Molecular Characterization and Expression of ... - Wiley Online Library

15 downloads 5852 Views 2MB Size Report
... University of Toronto; and §Addiction Research Foundation, Toronto, Ontario, Canada ..... systems, Gaithersburg) controlled by custom software writ- ... 2.0 (GraphPad Software). ..... development of subtype-specific selective agonists and.
Journal ofNeurochemistry Lippincott—Raven Publishers, Philadelphia © 1998 International Society for Neurochemistry

Molecular Characterization and Expression of Cloned Human Galanin Receptors GALR2 and GALR3 Lee F. Kolakowski, Jr., *GaJ.y P. O’Neill, ~Andrew D. Howard, Suzanne R. Broussard, tKathleen A. Sullivan, tScott D. Feighner, tMarek Sawzdargo, ~Tuan Nguyen, *Stacia Kargman, itin-Lin Shiao, lDonna L. Hreniuk, tCarina P. Tan, *Jilly Evans, *Mark Abramovitz, *Anne Chateauneuf, *Nathalje Coulombe, *Gordon Ng, Michael P. Johnson, Anita Tharian, Habibeh Khoshbouei, 1~SusanR. George, IRoy G. Smith, and 1~BrianF. O’Dowd Departments of Pharmacology and Biochemistry, University of Texas Health Science Center at San Antonio, San Antonio, Texas; ~Department of Biochemistry and Physiology, Merck Research Laboratories, Rahway, New Jersey, U.S.A.; * Department of Biochemistry and Molecular Biology, Merck-Frosst Centerfor Therapeutic Research, Kirkland, Quebec; tDepartments of Pharmacology and Medicine, University of Toronto; and § Addiction Research Foundation, Toronto, Ontario, Canada

Abstract: Galanin is a 29- or 30-amino acid peptide with wide-ranging effects on hormone release, feeding behavior, smooth muscle contractility, and somatosensory neuronal function. Three distinct galanin receptor (GALR) subtypes, designated GALR1, 2, and 3, have been cloned from the rat. We report here the cloning of the human GALR2 and GALR3 genes, an initial characterization of their pharmacology with respect to radioligand binding and signal transduction pathways, and a profile of their expression in brain and peripheral tissues. Human GALR2 and GALR3 show, respectively, 92 and 89% amino acid sequence identity with125l-galanin their rat homologues. Radioligand show that recombinant binding studies with human GALR2 binds with high affinity to human galanin (K 0 = 0.3 nM). Human GALR3 binds galanin with less affinity (IC50 of 12 nM for porcine galanin and 75 nM for human galanin). Human GALR2 was shown to couple to phospholipase C and elevation of intracellular calcium levels as assessed by aequorin luminescence in HEK293 cells and by Xenopus melanophore pigment aggregation and dispersion assays, in contrast to human GALR1 and human GALR3, which signal predominantly through inhibition of adenylate cyclase. GALR2 mRNA shows a wide distribution in the brain (mammillary nuclei, dentate gyrus, cingulate gyrus, and posterior hypothalamic, supraoptic, and arcuate nuclei), and restricted peripheral tissue distritution with highest mRNA levels detected in human small intestine. In comparison, whereas GALR3 mRNA was expressed in many areas of the rat brain, there was abundant expression in the primary 01factory cortex, olfactory tubercie, the islands of Calleja, the hippocampal CA regions of Ammon’s horn, and the dentate gyrus. GALR3 mRNA was highly expressed in human testis and was detectable in adrenal gland and pancreas. The genes for human GALR2 and 3 were localized to chromosomes 17q25 and 22q12.2-13.1, respectively. Key Words: Galanin receptor—G protein-coupled receptor—Cloning. J. Neurochem. 71, 2239—2251 (1998). 2239

Galanin is a widely distributed peptide with a broad spectrum of biological effects (Bartfai et a!., 1993; Bedecs et al., 1995). Galanin was originally isolated from pig intestine in 1983 during a search for C-terminally amidated peptides (Tatemoto et a!., 1983) and in subsequent years has been found in the CNS and the PNS of numerous species. The major biological properties of galanin relate to its modulation of hormone and neurotransmitter function with respect to nociception, learning, and memory and in feeding behavior. Galanin’s proposed role with respect to satiety is the demonstration that it increases food intake in rodents with a preference for fat diets (Tempel et al., 1988). It is interesting that central administration of neuropeptide Y, another proposed mediator of body weight homeostasis, up-regulates galanin concentrations, potentially augmenting galanin’s stimulation of food intake (Crawley et al., 1990). Galanin peptide antagonists have been shown to be effective in blocking galanin’ s promotion of food intake (Crawley et al., 1993). Because many diverse biological effects can be attributable to galanin, it was postulated early on that galanin receptor (GALR) subtypes exist (Bartfai et al., 1991; Hedlund et al., 1992; Wynick et al., 1993). Received April 1, 1998; final revised manuscript received June 24, 1998; accepted June 24, 1998. Address correspondence and reprint requests to Dr. A. D. Howard at Department of Biochemistry and Physiology, Merck Research Laboratories, Building RY-80Y-265, P. 0. Box 2000, Rahway, NJ 07065, U.S.A. The first seven authors are co-first authors. Abbreviations used: CHO, Chinese hamster ovary; FISH, fluorescent in situ hybridization; GALR, galanin receptor; GPCR, G protein-coupled receptor; HTGS, high-throughput genomic sequence; SSC, standard saline citrate.

