Neuron,

Vol. 6, 421-430, March, 1991, Copyright

0 1991 by Cell Press

cDNA Cloning, Characterization, and Brain Region-Specific Expression of a Neuromedin-B-Preferring Bombesin Receptor Etsuko Wada,* James Way,* Hagit Shapira,* Kiyoshi Kusano,* Anne Marie Lebacq-Verheyden,+ David Coy,* Robert Jensen,5 and James Battey* *Laboratory of Neurochemistry National Institute of Neurological Disorders and Stroke National Institutes of Health Bethesda, Maryland 20892 +Cellular Genetics Unit Institute of Cellular and Molecular Pathology University of Louvain Brussels Belgium *Peptide Research Laboratories Department of Medicine Tulane University Medical Center New Orleans, Louisiana 70112 SDigestive Disease Branch National Institute of Diabetes and Digestive and Kidney Diseases National institutes of Health Bethesda, Maryland 20892

Summary Recent binding studies in the central nervous system and other tissues provide evidence that the mammalian bombesin-like peptides, gastrin-releasing peptide (CRP) and neuromedin-B (NMB), exert their numerous physiological effects through at least two different receptors. We describe the structure and expression of a cloned NMB-preferring bombesin receptor (NMB-R) with prop erties distinct from a GRP-preferring bombesin receptor (CRP-R) reported previously. In particular, the NMB-R shows higher affinity binding to NMB than to GRP in BALB 3T3 fibroblasts expressing the cloned NMB-R. The distinct regional distribution of NMB-R and CRP-R mRNA in the brain suggests that both bombesin receptor subtypes play independent roles in mediating many of the dramatic effects of bombesin-like peptides in the central nervous system. Introduction Bombesin (BN) is a tetradecapeptide originally purified from the skin of the European frog Bombina bombina (Anastasi et al., 1971). More than ten BN-related peptides were subsequently isolated from various sources and classified into three subfamilies according to their C-terminal tripeptides (bombesin, ranatensin, and litorin subfamilies) (Erspamer et al., 1988). The two known mammalian BN-like peptides are neuromedin-B (NMB), in the ranatensin subfamily (Minamino et al., 1983), and gastrin-releasing peptide (CRP), in the BN subfamily (McDonald et al., 1979).

Mammalian BN-like peptides show a wide spectrum of biological and pharmacological activities, including regulation of smooth muscle contraction and secretion of other gastrointestinal peptide hormones. In addition, GRP and BN can function as growth factors in Swiss 3T3 murine embryonal fibroblasts (Rozengurt and Sinnett-Smith, 1983) and have been implicated as autocrine growth factors in the pathogenesis of some human small cell lung carcinomas (Cuttitta et al., 1985). In thecentral nervous system (CNS), these neuropeptides are thought to play a role in the regulation of homeostasis, thermoregulation, metabolism, and behavior (for reviews see Tache and Brown, 1982; Lebacq-Verheyden et al., 1990). The varied responses of target cells to mammalian EN-like peptides are mediated by binding to highaffinity receptors found on many cells, including Swiss 3T3 fibroblasts (Zachary and Rozengurt, 1985), a rat pituitary cell line (Westendorf and Schonbrunn, 1983), pancreatic acinar cells (Jensen et al., 1978), esophageal muscularis mucosa smooth muscle (von Schrenck et al., 1989), and rat brain membranes (Moody et al., 1978, 1988). More detailed analysis of the relative affinity of rat brain receptors for BN-like peptides suggests that there are at least two distinct BN receptor subtypes. For example, receptor autoradiographic studies using labeled BN and NMB show that only a small subset of CNS BN-binding sites mapped previously are GRP preferring and that the majority of CNS BN-binding sites are NM6 preferring (Ladenheim et al., 1990; Lee et al., 1991). Similarly, at least two BN receptor subtypes are distinguished in binding studies characterizing receptors found in tissues other than brain. The order of potency of a series of BN-related peptides is clearly different in eliciting responses from stomach strip preparations than from carotid artery preparations, suggesting the existence of at least two receptor subtypes (Regoli et al., 1988). Subsequently, a BN receptor with higher affinity for NMB than either GRPor BN (NMB-preferring bombesin receptor; NMB-R) has been demonstrated in esophageal smooth muscle, while pancreatic acinar cells express a BN receptor with higher affinity for GRP and BN than NMB (GRP-preferring bombesin receptor; GRP-R) (von Schrenck et al., 1990). Recently, we (Battey et al., 1991) and others (Spindel et al., 1990) reported the isolation and characterization of cDNA clones for the Swiss 3T3 GRP-R, whose properties are nearly identical to the GRP-R expressed by pancreatic acinar cells. When this cloned GRP-R is expressed in Xenopus oocytes, GRP is a more potent ligand than NMB in eliciting an electrophysiologic response. In addition, the response is abolished by a BN receptor antagonist [o-Phe61BN(6-13) ethyl ester SpecificfortheGRP-preferringsubtype(von Schrenck et al., 1990; Battey et al., 1991). Binding studies performed in transfected mouse fibroblasts (BALB 3T3)

Neuron 422

lo@

I 0

.

131

200 AACTCCCGCGAATCCCCTGGGAAAGGACACCATGCCCCCCAGGTCTCTCCCCAACCTCTCCTTGCCCACCGAGGCGAGCGAGAGCGAGTTGGAACCCGAG MetProProArgSerLe"ProAsnLeuSerLe"PKoThrGluAlaSerGl"SerGl"LeuGluProGlu

201

300 GTGTGGG-TGATTTCCTGCCTGACTCAGACGGGACCACCGCGGAGTTGGTAATCCGCTGTGTGATACCATCCCTCTACCTAATCATCATCTCGGTGG ValTrpGluAsnAspPheLe"ProAspSerAspGlyThrThKAlaGl"Le"ValIleArgCysValIleProSerLe"Tyr~"IleIleIleSerValG

.l.

301

.

