect to "Ca" (Amersham Corp.) uptake or CAMP accumulation (by radioimmunoassay from Advanced Magnetics) as detailed previously (5,7). Peptides,which were porcine species unless otherwise stated, were obtained from Bachem California, Peninsula Laboratories, Richelieu, or Marion Merrell-Dow.

10937

250 I

mu.

VI net VI

80" WPVR

Hum SKU MU.

SSR1

210

2bO

11no

............1.........................

280

I..............

290

300

I unHOtInlcmH

310

In1.................1.......................

320

I..........

180 110 190 350 160 I I I I 1 I ~fonoLo~rFmfco~nsnooovf~~nns~nn~ousc~~Lcons~un~rt~waaoo

310

'p T ........9 ..q ' u ....... ------------- . . .. ....... ............ . . ...... . . . . ............ --------nnnnnoKnnomKvnssclK I

ImlnLLsIuun-FnucuLPL1ITI(U

LFLLcnLInnlslcumPIrvcfL

5

sn

-

S K C V O K ~ K ~ L K ~ I U ~ l l L ~ F F ~ C U l P V V l C ~ S ~ O S F l L l l l I O ORGL CI F~FFH EC SC L~ M U P I L V R F L C R K P K I S ~ O H ~ L I S U S ~ C S.S . .

... ...

~~UP~HO~HG~ML~HLO~KKKfUKl~ULUULrTRICYLPYHLVflLCSF~fOlVCHKFl~~UVl~LFUl~~SS~~VMPlIVCCL~H~f~SGF~L~~~CCPUUIP~t~OCL URLKRGUOORKRSERKltLnUnnUUnUFUlCYnPFVUU~lUMUF~f~OO~l

U S ~ L S U l l C V ~ M S C ~ ~ P l L V C F L S O ~ f K ~ S ~ O ~ l l C L S U ~ O ~ ~ ~ f l ~ U O

FIG.2. Amino acid sequence alignment of C-protein-coupled receptors. The human Y1 (Hum Y I ) receptor serves as master sequence. In the proposed rat Y 1 (Rat Y 1 ) sequence (see text), only positions which differ from the human sequence are indicated. Dashes represent gaps introduced to optimize alignment. Three underlined tripeptides at the N terminus conform to the consensus sequence for .Vlinked glycosylation. The hydrophohic segments likely to he embedded in the cell membrane are bordered with vertical h r s and ouerlinrd with dotted lines. TM,transmemhrane. For comparison the sequences are also shown for the proposed bovine N P Y receptor (Boo NI'YR, LCl), the human substance K receptor (HumS K R ) . and the human type 1 somatostatin receptor (HumSSRI).Amino acid residues which are identical to the human Y 1 sequence are indicated with dots. i.e. '2sI-PYY, binding, and 2) second messenger, i.e. Ca'+ and CAMP, analyses.As a negative control for all these assays,we u used identicalCOSl cells transfectedwiththerattype-1 f angiotensin receptor in the same pCDM8 expression vector; in such controlcells we observed little or no specific "'I-PYY m E or PYY. binding, and no second messenger responses to NPY Radioligand binding assays in membranes prepared from 14 the hYl-5-transfected cells indicate that the clone encodes a 7.2 protein with the pharmacological characteristics t-ypical of a 5.7 Y1 receptor type. The dissociation constant ( K d ) was 0.86 2 28s 0.09 nM ( n = 4; mean & S.E.), assuming a single-site fit and 4.3 equal affinity of (porcine) "'1-PYY and unlabeled (porcine) I, 3.7 PYY (16). This Kd is similar to that observed in SK-N-MC and othercell types (5,7-11). The pharmacological profile of 185 ligands competing for '"I-PYY binding to the expressed clone, illustrated in Fig. 4, isalsoconsistentwiththat of a Y1 2.3 of PYY 2 NPY 2 receptor (5-11). The potencyseries 1.9 [LeU",Pro"']NPY >> NPY2-36 > C2-NPY > (human) PP > NPY13-36> NPY18-36 was determined (Fig. 4); similar rank orders of potencyhave beenobservedin variousvascular (15) and SK-N-MC (7). Human NPY smooth muscle cells 1.4 was equipotent with porcine NPY (not shown). 1.3 Two second messenger responses frequently associated with FIG.3. Northern and Southern hybridizations. A, Northern Y1 receptorsare influx ofCa", which is not necessarily blot of human neuroblastoma cell lines prohed with a human Y1 of phosphoinoaitidase C, and inhiassociated with activation fragment. Each lane contains 15 pg of total RNA. R. Southern blot of human genomic DNA under conditions of high stringency. Each bition of cAMP accumulation (5. 7). Thus, we found 100 nM of NPY and PYY to accelerate influx of '"Ca'+, as studied lane contains 10 ~g of genomic DNA. over 2 min, by 135 f 17% and 157 +- 23% of control, respec(350, 520, and 850 bp) were also detected in human cultured tively (mean f S.E.; R = 6; two different experiments; p < 0.001) in hYl-5-transfected COSl cells; this is similar to the circumflex coronary artery smooth muscle cells. The latter observation is in agreement with our previous suggestion that case for endogenous Y1 receptors in SK-N-MC ( 7 ) . Control or PYY ( 100 t h e Y1 receptor is expressed in vasculature (5,6,15). Southerntransfected cells did not respond to either NPY hybridizations tohuman genomic DNA followed by high nM). Another well established characteristic of Y1 receptors, stringency washes (Fig. 3) suggest that the genome contains e.g. in SK-N-MC (5, 7), is the coupling to reduced cAMP accumulation. Likewise, stimulation of the de n o w expressed a single Y1 receptor gene. Y1 receptor by 100 nM NPY inhibited forskolin (5 p ~ ele) The insert of hY1-5 was transferred to the mammalian expression vector, pCDM8, and used to transfect COSl cells. vated accumulation of cAMP in the COSl cells by 47 f 5% Such transfectedcells were usedfor studieson: 1)radioligand, (mean 2 S.E.; n = 6; similar results obtainedin two different Northern blot

