Short sequence-paper

Y-receptor-like genes GPR72 and GPR73: molecular cloning, genomicorganisation and assignment to human chromosome 11q21.1 and 2p14

and mouse chromosome 9 and 61

Rachel Parker a, Marjorie Liu a, Helen J. Eyre b, Neal G. Copeland c,Debra J. Gilbert c, Joanna Crawford b, Grant R. Sutherland b, Nancy A. Jenkins c,

Herbert Herzog a;*a Garvan Institute of Medical Research, Neurobiology Program, St. Vincent's Hospital, Darlinghurst, Sydney, NSW 2010, Australia

b Centre for Medical Genetics, Department of Cytogenetics and Molecular Genetics, Women's and Children's Hospital,Adelaide, S.A. 5006, Australia

c Mammalian Genetics Laboratory, ABL-Basic Research Program, Frederick Cancer Research and Development Centre,Frederick, MD 21702-1201, USA

Received 17 November 1999; received in revised form 8 December 1999; accepted 24 January 2000

Abstract

Two novel G-protein-coupled receptors, one from human, GPR72, and one from mouse, GPR73 have been isolated,sequenced and their genomic organisation determined. Non-isotopic in situ hybridisation and radiation hybrid mapping haveidentified GPR72 to be localised on human chromosome 11q21.1, and GPR73 on human chromosome 2p14. Interspecificmouse backcross mapping has localised the genes to mouse chromosomes 9 and 6, respectively. Northern analysis revealsGPR72 mRNA expression only in brain tissue. However, GPR73 mRNA can be found in heart, skeletal muscle andpancreas. Both receptors are closely related with 36 and 33% overall amino acid identity, respectively, to the Y-receptorfamily. However, although successful cell surface expression in a heterologous expression system can be achieved no specificbinding to this ligand family can be detected, indicating that perhaps additional factors are required for binding. ß 2000Elsevier Science B.V. All rights reserved.

Keywords: G-protein coupled receptor; Neuropeptide Y; Ligand binding; Cell-surface expression

The neuropeptide Y (NPY) gene family, which in-cludes peptide YY (PYY) and pancreatic polypeptide

(PP) mediates its important physiological e¡ects onappetite, regulation of blood pressure, reproductionand anxiety through interaction with speci¢c G-pro-tein-coupled receptors. The use of various cloningtechniques has so far identi¢ed ¢ve NPY receptorsubtypes (Y1, Y2, Y4, Y5 and y6) in human [1].However, several reports suggest the existence of fur-ther members of this family. For example, the Y3subtype, a receptor recognising NPY, but not PYY,has been proposed [2]. Binding studies also suggest

0167-4781 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 7 8 1 ( 0 0 ) 0 0 0 2 3 - 3

* Corresponding author. Garvan Institute of Medical Re-search, 384 Victoria Street, Darlinghurst, Sydney, NSW 2010,Australia. Fax: +61-2-9295-8281;E-mail : h.herzog@garvan.unsw.edu.au

1 Sequence data from this article have been deposited with theEMBL/GenBank Data Libraries under accession nos. AF236081and AF236082.

BBAEXP 91367 30-3-00

Biochimica et Biophysica Acta 1491 (2000) 369^375www.elsevier.com/locate/bba

Y2-like and PYY-preferring subtypes, mainly ex-pressed in the peripheral nervous system [3]. Se-quence analyses of the cloned human Y1, Y2, Y4,Y5 and y6 receptors reveals substantial di¡erences inthe primary amino acid sequence between the recep-tor subtypes with identities as low as 30% [1]. Thismakes the Y-receptor family one of the most diver-gent families within the G-protein-coupled receptorsuperfamily and cloning attempts to isolate relatedfamily members by low stringency screening a verydi¤cult task.

