Regulatory Peptides, 47 (1993) 247-258 © 1993 Elsevier Science Publishers B.V. All rights reserved 0167-0115/93/$06.00

247

REGPEP 01527

A proposed bovine neuropeptide Y (NPY) receptor cDNA clone, or its human homologue, confers neither NPY binding sites nor NPY

responsiveness on transfected cells

Elena E. Jazin a, Heahyun Yoo b, Anders G. Blomqvist a, Frances Yee b, Gezhi Weng b, Mary W. Walker ~, John Salon ¢, Dan Larhammar a and Claes Wahlestedt b

a Department of Medical Genetics, Uppsala University, Uppsala (Sweden), bDivision of Neurobiology, Department of Neurology and Neuroscience, Cornell University Medical College, New York, NY (USA) and c Synaptic Pharmaceutical Corporation, Paramus. NJ (USA)

(Received 16 November 1992; revised version received 8 March 1993; accepted 10 April 1993)

Key words: Neuropeptide Y receptor; Binding site; Responsiveness; Cloning; Gene expression; Human

Summary

Receptors with seven transmembrane domains (7TM) constitute a large family of structurally and functionally related proteins which respond to various types of ligands. We describe here the cloning and expression of a human 7TM receptor, denoted hFB22 (human Fetal Brain 22), which is the homologue (92~o amino acid identity) of a bovine receptor (LCR 1) reported by others to bind neuropeptide Y (NPY) with a pharmacological profile of the Y3 receptor subtype. However, upon expression in COS 1 (confirmed by Northern analysis), COS7 or CHO-K1 cells, the hFB22 receptor did not confer specific ~25I-Bolton-Hunter-NPY, 3H-propionyl-NPY or 125I-peptide YY (PYY) binding sites, in either intact cells or in membrane preparations. Similarly, cells trans- fected with the corresponding bovine clone (LCR1) did not show specific NPY/PYY binding exceeding that resulting from endogenous binding sites; mock-transfected COS7 cells, used frequently for heterologous ex- pression of receptors, were found to have endogenous specific 125I-NPY binding sites (Bma x = 112 fmol/mg protein; K d = 0.25 nM). Moreover, the hFB22 transfected cells, when compared to control transfected cells, did not display de novo NPY- or PYY-induced second messenger responses, i.e., (1) inhibition of forskolin- stimulated cAMP accumulation or (2) 45Ca2 ÷ influx. The presence of hFB22 mRNA was detected in several human neuroblastoma cell lines, none of which was found to express Y3-1ike NPY binding sites, hFB22 dis- plays 39~o amino acid sequence identity (in the transmembrane regions) to the human interleukin-8 receptor, and 32-36~o amino acid identity to the human receptors of angiotensin II, bradykinin, and n-formylpeptide,

Correspondence to." C. Wahlestedt, Division of Neurobiology, Cornell University Medical College, 411 East 69th Street, New York, NY 10021, USA.

248

but only 23 ~o amino acid identity to the previously described human NPY/PYY receptor of the Y1 receptor subtype. Our results show that hFB22 and LCR1 do not encode NPY receptors, and their true ligand(s) re- mains to be identified.

Introduction

G-protein-coupled receptors are cell surface re- ceptors, possessing seven membrane-spanning do- mains, which bind a variety of hormones, neurotrans- mitters and other ligands that function in the endocrine, neural, sensory or immune system [1]. Neuropeptide Y (NPY) is an amidated 36-amino acid peptide that belongs to a family of peptides which includes peptide YY (PYY) and pancreatic polypep- tide (PP) [2-5]. Pharmacological studies have iden- tified at least three distinct NPY receptors [3-8] which are also members of this 7TM receptor fam- ily. The Y1 receptor binds NPY and PYY with equal affinity and is the only receptor subtype which binds the NPY analogue [Leu 31, Pro34]NpY. The Y2 re-

ceptor also binds NPY and PYY with similar affini- ties, but it also binds long C-terminal fragments of either peptide, while the Y1 receptor does not. The main characteristic of the Y3 receptor is that it binds NPY with at least 100-fold higher affinity than PYY [6-8].

Recently, a cloned bovine 7TM receptor was re- ported to bind NPY and it was proposed to corre- spond to the Y3 subtype [9]. As we have recently cloned the human Y1 receptor [10], we decided to isolate clones encoding the human homologue of the suggested bovine Y3 receptor [9] in order to further characterize the properties of the NPY receptor sub- types. We describe here one of these human clones, denoted hFB22. We show that neither the human nor the described bovine clone [9], when transfected into mammalian cells, gives rise to receptors which bind NPY. Furthermore, neither receptor responds func- tionally upon addition of NPY. The rat homologue

of this receptor is expressed within both the immune and nervous system [9,11] perhaps suggesting a functional role in both.

Materials and Methods

PCR amplification of a human fetal brain cDNA library The cDNA library was constructed from mRNA

of a human female fetal (17-18 week gestation) brain in the bacteriophage vector ZapII (Stratagene) using both oligo(dT) and random-sequence primers. De- generate oligonucleotides were designed from the se- quence of a bovine receptor clone LCR1 [9]. The 5' primer was a 32-mer with the sequence CAGAA- (G/A)AA(G/A)CTAAG(G/A)AG(C/T)ATGACG- GA(C/T)AA(G/A)TA corresponding to positions 199-230 of the bovine receptor cDNA clone. The 3' primer was a 30-mer complementary to positions 823-852 and had the sequence CCA(C/T)TT(G/A)- TGCACAGT(G/A)CT(C/T)TC(G/A)AA(C/T)TC- (G/A)CA. DNA prepared from the human bulk- grown cDNA library was used as template for the PCR reactions with the following conditions: 2 min at 94°C, 3 min at 55°C and 3 min at 72°C for 35 cycles.

Screening of a human fetal brain cDNA library The library was screened with a 32p-labeled 650-bp

human PCR fragment obtained as described above. Hybridization was carried out for 16 h at 42°C in 25~o formamide, 6 × S S C , 10~o dextran sulfate, 5 x Denhardt's solution and 0.1 ~o SDS. Filters were washed twice for 5 min at room temperature and twice for 30 min at 42°C in 2 x SSC/0.1~o SDS.