3.:

bouthern blot

Human Y l Receptor

10938

cDNA clones corresponding to the proposed bovine NPY receptor (LCR1) and arepresently examining whether it confers NPY and/or PYY binding sites on transfected cells, and, if so, whether it corresponds to a previously postulated NPY/PYY receptor type. In summary, we have demonstrated the cloning and identification of the human Y1-type NPY/PYY receptor. This receptor is thought tobe instrumental for the ability of NPY/ PYY to induce vasoconstriction as well as several behavioral effects (5).

log[competiior] (M)

FIG. 4. Ligand competition for '2sI-PYY binding in hY1-5 ( Y 1 receptor)-transfected C O S l cell membranes. Competition data are presented as a percentage of specific binding in the absence of competitor where each point is the mean k S.E. of two triplicate experiments. The concentration of '251-PYY was 0.1 nM. Each tube contained membranes (crudeparticulate fractions) from 2 X lo6COSl cells. Nonspecific binding was defined as binding in the presence of 1 WM unlabeled NPY.

experiments; p < 0.01). In the latter experiments, in which the phosphodiesterase inhibitor methylisobutylxanthine (500 PM) was present throughout, 100 nM NPY also reduced basal CAMPconcentrations from 240 f 14 to 123 f 4 pmol of cAMP/35-mm well (means f S.E.; n = 6; p < 0.001). The heterologously expressed human Y1 receptor described herein is thus similar to the endogenous Y1 receptor in brain (5, 13),and neuroblastoma (5, 7) and vascular smooth muscle cells (5, 15) with respect to binding and second messenger properties. Sequence analysis strongly indicated that the Y1 receptor belongs to theG-protein-coupled receptor superfamily. The human Y1 sequence is, however, only distantly related to thetwo proposed NPY receptors that have appeared in the literature very recently (17, 18). The portion of the sequence spanning theTM regions of hY1-5 shows only 21% and 23% identity with proposed bovine (LCR, Ref. 17; see Fig. 2) and Drosophila (PR4; Ref. 18) NPY receptors, respectively. In fact, as illustrated in Fig. 2, the Y1 sequence appears more closely related to the tachykinin receptors (29% sequence identity; Ref. 19) and is as similar to thehuman somatostatin type 1 receptor (23% identity; cf. Ref. 20) as to the bovine and Drosophila NPY receptors. Highly divergent sequences within ligand receptor families have also been reported for subtypes of amine receptors (21). However, no other peptide has previously been found to have receptor subtypes which display the degree of sequence divergence that exists between human Y1 (hY1-5) and bovine LCRl (and Drosophila PR4). For example, the three mammalian tachykinin receptors (19) and the two human somatostatin receptors (20) are 58-67% and 55% identical, respectively, over the regions spanning the T M segments. We have recently isolated human full-length