In an attempt to search for additional Y-receptorsubtypes we have utilised the limited sequence con-servation within the extra cellular loop region I andthe transmembrane regions TMVI of this receptorfamily for the design of degenerate primers in orderto amplify DNA from pooled human and mousecDNA libraries. Oligonucleotides, (5P-TGGRTNWT-YGGNRANGKNMTGTGY-3P and 5P-AANAVN-KKNARNGGNADCCARCA-3P) where used undervarying annealing conditions to amplify cDNA se-quences from several di¡erent pooled cDNA libra-ries. The cDNA libraries were made from mRNAsoriginating from human or mouse tissues includingbrain, heart, kidney, pancreas and testis. Analysis of

the PCR ampli¢cation products by agarose gel elec-trophoresis showed bands of the expected size (450^550 bp) in all of the library samples. Subsequentsubcloning and sequencing revealed that six of the32 clones analysed corresponded to G-protein-coupled receptor sequences. Four of these were iden-tical to known Y-receptor subtypes. Two additionalindependent clones (Hz3 from a human heart libraryand Hz8 from a mouse brain library) showed signi¢-cant sequence identity to Y-receptor subtypes, where-as the remaining clones did not show any amino acidsequence features identifying them as G-protein-coupled receptors. The partial cDNAs were labelledand used to screen the cDNA libraries from whichthey originated in order to obtain full-length clones.In both cases, positive hybridising clones were iden-ti¢ed and isolated.

A 2.8-kb cDNA corresponded to the PCR productHz3 encoding a 423 amino acid long protein nowcalled GPR72. When compared to the GenBank Da-tabase, GPR72 was identi¢ed as the human homo-logue of the mouse orphan receptor GIR showing85% amino acid identity [4]. However, GPR72 alsoexhibits an overall identity of 36% to the human Y2receptor subtype. A human P1 genomic library (Ge-

Fig. 1. GPR72 and GPR73 gene structure. (A) A schematic representation of the 22-kb region containing the human GPR72 genewith a complete restriction map for the enzymes BamHI, EcoRI, and HindIII and (B) of the 10-kb region containing the mouseGPR73 gene with a complete restriction map for the enzymes ApaI, BamHI and PstI is shown. Coding sequences of the exons are in-dicated as black boxes. The position of the translation initiation site ATG and the stop signal are shown. The intron positions for theGPR72 and GPR73 genes in relation to the putative transmembrane domains are shown below each gene.

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375370

nome Systems) was screened with the entire GPR72cDNA probe under high stringency. Two positiveclones were identi¢ed and isolated. The P1 DNAwas digested with the restriction enzyme EcoRIthat has no recognition site within the coding regionof the corresponding GPR72 cDNA. SubsequentSouthern analysis identi¢ed four EcoRI fragments2, 4, 5 and 6 kb in size in the human P1 clones,indicating the presents of large introns in theGPR72 gene. Subcloning, sequencing with internalprimers and mapping of the gene revealed that theGPR72 gene consists of four exons separated bythree introns of 4.2, 2.7 and 12.6 kb in size whichare located after the putative transmembrane sequen-

ces II, III and IV, respectively (Fig. 1A).The full-length cDNA corresponding to the PCR

fragment Hz8 is 3.6 kb in size and encodes a 392amino acid receptor protein. Comparison of thisnovel mouse receptor protein GPR73 with the Gen-Bank Database identi¢ed again the Y2 receptor asbeing the most closely related sequence with 33%overall amino acid identity. A mouse BAC genomiclibrary (Genome Systems) was screened with the en-tire GPR73 cDNA probes under high stringency.One positive clone was identi¢ed and isolated. TheBAC DNA was digested with the restriction enzymePstI and Southern analysis identi¢ed a 3.7- and a4.5-kb hybridising PstI fragment also indicating thepresence of intronic sequence in that gene. Sequenc-ing and complete mapping of the region containingthe GPR73 gene revealed that it consists of twoexons separated by a 7.1-kb intron located at theborder of TM III within the common DRY sequencemotive (Fig. 1B). In both genes, the nucleotide se-quences of the introns adjoining the splice junctions(Tables 1 and 2) are consistent with the recognisedconsensus sequence GT/AG [5].