DNA sequencing Phagemids were rescued from the lambda vector

and transfected into XL-1 Blue bacteria as described by Stratagene. Three overlapping clones and two in- ternal PCR products were sequenced. Two internal synthetic sequencing primers were used to confirm the sequences of the PCR products. Sequence deter- minations were performed with dideoxy chain termi- nation [12] using an automated fluorescent-dye DNA sequencer (Applied Biosystems) or [~-35S ]dATP followed by autoradiography. The se- quence was determined manually and/or by auto- mated fluorescent sequencing in at least three differ- ent clones in each region.

Northern hybridization COS 1 cells were transfected by a DEAE dextran

procedure (see below). 48 h later, the cells were har- vested and total RNA was extracted by use of guani- dinium isothiocyanate and 18 h ofultracentrifugation on cesium chloride gradients. This RNA was run on a 6~o formaldehyde gel and transferred to a nitro- cellulose filter. Hybridization was with the random primed, [32p]dCTP labeled, human PCR fragment, described above, in the presence of 50~o formamide at 42°C. Washes involved use of 1 × SSC and 0.1~o SDS at 42 ° C for 10 + 30 min, followed by 0.1 × S SC and 0.1~o SDS at 55°C for 30 min. The filter was then dried and exposed to film for 60 min.

Reverse transcriptase-PCR (R T-PCR) RNAs, total or messenger, were prepared from

several cell types and first strand cDNAs were syn- thesized by use of RT (MMLV; Gibco/BRL). The conditions for PCR were essentially the same as those described above.

Southern hybridization Genomic DNA was purified from leucocytes and

digested with restriction enzymes. The probe and washes were as described for the Northern hybrid- ization. Hybridization was done for 5 h at 65°C in

249

QuickHyb (Stratagene) solution as described by the manufacturer.

Functional expression The entire cDNA inserts of clone hFB22 (human

Fetal Brain 22) and hFB 18, excised with EcoRV and SmaI, were transferred in both orientations to the eukaryotic expression vector pCDM8 (Invitrogen) digested with Xho I and blunt-ended. The orientation of the clones was confirmed by PCR using combi- nations of vector and insert primers. The resulting pCDM8/hFB22 and pCDM8/hFB18 constructs were used for transfection into COS 1 or COS7 mon- key kidney cells, or Chinese hamster ovary (CHO- K1) cells, grown either in 150 mm (for radioreceptor binding studies on cell membranes or whole cells in suspension) or 35 mm (for radioreceptor binding to intact, attached cells or for second messenger analy- ses) dishes. In addition to the human clones, we also transfected the bovine LCR1 clone (a kind gift of Dr. Ronald S. Duman, Yale University). A cDNA with the identical nucleotide sequence as that reported for LCR1 was isolated at Synaptic Pharmaceuticals and used for some of the present experiments. Both bo- vine clones used in our transfection experiments had been subcloned into the pcDNAI vector (Invitro- gen).

Plasmid DNA was transfected into COS 1, COS7 or CHO-K1 cells over 3.5 h using DEAE/dextran in D MEM with 10~o Nuserum/5~o CO2 as described [10]. In a separate series of experiments, cells (only COS1) were transfected with Lipofectin R (Gibco/ BRL) according to the manufacturer's instructions. Throughout this study, control cells were either mock-transfected (use of no plasmid or plasmid without insert) or transfected with rat type I angio- tensin II receptor, pBa23.i4, in pCDM8 (a kind gift of Dr. T.J. Murphy, Emory University); the latter receptor clone was used as a negative control be- cause of its relatively high homology with the herein described 7TM receptor clones (see below). As a positive control for NPY/PYY binding sites and sec-

250

ond messenger responsiveness we have used the human Y1 receptor, in a pCDM8 construct [ 10].

Within 60 h, the cells were studied with respect to NPY/PYY binding sites, either in situ or following harvesting with subsequent preparation of mem- branes. Radioligand binding assays were based on previously described methods [4,8,10,13], described here briefly. For intact cell binding assays, cultured cells were washed three times with binding buffer (141 mM NaC1, 20 mM Hepes, 5.4 mM KC1, 0.44 mM KH2PO4, 1.26 mM CaCI2, 0.81 mM MgSO4; pH 7.4). Incubations of the intact cells with radioli- gand were carried out for 45 min at 37°C in 1 ml supplemented binding buffer (binding buffer plus 1 mM dithiothreitol, 0.1 ~o bacitracin and 0.1 ~o bovine serum albumin). Incubations were terminated by washing three times with 3 ml ice-cold binding buffer and exposure to 1 + 0.5 ml of lysis buffer (3 M ace- tic acid with 8 M urea and 2 ~o, v/v, Nonidet P-40). For membrane binding assays, cultured cells were washed twice in ice-cold phosphate buffered saline (pH 7.4) and lysed by sonication in ice-cold buffer (20 mM Tris-HC1 and 5 mM EDTA; pH 7.7), fol- lowed by low-speed centrifugation (200 g, 5 min, 4°C). Membranes were collected from the super- natant fraction by high-speed centrifugation (35,000 g, 20 min, 4 ° C). The final membrane pellet was sus- pended into supplemented binding buffer as de- scribed above, except that [NaC1] = 10 mM for ex- periments with COS7 membranes. Incubations with radioligand were carried out for 120 min at 22°C or for 60 min at 30 °C for experiments with COS7 mem- branes. Incubations were terminated either by filtra- tion over Whatman GF/C filters (pre-coated with 0.5~o polyethyleneimine and air-dried) or by centri- fugation for 5 min in an Eppendorf microfuge (10,000 g, 5 min). Isolated membranes were washed with ice-cold buffer.

As radioligands, we have used (1) [125I]Bolton- Hunter-NPY (New England Nuclear, NEN or Amersham), (2) [3H ]propionyl-NPY (Amersham) or (3) [125I]PYY (NEN). The radioiodinated com- pounds were present in final concentrations ranging

from 10-12 M to 2 .10-9 M while [3H]propionyl-

NPY was used in concentrations up to 7.5.10- s M. Non-specific binding was defined in the presence of 0.1-1 /~M porcine NPY and/or porcine PYY. The binding data were analyzed by Graphpad or Lundon Software.