L F. KOLAKOWSKI, JR., ET AL

2240

The first GALR, termed GALRI, was cloned from rat insulinoma, rat brain, and human melanoma cells (Burgevin et a!., 1995; Habert-Ortoli et a!., 1994; Parker et a!., 1995). Subsequently, cDNAs encoding a second rat GALR subtype, termed GALR2, was isolated and characterized pharmacologically (Fathi et al., 1997; Howard et al., 1997; Smith et a!., 1997; Wang et al., 1997a). Recently a third rat GALR, designated GALR3, has also been identified from a rat hypothalamic cDNA library (Wang eta!., 1997b). In the present report, we describe the cloning of human GALR2 and GALR3 receptors and comparative analysis of the gene structure, functional characterization, and tissue distribution of these receptors. MATERIALS AND METHODS Cloning of human GALR2 gene Human genomic DNA was amplified by PCR using degenerate oligonucleotides based on the sequences encoding transmembrane domain 3 (5’-CTG ACC GYC ATG RSC ATTGACSGCTAC;Y=CorT,R=AorG,S=Cor G) and transmembrane domain 7 (5 ‘-GGG GTT GRS GCA GCT GTT GG CRT A) of somatostatin G protein-coupled receptors (GPCRs) and the related orphan GPCR known as SLC-1 (Jung et a!., 1997). The PCR conditions were as follows: denaturation at 95°Cfor I mm, annealing at 55, 45, or 38°Cfor 1 mm, and extension at 72°Cfor 2.5 mm for 30 cycles, followed by a 7-mm extension at 72°C. The PCR products were phenol/chloroform-extracted, precipitated with ethanol, phosphorylated with T4 polynucleotide kinase, and blunt-ended with Klenow enzyme. Subsequently, the PCR products were electrophoresed on a 0.5% low-melting point agarose, and a fragment of the expected size was excised from the gel, purified, and subc!oned into the EcoRV site of pBluescript SK(—) (Stratagene, La Jolla, CA, U.S.A.). Colonies were selected, plasmid DNA was purified, and the inserts were sequenced. A partial DNA sequence displaying high sequence identity to the rat GALR2 transmembrane domain 3 to transmembrane domain 7 region was isolated and tentatively termed human GALR2. The partial human GALR2 DNA fragment was radiolabeled with [a2P] dCTP by nick translation (Amersham) and used as a probe to screen an EMBL3 SP6/T7 human genomic library (Clontech, Palo Alto, CA, U.S.A.). Positive phage clones were plaque-purified, and DNA was prepared, restriction enzyme-digested, electrophoresed on an agarose gel, transferred to nylon membrane, and hybridized with the same probe used to screen the library, as described previously (Marchese et a!., 1994). Positive phages were subcloned by digesting phage DNA and subcioning the resultant fragment into the pB!uescript vector. The DNA sequence of one clone called hGALR2-#42 was determined using standard methods on an ABI 372 automated sequencer (Perkin-Elmer-Applied Biosystems, Foster City, CA, U.S.A.).

ditions using primers flanking the GALR2 gene. The forward and reverse primers for exon I were (5 ‘-ATG AAC GTC TCG GGC TGC CCA GGG GCC-3’) and (5 ‘-GCG GAT

GGC CAG ATA CCT GTC TAG AGA GAC GGC-3’), respectively. The forward and reverse primers for exon II

were (5 ‘-GTC TCT CTA GAC AGG TAT CTG GCC ATC CGC TAC-3’) and (5’-TCA GGC CAC ATC AAC CGT CAG GAT GCT GTC-3’), respectively. The PCR products of expected size were subcloned into the EcoRV site of the pCI-neo mammalian expression vector (Promega). Cloning of human GALR3 gene A homology search using the algorithm TFASTX of GenBank sequence data (National Center For Biotechnology Information, Bethesda, MD, U.S.A.) and a set of 60 rhodopsin family GPCR amino acid sequences yielded a DNA sequence alignment match of —~300bp to a portion of a large

(—.1 00,000-bp) bacteria! artificial chromosome clone of human genomic DNA with accession no. Z97630. Using overlapping sequence data from a second bacterial artificial chromosome clone of human genomic DNA (GenBank accession no. Z8224!), a putative gene encoding a galanin-like GPCR gene was assembled electronically and tentatively designated human GALR3. A human genomic DNA library (Stratagene) was screened to isolate the putative GALR3 gene. Primary screening under medium stringency (Marchese et al., 1994) resulted in six positive plaques using an exon 2 probe. One hybridizing phage plaque was obtained on secondary screening. A 13-kb EcoRI/EcoRV fragment was identified from the genomic clone by Southern blotting, transferred into pB!uescript vector (Stratagene), and se-

quenced. Construction of human GALR3 expression plasmid An intronless human GALR3 expression construct was assembled in a stepwise manner from PCR products amplified from the genomic DNA clone. Based on the DNA sequence of the human GALR3 genomic clone, PCR primers were designed to amplify exon I and exon II using PCR amplification conditions as described above for human GALR2. The primers for exon I were as follows: forward exon I (5 ‘-GCG AAT TCG GTA CCA TGG CTG ATG CCC AGA ACA-3’) and reverse exon 1(5 ‘-CGC CTG TCG

ACA GAT ACA GCA GC-3’). The primers for exon II were as follows: forward (5 ‘-TGT ATC TGT CGA CAG GTA ACC TGG CCG TGC GGC ACC C-3’) and reverse (5’-GCG CGG CCG CTT ATT CCG GTC CTC GGG C3’). PCR products were subcloned into the plasmid pCRII (Invitrogen) and characterized by DNA sequencing. For ex-

pression in mammalian cells, an intronless human GALR3 open reading frame was reconstructed by ligation of human GALR3 exon I and exon II and subcloning into pcDNA3

(Invitrogen) to yield the expression plasmid termed pcDNA3-hGALR3. Cloning of a partial rat GALR3 gene A partial clone of the rat GALR3 gene was isolated for

Construction of human GALR2 expression plasmid An intronless version of the human GALR2 receptor was constructed by sequential PCR amplification of the two cxons from a human genomic DNA clone. The two PCR products, each containing an exon with an overlapping sequence, were combined and reamplified by PCR under standard con-

J. Neurochem., Vol. 71, No. 6, 1998

use as a probe for in situ hybridization experiments. Primers based on the intronless human GALR3 sequence from transmembrane domains 4 and 7 were designed and used to amplify by PCR using Pfu DNA polymerase the rat GALR3 orthologue from rat genomic DNA. A PCR product of the appropriate size (—400 bp) that hybridized with an exon 2 probe from the human GALR3 gene was subc!oned into

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3 pBluescript vector (Stratagene). DNA sequence analysis revealed significant homology with human GALR3, showing —95% amino acid sequence identity for 129 amino acids spanning transmembrane domains 4—7.

Expression of GALRs in mammalian cells Mammalian COS-7 cells were transfected by electroporation. COS-7 cells (I X I0~)were suspended in 0.85 ml of Ringer’s buffer, and 20 jig of the pcDNA3-hGALR3 expression p!asmid was added to a 0.4-mm electroporation cuvette (Bio-Rad, Hercules, CA, U.S.A.). Current was applied (960 mF, 260 V) using a Bio-Rad electroporator device, and the cells were transferred to a T- 180 flask (Corning). Expression was allowed to proceed for 72 h. Chinese hamster ovary (CHO) cell lines stably expressing human GALR2 were isolated as previously described (Sullivan et a!., 1997). In brief, CHO cells were electroporated with the pClneohGALR2 p!asmid, and cells were transferred to growth medium containing 250 jig/ml G4!8 48 h later. Individual clones were isolated by limiting dilution and selected by radioreceptor binding assay.