0 . .2 . 400 GCTTGCTGGGCAACATCATGCTGGTGMGATATATTCCTCACCAACAGCACCATGCGGAGTGTCCCCAACATCTTCATCTCTAACCTGGCTGCGGGAGACCT lyLeuLeuGlyAsnIl~etLeuValLysIlePhe~"ThrAS~SerThrMetAr~SerValProAsnIlePheIleSerAsnLe"AlaAlaGlyAspLe

500

401

.3......... 501

600 CTCACCTCGGTGGGGCTTTCCGTGTTCACTCTCACGGCCCTCAGCGCTGACAGGTACAGAGCTATCGT~CCCCATGGACATGCAGACGTCTGGTGTG~ LeuThrSerValGlyValSerVslPhaThrLeuThrAlaLe"SerAl~s~rgTyrAr~AlaIleValAsnProMetAspMetGlnThrSerGlyValV

601

700

701

800

801

5. . . . . . . . . . CCCCTTGTTATCATCAGCATTTATTATTATCACATTGCGMGACTTTMTTAGMGTGCACACMTCTTCCTGGAGMTACMTGMCATACC~GC ProLeuVa1I1eI1~ScrIlcTyrTyrTyrHisIlcAlaLysThrLe"IleArgSerAlaHlsAsn~uProGlyGl"TyrAsnGluHlsThrLysLysG

901

.6 . . . :occ AGATGGAGACACGGRIV\CGCCTGGCCAAGATCGTTCGTTCTGGTGTTTGTGGGCTGCTTTGTCTTCTGCTGGTTTCCCMCCACATCCTCTACTTGTATAGGTC lnHetGl"ThrArgLysArgLeuAlaLysIleValLe"ValPheValGlyCysPheValPheCysT~pPheProAsnHiSIleLe"TyrLeuTyrArqSe

.7.

.

.

900

. ;100

1001

TTTCMCTACMGGAGATCWTCCTTCTCTTGGACACATGATTCTCCTGTGTCMCCCGTTTGCT rPhaAsnTyrLysGluIleAspProSerLe"GlyHisMetIleV~lThrLe"ValA~~ArgValLe"SerPheSerAsnSerCysValAsnProPheAla

1101

: ice CTTTACCTGCTCAGTGAllAGCTTCAGGAAGCATTTCATTTCMCAGCCAGCTCTGTTGTGGGCAGAAGTCCTATCCTGAGAGGTCTACCAGCTACCTCCTCAGCT LeuTyrLe"LeuSerGluSerPheArgLysHisPheAsnSerGlnLe"CysCysGlyGlnLysSerTyrProGluAr~SerThrSerTyrLeuLeuSerS

1201

;30:

1301

!403 STGATCGGAGACCATCCMTTCATCCTCGGGAAATACCATTTTCACMCTTTTCCATTATTATTW\GCGAAGCAGAGCTIW\TMTCACCACATTTACAC uEnd

1401

TGCTCCCCAGCTMTTCATGATTGACTCAAGCGCGCACTCACACG

1501

TAATTCACATATATCTCCTGCTAACATCGGTTTTACATT

:530

1584

NMB-Preferring 423

Bombesin

Receptor cDNA Cloning

expressing the cloned Swiss 3T3 receptors confirm that CRP binds this BN receptor at higher affinitythan NMB. By RNA blotting, GRP-R mRNA is found in mouse pancreas, a rat pancreatic acinar cell line, rat colon, as well as rat and mouse brain (Battey et al., 1991). In contrast, GRP-R mRNA is not detectable in samples from either the esophagus or the olfactory region of the brain, two regions where NMB-Rs are reported in previous receptor autoradiographic studies (von Schrenck et al., 1989; Ladenheim et al., 1990; Lee et al., 1991). These results provide strong evidence for at least two functionally distinct BN receptor subtypes, only one of which corresponds to the GRP-R cDNA clone reported previously. Our strategy to obtain cDNA clones for the NMB-R involved the use of the Swiss 3T3 GRP-R cDNA as a probe for low-stringency hybridization, based on the prediction that the two BN receptor subtypes would cross-hybridize under low-stringency hybridization and wash conditions. The ideal source for cDNA library construction would therefore contain relatively high levels of mRNA for the NMB-R and not contain significant levels of CRP-R mRNA. Previous binding studies suggested that smooth muscle cells in the rat esophagus would have these properties (von Schrenck et al., 1989). In this study, we report the characterization of acDNAclone encoding an NMB-preferring BN receptor subtype (NMB-R) isolated from a rat esophagus cDNA library by low-stringency cloning using a mouse GRP-R probe. Precise localization of brain regions expressing GRP-R and NMB-R mRNA using in situ hybridization indicated that thetwo receptor subtypes are expressed in different subsets of brain regions that possess high-affinity BN receptors. Results Isolation of Candidate NMB-R cDNA Clones A hexamer-primed cDNA library was constructed from rat esophagus and screened by hybridization at low stringencywith the Swiss 3T3 GRP-R cDNA probe. Several candidate clones were isolated, two of which contained the entire coding region of a long open reading frame. Several criteria were used to establish that the cDNA clones encode an NMB-R protein distinct from the GRP-R described previously (Battey et al., 1991). The properties distinguishing these two BN receptor subtypes include protein structure, sensitivity of receptor function to specific antagonists, relative binding affinity for BN peptide ligands, and tissue