Intact, attached COS 1 cells were also studied with respect to 45Ca2 + (Amersham) uptake or cAMP ac- cumulation (by radioimmunoassay from Advanced Magnetics) as detailed previously [4,8,10]. In these second messenger studies, the human Y1 receptor clone, hY1-5, was used as a positive control.

Peptides, which were either human or porcine spe- cies, were obtained from Bachem California or Pen- insula.

Results

PCR amplification of the human homologue of the bo- vote LCR1 clone

A human fetal brain cDNA library was bulk-grown and DNA was prepared to be used as template for the PCR reactions. Degenerate oligonucleotides de- signed from the bovine LCR 1 receptor sequence were used as primers for PCR amplification. This ap- proach yielded a 650 bp fragment which was sub- cloned into pBluescript SK + II and sequenced. This fragment coded for the central part of the human homologue of the bovine clone, LCR1, as it displayed approx. 90 ~o nucleotide sequence identity to LCR1.

Isolation of full length human receptor cDNA clones The PCR product was used to screen 200,000

plaques of the human fetal brain cDNA library. 24 human clones were isolated and rescued from the phage vector and of these, three independent over- lapping clones, hFB22 (human Fetal Brain 22), hFB 18, and hFB 14, were used for DNA sequencing. Clone hFB22 has an insert of 1.3 kb that includes the entire coding region of the human homologue of LCR1 [9]. It contains an open reading frame of 353 amino acids (Fig. 1), and the predicted hFB22 se-

251

1 M E G I S I Y T S D N Y T E E M G S G D Y D S M K E P C F R E E 32 1 CGCATCTGGAGAACCAC~ACCATGGAGGGGATCAGTATATACACTTCAC~TAACTACA CCCaAGGAAA CTATGACTCCATGA~CCC~CC~A 120

TMI 33 N A N F N K I ~[~.~£~.~.~/....~Y....~.~...~....[...~...~..~..G....~....V..~..G....~.~...G....~..~..V..~..~.....V..~.M G Y Q K K L R S M 72

121 AAATC, CTAATI~TAAAA CCATCTACT~TCATCTTCTTAACTGGCA~I, GT~TGGAq-~TCATCCT~CATGGGTTACCAGAAC~CTGAGAAC~T 240 TM2 TM3

73 T D K Y R r. H I~...~....M...K...D...~I~...L...E....V....~...~..~./~...~....E...~.~/x...~V....D....~....V A N W Y F G N F L C K Ik....V.. 112 241 GACGGACAAGTACAGG~CCTGTCAGT~CCTCCTCT~CATCACGCTTCC~GTTGATGC(~GTC~CTGGTAtq~-~~A~T 360

113 ..~i....V....~...X....~.....V....F~...L...X....~....Z"..~%L~..~...~i...~...~̀ ...IL...I~..~...~ D R Y L A I V H A T N S Q R P R K L 5 A 152 361 CCA"/~TCATCTACACAGTCAACCTCTACAGCAGT~CCTCATC~CTTCATCAGTCTGGAC~ACCT~TC~TC(~%CC~CCACCAACAGTCAG~~ 480

TM4 153 E K y.....V....X....y.....G.....V....~.....~...E...~k...b....~....L....~....I....E...D....E...~-..E...~ N V S E A D D R Y I C D R F Y P N 192 481 TGAAAAGGTGGTCTATG~~CCTCCTGCTC4%CTATrCCCGACTrCATt~I'I-It~CCAA(2GTCAG~TG ACAGATATATCTGTGA~CTACCCCAA 600

TI6 193 D L W V V V F Q E...~...K...~...~...y.....q...~...~...L...E...G.-..~...~...~...I~...~.....~....X.....~ I I I S K L S H S K G H 232 601 TC~CTT~TGGGTGGTTGTGTTCCAGTTTCAGCACATCATGGT~CITATCCTGCCTGGTA%~`TCATCCTGT~A~AT~ CACTCCAAC~ 720

TM6 233 Q K R K A L K T Z....M...~...I~-..~...~.../k...E...L..~̀ ....~....~.....I~...E...2(..-.~....~I-....G....~...Z....I D S F I L L E I I K Q 272 721 CCAGAAG~GGCCC~C~C~GTCATCCTCATCCTGGCT~CTTCGC ACTA(I~TI~TCAGCATCC, ACTCC~TCCTC~TCATCAAGC_A 840

TM7 273 G C E F E N T V H K w I S ~...Z.-..~...~x...L...i~...E...L../~....~.....~....I*...~..-.~..../....~*...x....~...E...~ G A K F K T S 312 841 AGGGTG'TCaA~GAACACTG'TGCACAA~TI'IXTCATCA CL-V_g~GGCC CT~CT~I-I~'~L"/V, AA CCCV.ATCCTCTA TI'TAAAACCTC 960

313 A Q H A L T S V S R G S S L K I L S K G K R G G H S S V S T E S E S S S F H S S 352 961 TGCCCAGCA(3C~CTCACCTC~GCAGAGGGTCCAGCCTCAAGATCCTCTCCAAA CATTCATCTGT~CCAC~ U-I-I'FICACTCCAG 1080

353 * 1081 CTAACACAGATGTAAAAG~c~T~.F~.F~.~-~TACGATAAATAA~FFt.FF~.Ft~AGTTACACA.FF~-1.~CAGATATAAAAGAC~GA~TAT~TA~u~F~F~.~A~~F~.F~-~ 1200

1201 ~Tr'IL-'TITAG'I-I-I'FIuTG 1225

Fig. 1. Nucleotide sequence and deduced amino acid sequence of the human 7TM receptor clone hFB22. The predominantly hydropho- bic segments assumed to span the cell membrane are underlined. Two potential sites for extracellular N-linked glycosylation are under-

lined.

quence shows 92 ~o identity to that deduced from the bovine LCR1 clone (Fig. 2). 60~o of the 28 amino acid replacements occur in the amino-terminal ex- tracellular part, in t ransmembrane region 4 and the following extracellular loop. These regions are among the most variable between species homologues in the 7TM receptor family. One potential site for N-linked glycosylation is conserved between the bovine and

human sequences in the amino-terminal part. The human sequence has one amino acid less in the amino terminus as compared to the bovine sequence. We have also analyzed another clone, hFB18, with an insert of 1.8 kb. Both clones had identical sequences but they differ from one another with respect to the length of the 5' UT (untranslated) and 3' U T re- gions.