Membrane preparation and radioligand binding assays Membranes were prepared from transfected cells following dissociation in enzyme-free dissociation solution (Specialty Media, Lavallette, NJ, U.S.A.) by disruption in a Dounce homogenizer in ice-cold membrane buffer [10 mM Tris (pH 7.4), 10 mM phenylmethylsu!fony! fluoride, 10 mM phosphoramidon, and 40 jig/rn! bacitracin]. After a low-speed (1,100 g for 10 mm at 4°C)and a high-speed (38,700 g for 15 mm at 4°C)centrifugation, membranes were resuspended in buffer, and their protein concentration 1251-human was determinedgalanin (Bio-Rad assayactivity, kit). Binding or ‘251-porcine (specific 2,200 of Ci/mmol; Du Pont NEN) was measured in membranes using a buffer of 25 mM Tris (pH 7.4), 0.5% bovine serum albumin, 2 mM MgC1 2, 40 jig/ml bacitracin, 4 jig/rnl phosphoramidon, and 251-human or 10 ‘251-porcine jiM leupeptin galanin in a total (70 pM) volume wasofused. 250 Reactions jil. ‘ were initiated by addition of membranes, and the incubation was allowed to proceed at room temperature for 1 h. Nonspecific binding was defined as the amount of radioactivity remaining bound in the presence of 1 jiM unlabeled human galanin and was generally not >200 cprn (< i0~cells plated 18 h before transfection in a T-75 flask) were transfected with 22 jig of pClneo-ratGALR2 or pClneo-hGALR2 plasmid DNA and 264 jig of LipofectAMINE (Life Technologies). At 40 h after transfection the apoaequorin in the cells was charged for 1 h with coelenterazine CP (10 jiM) under reducing conditions (300 mM reduced glutathione) in ECB buffer [140 mM NaC!, 20 mM KC1, 20 mM HEPES-NaOH (pH 7.4), 5 mM glucose, 1 mM MgC12, 1 mM CaC12, and 0.1 mg/ml bovine serum albumin]. The cells were harvested, washed once in ECB medium, and resuspended to 5 >< i0~cells/ml. Cell suspension (100 jil; corresponding to 5 x i0~cells) was then injected into each well of a 96-well microtiter test plate, and the integrated light emission was recorded over 30 s, in 0.5s units. Twenty microliters of lysis buffer (0.1% final Triton X-100 concentration) was then injected, and the integrated light emission was recorded over 10 s, in 0.5-s units. The “fractional response” values for each well were calculated by taking the ratio of the integrated response to the initial challenge to the total integrated luminescence, including the Triton X-100 lysis response. Data were analyzed using Prism software ver. 2.0 (GraphPad Software).

Northern blot analysis Human northern blots from Clontech and BioChain Institute (San Leandro, CA, U.S.A.) were hybridized with probes prepared from the coding regions of the human GALR2 and human GALR3 cDNAs. The probes were radiolabeled using the 32P] RadPrirne UTP (NEN), radiolabeling and unincorporated system (Life Technologies) nucleotides were and [aremoved by spin column purification (Qiagen). The blots were hybridized at 68°Cin ExpressHyb solution (Clontech) for 16 h. Northern blots were washed at high stringency in 0.lX standard saline citrate (SSC) for 30 mm at 60°Cand exposed to x-ray film at —80°C for 10 (human GALR2) or 14 days (human GALR3).

RNase protection assay Primate tissues were obtained from a 15-year-old female baboon (designated 1x4344, harbored at Southwest Foundation for Biomedical Research) that had suffered a stroke 2 days before necropsy. The pituitary was obtained from a 17-year-old female designated 1x3790. Tissues were snapfrozen in liquid nitrogen at the time of necropsy, and RNA was isolated using TRIzol (Life Technologies). Human GALR2 exon I in pBluescript and human j3-actin in pCRII (Invitrogen) were linearized by restriction enzyme digestion, and the antisense cRNA probe was transcribed with RNA polymerase for 1 h at 37°Cin the presence of [a-32P]UTP (NEN). The antisense cRNA probe was gel-purified and hybridized with 10 jig of primate RNA using the RPA II Kit(Ambion, Austin, TX, U.S.A.). The samples were treated with RNase A and Ti, electrophoresed on 6% polyacrylamide gels, and autoradiographed at —80°Cfor 36 h.

J. Neurochem., Vol. 71, No. 6, /998

In situ hybridization analysis Male rats (Jackson Laboratories) were killed by decapitation, and brains were removed in 30 s and frozen in crushed dry ice. Frozen brains were sectioned at 14 mm thickness on a Reichert—Jung cryostat at —20°C and thaw-mounted onto microscope slides. Sections were fixed in freshly prepared 4% paraformaldehyde in 0.02% diethylpyrocarbonatetreated water for 20 mm at 4°Cin an ice bath and then washed for 5 mm in cold phosphate-buffered saline (pH 7.4) before dehydration in an alcohol series (Zastawny et al., 1994). Fixed sections were stored at —70°C.Fragments encoding rat orthologues of either human GALR2 or human GALR3 were labeled by random priming using a- 35S-cICTP (Du Pont NEN). Rat brain sections were prehybridized for 2 h in buffer containing 50% deionized formamide, 0.6 M NaC1, 10 mM Tris-HC1 (pH 7.5), 10% dextran sulfate, 1% polyvinylpyrrolidone, 2% sodium dodecyl sulfate, 100 mM dithiothreitol, and 200 jig/rn! herring sperm DNA, hybridized with the labeled probe (106 cpm per slice) for 16 h, and washed in conditions of increasing temperature and decreasing salt concentrations. The hybridized sections were dehydrated in a graded alcohol series, exposed to x-ray film (Du Pont MRF-34) for 4—6 weeks at —70°C,and developed. For use as controls, adjacent sections were hybridized following treatment with RNase, to confirm the specificity of hybridization.