To determine the chromosomal localisation of thehuman GPR72 receptor gene, the complete 100-kb

Table 1Exon/intron organisation of human GPR72 gene

Fig. 2. Human chromosomal localisation. Metaphase spread showing FISH with the GPR72 probe. Normal male chromosomesstained with DAPI. Hybridisation sites on chromosome 11 are indicated by arrows.

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375 371

P1 clone was nick-translated with biotin-14-dATPand hybridised in situ at a ¢nal concentration of 20ng/Wl to metaphase chromosomes from two normalmales [6]. Twenty metaphases were examined for£uorescent signal. All of these metaphases showedsignals on one or both chromatids of chromosome11 in the region q14^q22 and 55% of this signal wasat 11q21 (Fig. 2). There are only four non-speci¢cbackground dots in these 20 metaphases. Localisa-tion to chromosome 11q21 was con¢rmed by radia-tion hybrid analysis identifying the markerD11S4176 as most closely associated with the genefor GPR72 with a LOD score of s 1000 (data notshown). Interestingly, this chromosomal region hasrecently being linked with increased body fat in

Pima Indians [7] suggesting a possible role for thisreceptor in the regulation of energy balance.

The chromosomal localisation of the human ho-mologue of GPR73 was also determined by radiationhybrid mapping. The analysis indicated GPR73 ismost closely associated with the Stanford chromo-some 2 marker SHGC-3686 (with a LOD score of8.8) which is located between markers D2S147 andD2S2171. These markers have been mapped previ-

Table 2Exon/intron organisation of mouse GPR73 gene

Fig. 3. Murine chromosomal location of Gpr72 and Gpr73. The segregation patterns of Gpr72 and Gpr73 and their £anking genes in128 and 115 backcross animals, respectively, that were typed for all loci are shown at the top of the ¢gure. For individual pairs ofloci, more animals were typed (see text). Each column represents the chromosome identi¢ed in the backcross progeny that was inher-ited from the (C57BL/6JUM. spretus)F1 parent. Black boxes represent the presence of a C57BL/6J allele and white boxes representthe presence of an M. spretus allele. The number of o¡spring inheriting each type of chromosome is listed at the bottom of each col-umn. Partial chromosome 9 and 6 linkage maps showing the location of Gpr72 and Gpr73 in relation to linked genes are shown atthe bottom of the ¢gure. Recombination distances between loci in centiMorgans are shown to the left of the chromosome and the po-sitions of loci in human chromosomes, where known, are shown to the right. References for the human map positions of loci cited inthis study can be obtained from (Genome Data Base), a computerised database of human linkage information maintained by TheWilliam H. Welch Medical Library of The Johns Hopkins University (Baltimore, MD).

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375372

ously to a YAC contig which lies in the region span-ning 2p13.3 to 2p14 [8].

The mouse chromosomal location of Gpr72 andGpr73 was determined by interspeci¢c backcrossanalysis using progeny derived from matings of[(C57BL/6JUMus spretus)F1UC57BL/6J] mice [9].This interspeci¢c backcross mapping panel has beentyped for over 2900 loci that are well distributedamong all the autosomes as well as the X chromo-some. C57BL/6J and M. spretus DNAs were digestedwith several enzymes and analysed by Southern blothybridisation for informative restriction fragmentlength polymorphisms (RFLPs) using mouse probes[10]. The 2.1-kb BglII M. spretus RFLP was used tofollow the segregation of the Gpr72 locus in back-cross mice. The mapping results indicated that Gpr72is located in the proximal region of mouse chromo-some 9 linked to Mmel, Fut4, and Ldlr [11,12] (Fig.3). The ratios of the total number of mice exhibitingrecombinant chromosomes to the total number ofmice analysed for each pair of loci, the recombina-tion frequencies (expressed as genetic distances incentiMorgans (cM) þ the standard error) and themost likely gene order are: centromere^Mmel^(3/161); 1.9 þ 1.1^[Gpr72^0/158^Fut4^4/150]; 2.7 þ 1.3^Ldlr. No recombinants were detected between

Gpr72 and Fut4 in 158 animals typed in commonsuggesting that the two loci are within 1.9 cM ofeach other (upper 95% con¢dence limit).