Figure 1. Nucleotide Hydropathy Plot

Sequence

and Predicted

Amino

Acid Sequence,

distribution of expression. These properties were studied using the cDNA clones isolated at low stringency from the esophageal cDNA library. The Nucleotide Sequence and Amino Acid Sequence of NMB-R cDNA The nucleotide sequence and predicted amino acid sequence of a single long open reading frame present in two independent clones encoding the putative NMB-R is shown in Figure 1. These cDNAs derive from mRNAs that encode a protein 390 amino acids in length, with a calculated molecular mass of 43 kd. A hydropathy analysis of the predicted NMB-R protein reveals seven stretches of hydrophobic amino acids, consistent with a seven transmembrane structure typical of G protein-coupled receptors. There are three potential sites for N-linked glycosylation (As+, Asn7’, and Asn1v2), consistent with the prediction that the NMB-Rprotein, IiketheGRP-R, maybeaglycoprotein. In Figure 2, the predicted amino acid sequences of the mouse Swiss 3T3 GRP-R and the putative rat NMB-R protein are compared. The putative NMB-R amino acid sequence has greater similarity to the GRP-R than any other sequence reported to date (56% identity). A previously reported comparison of the rat substance P and substance K receptors shows comparable amino acid sequence identity between these two tachykinin receptor subtypes (48% identity) (Yokota et al., 1989). In contrast, the sequence identity between the putative rat NMB-Rand the mouse GRP-R is considerably lower than that observed when the bovine (Masu et al., 1987) and rat (Yokota et al., 1989) substance K receptors are compared (85%). Thus, the sequence identity observed between the mouse Swiss 3T3 CRP-R and putative rat NMB-R protein (56%) is in the range typically observed when two receptor subtypes are compared and is lower than that typically observed when the same receptor subtype is compared between two different mammalian species. Taken together, the sequence identity comparisons were most consistent with the idea that the putative NMB-R protein was a BN receptor subtype distinct from the Swiss 3T3 GRP-R subtype characterized previously. A comparison between the amino acid sequence predicted for the putative NMB-R and other members of the G protein-coupled receptor superfamily shows that many amino acid residues conserved in this family are present at corresponding positions in the NMB-R sequence. Two cysteine residues that may

Derived

from Two Independent

Rat NMB-R cDNA

Clones,

and

(Upper panel) Lines between nucleotide and amino acid sequences indicate the location of seven predicted transmembrane domains (numbered sequentially) based on homologytoother G protein-coupled receptor superfamily members and the hydropathy plot shown below. Heavy dots show the location of potential sites for N-linked glycosylation. (Lower panel) Hydropathy plot generated using the Pepplot program (window = 20 amino acids) in the Sequence Analysis Software Package of the University of Wisconsin Genetics Group (Devereux et al., 1984). Positive regions are relatively hydrophobic, and negative regions are hydrophilic. Putative transmembrane domains are numbered sequentially. Solid line: Kyte-Doolittle criterion. Dotted line: Goldman criterion.

Neuron 424

A

NM6

B

GRP

NMB

GRP

tt GRP-R

c

GRP

NMG

D

GRP

NMB

Figure 3. Effects of [o-Phe6]BN(6-13) Ethyl Ester on AgonistEvoked Cl- Currents in Xenopus Oocytes Expressing the NMB-R or the GRP-R Electrophysiologic responses to NMB and GRP were recorded under voltage clamp at -60 mV. Addition of [o-Phe6]BN(6-13) ethyl ester (IO PM) blocks the responses of the CRP-R to NMB and CRP (1 uM). (A) GRP-R response to agonists alone; (B) GRP-R response to agonists together with antagonists. In contrast, the antagonist has no effect on NMB-R responses. (C) NMB-R response to agonists alone; (D) NMB-R response to agonists together with antagonist. Figure 2. Comparison of the NMB-R Protein

of the Predicted Amino Acid Sequences with that of the GRP-R Protein

The predicted amino acid sequences of the NMB-R (Figure 1) and the mouse Swiss 3T3 CRP-R (Battey et al., 1991) are aligned to maximize homology using the GAP Program in the Software Package of the University of Wisconsin Genetics Computer Group (Devereux et al., 1984). Solid lines between amino acid residues designate amino acid identities in the two sequences. Conserved amino acid residues that are typically conserved in many other known G protein-coupled receptor superfamily members are enclosed in boxes.

form a disulfide linkage (Dixon et al., 1987), situated in the first and second extracellular loop, are conserved in the predicted NMB-R sequence at positions 116 and 198. Another well-conserved cysteine residue that is thought to be important in anchoring the bdrenergic receptor to the plasma membrane (O’Dowd et al., 1989) is also present in the predicted sequence of NMB-R, 14 amino acid residues downstream from the end of the seventh transmembrane domain. In addition, numerous other amino acid residues typicallyconserved in members of the G protein-coupled receptor superfamily are also found in the predicted amino acid sequence of the NMB-R (Figure 2, boxed residues). These similarities indicate that, like the GRP-R, the putative NMB-R is a member of the G protein-coupled receptor superfamily. Analysis of the Functional Properties of the NMB-R To confirm the functional identity of the putative NMB-R cDNA, Xenopus oocytes were injected with RNA transcribed in vitro from cDNA clones containing the entire NMB-R protein-coding domain. After allowing sufficient time for the injected mRNA to be expressed, oocytes were voltage clamped at - 60

mV, and ligand-evoked responses were measured. As shown in Figure 3C, either NMB (10m6M) or GRP (10m6 M) causes a depolarizing current that is typical for IPx- and Ca2+-mediated Cl- channel opening. At lower agonist concentrations (low9 M), only NMB could elicit a detectable response (data not shown). These data establish that the cDNA clones isolated from the esophagus library encode a functional NMB-R that, in contrast to the GRP-R, responds to lower concentrations of NMB than GRP. We then tested the effect of a specific antagonist shown previously to block selectively responses mediated by the GRP-R and not the NMB-R. The des-Met BN analog ([o-Phe6]BN(6-13) ethyl ester) functions as a specific antagonist for the pancreatic GRP-R (Wang et al., 1990), but not the esophageal NMB-R (von Schrenck et al., 1990). This antagonist completely blocks the electrophysiologic response of oocytes expressingthecloned Swiss3T3GRP-Rwhen it isapplied at a IO:1 molar ratio with micromolar concentrations of either GRP or NMB agonists (Figure 3A, GRP-R responses to GRP and NMB agonists; Figure 3B, GRP-R responses to agonists together with antagonist). In contrast, addition of the antagonist along with either NMB or GRP agonist (IO:1 molar ratio) did not diminish the response of the cloned NMB-R expressed in Xenopus oocytes (Figure 3C, NMB-R responses to GRP and NMB agonists alone; Figure 3D, NMB-R responses to agonists together with antagonists). To establish that the differences in physiological response of the receptor to NMB and GRP were due to relative binding affinities, we examined the ligandbinding properties of the cloned receptor expressed