Fluorescent in situ hybridization (FISH) Human lymphocytes were cultured in a minimal essential medium supplemented with 10% fetal calf serum and phytohernagglutinin at 37°Cfor 68—72 h. The lymphocyte cultures were treated with 5 ‘-bromodeoxyuridine (0.18 mg/ml; Sigma) for an additional 16 h to synchronize the cell population. The synchronized cells were washed three times with serum-free medium to release the block and recultured at 37°Cfor 6 h in a-minimal essential medium with thymidine (2.5 mg/ml; Sigma). Cells were harvested, and slides were made by using standard procedures including hypotonic treatment, fixation, and air-drying. The procedure for FISH was performed as described (Heng and Tsui, 1994). In brief, slides aged a few days were baked at 55°Cfor 1 h. After RNase A treatment, the slides were denatured in 70% formamide in 2>< SSC for 1 mm at 70°Cfollowed by dehydration with ethanol. Probes were biotinylated with [a32P]dATP using the BRL BioNick labeling kit, and the labeled probe was denatured at 75°Cfor 5 mm in a hybridization mix consisting of 50% formamide, 10% dextran sulfate, and human cot I DNA. Denatured probe was loaded on the denatured slides after a 15-mm incubation at 37°Cto block repetitive sequences. After overnight hybridization, detection, and amplification, the FISH signals and the (4,6-diamidino-2-phenylindole) banding pattern were visualized with fluorescence microscopy. FISH signals and the 4,6diamidino-2-phenylindole banding pattern was recorded separately by taking photographs, and the assignment of the FISH mapping data with chromosomal bands was achieved by superimposing FISH signals with 4,6-diamidino-2-phenylindole-banded chromosomes.

RESULTS AND DISCUSSION Cloning of human GALR2 gene We have previously identified novel GPCRs using a PCR-based search strategy and oligonucleotide prim-

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3

2243

FIG. 1. Amino acid sequence comparison of rodent and human GALR subtypes 1, 2, and 3. Mouse GALR1 = mGALR1 (GenBank accession no. Vi 5004); human GALR1 = hGALR1 (GenBank accession no. U5351 1); rat GALR1 = rGALR1 (GenBank accession no. U30290); human GALR2 = hGALR2 (GenBank accession number to be assigned); rat GALR2 = rGALR2 (GenBank accession no. AFO1 031 8); human GALR3 = hGALR3 (GenBank accession number to be assigned); rat GALR3 = rGALR3 (GenBank accession no. AFO31 522). Predicted transmembrane domains are numbered 1—7. Conserved amino acid residues are boxed, with those residues conserved in at least two of the six GALRs indicated by shading. Gaps are indicated by dashes. The PileUp alignment was generated using the Wisconsin Package ver. 9.1 (Genetics Computer Group, Madison, WI, U.S.A.) with gap extension of 4 and gap creation of 12.

ers based on conserved transmembrane domains of known GPCRs. In a search for new GPCRs, PCR was carried out on human genomic DNA using degenerate

oligonucleotides based on the sequences encoding transmembrane domains 3 and 7 of somatostatin GPCRs and the related orphan GPCR known as SLC-

J. Neurochem., Vol. 71, No. 6, 1998

2244

L. F. KOLAKOWSKI, JR., ET AL.

FIG. 2. Comparison of the genomic structures of the human (h) GALR1, GALR2, and GALR3 genes. The genomic structure of hGALR1 is as described by Jacoby et al. (1997). The exons are shown as open boxes; the numbers in the boxes indicate the number of amino acids (AA) encoded within that exon. The numbers over the boxes indicate the position of the exons with the numbering for each gene beginning at the first known nucleotide most 5’ to the ATG initiating codon of each gene.

1 (Jung et al., 1997). Using this strategy a PCR product was isolated that showed -—40% sequence identity to the transmembrane domain 3—7 coding region of human GALR 1 and >80% sequence identity to homologous region in rat GALR2 (Howard et a!., 1997). The putative transmembrane domain 3—7 coding region of human GALR2 PCR product was then used as a probe to isolate a full-length human GALR2 genomic clone by hybridization screening of a )~.phage EMBL genomic library. As observed with rat GALR2, the human GALR2 gene contains a single intron of 1,390 bp and divides the human GALR2 open reading frame into two exons with perfectly conserved splice donor/acceptor consensus junctions. Based on the conceptual translation of the human GALR2 gene, the first exon encodes the methionine initiator codon to the end of transmembrane region 3, with the second exon encoding the second intracellular domain to the stop codon. The deduced amino acid sequence of the contiguous open reading frame gives a protein of 387 amino acids with hallmark features typical of GPCRs, including seven predicted transmembrane a-helical domains and the aromatic triplet signature sequence [(D/E)RYJ adjacent to transmembrane domain 3 (Figs. 1 and 2). The human GALR2 protein bears strong sequence identity (87%) and similarity (92%) to the rat GALR2 orthologue. The GALR2 sequences shown in Fig. 1 illustrate one notable difference between the human and rat forms, which, is that the C-terminal intracellular domain of human GALR2 is extended by an additional 15 amino acids.

Cloning of human GALR3 gene A homology search of the high-throughput genomic sequence (HTGS) division of Genbank using the algorithm TFASTX and a set of 60 known GPCR coding region sequences led to the identification of a GALRlike sequence that was tentatively identified as human GALR3. The putative human GALR3 gene was contained within two large bacterial artificial chromosome sequence contigs of human genomic DNA (GenBank accession nos. Z82241 and Z97630) that had been deposited as part of the human chromosome 22 sequencJ. Neurochem., Vol. 71, No. 6, 1998

ing project (Sanger Centre, Cambridge, U.K.). The deduced amino acid translation of the putative human GALR3 and its alignment to the human GALR2 gene revealed a similar genomic organization (Figs. 1 and 2). The human GALR3 gene is divided by a single intron located at a splice site identical to that in human GALR2. The human GALR3 gene contains a single intron of 958 bp and divides the human GALR3 open reading frame into two exons, with the first exon encoding the methionine initiator codon to the end of transmembrane domain 3 and the second exon encoding the second intracellular domain to the stop codon. The deduced amino acid sequence of the contiguous open reading frame gives a protein of 368 amino acids with hallmark features typical of GPCRs, including seven predicted transmembrane a-helical domains and the aromatic triplet signature sequence (DRY) adjacent to transmembrane domain 3 (Fig. 1). The human GALR3 protein shows amino acid sequence identity to human GALR1 (36%), human GALR2 (58%), and rat GALR3 (89%). To isolate the human GALR3 gene, a hybridization probe was first generated by PCR of human genomic DNA using oligonucleotide primers encompassing exon II of the predicted human GALR3. Using this human GALR3 exon II probe, a X phage human genomic library was screened by hybridization. From this screen a single phage plaque was isolated that contained the human GALR3 gene localized to a 13-kb EcoRI/EcoRV restriction fragment. An intronless human GALR3 construct was prepared by stepwise PCR amplification of exon land exon II, followed by their assembly into a complete open reading frame in the plasmid expression vector pCDNA3. There are several features that are conserved among a!! the known rodent and human GALR subtypes, including a single potential N-linked glycosylation site (Asn-X-Ser) at residues 11 and 6 in human GALR2 and GALR3, respectively, and two cystemne residues in the first and second extracellular 1oops possibly involved in forming a disulfide bridge (residues 98 and 175 in human GALR2 and residues 95 and 172 in human GALR3) (Wang et a!., 1997a).