As shown in Fig. 2, comparisons of hFB22 with the NPY Y1 receptor revealed only 23 ~o amino acid identity in the segment that spans the seven trans- membrane regions. In contrast, hFB22 shared sig- nificantly higher, 32-39~o, amino acid identity with

a group of receptors for neutrophil chemoattractants. Finally, hFB22 was found to share 36~o identity with human angiotensin II type-1 receptor (Fig. 2).

Southern blot analysis of human genomic DNA Southern hybridizations with a hFB22-derived

probe to genomic D N A followed by high-stringency washes suggested that the human genome contains only one hybridizing gene (Fig. 3). D N A from two unrelated individuals displayed identical hybridiza- tion patterns for the three restriction enzymes used. Hybridizations done at lower stringency did not re- veal any closely related receptor genes in the genome (not shown).

Expression of the human clones hFB22 and hFBI8 and the bovine clone LCR1 in mammalian cells

The inserts of clones hFB22 and hFB18 were transferred to the mammalian expression vector, pCDM8, and used for transfection of mammalian COS1, COS7 or CHO-K1 cells. In some experi-

b~

% identity

to

H FB22

in TM region

H FB22

MEGISIYTSDNYTE

EMGSGDYDSMKEPCFREENANFNKIFLPTIYSIIFLTGIVGNGLVILVMGYQKKLRSMTDKYRLHLSVADLLFVIT

LPFWAVDAVAN

WYFGNFLCKEVHVIYTVN

B LCRI

92

.... R

-F ....... DDL ..................

H--R ..... V ............................................

~ ................

K .... A .

.......

H IL-8

39

MSNITDPQMWDFDDLN

FT-MPPA-EDYS--ML.TET

L--YVVIIA.ALV--LSLL--S--M..IL.SRVG..V..V.L.N.AL

..... ALo

..I-.ASK.NG

-I--T ....

HANG

36

MILN

SSTEDGIKRIQDD.PKAGRH.YIFVMI..L

..... VV.-F.-S.oVI.IYFYM..KTVASVFL.N.AL...C.LL.

.-L---YTAMEYR-P...Y.-.IASAEVSF.

H RMLP I

32

METNS-LPTNISGGTPAVSAGYL-LD.ITYLVFAVT-VL-VL

...... W-A.FRMTHTVT-IS-

-N-A---FC-TS

....

FM-RKAMGGH.P..W

.... F

-FT-VDI.

H Y1

23

.NSTLFSQVE~HSVHSNF~EKNAQLLAFENDDCHLPLAM.~TALA~GAVIIL~VS~LAL~IIILK~EM.NV~NILIVN~FS~VA~MC~TF.YTI~MDH

-V..EAM--LNPFVOC-S

D NPYrec

24

()TL°GLQFET.NITV~R~FSC.DYDLLSEDMWSSAYF~IVYML~IP~IFALI~T~CYIVYSTPRM~TV~NYFIAS.AIG~I~M~FFcE~SSFISLFILNY~P~LA.~HF~NY~A~S

H FB22

B LCRI

H IL-8

HANG

H RMLP

I

H Y1

D NPYrec

~DRYLAIVHATNSQRPRKLLAEKVVYVGVWIPALLLTIPDFIF

ANVSEADDR

YICDRFYPNDL

WVVVFOFOHIMVGLILPGIVILSCy¢IIISKLSHSKGHQKRK

• -...-.-°°°.---.-..-...-.-K.o.--...-.-

.... L--V°--..-L--

.DIK-V-E

.......

.-S..

.L-. ...... V---L-...,

...............

..y....

F..GI°L-°C°.V-..o..-..-RTLTQKRH-V

.~oCL-C.GLEMN.SL.F-L.

RQAYHPNNSSPV-YEVLG-,TAK,RM-LRILPHTF.F.V.LF.M-F.-GFTLRT.FKAHMG,.HR

• -A..FL-TCL-I-°

.... ~PMK.RLR~TM~VA~C~LL~G~ASL.A~HRNvFFIENTNITv.AFH~ESQNSTL~IGLGLTKN~L.FLF~FLI..TE~TL~WKA.KKAYEI..N~

• FG°.FLI.L.A..°CVCVL.PVWT.NHoTVSLA.K.II.P.V/4

..... L.VI-RVTTopGKTGT

VA(---)INVAV/L~ILT-RGIIRFII-FSA.M~IVAVE-GL-AT-IHKQGLIKSSR

ITV-IFS-VL-AVE-HQL-INPRGW

°.NNRH°

Y~GIAVI°VL~VAEEL~FL~Y~VMTD~PFQNVTL(---)QFPSDSHRLSYTTLLLVL~YFG~LCF~F~FK.YIR.KR~KM

VLV-AYT-VA-.I..-I.-MWPLKP

.IT-RY-TF

IIA.••FI..ATAL.IP•VSGLDIPMSpwHTK.(---)WPSRSQEYYYTLELFAL•FVV.LG•LIFT.AR.TIRVWAKRPPGEAE

H FB22

ALKTTVILI

LAFFACWLPYYI

GI S IDSF I LLE I I KQGCEFENTVHKWI

S GAKFKTSAQHALTSVSRGSSLKI

LSKGKRGGHSSVSTESESSSFHSS

B LCRI

..........

T .......................

Q ...... s ..........................................................................

H IL

8 .MRVIFAW-

r. LL .

.... NLVLLA. TLMRTQV.

QET, - RR- N

IGRALDA-

• I- G- L- S .

.... I • • • I • ON • RHGFLKI

• AM}IGLV • KEFLARHRVTSYT-

• SVNV. SNL

H ANG

PRNDDIFoIIMAIVoF..FS.I.HO-FTFL-VL-QoG-oRD

-RIADI,DTAMP-°ICI.Y-NN

.... LF.G°-°K-°oRYFLQLLKYIPPKAKSHSNLSTKMSTL.YRPSDNV-,-TKKPAPCFEVE

H RMLP

I p. RVLEFVAA.