NMB-Preferring 425

Bombesin

Receptor cDNA Cloning

Table 1. Displacement of (‘251-Ty+)BN Binding by CRP, NMB, and [o-Phe6]BN(6-13) Ethyl Ester Antagonist in Different BN Receptor Subtypes

K, (nM)

CONCENTRATION

(Molar1

Figure 4. Displacement of [1251-Tyr4]BN Binding by Either NMB or GRP Agonists to the Cloned NMB-R Expressed in BALB 3T3 Fibroblasts The methods used to analyze whole-cell binding to transfected BALB 3T3 fibroblasts are essentially as described previously (Kris et al., 1987). The binding of labeled BN (40 pM) was determined in the presence of different concentrations of unlabeled GRP1.27 or NMB agonist peptides. Unlabeled NMB displaces binding of labeled BN at lower concentrations than unlabeled GRP (K, NMB = 2 nM; K, CRP = 43 nM), indicating that the NMB peptide binds the cloned NMB-R at higher affinity than GRP peptide.

in BALB 3T3 fibroblasts. Preliminary binding studies showed that BALB 3T3 cells would be an appropriate hostforexpressingthecloned NMB-R,sincetheyhave very low levels of endogenous displaceable BN binding (data not shown). A fragment containing the protein-coding domain was subcloned in a modified pCD2 vector (Wada et al., 1989), which also encodes a gene conferring resistance to the aminoglycoside G418 (Southern and Berg, 1982), allowing selection of stably transfected cells. Ten stably expressing cell lines were selected and screened for NMB-binding activity. One cell line showing high levels of specific binding was selected for detailed analysis. The relative ligand affinity of the transfected NMB-R was assessed by quantitative displacement of [1251Tyr4]BN binding by unlabeled NMB or GRP (Figure 4). NMB was more potent than GRP in displacing labeled BN (K, for NMB = 2 nM; Ki for GRP = 43 nM). Ligand displacement properties determined for the transfected cells are compared in Table 1 with those obtained from esophageal tissue sections, known to express an NMB-R (von Schrenck et al., 1990), as well as the pancreatic acinar cell line AR42) and pancreatic tissue sections, known to express a GRP-R with properties similar to the Swiss 3T3 GRP-R (Wang et al., 1990). NMB was more potent than GRP in displacing [1251-Tyr4]BN bound to transfected BALB 3T3 cells expressing the NMB-R, as was observed in esophagus tissue sections. In contrast, GRP is more potent than NMB in displacing [‘*%Tyr4]BN binding to pancreatic acinar cells, AR42), or Swiss 3T3 cells. These results showthatthecDNAunderstudyencodesafunctiona1 NMB-R with binding properties resembling those of the esophageal NMB-R reported previously (von Schrenck et al., 1989,199O). As expected, the specific GRP-R antagonist [o-Phe6]BN(6-13) ethyl ester binds

Cell Type

NMB

GRP

Antagonist

BALB 3T3/NMB-R Esophagus Pancreas AR421 Swiss 3T3

2 0.3 351 287 62

43 30 15 2 2

>I000 >looo 5.3 2.1 1.6

Displacement of (‘“I-Ty4)BN binding by GRP1.27, NMB, and the GRP-R antagonist [o-Pheb]BN(6-13) ethyl ester (Wang et al., 1990) was analyzed in tissues and cultured cells expressing different BN receptor subtypes. Whole-cell binding studies on cell lines (BALB 3T3/NMB-R transfectants and Swiss 3T3) were performed essentially as described (Kris et al., 1987) using 40 pM labeled BN tracer. Binding displacement analysis of tissue sections and AR42J cells was performed in a very similar manner with a few modifications (van Schrenck et al., 1990). Binding properties of the NMB-R expressed on transfected BALB 3T3 fibroblasts most closely resemble the esophagus NMB-R and are clearly different from CRP-R subtypes found on pancreatic acinar cells, Swiss 3T3 cells, and AR42) cells.

GRP-Rs (pancreas, AR42), and Swiss 3T3) at high affinity (Ki = 1.6-5.3 nM), but has much lower affinity for NMB-Rs on either esophagus or BALB 3T3 cells expressing the cloned NMB-R (K, > 1000 nM) (Table 1).

Tissue Distribution

of NMB-R mRNA

BN receptors have been described in both neural and nonneural tissues as well as various cell lines (for review see Lebacq-Verheyden et al., 1990). To determine which cells express the NMB-R subtype encoded in the cDNA clone, mRNA was examined in various tissues and cell lines using Northern blot hybridization analysis. Poly(A)+ RNA isolated from the rat brain, olfactory region, esophagus, and C6 glioma cell line contains two hybridizing mRNA species present after a high-stringency wash, with estimated sizes of approximately 3.2 kb and 2.7 kb (Figure 5). Both bands were observed together in all expressing tissues and were still present after high-stringency washing, indicating that in all likelihood they are transcripts from the same gene. It is interesting to note that the relative amounts of the 3.2 and 2.7 kb mRNA species vary somewhat in different tissues; the explanation for this observation is not known at present. No NMB-R mRNA was detected in poly(A)+ mRNA samples isolated from pancreas, the AR42J rat pancreatic acinar cell line, and Swiss 3T3 cells, shown previously to express GRP-R mRNA (Batteyet al., 1991). No hybridizing mRNA species were detected by either the GRP-R (Battey et al., 1991) or the NMB-R probe (Figure 5) in mRNA samples from lung, thymus, and BALB 3T3 cells. These results show that the cloned NMB-R mRNA reported in this study is expressed in the brain as well as in the esophagus. In addition, we localized NMB-R mRNA within the brain to the olfactory bulb, a brain region reported to express relatively high levels of binding

Neuron 426

:.