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3

2245

specific binding of radioligand was observed. Low lev-

els of expression of this receptor in COS-7 cells precluded a detailed pharmacological characterization of this receptor; however, binding was saturable. The IC50 values for porcine and human gaianin were 12 and 75 nM, respectively, as determined by competition analysis (Fig. 3B). This reflects a possible sensitivity of

GALR3 to changes in galanin itself; at these low levels of receptor expression, consistent saturable binding was difficult to measure with human galanin as the radioactive ligand. However, this was not1251-porcine a problem when binding studies areisperformed with the galanin galanin. Porcine galanin unique among peptide family in possessing an additional tyrosine at

amino acid 26, which affords an additional iodination site. The difficulty in using ‘25I-human galanin as a

reagent to probe human GALR3 suggests that perhaps regions encompassing Tyr9 in galanin form part of an important recognition site for galanin on this receptor.

This distinguishes human GALR3 from either human GALR1 or human GALR2. The ability to measure downstream activation of effector pathways mediated by galanin activation of GALR3 should help resolve this issue.

FIG. 3. Pharmacological characterization of human GALR2 and GALR3. A: Saturation isotherm and Scatchard analysis (inset) for binding of 1251-human galanin to the cloned human GALR2 expressed in CHO cells. Total binding, •; nonspecific binding, A; specific binding, V. B: Competition of 1251-porcine galanin for the cloned human GALR3 expressed in COS-7 cells by porcine and human galanin. Data shown are of a representative experiment repeated three times with similar results (±10%). For competitions, 0.07 nM radiolabeled ligand was used in the presence of different concentrations of unlabeled competitor with 20 jig of membrane protein from GALR3-transfected cells. Specific binding was 4,172 cpm, whereas nonspecific binding (in the presence of 1 1iM unlabeled porcine galanin) was 284 cpm. The binding capacity (B,,,~,,)for GALR3 transfectants was 0.2 pmol! mg of protein.

Ligand binding properties of human GALR2 and human GALR3 Membranes were prepared from CHO cells human stably 25I-labeled expressing humanAs GALR2 galanin binding. showntoinassay Fig. ‘3A, saturable and specific high-affinity binding was observed to a single class of noninteracting sites, showing a KD value for ‘251-human galanin of 0.3 nM and Bmax of 2.9 pmol/ mg of membrane protein. The radioligand binding of human GALR3 was sim-

ilarly studied. Specific binding of radioligand was observed, and this increased as a function of membrane protein concentration. In COS-7 cells mock-transfected with expression vector only (no GALR3 gene), no

Signal transduction pathway of human GALR2 and GALR3 Rat GALR2 has been reported to couple to the inhibition of forskolin-stimulated intracellular cyclic AMP production (Wang et a!., 1 997a), to elicit increased inositol phospholipid turnover and intracellular calcium levels (Fathi et a!., 1997; Smith et a!., 1997), and to generate calcium-activated chloride channels in Xenopus oocytes (Smith et a!., 1997). Because rat GALR2 would appear to couple to both G 1- and Gq~ coupled signaling pathways, we chose to investigate the signaling mechanism of human GALR2 in Xenopus melanophores (Daniolos et a!., 1990; Potenza et al., 1992; Lerner, 1994). The melanophore assay system is based on the dispersion and aggregation of intracellular

pigment granules in response to changes in intracellular second messenger molecules. Thus, as has been shown for >40 different GPCRs, agonist activation of a recombinant G~-or Gqcoupled GPCR expressed heterologously in melanophores will lead to pigment dispersion. Conversely, agonist activation of a recombinant G~-coupledGPCR expressed heterologously in melanophores will lead to pigment aggregation. Changes in melanophore pigmentation show a dosedependent correlation with the level of specific receptor activation and can be quantified by the change in absorbance at 600 nm between the nonactivated and agonist-activated cells (Daniolos et al., 1990; Potenza

et a!., 1992; Lerner, 1994). Melanophores that had been transfected transiently with pClneo-hGALR2 showed a dose-dependent dis-

persion ofpigment in responseto increasing concentrations of galanin and galanin-related agonists in contrast to control vector-transfected cells (Fig. 4A and B). J. Neurochem., Vol. 71, No. 6, /998

2246

L. F. KOLAKOWSKI, JR., ET AL.

FIG. 5. Aequorin bioluminescence assay of human GALR2 (huGAL) in HEK-293 cells. HEK-293-AEQ17 cells (8 x io~cells) were transfected by the LipofectAMlNE procedure with huGAL expression plasmid DNA. At 40 h after transfection the apoaequorin in the cells was charged for 4 h with coelenterazine. In a 96-well plate assay format, 5 >< i0~cells were then injected into the test plate containing galanin peptides, and the integrated light emission was recorded over 30 s followed by an injection of lysis buffer (0.1% final Triton X-100 concentration) recorded over 10 s. The “fractional response” values for each well were calculated by taking the ratio of the integrated response to the initial challenge to the total integrated luminescence including the Triton X-1 00 lysis response. rGAL, rat GALR.

FIG. 4. Functional coupling of human GALR2 and GALR3 in Xenopus melanophores. Pigment dispersion and aggregation responses were assayed in melanophores transfected with human GALR2 and GALR3 expression plasmids in response to galanin and related peptides. Melanophores were transiently transfected by electroporation with expression plasmids and plated in a 96well format. Two days following transfection the cells were assayed for agonist activation as described in Materials and Methods. Before addition of galanin peptides, transfected melanophores were preincubated with melatonin for detection of G~/ Gqmediated responses (pigment aggregation) or 70% L-i5 medium containing 2% fibroblast-conditioned growth medium and 2 mM glutamine for G-mediated responses (pigment dispersion). After initial A 600 readings, final A600 readings were determined after a 1.5-h incubation in the dark. Absorbance data were converted to transmission values as outlined in Materials and Methods where 7 = initial transmission and Tf = final transmission. A: Dose—response curves of melanophores transiently expressing human GALR2 receptor and activated by human galanin, porcine galanin, rat galanin 2—29, M15, and C7 for 1 h at room temperature. Data are mean ± SEM (bars) values. B: Dose—response curves of vector control-transfected melanophores challenged with human galanin, porcine galanin, rat galanin 2—29, M15, and C7 for 1 h at room temperature. Data are mean ± SEM (bars) values. C: Pigment aggregation assay in melanophores transiently expressing human GALR3 (hGALR3) in response to porcine galanin. Melanophores were tran-