The V21.0- and 1.8-kb HindIII M. spretusRFLPs, which co-segregated, were used to followthe segregation of the Gpr73 locus in backcrossmice. Gpr73 mapped to the central region of mousechromosome 6 linked to Mad and Mitf [13,14] (Fig.3). The most likely gene order is : centromere^Mad^0.9 þ 0.8^Gpr73^6.3 þ 2.1^Mitf. The proximal regionof mouse chromosome 9 and the central region ofmouse chromosome 6 share homology with humanchromosomes 11q and 19p as well as human chro-mosome 2p and 3p, respectively. This is in goodagreement with our human localisation data whichplaces GPR72 to 11q21.1 and GPR73 to 2p14.

Northern blot analyses indicate that GPR72 is ex-clusively expressed in brain (Fig. 4A). This is inagreement with the expression pattern found forthe mouse homologue GIR for which levels ofmRNA are also highest in brain tissue [4]. ThemRNA for GPR73 is found predominantly in pe-ripheral tissues with heart, skeletal muscle and pan-creas showing the highest level of expression. Butlower levels of mRNA are also seen in brain, lung,liver and kidney (Fig. 4B).

Fig. 4. Northern analysis of mRNA expression in di¡erent human tissues. (A) Left panel, GPR72; right panel, GPR73. Lanes:1, heart; 2, brain; 3, placenta; 4, lung; 5, liver; 6, skeletal muscle; 7, kidney; 8, pancreas. (B) A control (L-actin) for loading isshown.

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375 373

The mRNA expression pattern for the GPR72 andGPR73 receptor mRNAs overlaps with the distribu-tion pattern of the known Y-receptors and their li-gands [15]. Therefore it is possible that these novelreceptor proteins may be stimulated endogenously byany of the three peptides of the NPY family withmultiple physiological e¡ects. However, both recep-tors when expressed transiently or stably in a heter-ologous expression system like HEK293 or COS6cells, fail to show speci¢c binding to neuropeptideY (NPY), peptide YY (PYY) or pancreatic polypep-tide (PP) under standard radioligand binding assayconditions [16]. In addition, neither of the receptorsshows an e¡ect on the activation of second-messen-ger systems, including the inhibition of cAMP accu-mulation, common to all other known Y-receptorfamily members [1] (data not shown). In situ hybrid-isation histochemistry with speci¢c riboprobes dem-onstrated signi¢cant receptor mRNA levels in thesecell lines indicating correct transcription of the trans-gene (data not shown). In order to eliminate thepossibility that the lack of binding and signalling inthese cell lines is due to the inability of the recombi-nant receptor proteins to reach the cell surface green-£uorescence-protein-fusion constructs for the two re-ceptor molecules were generated and tested in tran-sient and stable transfected cell lines. Such G-pro-tein-coupled-GFP-fusion proteins have been shownbefore to form correctly integrated and fully func-tional membrane proteins [17]. In both instances,strong £uorescent signals for both receptor-fusionproteins were visualised in the cell membrane ofthese cells by confocal microscopy, proving the pres-ence of the fusion protein. However, the ability tobind NPY ligands or to activate signalling pathwaysin these cell lines is still absent.

Considering the high sequence identity betweenGPR72 and GPR73 and the Y-receptor family, it israther surprising that these novel receptors are notactivated by known NPY family ligands. In addition,extensive evolutionary analysis, investigating the ge-nomic localisation of Y-receptors across a wide rangeof species, including Zebra ¢sh, indicates that themost likely position for further Y-receptor familymembers is next to the UCP3 locus on human chro-mosome 11q21 (D. Larhammar, personal communi-cation) the exact position where GPR72 is located.Furthermore, the fact that this chromosomal region

is also associated with increased body fat in PimaIndians [7], together with the facts that NPY is thestrongest stimulus of food intake known and is ableto induce obesity, makes GPR72 a strong candidatefor a novel Y-receptor subtype. However, the lack ofspeci¢c binding of NPY, PYY and PP to these novelreceptors could suggest that additional members ofthe NPY ligand family exist or it might be that as yetunknown factors are required for speci¢c activationof these receptor molecules, similar to RAMPs in thecase of CGRP receptors [18].