28%

18S-

Figure5. RNABlotAnalysisof mRNASamples Isolated fromvarious Rat Tissues as well as Selected Mouse and Rat Cell Lines Two micrograms of poly(A)+ RNA was resolved by formaldehydeagarose gel electrophoresis, blotted, hybridized to a NMB-R cDNA probe, and washed at high stringency. Samples isolated from rat brain, olfactory region, esophagus, and a rat glioma cell line (C6) show two hybridizing species with estimated sizes of 3.2 kb and 2.7 kb. In contrast, no hybridization signal was observed in samples isolated from ratthymus, lung, a rat pancreatic acinar cell line (AR42)), and the mouse Swiss 3T3 fibroblast cell line.

sites for NMB-Rs 1991).

(Ladenheim

et al., 1990; Lee et al.,

NMB-R and CRP-R mRNA Is Expressed in Different Brain Regions RNA blot hybridization studies on rat brain mRNA using both the NMB-R probe (Figure 5) and the Swiss 3T3 GRP-R probe (Battey et al., 1991) indicated that both BN receptor subtypes are expressed in the brain. NMB-R and GRP-R mRNA expression in the rat CNS was examined in more detail using in situ hybridization histochemistry to determine the correlation between regions expressing the specific cloned NMB-R and GRP-R genes and regions shown in previous ligand-binding autoradiographic studies to express brain BN-binding sites (Wolf et al., 1983; Zarbin et al., 1985; Ladenheim et al., 1990; Lee et al., 1991). Labeled cRNA probes were generated by in vitro transcription of plasmid templates containing a 2 kb EcoRl insert fragment encoding either the rat NMB-R cDNA reported here (Figure 1) or the rat homolog to the Swiss 3T3 GRP-R (showing 95% nucleotide sequence identity in the coding region to the Swiss 3T3 GRP-R) cloned from an AR42) rat pancreatic acinar cell cDNA library (unpublished data). These probes were hybridized to coronal rat brain sections from the olfactory regions (Figures 6A and 6D) as well as thalamic and hypothalamic regions (Figures 6B, 6C, 6E, and 6F),

where labeled BN and NMB binding was prominent in previous studies. Overall, NMB-R expression was most striking in the olfactory and central thalamic regions, while GRP-R expression was most prominent in the hypothalamus. More detailed analysis of the sections showed that NMB-R mRNA expression was highest in the anterior olfactory nucleus, tenia tecta, and piriform cortex. In addition, many other regions, including the accessory olfactory bulb, frontal cortex, thalamic nuclei (paraventricular, anterodorsal, central medial, central lateral, and rhomboid), dentate gyrus, amygdalopiriform nucleus, and dorsal raphe also expressed NMB-R mRNA. GRP-R mRNA expression was highest in the suprachiasmatic nucleus, paraventricular nucleus, nucleus of the lateral olfactory tract, magnocellular preoptic nucleus, and lateral mammillary nucleus. Moderate expression was seen in the bed nucleus of the accessory olfactory tract, lateral hypothalamicarea, supraoptic nucleus, dentategyrus, field CA3 of Ammon’s horn, isocortex, medial amygdaloid nucleus, and nucleus ambiguus. These results show that NMB-R and GRP-R mRNAs are selectively expressed in different rat brain regions. Discussion In this study, we report the isolation and characterization of cDNA clones encoding an NMB-R with properties different from those of the GRP-R described previously (Battey et al., 1991). The two receptors differ in their predicted amino acid sequences, sensitivities to a specific antagonist, ligand-binding preferences, and patterns of expression. Each receptor is expressed in a different subset of brain regions determined previously to express high-affinity BN-binding sites. Analysis of the predicted amino acid sequence of the cloned NMB-R predicts seven hydrophobic membrane-spanning domains, indicating that the NMB-R is another member of the G protein-coupled receptor superfamily. The cloned NMB-R has numerous amino acid residues at corresponding positions in common with a majority of known members in this family. In addition, there is a potential site for N-linked glycosylation in the amino-terminal extracellular domain often found at this position in members of this rapidly expanding gene family. It is not known whether or not the Swiss 3T3 GRP-R and the NMB-R use the same or different G proteins and second messenger signal transduction pathways. This issue is of great interest, since BN receptors mediate signals in many diverse biological responses. Future studies using either the C6 rat glioma cell line or transfected fibroblasts expressing the NMB-R at high levels should allow identification of specific G proteins and other components of the signal transduction pathway activated by the NMB-R. Subtype-specific antagonistsarea powerful tool for separating the functional properties of different receptor subtypes, especially when multiple subtypes are found in the same tissue. Functional analysis of

NMB-Preferring 427

Bombesin

Receptor cDNA Cloning

GRP-R

NMB-R

LH MCPC

Figure 6. Comparison

of the Distribution

of NMB-R and CRP-R mRNAs

in Selected

Regions of the Rat CNS Using In Situ Hybridization

Adjacent coronal sections through the olfactory regions (A and D) or hypothalamus and thalamus (B, C, E, and F) of rat brain were hybridized to either a NMB-R (A-C) or GRP-R (D-F) antisense 35S-labeled cRNA probe. NMB-R-expressing regions: AO, anterior olfactory nucleus; AD, anterodorsal thalamic nucleus; C, central thalamic nucleus; DC, dentate gyrus; IM, intercalated amygdaloid nucleus; ISO, isocortex; Pe, periventricular hypothalamic nucleus; Pi, piriform cortex; PV, paraventricular thalamic nucleus; R, rhomboid thalamic nucleus; TT, tenia tecta. GRP-R-expressing regions: BAOT, bed nucleus of the accessory olfactory tract; CA, field CA3 of Ammon’s horn; DC, dentate gyrus; ISO, isocortex; LH, lateral hypothalamic nucleus; LOT, nucleus of the lateral olfactory tract; MeA, medial amygdaloid nucleus; McPC, magnocellular preoptic nucleus; PVN, paraventricular hypothalamic nucleus; Sch, suprachiasmatic nucleus; SON, supraoptic nucleus. Scale bar in (F) indicates 2.5 m m length.