• -L- -S- .OVVAL-ATVRIR.

LLQGMYK

• IGIA

VDV-E .... .NS- • • .M- -

V.~.QD.RERLI-

• • PASLERALTEDSTQTSDTATNSTLPSA-VELQAK

H Y1

RDNKYRSSETKR

INIMLLS TW.

• AV • • • • LT - FNTVFDWNHQI

- ATCNHNLLFLLC~LT.M~T~V°~F°~.~NKN.QRDL~FFFNFCDFR.RDDDYETIAMSTM}~TDVSKTSLKQASPVAFK(---)

D NPYrec

(---)RSKR

..

..

..

•

LQ--LNDEEFAHWDPLPYV.FAFHW

--MSH--Y.--I-CYMN-R-RSGFVQLMHRMPGLRRWCC-RSVGDRMNATSG.GPALPLNRMNTST(---)

Fig.

2.

Am

ino

acid

seq

uenc

e al

ignm

ent

of s

elec

ted

G-p

rote

in-c

oupl

ed r

ecep

tors

. T

he h

uman

hF

B22

rec

epto

r cl

one

serv

es a

s m

aste

r se

quen

ce.

Onl

y po

siti

ons

whi

ch

diff

er fr

om t

he h

uman

hF

B22

seq

uenc

e ar

e di

spla

yed

whi

le d

ots

indi

cate

am

ino

acid

s id

entic

al t

o th

e m

aste

r se

quen

ce.

Das

hes

wit

hin

pare

nthe

ses

and

spac

es i

n-

dica

te g

aps

intr

oduc

ed t

o op

tim

ize

alig

nmen

t. H

ydro

phob

ic s

egm

ents

, lik

ely

to b

e em

bedd

ed in

the

cel

l m

embr

ane,

are

und

erlin

ed.

For

com

pari

son,

the

seq

uenc

es

are

also

sho

wn

for

the

sugg

este

d bo

vine

NP

Y r

ecep

tor

(B L

CR

1; R

ef.

9) a

nd h

uman

rec

epto

rs f

or i

nter

leuk

in-8

(H

IL

-8;

Ref

. 16

), a

ngio

tens

in I

I ty

pe-1

(H

AN

G;

ref.

18),

n-f

orm

ylpe

ptid

e (H

RL

MP

I;

Ref

. 17

), N

PY

rec

epto

r of

the

Y1

subt

ype

(H Y

1; R

ef.

10)

and

a pr

opos

ed D

roso

phila

NP

Y r

ecep

tor

(D N

PY

rec

.; R

ef.

19).

The

per

cent

age

of a

min

o ac

id i

dent

ity,

in t

he s

egm

ent

span

ning

the

sev

en-t

rans

mem

bran

e do

mai

n re

gion

, be

twee

n ea

ch r

ecep

tor

and

the

hum

an h

FB

22 r

ecep

tor

cand

idat

e de

scri

bed

here

in i

s in

dica

ted.

253

HINDIII XBAI TAQI 1 2 l 2 1 2

A B C D

kb

14.1

7.2

5.7

4.3

0 . 7 n

Fig. 4. Northern blot analysis of total RNA of COS 1 cells trans- fected with: (A and B, corresponding to two independent experi- ments) a combination of the human hFB22 (transcript size: 1.8 kb) and hFB18 (1.4 kb) pCDM8 constructs; (C) the bovine LCR1 clone in pcDNAI; and (D) the control pCDM8 vector with an unrelated insert (rat angiotensin II type-1 receptor clone). The cells had been harvested 48 h following the transfection and the probe was that described in Fig. 3. The autoradiographic film was

exposed for 60 min.

Fig. 3. Southern blot analysis of human genomic DNA. Each lane contained 10 #g ofgenomic DNA. Two independent samples were digested with the indicated restriction enzymes. The blot was hybridized with a 32p-labeled human hFB22-derived 650 bp frag-

ment under conditions of high stringency.

ments the two clones, hFB22 and hFB18, were

cotransfected, in case either clone did not give rise to the mature receptor protein. These same cell types were also used for transfection of the bovine LCR1 clone, previously suggested to encode an N P Y re- ceptor [9]. Fig. 4 shows Northern analyses o f R N A from transfected COS1 cells; this film was exposed for 1 h only, indicating a very high concentrat ion of receptor m R N A .

Radioreceptor binding studies In 12 independent experiments performed over

eight months, we have not been able to observe that COS1, COS7 or C H O - K 1 cells transiently trans- fected with hFB22 and/or hFB18, or LCR1, exhibit specific binding sites labeled by [ 1251] Bolton-Hunter-

N P Y (used by Rimland et al., Ref. 9), [3H]propionyl- N P Y or [ ~25I]PYY. Table I summarizes our attempts

to detect such binding sites. Throughout this work, we used the human N P Y Y1 receptor clone, hY1-5 [ 10], as a positive control; transfection of the latter clone consistently resulted in robust ( > 1 pmol /mg of total cellular protein) de novo expression of N P Y / PYY binding sites using either of the three aforemen- tioned radioligands.

COS cells exhibit specific F251]NPY binding sites We have found COS cells to consistently exhibit

small numbers o f high affinity [125I]NPY binding sites. These results, obtained by use of COS7 cells,

254

TABLE I

Summary of conditions for binding studies on transfected cells

Cell types transfected: COSI, COS7, or CHO-KI. cDNA clones: hFB22 and/or hFB18 (human) or LCR1 (bovine). Transfection procedures: DEAE dextran or lipofection. Cell preparations: intact cells, whole cells in suspension or cell

membranes. Radioligands tested: [1251]BH-NPY, [3H]propionyl-NPY or

[ 125I]PYY. Positive control: human NPY/PYY Y1 receptor. Negative controls: rat angiotensin II type-1 receptor or mock-

transfection.

are illustrated in Fig. 5. The Bma x values in COS7 cells have been in the range of 50-150 fmol/mg (Kd: 0.25-0.35 nM) of cellular protein, with lower ( < 50

fmol/mg protein) Bma x values found in C H O - K 1 cells.