J. Neurochem., Vol. 71, No. 6, 1998

The apparent BC50 for human galanin in pclneohGALR2-transfected melanophores was 20 nM. None of the galanin messenger-associated peptide fragments (1—41, 16—41, 25—41, and 44—59) was able to cause pigment dispersion in pClneo-hGALR2-transfected melanophores (data not shown). There was no detectable aggregation of the pigment in either pclneohGALR2-transfected or mock-transfected melanophores following incubation in the presence of 0.001 1,000 nM human galanin (Fig. 4B). These results indicate that in the melanophores human GALR2 couples to Gq and/or G. but not to G1-mediated signaling pathways. Because both G~-and Gq-coupled signaling leads to pigment dispersion in the melanophore assay, we sought to determine through which of these pathways human GALR2 was signaling using the aequorin biolumines—

siently transfected by electroporation with pcDNA3.1-hGALR3 or mock-transfected with pcDNA3.i expression plasmids followed by challenge with porcine galanin as described in Materials and Methods. Data shown are the means of a representative experiment (two to five separate determinations) performed in duplicate.

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3

2247

FIG. 6. Expression profile of GALR2 and GALR3 shown by northern blot analyses in human tissues of human GALR2 (a) and probes human 32P-labeled GALR3 (b). Northern blots containing 2 ~igper lane of poly(AY mRNA from human tissues were hybridized with aencompassing the complete open reading frames of human GALR2 or human GALR3, washed at high stringency (0.1 x SSC at 60°C), and autoradiographed at —80°Cfor 10 (human GALR2) or 14 days (human GALR3). PBL, peripheral blood lymphocytes. C: RNase protection assay of GaIR2 and GaIR3 mRNA levels in baboon tissues. Poly(A)~mRNA was isolated from brain regions and peripheral tissues from two baboons. An antisense probe corresponding to the first exon of human (A) GALR2 and (B) GALR3 was synthesized in the presence of [32P]UTP,hybridized to 10 ~g of monkey poly(Ay mRNA, and digested with RNases. 32P-labeled protected probe fragments were electrophoresed on 6% polyacrylamide gels and autoradiographed at —80°Cfor 36 h.

cence assay (Button and Brownstein, 1993). This assay primarily identifies QPCRs that couple through the (l~ subfamily, leading to the activation of phospholipase C, mobilization of intracellular calcium, and activation of protein kinase C. Expression of human or rat GALR2 in the aequorin-expressing HEK-293 cell line (293AEQ17) gave a dose-dependent increase in aequorin bioluminescence in response to challenge by galanin and several related peptides (Fig. 5). Transfection of human GALR1, which signals through G, and the inhibition of adenylyl cyclase, failed to elicit a response. Responses observed for human or rat GALR2 activation were saturable, and the rank order of potency was similar to that observed for competition studies for ‘251-human galanin

binding (Fig. 5). The chimeric galanin peptides C7 and M40, which have been shown to act as antagonists in some in vivo assays of galanin function (Crawley et al., 1993), are agonists at human GALR2. These data indicate that the primary signaling mechanism for GALR2 is through the phospholipase C/protein kinase C pathway, in contrast to GALR1, which communicates its intracellular signal by inhibition of adenylyl cyclase through G~.

To determine the functional signaling pathway of human GALR3, the plasmid pcDNA3.1-hGALR3 was transfected transiently into melanophores followed by challenge of the cells with galanin and assaying for both aggregation and dispersion of the cellular pigJ. Neurochem., Vol. 71, No. 6, 1998

2248

L. F. KOLAKOWSKI, JR., ET AL.

FIG. 7. In situ hybridization of GALR2 in rat brain. Sections 14 p~mthick from frozen and fixed whole rat brain were thaw-mounted on 355-dCTP microscope slides and with awashed probe derived from the total open reading frame of rat GALR2 with by random priming. Thehybridized sections were at high stringency, dehydrated in a graded alcohol series,radiolabeled and exposed to x-ray film (Du Pont MRF-34) for 4—6 weeks at —70°C. ACg, anterior cingulate cortex; Arc, arcuate nuclei; CA, field of Ammon’s horn; CA3, hippocampus; Cg, cingulate gyrus; Ch, choroid plexus; Cl, claustrum; DG, dentate gyrus; Dk, nucleus of Darkschewitsch; Fr, frontal cortex; lCj, islands of Calleja; LM, lateral mammillary nuclei; MHb, medial habenular nucleus; ML, medial mammillary nucleus, lateral; MM, medial mammillary nucleus; Pa, paraventricular hypothalamic nucleus; Par, parietal cortex; PH, posthypothalamic nucleus; PMD, premammillary nuclei, dorsal; PMV, premammillary nuclei, ventral; P0, primary olfactory cortex; SNC, substantia nigra compacta; SO, supraoptic hypothalamic nucleus; SuM, supramammillary hypothalamic nucleus; U, taenia tecta; ZI, zona incerta.

ment. Increasing doses of porcine galanin from 0.01 to 10 nM resulted in a dose-dependent aggregation of pigment in human GALR3-transfected melanophores (Fig. 4C), suggesting G, coupling and signaling through inhibition of adenylate cyclase. There were no significant changes in the pigmentation of the vector control-transfected melanophores to porcine galanin. It is interesting that human galanin and several galanin messenger-associated peptides did not elicit any pigment changes over a dose range of 0.01—10,000 nM

(data not shown). The apparent EC 50 for porcine galanin in pcDNA3.1 -hGALR3-transfected melanophores was ‘—1 nM. In response to galanin concentrations >10 nM, a dose-dependent reversal of the pigment aggregation response was observed (Fig. 4C). No functional Gq or G. coupling could be detected following expression of human GALR3 in either the J, Neurochem., Vol. 71, No. 6, 1998

HEK-293 aequorin assay or the melanophore pigment dispersion assay. Tissue distribution of GALR2 and GALR3 Several approaches were used to determine the expression profile of the GALR2 and GALR3 genes, including northern blot analysis using various human tissue mRNA samples, RNase protection assays conducted on monkey brain poly(A) + mRNA, and in situ hybridization in rat brain. Collectively, the data suggest a widespread expression profile for both GALR2 and GALR3 (Fathi et a!., 1997; Wang et a!., 1997a,b). Northern blot and RNase protection analyses Northern blot analysis of human GALR2 mRNA expression revealed a prevalent transcript of ‘=2.0 kb in tissues from the gastrointestinal tract with the highest expression in the small intestine (jejunum and il-