the cloned NMB-R showed that the potent GRP-R antagonist[o-Phe6]BN(6-13)ethyl ester(Wangetal., 1990) had no effect on the ligand-activated response of the NMB-R expressed in Xenopus oocytes. Ligand displacement studies using unlabeled antagonist as the competitor showed that the Ki for [lz51-Tyr4]BN displacement in the NMB-R was greater than lO-‘j M (Table I), while the K, for the GRP-R is three orders of magnitude lower (about 10Tg M) (Wang et al., 1990). This difference in affinity is presumably the basis for the subtype specificity observed in functional assay systems. This compound may prove useful for studies of individual receptor subtype function in biological systems in which both receptors are present, such as the nervous system. At present, no anaiogous antagonist is available for functional studies involving NMBR-mediated responses. Model systems expressing the cloned NMB-R and GRP-R at high levels, such as in fibroblasts or Xenopus oocytes, should prove helpful in the search for a specific NMB-R antagonist as well as for the development of other pharmacologic agents that specifically modulate the function of each receptor subtype. In situ hybridization studies showed that the brain regions expressing NMB-R mRNA were different from the regions expressing GRP-R mRNA (Figure 6). These results are in very good agreement with previous binding studies defining the location of high-affinity BN-binding sites (Wolf et al., 1983; Zarbin et al., 1985; Ladenheim et al., 1990; Leeet al., 1991). The majority of CNS BN-binding regions are reported to bind labeled NMB selectively (Ladenheim et al., 1990; Lee et al., 1991). As expected, these regions hybridize to the NMB-R probe rather than the GRP-R probe. Previous studies have shown that, in some cases, specific brain regions with high-affinity BN-binding sites are associated with the biological responses to BN peptides in the CNS (reviews by TachC and Brown, 1982; Moody et al., 1988). For example, the effects on temperature regulation are observed after direct injection of BN into the preoptic area, a region where GRP-R mRNA is localized. Injection of BN into the paraventricular nucleus of the rat hypothalamus, where GRP-R mRNA was localized, increases gastric pH as a result of decreased gastric acid secretion and is shown to reduce food intake. In future functional studies in the CNS, it will be critical to determine which BN peptide ligands and receptor subtypes are involved in a given response and to use agonists and antagonists that are appropriate for the different receptor subtypes. Experimental

Procedures

Construction and Screening of Esophagus cDNA library Total RNA was purified from rat esophagus, the polyadenylated fraction was selected by oligo(dT)-cellulose chromatography, and a hexamer-primed cDNA library was constructed using established methods (Davis et al., 1986), except that the cDNA was size-fractionated by Sepharose 46 chromatography rather than polyacrylamide gel electrophoresis prior to ligation to the XgtlO vector. Approximately 7.5 x IO5 plaques from the cDNA library

were screened with ag*P-labeled mouse CRP-R cDNA probe (Battey et al., 1991). Filter hybridization was performed overnight at 37OC using methods described previously (Davis et al., 1986). Filters were washed twice at room temperature for 15 min in 300 m M NaCI, 30 m M sodium citrate, 0.1% SDS and at higher stringency twice for 15 min at 37°C in 15 m M NaCI, 1.5 m M sodium citrate, 0.1% SDS. After autoradiography for 2 days, positive clones were plaque purified, subcloned into plasmid vectors, and sequenced. After identifying a NMB-R candidate in the initial screen, an additional 7.5 x 105 library members were screened at high stringency using a candidate isolated in the first screen as a probe to obtain clones containing the entire open reading frame of the NMB-R. Two positive clones were plaque purified, subcloned into plasmid vectors, and sequenced using standard methodology (Davis et al., 1986). Functional Expression in Xenopus Oocytes RNA was transcribed and capped in vitro from either the NMB-R or GRP-R cDNA clones using T7 RNA polymerase as recommended by the manufacturer (Promega). Defolliculated oocytes were microinjected with about 10 ng of mRNA per oocyte and kept at 20°C in ND 96 solution (96 m M NaCI, 2 m M KCI, 1 m M MgCI, 5 m M Na+ HEPES, 1.8 m M CaCb) (Lupu-Meiri et al., 1989). After 24-48 hr, oocytes were placed in a perfusion chamber and voltage clamped at a holding potential of -60 mV. Ligands were added directly to the chamber, and liganddependent Cl- currents were measured. The GRP1.27 and NM6 peptides were purchased from Peninsula (Burlingame, CA), and the [o-Phe6]BN(613) ethyl ester antagonist was synthesized as described (Wang et al., 1990). Ligand Displacement Analysis An EcoRl fragment from the longest NMB-R cDNA clone encodingtheentireopen readingframewassubcloned intoamodified version of the pCD2 plasmid (Wada et al., 1989). BALB 3T3 cells were transfected with 40 Pg of the NMB-R expression plasmid construct using the calcium phosphate precipitation method (Graham and Van der Eb, 1973) with a few modifications (Davis et al., 1986). Stably transfected cells were selected for resistance to the aminoglycoside C418 (800 ug/ml). After a 3 week selection period, ten clones were selected and screened for high-affinity binding. Binding and displacement studies on the transfected BALB 3T3 cells were performed as described previously (Kris et al., 1987) in 24well tissue culture dishes using 40 pM [Y-Tye]BN purified after labeling by reverse phase high-pressure liquid chromatography (van Schrenck et al., 1990). Each point on the displacement curve was determined four times and the average value was plotted. The BN displacement studies done to determine the K, values for NMB, GRP, and the ethyl ester antagonist on pancreatic and esophagus tissue sections were performed as described previously (van Schrenck et al., 1990). RNA Blot Analysis Poly(A)+ mRNA was prepared from either cell lines or tissues, resolved by electrophoresis on agarose-formaldehyde gels, and blotted to nitrocellulose membranes using standard methodology (Davis et al., 1986). After baking at 80°C, membranes were hybridized to a “P-labeled NMB-R cDNA fragment purified from the longest cDNA clone obtained (2 kb) and washed at high stringency (65OC in 15 m M NaCI, 1.5 m M sodium citrate, 0.1% SDS, twice for 15 min) as described previously (Davis et al., 1986). In Situ Hybridization The method used for in situ hybridization in this study is described in detail elsewhere (Wada et al., 1990). Briefly, adult male rats were fixed by perfusion with 4% paraformaldehyde, 0.05% glutaraldehyde. After perfusion, the brain was removed and placed in postfix solution (4% paraformaldehyde plus 10% sucrose) overnight at 4’C. Sections (25 urn thick) were mounted on polylysine-coated slides and then treated with proteinase K (10 pg/ml, 37OC, 30 min), acetic anhydride and dehydrated by successive immersion in SO%, 70%,95X, and 100% ethanol. 3SS-labeled