hFB22/hFB18 or LCR1 transfected cells do not exhibit

de novo second messenger responses to N P Y

COS1 cells were transfected with the hFB22 and

120

lo01

8c

o = m 60

z 4(3 t

2(3

0.00

o

/

o

i i i i i i i

0 . 2 5 0 . 5 0 0 . 7 5 1 . 0 0 1 . 2 5 1 . 5 0 1 . 7 5 2 . 0 0

[1251]-NPY FREE (nM)

Fig. 5. Specific [ 12~l]Bolton-Hunter-NPY binding to membranes from COS7 cells transfected with the bovine LCR1 clone or mock- transfected cells. Data were fit to a 1-site adsorption isotherm model. For LCRl-transfected cells (filled circles, solid line), B m a x = 122 fmol/mg membrane protein, K~ = 0.34 nM. For mock- transfected cells (open circles, dashed line), B ..... = 112 fmol/mg membrane protein, Kd = 0.25 nM. Non-specific binding was mea-

sured with 10-7 M unlabeled NPY.

hFB18, or the LCR1, constructs. These cells were then studied with respect to the ability of N P Y to (1)

inhibit forskolin-stimulated c A M P accumulation and (2) elicit influx of 45Ca2+. Both these parameters

were affected by N P Y in positive control cells trans- fected with the human Y1 receptor clone, hY1-5 , as

previously reported [ 10]. However, cells transfected with hFB22/hFB 18 or LCR1 did not differ in second

messenger responses associated with N P Y (up to 1 #M) when compared to negative control cells trans- fected with the rat angiotensin II type-1 receptor clone, pBa23.i4, nor when compared to mock-

transfected cells. Inhibition o f c A M P accumulation. COS 1 cells,

transfected with h Y 1 - 5 or hFB22/hFB18, were treated with 500 nM N P Y with or without subse- quent 15 min stimulation with 5 # M forskolin. Our positive control, h Y 1 - 5 transfected cells, responded

to N P Y with a 57 + 6~o (with forskolin) and 42 + 7~o (without forskolin) reduction of c A M P concentra- tions (means+ S.E.M.; n = 4 - 6 ; P < 0 . 0 0 1 in both instances). In hFB22/hFB 18 transfected cells, c A M P accumulation values were not affected by N P Y (500

nM): 92 + 11 ~o (with forskolin) and 95 + 8 ~o of con- trol cells not receiving N P Y (n = 4-8) .

45Cae + influx. In hY1-5 transfected COS 1 cells, N P Y (1 #M) caused a 186 + 15~o increase of 45Ca 2+

influx over 4 min (n = 6; P<0 .001) . In parallel ex-

periments, the hFB22/hFB18 transfected cells did not respond to NPY, the influx of 45Ca2+ being

15 + 8 ~o lower than in mock-transfected COS 1 cells (n = 6-9 ; not significant).

Cultured human or bovine cells expressing the hFB22

or LCR1 m R N A do not exhibit N P Y Y3-type binding

sites

Many cell lines and primary cultures express N P Y receptors (e.g., Refs. 3,4). We therefore examined whether a panel of cultured human or bovine cells express, respectively, the m R N A corresponding to hFB22 or LCR1, and whether these same cells ex- hibit [125I]Bolton-Hunter-NPY and/or [125I]PYY

binding sites. The results of these experiments are

TABLE II

Presence of hFB22/LCR1 mRNA and/or NPY/PYY binding slites in cultured cells

255

hFB22/LCR1 mRNA Specific [125I]BH-NPY Specific [IzsI]PYY Y3-1ike-profile 3 binding I binding I

Human NB: LA-N-5 yes 3 yes yes no Human NB: SH-SY5Y yes 3 yes yes no Human NB: IMR-32 yes 3 no 5 no 5 no Human NB: SK-N-BE(2) yes 4 yes yes no Bovine adrenal chromaffin yes 4 yes no 5 yes Bovine endothelial (CAB) yes 4 no 5 no 5 no

Binding studies were performed using intact cells. 2 Specific binding obtained with [~25I]Bolton-Hunter-NPY but not with [ 125I]PYY (> 100-fold difference in potency; Ref. 8). 3 mRNA studied by Northern analysis (see Fig. 4). 4 mRNA studied by reverse transcriptase PCR using primers corresponding to the LCR1 sequence (data not shown). 5 Not detectable at 0.3 nM radioligand concentration. NB = neuroblastoma cells.

shown in Table II. Presence of m R N A was deter-

mined by either Nor thern analysis (see Fig. 6 in which

a range of human neuroblas toma cells are shown) or

reverse transcriptase PCR for the bovine m R N A (re-

suits not shown). It was found that no correlation

seems to exist between expression of hFB22/LCR1

1.8 kb !:>-

• ~ 2 8 S

B - - 18 s

o

m R N A and the presence of N P Y binding sites. With

respect to Y3-type binding sites (defined by high af-

finity to [~25I]Bolton-Hunter-NPY but not to

[125I]PYY), of the six cell types analyzed, only bo-

vine adrenal chromaffin cells appear to co-express

these binding sites with the LCR1 m R N A . Several

human neuroblas toma cell lines expressing the

hFB22-related m R N A did not display Y3-1ike bind-

ing sites. Moreover, IMR-32 did not show specific

N P Y or PYY binding sites while it expresses hFB22-

related m R N A (Fig. 6). Finally, bovine aortic endot-

helial cells (CAB) were found to contain the L C R I -

related m R N A while not expressing detectable [125I]Bolton-Hunter-NPY or [125I]PYY binding

sites (cf. Table II).

GAPDH 1.Skb D D O D Q •

Fig. 6. Northern anlysis of total RNA isolated from the five in- dicated human neuroblastoma cell lines. The same probe as that described in Fig. 3 was used in these hybridizations and another

probe, that for GAPDH, served as an internal standard.