Acknowledgments-We are grateful to Professors Donald J. Reis and Ulf Pettersson for having provided excellent working facilities and for helpful discussions. We acknowledge Dr. Herbert Herzog for communication of resultsprior to publication. We thank Ingrid Lundell and Liubov Lyandvert for expert technical help, Sven Pihlman and Irja Johansson for mRNA filters, Staffan Svard for some of the oligonucleotides, and Elwood Reynolds for human smoothmuscle cells. D. L. and A. G. B. have contributed equally to this work. REFERENCES 1. Tatemoto, K., Carlquist, M. & Mutt, V. (1982) Nature 2 9 6 , 659660

2. Blomqvist, A. G., Soderberg, C., Lundell, I., Milner, R. J. & Larhammar, D. (1992) Proc. Natl. Acad. Sci. U. S. A,, in press 3. Wahlestedt, C., Ekman, R. & Widerlov, E. (1989) Prog. Neuropsychopharmacol. Biol. Psychiatry 1 3 , 31-54 4. Sundler, F., HBkanson, R., Ekblad,E., Uddman, R. & Wahlestedt, C. (1986) Znt. Reu. Cytol. 1 0 2 , 243-269 5. Wahlestedt, C., Grundemar, L., Hlkanson, R., Heilig, M., Shen, G. H., Zukowska-Grojec, Z. & Reis, D. J. (1990) Ann. Ai. Y. Acad. Sci. 6 1 1 , 7-26 6. Wahlestedt, C., Yanaihara, N. & HBkanson, R. (1986) Regul. Pept. 13,317-328 7. Wahlestedt, C., Regunathan, S. & Reis, D. J. (1992) Life Sci. 50, PL7-PL12 8. Sheikh, S.,O'Hare. M. M. T., Tortora, 0. & Schwartz, T. W. (1989) J . Biol. Chem. 2 6 4 , 6648-6654 9. Michel, M. (1991) Trends Pharmacol. Sci. 1 2 , 389-394 10. Fuhlendorff, J., Gether, U., Aakerlund, L., Johansen, N. L., Thogersen, H., Melberg, S., Olsen, U. B., Thastrup, 0. & Schwartz, T. W. (1990) Proc. Natl. Acad. Sci. U. S. A. 8 7 , 182186 11. Schwartz, T. W., Fuhlendorff, J., Kjems, L.L., Kristensen, M. S., Vervelde, M., O'Hare, M., Krstenansky, J. L. & Bjornholm, B. (1990) Ann. N. Y. Acad. Sci. 6 1 1 , 35-47 12. Eva, C., Keinanen, K., Monyer, H., Seeburg, P. & Sprengel, R (1990) FEBS Lett. 2 7 1 , 81-84 13. Aicher, S.A., Springston, M., Berger, S. B., Reis, D. J. & Wahlestedt, C. (1991) Neurosci. Lett. 1 3 0 , 32-36 14. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A. 7 4 , 5463-5467 15. Grundemar, L., Morner, J., Hogestatt, E. D., Wahlestedt, C. & HBkanson, R. (1992) Br. J. Pharmacol. 105,45-50 16. Walker, M. W. & Miller, R. J. (1988) Mol. Pharmacol. 3 4 , 779792 17. Rimland, J., Xin, W., Sweetnam, P., Saijoh, K., Nestler, E. J . & Duman, R. S. (1991) Mol. Pharrnacol. 4 0 , 869-875 18. Li. X.-J.. Wu. Y.-N.. North. A. & Forte.. M. (1992) . . J . Biol. Chem. 267,9-12 ' 19. Okhubo. H. & Nakanishi.. S.(1991) . . Ann. N. Y. Acad. Sci. 632. 53-62' 20. Yamada, Y., Post, S. R., Wang, K., Tager, H. S., Bell, G. I. & Seino, S.(1992) Proc. Natl. Acad. Sci. U. S. A . 8 9 , 251-255 21. Linden, J., Tucker, A. L. & Lynch, K. R. (1991) Trends Pharm a d . Sci. 1 2 , 326-328