NMB-Preferring 429

Bombesin

Receptor cDNA Cloning

sense or antisense cRNA probes (specific activity about 2 x IO9 cpm/ug) were synthesized from a pGEM-4 plasmid vector (Promega) containing a 2.0 kb cDNA fragment encoding either the rat NMB-R (described here) or rat CRP-R (cloned from a rat pancreatic acinar cell line AR42); unpublished data) subcloned in the polylinker region between the SP6 and T7 RNA polymerase promoters. Hybridizations were performed in 50% formamide, 0.3 M NaCI, 10% dextran sulfate, 10 m M dithiothreitol at 55OC overnight with a probe concentration of 5 x IO6 cpm per ml of hybridization buffer. Sections were then washed in a solution containing 4x SSC (lx SSC = 150 m M NaCI, 15 m M sodium citrate [pH 7.01) and 1 m M dithiothreitol at room temperature, incubated with RNAase A (20 ug/ml at 37OC for 30 min), and washed at room temperature with solutions containing progressively lower concentrations of SSC and 1 m M dithiothreitol, beginning with 2x SSC and ending with 0.5x SSC. A final highstringency wash was performed in a solution containing 0.1 x SSC and 1 m M dithiothreitol at 55°C for 30 min. Slides were dehydrated in 50%, 70%, 95%, and 100% ethanol and exposed to 8max film (Amersham) at room temperature for 3-7 days. Acknowledgments We thank Drs. Ellen Ladenheim, Tim Moran, and Terry Moody for providing their data on the localization of ‘“I-NMB-binding sites in the brain before publication. We thank Dr. Keiji Wada for many helpful discussions and Drs. James Patrick, Ben Szaro, Martha Corjay, Zahra Fathi, and Harold Gainer for their careful review of the manuscript. Finally, we thank Samuel Mantey for expert technical assistance in preparation of the ‘“I-BN ligand used in binding displacement sites. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 USC Section 1734 solely to indicate this fact. Received

November

20, 1990; revised

January

Anastasi, A., Erspamer, V., and Bucci, M. (1971). Isolation and structure of bombesin and alytesin, two analogous active peptides from the skin of the European amphibians Bombina and Alytes. Experientia 27, 166167. Battey, I. F., Way, J., Corjay, M. H., Shapira, H., Kusano, K., Harkins, R., Wu, J. M., Slattery, T., Mann, E., and Feldman, R. I. (1991). Molecularcloningof the bombesinlGRP receptor from Swiss 3T3 cells. Proc. Natl. Acad. Sci. USA 88, 395-399. Cuttitta, F., Carney, D. N., Mulshine, J., Moody, T. W., Fedorko, J., Fischler, A., and Minna, J. D. (1985). Bombesin-like peptides can function as autocrine growth factors in human small-cell lung cancer. Nature 376,823-826.

Devereux, J., Haeberli, P., and Smithies, sive set of sequence analysis programs Res. 12, 387-395.

acinar cells. Proc. Natl. Acad. Sci.

Kris, M. R., Hazan, R., Villines, J., Moody, T. W., and Schlessinger, J. (1987). Identification of the bombesin receptor on murine and human cells by cross-linking experiments. J. Biol. Chem. 262, 11215-11220. Ladenheim, E. E., Jensen, R. T., Mantey, S. A., MC Hugh, P. R., and Moran, T. H. (1990). Receptor heterogeneity for bombesin-like peptides in the rat central nervous system. Brain Res. 537,233240. Lebacq-Verheyden, A.-M., Trepel, J., Sausville, E. A., and Battey, J. F. (1990). Peptide growth factors and their receptors II. In Handbook of Experimental Pharmacology, Vol. 95/ll, M. N. Sporn and A. B. Roberts, eds. (Berlin: Springer-Verlag), pp. 71-124. Lee, M. C., Jensen, R. T., Coy, D. H., and Moody, T. W. (1991). Autoradiographic localization of neuromedin B binding sites in rat brain. Mol. Cell. Neurosci., in press. Lupu-Meiri, M., Shapira, H., and Oron,Y. (1989). Dual regulation by protein kinase C of the muscarinic response in Xenopus oocytes. Pflugers Arch. 473, 498-504. Masu, Y., Nakayama, K., Tamaki, H., Harada, Y., Kuno, M., and Nakanishi, S. (1987). cDNA cloning of bovine substance-K receptor through oocyte expression system. Nature 329, 836-838. McDonald, T. J., Jornvall, H., Nilsson, C., Vagne, M., Chatei, M., Bloom, S. R., and Mutt, V. (1979). Characterization of a gastrin releasing peptide from porcine non-antral gastric tissue. Biothem. Biophys. Res. Commun. 90, 227-233. Minamino, N., Kanagawa, B: a novel bombesin-like cord. Biochem. Biophys.