Discussion

A bovine receptor clone called LCR1, which was

recently isolated from a locus coeruleus c D N A li-

brary, has been proposed to encode an N P Y receptor

[9], possibly of the Y3 subtype (Ref. 9, note added

in proof). Here we report the cloning and expression

256

of hFB22, the human homologue of LCR1. The pre- dicted amino acid sequence of hFB22 is 92~o iden- tical to that deduced from LCR1. Southern blot analysis performed at conditions of high and low stringency indicate that a single gene corresponds to hFB22 and that there exist no other closely related genes in the human genome. The ovine homologue of LCR1 has also been reported recently [11] and it differs in only two positions from the bovine amino acid sequence.

Four lines of evidence make it highly unlikely that hFB22, or its bovine or ovine homologues, encodes an NPY receptor. In summary, these are: (i) trans- fected cells did not bind or respond to NPY; (ii) in a panel of cultured cells, there was no correlation between expression of hFB22/LCR1 mRNA and NPY binding sites; (iii) the in vivo distribution of the rat homologue of hFB22 does not parallel that of any known NPY receptor; (iv) structurally, hFB22 and its species homologues are only distantly related to the well established NPY/PYY Yl-type receptor.

First, after transient transfection of the human and bovine clones into COS1, COS7 or CHO-K1 cells, neither clone gave rise to receptors which bind ra- diolabeled NPY (or PYY). Moreover, these trans- fected cells did not respond functionally to NPY, as shown by the absence of de novo second messenger responses. As a positive control throughout this study we have used a human NPY receptor clone of the Y1 subtype, which we recently cloned and characterized [10].

The second line of arguments against hFB22 or LCR1 encoding an NPY receptor of any subtype derives from our studies on a variety of cultured cells. We thus found that several human neuroblastoma cell lines express considerable amounts of the hFB22 mRNA. While some, but not all, of these cell lines expectedly were found to possess NPY/PYY binding sites, none of the lines showed a Y3-1ike profile, i.e., binding NPY, but not PYY, with high affinity.

Two bovine cultures, adrenal chromaffin and aor- tic endothelial cells, were shown to contain LCR1 mRNA. As previously reported [8] the chromaffin

cells possess Y3-1ike binding sites, and thus consti- tuted the only example of a cell type, used in this study, whose expression patterns would be consis- tent with the proposal [9] that LCR1 encodes a Y3- type NPY receptor. However, the endothelial cells did not exhibit any detectable NPY binding sites at all.

Third, with respect to the in vivo distribution of the mRNA corresponding to hFB22 and its species ho- mologues many discrepancies to the known distri- butions of NPY receptors also seem to exist. For example, species analogs of the ovine clone (A J-13) are reported to be highly expressed in mouse lym- phocytes and proliferative areas of developing rat brain [11 ]; this distribution pattern is not consistent with A J-13 encoding an NPY receptor. Moreover, in our own laboratory (Jazin et al., unpublished data), mRNA for the rat homologue ofhFB22 was detected by in situ hybridization in multiple neural tissues of neonatal rats including cerebellum, olfactory epithe- lium, retina, olfactory bulb and hippocampus. We also detected abundant hFB22-associated mRNA expression in rat thymus, i.e., another example of an organ in which NPY receptors have not been ob- served.

It should be mentioned that Rimland et al. [9] also pointed out that there are 'discrepancies between the regional distribution of LCR1 mRNA and that of total NPY ligand binding sites'.

Fourth, amino acid sequence analysis does not support the possibility of hFB22 being an NPY re- ceptor. Comparisons of hFB22 with the NPY Y1 receptor reveals only 23 ~o amino acid identity in the segment that spans the seven transmembrane re- gions. In most cases, receptors that bind the same or related ligands form a receptor family with 50-60 ~o amino acid identity. Thus, for example, the two human somatostatin receptors share 55~o identity [ 14] and the three tachykinin receptors are 58-67~o identical to each other in the part spanning the TM regions [ 15].

On the other hand, hFB22 shares 32 to 39~o amino acid identity in the seven transmembrane regions with

a group of receptors for neutrophil chemoattractants including interleukin-8 (IL-8), n-formylpeptide, and anaphylotoxin (Fig. 2). It is noteworthy that these chemoattractant receptors share between 29 to 35 ~o identity [ 16,17]. Therefore, human IL-8 receptor is more similar (39~o identity) to human hFB22 than to the other chemoattractant receptors. However, hFB22 also shares 367o identity with human angio- tensin II type-1 (A-II) receptor [18]. When analyz- ing the full length amino acid sequence, the IL-8 receptor still shows 39~o identity to hFB22, whereas the corresponding score for the A-II receptor drops to 34~o, perhaps suggesting that hFB22 is more closely related to the IL-8 receptor.

Thus, the identity of the gene product of hFB22, and its species homologues, remains unknown. In- terestingly, the mRNAs are extremely widely distrib- uted and exhibit both neuroendocrine and immune system expression patterns. The ligand, perhaps being a peptide other than NPY, is therefore likely to exert a wide variety of effects, modulating the activ- ity of neurons as well as endocrine and immune cells.

How is it then possible that LCR1 was identified as an NPY receptor [9]? The most conceivable ex- planation is that the cells these authors used for het- erologous expression studies possess small numbers of endogenous NPY receptors detected by the Novascreen R approach [9]. In the present work, we have studied the same cell lines as Rimland et al. [9], that is COS and CHO cells, and detected the pres- ence of [ 125I]NPY binding sites also in control cells. Since no second messenger responses were detected in the COS1 cells, it is unclear whether these (small numbers of) endogenous binding sites correspond to functional G-protein-coupled NPY receptors.

The Bma x for [125I]NPY binding of 64 fmol/mg total cellular protein reported by Rimland et al. [9] in LCR1 transfected CHO cells is quite low, indeed much (1-3 orders of magnitude) lower than that ob- served with most other cloned receptors. In fact, we found similar or even higher levels (> 100 fmol/mg of membrane protein) of endogenous [125I]NPY bind- ing sites in mock-transfected cells, cf. Fig. 5. Also,

257

synaptosomes prepared from rat forebrain harbor considerably greater densities of [ 125I]NPY binding sites [13] than the transfected CHO cells of Rimland et al. [ 9]. Importantly, however, unlike Rimland et al. [9], we have not been able to produce and select stably LCR1 transfected CHO cells, and direct com- parisons between this aspect present work and that of Rimland et al. [9] can therefore not be made.