Acknowledgements

This work was supported by the National Healthand Medical Research Council of Australia and inpart, by the National Cancer Institute, DHHS, undercontract with ABL. We thank Deborah B. House-holder and Olga Antinova for excellent technical as-sistance.

References

[1] A. Blomqvist, H. Herzog, Trends Neurosci. 20 (1997) 294^298.

[2] S.R. Glaum, R.J. Miller, H. Rhim, D. Maclean, L.M. Geor-gic, R.G. MacKenzie, L. Grundemar, Br. J. Pharm. 120(1997) 481^487.

[3] S.J. Wimalawansa, J. Biol. Chem. 270 (1995) 18523^18530.[4] M.T. Harrigan, N.F. Campbell, S. Bourgeois, Mol. Endo-

crinol. 5 (1991) 1331^1338.[5] R. Breathnach, P. Chambon, Annu. Rev. Biochem. 50

(1981) 349^383.[6] D.F. Callen, E. Baker, H.J. Eyre, J.E. Chernos, J.A. Bell,

G.R. Sutherland, Ann. Genet. 33 (1990) 219^221.[7] R.A. Norman, D.B. Thompson, T. Foroud, W.T. Garvey,

P.H. Bennett, C. Bogardus, E. Ravussin, Am. J. Hum. Gen-et. 60 (1997) 166^173.

[8] M.D. Adams, A.R. Kerlavage, R.D. Fleischmann, R.A.Fuldner, C.J. Bult, N.H. Lee, E.F. Kirkness, K.G. Wein-stock, J.D. Gocayne, O. White et al., Nature 377 (6547suppl.) (1995) 3^174.

[9] N.G. Copeland, N.A. Jenkins, Trends Genet. 7 (1991) 113^118.

[10] N.A. Jenkins, N.G. Copeland, B.A. Taylor, B.K. Lee,J. Virol. 43 (1982) 26^36.

[11] D.M. Kingsley, N.A. Jenkins, N.G. Copeland, Genetics 123(1989) 165^172.

[12] K.T. Vaughan, A. Mikami, B.M. Paschal, E.L.F. Holzbaur,S.M. Hughes, C.J. Echeverri, K.J. Moore, D.J. Gilbert,

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375374

N.G. Copeland, N.A. Jenkins, R.B. Vallee, Genomics 36(1996) 29^38.

[13] S. Edelho¡, D.E. Ayer, A.S. Zervos, E. Streingr|̈msson, N.A.Jenkins, N.G. Copeland, R.N. Eiseman, R. Brent, C.M. Dis-tech, Oncogene 9 (1994) 665^668.

[14] C.A. Hodgkinson, K.J. Moore, A. Nakayama, E.Steingr|̈msson, N.G. Copeland, N.A. Jenkins, H. Arnheiter,Cell 74 (1994) 395^404.

[15] R.M.C. Parker, H. Herzog, Eur. J. Neurosci. 11 (1999)1431^1448.

[16] H. Herzog, Y.J. Hort, H.J. Ball, G. Haynes, J. Shine, L.A.Selbie, Proc. Natl. Acad. Sci. USA 88 (1992) 10931^10935.

[17] T. Awaji, A. Hirasawa, M. Kataoka, H. Shinoura, Y. Na-kayama, T. Sugawara, S. Izumi, G. Tsujimoto, Mol. Endo-crinol. 12 (1998) 1099^1111.

[18] L.M. McLatchie, N.J. Fraser, M.J. Main, A. Wise, J.Brown, N. Thompson, R. Solari, M.G. Lee, S.M. Foord,Nature 393 (1998) 333^339.

BBAEXP 91367 30-3-00

R. Parker et al. / Biochimica et Biophysica Acta 1491 (2000) 369^375 375