K., and Matsuo, H. (1983). Neuromedin peptide identified in porcine spinal Res. Commun. 774, 541-548.

Moody, T. W., Pert, C. B., Rivier, J., and Brown, M. R. (1978). Bombesin: specific binding to rat brain membranes. Proc. Natl. Acad. Sci. USA 75, 5372-5376. Moody, T. W., Cetz, R., O’Donohue, T. L., and Rosenstein, J. M. (1988). Localization of receptors for bombesin-lake peptides in the rat brain. Ann. NY Acad. Sci. 547, 114-129.

4,lYYl.

Davis, L., Dibner, M., and Battey, J. F. (1986). Basic Methods Molecular Biology (New York, NY: Elsevier), pp. l-388.

brane receptors on pancreatic USA 75, 6139-6143.

in

0. (1984). Acomprehenfor the VAX. Nucl. Acids

Dixon, R. A., Sigal, I. S., Candelore, M. R., Register, R. B., Scattergood, W., Rands, E., and Strader, C. D. (1987). Structural features required for ligand binding to the 8-adrenergic receptor. EMBO J. 6, 3269-3275. Erspamer, C. F., Severini, C., Erspamer, V., Melchiorri, P., delle Fave, C., and Nakajima, T. (1988). Parallel bioassay of 27 bombesin-like peptides on 9 smooth muscle preparations. Structure-activity relationships and bombesin receptor subtypes. Regul. Pept. 27, I-II.

O’Dowd, Bouvier, receptor.

B. F., Hnatowich, M., Caron, M. G., Lefkowitz, R. J., and M. (1989). Palmitoylation of the human BL-adrenergic J. Biol. Chem. 264, 7564-7569.

Regoli, D., Dion, S., Rhaleb, N.-E., Drapeau, C., Rouissi, N., and D’Orleans-Juste, P. (1988). Receptors for neurokinins, tachykinins, and bombesin: a pharmacological study. Ann. NY Acad. Sci. 547, 158-173. Rozengurt, E., and Sinnett-Smith, J. (1983). Bombesin stimulation of DNA synthesis and cell division in cultures of Swiss 3T3 cells. Proc. Natl. Acad. Sci. USA 80, 2936-2940. Southern, P. J.,and Berg, P. (1982J.Transformation of mammalian cells to antibiotic resistance with a bacterial gene under the control of the SV40 early region promoter. I. Mol. Appl. Gen. 7,327341. Spindel, E. R., Giladi, E., Brehm,T. P., Goodman, R. H., and Segerson, T. P. (1990). Cloning and functional characterization of a complementary DNAencoding the murinefibroblast bombesin/ gastrin-releasing peptide receptor. Mol. Endocrinol. 43, 19501963. Tache, Y., and Brown, M. (1982). On the role of bombesin homeostasis. Trends Neurosci. 5, 431-433.

in

von Schrenck,T., Heinz-Erian, P., Moran, T., Mantey, S. A., Gardner, J. D., and Jensen, R. T. (1989). Neuromedin B receptor in esophagus: evidence for subtypes of bombesin receptors. Am. J. Physiol. 256, C747-C758.

for 52,

von Schrenck, T., Wang, L.-H., Coy, D. H., Villanueva, M. L., Mantey, S., and Jensen, R. T. (1990). Potent bombesin receptor antagonists distinguish receptor subtypes. Am. 1. Physiol. 259, G468-G473.

Jensen, R. T., Moody, T., Pert, C., Rivier, J. E., and Gardner, J. D. (1978). Interaction of bombesin and litorin with specific mem-

Wada, K., Dechesne, C. J., Shimasaki, S., King, R. G., Kusano, K., Buonanno, A., Hampson, D. R., Banner, C., Wenthold, R. J., and Nakatani, Y. (1989). Sequence and expression of a frog brain

Graham, F. L., and van der Eb, A. J. (1973). A new technique the assay of infectivity of human adenovirus 5 DNA. Virology 456-467.

NellrOll

430

complementary 342, 684-689.

DNA encoding

a kinate-binding

protein.

Nature

Wada, E., Way, J., Lebacq-Verheyden,A.-M, and Battey, J. (1990). Neuromedin B and gastrin releasing peptide mRNAs are differentiallydistributed in the rat central nervous system. j. Neurosci. 70, 2917-2930. Wang, L.-H., Coy, D. H., Taylor, J. E., Jiang, N.-Y., Moreau, J.-P., Huang, S. C., Frucht, H., Haffar, B. M., and Jensen, R. T. (1990). Des-Met carboxyl-terminally modified analogues of bombesin function as potent bombesin receptor antagonists, partial agonists, or agonists. J. Biol. Chem. 265, 15695-15703. Westendorf, J. M., and Schonbrunn, A. (1983). Characterization of bombesin receptors in a rat pituitary cell line. J. Biol. Chem. 258, 7527-7535. Wolf, S. S., Moody, T. W., O’Donohue, T. L., Zarbin, M. A., and Kuhar, M. J. (1983). Autoradiographic visualization of rat brain binding sites for bombesin-like peptides. Eur. J. Pharmacol. 87, 163-164. Yokota, Y., Sasai, Y., Tanaka, K., Fugiwara, T., Tsuchida, K., Shigemoto, R., Kakizuka, A., Ohkubo, H., and Nakanishi, S. (1989). Molecular characterization of a functional cDNA for rat substance P receptor. J. Biol. Chem. 264, 17649-17652. Zachary, I., and Rozengurt, E. (1985). High-affinity receptors for peptides of the bombesin family in Swiss 3T3 cells. Proc. Natl. Acad. Sci. USA 82, 7616-7620. Zarbin, M. A., Kuhar, M. J., O’Donohue, T. L., Wolf, S. S., and Moody, T. W. (1985). Autoradiographic localization of (‘*si-Tyr4) bombesin-binding sites in rat brain. J. Neurosci. 5, 429-437.