In conclusion, hFB22 and its bovine homologue do not appear to encode a receptor for which NPY is a ligand. The true ligand, perhaps a peptide with many and varied effects, remains to be identified.

Note added in proof (received 16 June 1993)

We have now produced a CHO-K1 cell line sta- bly transfected with LFB22-containing pc DNAI/ Neo plasmid. These cells do not differ from mock- translated CHO-K1 cells with respect to [3H]NPY binding.

Acknowledgements

We thank Ulf Pettersson and Donald J. Reis for providing excellent working facilities. We are grate- ful to Staffan Svftrd for some of the oligonucleotides, to Irja Johansson and Sven P~lhlman for neuroblas- toma mRNA filters, and to Liubov Lyandvert for assistance with tissue culture. Drs. Theresa Branchek, Paul Hartig and Richard Weinshank are acknowledged for helpful suggestions on this project. This work was supported by grants to D.L. from the Swedish Natural Science Research Council (No. B-BU 8524-308), and to C.W. from the National Heart and Lung Institute (HL 18974) and the Na- tional Institute for Drug Abuse (DA 06805). E.E.J. is the recipient of a fellowship from the Swedish Institute. The nucleotide sequence data reported in this paper have been submitted to the GenbankTM/ EMBL Data Bank under the accession number L01639.

258

References

1 Strosberg, A.D., Structure/function relationship of proteins belonging to the family of receptors coupled to GTP-binding proteins, Eur. J. Biochem., 196 (1991) 1-10.

2 Larhammar, D., S6derberg, C. and Blomqvist, A.G., Evolu- tion of the neuropeptide Y family of peptides. In W.F. Colmers and C. Wahlestedt (Eds.), Neuropeptide Y and Related Pep- tides. Contemporary Neuroscience Series, Humana Press, Clifton, 1993, pp. 1-41.

3 Wahlestedt, C. and Reis, D.J., Neuropeptide Y-related pep- tides and their receptors: are these receptors potential thera- peutic drug targets?, Annu. Rev. Pharmacol. Toxicol., 32 (1993) 309-352.

4 Wahlestedt, C., Grundemar, L., H~tkanson, R., Heilig, M., Shen, G.H., Zukowska-Grojec, Z. and Reis, D.J., Neuropep- tide Y receptor subtypes, Y1 and Y2, Ann. NY Acad. Sci., 611 (1990) 7-26.

5 Michel, M.C., Receptors for neuropeptide Y: multiple sub- types and multiple second messengers, Trends Pharmacol. Sci., 12 (1991) 389-394.

6 Balasubramaniam, A., Sheriff, S., Rigel, D.F. and Fischer, J.E., Characterization of neuropeptide Y binding sites in rat cardiac ventricular membranes, Peptides, 11 (1990) 545-550.

7 Grundemar, L., Wahlestedt, C. and Reis, D.J., Neuropeptide Y acts at an atypical receptor to evoke cardiovascular depres- sion and to inhibit glutamate responsiveness in the brainstem, J. Pharmacol. Exp. Ther., 258 (1991) 633-638.

8 Wahlestedt, C., Regunathan, S. and Reis, D.J., Identification of cultured cells selectively expressing YI-, Y2-, or Y3-type receptors for neuropeptide Y/peptide YY, Life Sci., 50 (1991) 7-12.

9 Rimland, J., Xin, W., Sweetnam, P., Saijoh, K., Nestler, E.J. and Duman, R.S., Sequence and expression of a neuropeptide Y receptor cDNA, Mol. Pharmacol., 40 (1991) 869-875.

10 Larhammar, D., Blomqvist, A.G., Yee, F., Jazin, E.E., Yoo, H. and Wahlestedt, C., Cloning and functional expression of

a human neuropeptide Y/peptide YY receptor of the Y 1 type, J. Biol. Chem., 267 (1992) 10935-10938.

11 Reppert, S.M., Weaver, D.R., Stehle, J.H., Rivkees, S.A., Grabbe, S. and Granstein, R.D., Molecular cloning of a G-protein coupled receptor that is highly expressed in lym- phocytes and proliferative areas of developing brain, Mol. Cell. Neurosci., 3 (1992) 206-214.

12 Sanger, F., Nicklen, S. and Coulson, A.R., DNA sequencing with chain-terminating inhibitors, Proc. Natl. Acad. Sci. USA, 74 (1977) 5463-5467.

13 Walker, M.W. and Miller, R.J., [125I]neuropeptide Y and [ J25I]peptide YY bind to multiple receptor sites in rat brain, Mol. Pharmacol., 34 (1988) 779-792.

14 Yamada, Y., Post, S.R., Wang, K., Tager, H.S., Bell, G.J. and Seino, S., Cloning and functional characterization of a family of human and mouse somatostatin receptors expressed in brain, gastrointestinal tract and kidney, Proc. Natl. Acad. Sci. USA, 89 (1992) 251-255.

15 Okhubo, H. and Nakanishi, S., Molecular characterization of the three tachykinin receptors, Ann. NY Acad. Sci., 632 (199 l) 53-62.

16 Holmes, W.E., Lee, J., Kuang, W., Rice, G.C. and Wood, W.I., Structure and functional expression of a human interleukin-8 receptor, Science, 253 (1991) 1278-1280.

17 Boulay, F., Tardif, M., Brouchon, L. and Vignais, P., The human n-formylpeptide receptor. Characterization of two cDNA isolates and evidence for a new subfamily of G-protein- coupled receptors, Biochem. J., 29 (1990) 11123-11133.

18 Bergsma, D.J., Ellis, C., Kumar, C., Nuthulaganti, P., Ker- sten, H., Elshourbagy, N., Griffin, E., Stadel, J.M. and Alyar, N., Cloning and characterization of a human angiotensin II type 1 receptor, Biochem. Biophys. Res. Commun., 183 (1992) 989-995.

19 Li, X., Wu, Y., North, R.A. and Forte, M., Cloning, functional expression and developmental regulation of a neuropeptide Y receptor from Drosophila rnelanogaster, J. Biol. Chem., 267 (1992) 9-12.