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3

2249

FIG. 8. In situ hybridization of GALR3 in rat brain. The conditions for in situ hybridization and nomenclature are as described in the legend to Fig. 7. ACo, anterior cortical amygdaloid nucleus; LS, lateral septal nucleus; Me, medial amygdaloid nucleus; OT, nucleus of the optic tract; PMCo, posteromedial cortical amygdaloid nucleus; RSP, retrosplenial cortex; VMH, ventromedial hypothalamic nucleus.

eum) and lesser amounts in the colon and rectum (Fig. 6a). Using a more sensitive RNase protection analysis,

the relative levels of monkey GALR2 mRNA revealed highest expression in all regions of the brain tested in comparison with lower levels in several peripheral

tissues, including lung, heart, and striated muscle (Fig. 6c). The widespread peripheral distribution of the primate GALR2 mRNA is in general agreement with expression of GALR2 in the rat (Fathi et al., 1997; Howard et al., 1997; Wang et al., 1997a). Northern blot analysis for human GALR3 showed

medulla; signal was also detected in amygdala, choroid plexus, frontal cortex, pituitary, and cerebellum (Fig. 6c). In the peripheral tissues abundant levels of monkey GALR3 mRNA were observed in the striated mus-

cle, in comparison with lower levels detected in spleen, axillary lymph node, heart, lung, and kidney; signal was also shown in liver and adrenal cortex (Fig. 6c). Rat GALR3 mRNA was reported to be most abundant in heart and testis, with no apparent expression in other tissues, including brain. In contrast, human GALR3 is

distributed widely and is detected as a 1.4-kb transcript

several tissues expressing a heterogeneous population of mRNAs of 1.4, 2.4, and 5.0 kb in thyroid, adrenal

in various brain regions (Fig. 6b). The widespread peripheral expression patterns of both human GALR2

gland, testis, brain, skeletal muscle, pancreas, small

and GALR3 mRNAs are similar to that of human GALR1, which is detected at significant levels in brain but also at moderate levels in heart, small intestine, the prostate, and the testes (Sullivan et a!., 1997). In contrast, mouse GALR1 mRNA is expressed predominantly in the brain and spinal cord with limited expression in the periphery (Wang et a!., 1997c).

intestine, large intestine, rectum, and placenta; the

most abundant mRNA was a 1.4-kb transcript in testis (Fig. 6b). The RNase protection assay revealed the relative levels of monkey GALR3 mRNA centrally with the highest expression in the hypothalamus and lower expression levels in the occipital pole, pons, and

J. Neurochein., Vol. 71, No. 6, 1998

2250

L. F. KOLAKOWSKI, JR., ET AL.

In situ hybridization analysis In situ hybridization was conducted to map the distribution of GALR2 mRNA in rat brain using a labeled rat GALR2 open reading frame fragment as a hybridization probe (Fig. 7). Specific hybridization was detected in several brain nuclei and regions, mostly in the mammillary nuclei, the dentate gyrus, and posterior hypothalamic, supraoptic, paraventricular, and arcuate nuclei. Frontal, anterior cingulate, and parietal cortical regions were also labeled. A similar analysis was conducted for rat GALR3 (Fig. 8). There was moderate expression in many regions of cerebral cortex, including anterior cingulate, frontoparietal, and retrosplenial areas. There was particularly dense expression in the primary olfactory cortex, olfactory tubercle, and the islands of Calleja. The taenia tecta and the lateral septal nuclei were labeled, as was the choroid plexus throughout the brain. The caudate-putamen and nucleus accumbens were labeled only minimally. The hippocampal CA regions of Am-

mon’s horn and the dendate gyms expressed abundant GALR3 mRNA. Transcripts were also detected in the medial habenular nucleus and zona incerta with small amounts diffusely present in some of the thalamic nuclei. In the hypothalamus, GALR3 mRNA was detected in several nuclei: the ventromedial, arcuate, paraventricular, and supraoptic. Additional areas containing GALR3 transcripts included the mammillary nuclei, amygdaloid nuclei, substantia nigra compacta, nucleus of Darkschewitsch, and central gray regions. Taken together, the localization data for GALR2 suggest a more widespread distribution than for GALR I in the brain and periphery with many overlapping areas of expression. The expression patterns for GALR2 and GALR3 in the brain appear similar. Highresolution colocalization experiments are needed to map more precisely the cellular distribution of the GALR subtypes. Chromosomal localizations of human GALR2 and GALR3 FISH of metaphase-spread chromosomes derived from human lymphocytes and their 4,6-diamidino-2phenylindole banding patterns were used to map human GALR2 to its chromosome and to confirm and refine the assignment of human GALR2 to chromosome 17. FISH data localize the GALR2 gene to human chromosome 17q25. In the case of human GALR3, the bacterial artificial chromosome clones that were identified by searches of the HTGS dataset have been mapped by the Sanger Centre Genome research laboratory to human chromosome 22. FISH analysis conducted herein has confirmed this assignment and refined it to 22q12.2-13.1. In summary, we have described here the cloning and characterization of human GALR2 and human GALR3 receptors. This characterization has included radioligand binding, functional signaling, tissue distribution, and chromosomal localization. In addition, we have J. Neurochem., Vol. 71, No. 6, 1998

cloned the murine GALR2 and GALR3 receptors (au-

thors’ unpublished data) to create a gene knockout mouse to define further the roles of GALR subtypes. The study of receptor gene knockout mice and the development of subtype-specific selective agonists and antagonists of GALRs will allow the physiology and potential therapeutic value of this receptor family to be determined. Acknowledgment: This research was supported by grants from the Addiction Research Foundation (Ontario), the National Institute for Drug Abuse, the Medical Research Council of Canada, and the Smokeless Tobacco Research Council to B.F.O. and S.R.G. MS. was supported with a studentship from the Medical Research Council of Canada.

REFERENCES Bartfai T., Bedecs K., Land T., Langel U., Bertorelli R.. Birotti P., Consolo S., Xu S., Wiesenfeld-Hallin Z., Nilsson S., Pieribone V. A., and Hokfelt T. (1991) M-15, high-affinity chimeric peptide that blocks the neuronal actions of galanin in the hippocampus, locus coeruleus, and spinal cord. Proc. NatI. Acad. Sci. USA 88, 10961—10965. Bartfai T., Hökfelt T., and Langel U. (1993) Galanin—a neuroendocrIne peptide. Crit. Rev. Neurobiol. 7, 229—274. Bedecs K., Berthold M., and Bartfai T. (1995) Galanin—lO years with a neuroendocrine peptide. mt. .1. Biochem. Cell Biol. 27, 337—349. Burgevin M. C., Loquet I., Quarteronet D., and Habert-Ortoli E. (1995) Cloning, pharmacological characterization, and anatomical distribution of a rat eDNA encoding for a galanin receptor. J. Mol. Neurosci. 6, 33—41.

Button D. and Brownstein M. (1993) Aequorin-expressing 2~mobilization. Cell mammaCalcium han cell lines used to report Ca 14, 663—671. Crawley J. N., Austin M. C., Fiske S. M., Martin B., Consolo S., Berthold M., Langel U., Fisone G., and Bartfai T. (1990) Activity of centrally administered galanin fragments on stimulation of feeding behavior and on galanin receptor binding in the rat hypothalamus. J. Neurosci. 10, 3695—3700. Crawley J. N., Robinson J. K., Langel U., and Bartfai T. (1993) Galanin receptor antagonists M40 and C7 block galanin-induced feeding. Brain Res. 600, 268—272. Daniolos A., Lemer A. B., and Lerner M. R. (1990) Action of light on frog pigment cells in culture. Pigment Cell Res. 3, 38—43. Fathi Z., Cunningham A. M., Iben L. G., Battagliano P. 8., Ward S. A., Nicho] K. A., Pine K. A., Wang J., Goldstein M. E., usmaa T. P., and Zimanya I. A. (1997) Cloning, pharmacological characterization and distribution of a novel galanin receptor. Brain Res. Mol. Brain Res. 51, 49—59. Habert-Ortoli E., Amiranoff G., Loquet 1., Laburther M., and Mayaux J.-F. (1994) Molecular cloning of a functional human galanin receptor. Proc. Nati. Acad. Sci. USA 91, 9780—9783. Hedlund P. B., Yanaihara N., and Fuxe K. (1992) Evidence for specific N-terminal galanin fragment binding sites in the rat brain. Ear. J. Pharmacol. 224, 203—205. Heng H. H. and Tsui L. C. (1994) FISH detection on DAPI-banded chromosomes. Methods Mol. Biol. 33, 35—49. Howard A. D., Tan C., Shiao L. L., Palyha 0. C., McKee K. K., Weinberg D. H., Feighner S. D., Cascieri M. A., Smith R. G., Van Der Ploeg L. H. T., and Sullivan K. A. (1997) Molecular cloning and characterization of a new receptor for galanin. FEBS Lett. 405, 285—290. Jacoby A. S., Webb G. C., Liu M. L. L., Kofier B., Hort Y. 1., Fathi Z., Bottema D. K., Shine J., and Iismaa T. P. (1997) Structural organization of the mouse and human GALR I galanin receptor genes (Galnr and GALNR) and chromosomal localization of the mouse gene. Genomics 45, 496—508.

CLONING AND CHARACTERIZATION OF HUMAN GALR2 AND GALR3 Jung B., Nguyen T., Kolakowski L. F. Jr., Lynch K. R., Heng H. H. Q., George S. R., and O’Dowd B. F. (1997) Discovery of a novel human G protein-coupled receptor gene (GPR25) located on chromosome 1. Biochem. Biophys. Res. Commun.

230, 69—72. Lemer M. (1994) Tools for investigating functional interactions between ligands and G-protein-coupled receptors. Trends Neurosci. 17, 142—146. Marchese A., Docherty J. M., Nguyen T., Heiber M., Cheng R., Heng H. H. Q., Tsui L.-C., Shi X., George S. R., and O’Dowd B. F. (1994) Cloning of human genes encoding novel G proteincoupled receptors. Genomics 23, 609—618. Parker E. M., Izzarelhi D. G., Nowak H. P., Mahle C. D., Iben L. G., Wang J., and Goldstein M. E. (1995) Cloning and characterization of the rat GALRI galanin receptor from Rinl4B insulinoma cells. Brain Res. Mol. Brain Res. 34, 179—189. Potenza M. N., Graminski G., and Leaner M. R. (1992) A method for evaluating the effects of higands upon Gs protein-coupled receptors using a recombinant melanophore-based bioassay. Anal. Biochem. 206, 315—322. Shipetz D., O’Neill G. P., Favreau L., Dufresne C., Gallant M., Gareau Y., Guay D., Labelle M., and Metters K. (1995) Activation of the human peripheral cannabinoid receptor results in inhibition of adenylyl cyclase. Mol. Pharmacol. 48, 352—361. Smith K. E., Forray C., Walker M. W., Jones K. A., Tamm J. A., Bard J., Branchek T. A., Linemeyer D. L., and Gerald C. (1997) Expression cloning of a rat hypothalamic galanin receptor coupled to phosphoinositide turnover. J. Biol. Chem. 272, 24612— 24616. Sullivan K. A., Shiao L.-L., and Cascieri M. A. (1997) Pharmaco-

2251

logical characterization and tissue distribution of the human and rat GALR1 receptors. Biochem. Biophys. Res. Commun. 233, 823 —828. Tatemoto K., Rokeaus A., Jornvall H., McDonald T., and Mutt V. (1983) Galanin—a novel biologically active peptide from porcine intestine. FEBS Lett. 164, 124—128. Tempel D. L., Leibowitz K. J., and Leibowitz S. F. (1988) Effects of PVN galanin on macronutrient selection. Peptides 9, 309— 314. Wang S., Hashemi T., He C., Strader C., and Bayne M. (1997a) Molecular cloning and pharmacological characterization of a new galanin receptor subtype. Mol. Pharmacol. 52, 337—343. Wang S., He C., Hashemi T., and Bayne M. (1997b) Cloning and expressional characterization of a novel galanin receptor. J. Biol. Chem. 272, 31949—31952. Wang S., He C., Maguire M. T., Clemmons A. L., Burner R. E., Guzzi M. F., Strader C. D., Parker E. M., and Bayne M. (1997c) Genomic organization and functional characterization of the mouse GalRl galanin receptor. FEBS Lett. 411, 225— 230. Wynick D., Smith D. M., Ghatei M., Akinsanya K., Bhogal R., Purkiss P., Byfield P., Yanaihara N., and Bloom S. R. (1993) Characterization of a high-affinity galanin receptor in the rat anterior pituitary: absence of biological effect and reduced membrane binding of the antagonist Ml 5 differentiate it from the brain/gut receptor. Proc. Nati. Acad. Sci. USA 90, 4231— 4235. Zastawny R. L., George S. R., Nguyen T., Cheng R., Tsatsos J., Briones-Urbina R., and O’Dowd B. F. (1994) Cloning, characterization, and distribution of a ~.t-opioidreceptor in rat brain. J. Neurochem. 62, 2099—2105.

J. Neurochem., Vol. 71, No. 6, 1998