Identification and Cloning of an Orphan G Protein-Coupled Receptor of the Glycoprotein Hormone...

5
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 247, 266–270 (1998) ARTICLE NO. RC988774 Identification and Cloning of an Orphan G Protein-Coupled Receptor of the Glycoprotein Hormone Receptor Subfamily Terry McDonald,* Ruiping Wang,* Wendy Bailey,² Guochun Xie,² Fang Chen,* C. Thomas Caskey,* and Qingyun Liu* ,1 *Department of Human Genetics, Merck Research Laboratories, WP26A-3000, West Point, Pennsylvania 19486; and ²Department of Bioinformatics, Merck Research Laboratories, WP42-300, West Point, Pennsylvania 19486 Received May 4, 1998 fide bridge two cysteine residues located in the first and Mining of the EST database identified a human EST second extracellular loop (3). Within each subfamily, dis- that was predicted to encode a novel member of the tinct, highly conserved motifs have been identified . These glycoprotein hormone receptor subfamily. Based on motifs have been suggested to be critical for the struc- the sequence information, the full-length coding re- tural integrity of the receptor as well as for coupling to gion of this gene was isolated and sequenced. This G proteins. These GPCR-specific motifs, furthermore, can gene, designated HG38, is predicted to encode a poly- be utilized to predict whether a gene encoding a 7-TM peptide of 907 amino acid residues with a putative sig- protein is likely to be a receptor coupled to G protein or nal peptide sequence at its very N-terminus. HG38 is not (4). most closely related to members of the glycoprotein Historically, GPCR genes have been isolated either by hormone receptor subfamily with Ç35% overall iden- expression cloning, i.e., the use of high affinity ligands to tity at the protein sequence level. As with the glycopro- screen expression cDNA libraries, or by the synthesis of tein hormone receptors, HG38 contains a long extra- specific oligonucleotides based on amino acid sequence cellular domain with a total of 16 leucine-rich repeats. Northern blot analysis showed that HG38 was ex- from purified receptor protein. Both approaches have been pressed in skeletal muscle, placenta, spinal cord, and highly successful in cloning receptors for a large variety various regions of the brain. Radiation hybrid map- of ligands, including the cloning of the first b-adrenoceptor ping placed HG38 into human chromosome 12q22-23. and the first muscarinic acetylcholine receptor (5, 6). HG38 is most likely to be a receptor for a novel class Low stringency cross-hybridization and degenerate PCR of glycoprotein ligands. q 1998 Academic Press guided by cloned receptor sequences resulted in the clon- ing of many novel receptor genes that were closely related, as exemplified by the rapid expansion of the seven trans- membrane G protein-coupled receptor family. While these G protein-coupled receptors (GPCRs) are seven trans- approaches are still employed to clone receptors for specific membrane (7-TM) domain proteins that mediate signal ligands, the recent explosion in the number of ESTs (ex- transduction of a large, diverse number of ligands pressed sequence tags) by both public and private efforts through heterotrimeric G proteins (see ref. 1 for review). provides a rich source for the finding and cloning of novel Approximately 140 different subtypes GPCR genes have receptor genes (7). GPCR-encoding ESTs are found by us- been cloned even without counting those genes encoding ing computer search programs with known GPCR as olfactory and gustatory receptors. The GPCRs isolated so query sequences and converted to full-length genes by a far can be classified into three subfamilies based on their variety of experiments approaches. sequence similarity among each other: the rhodopsin sub- Glycoprotein hormones (lutropin or LH, thyrotropin family, the secretin receptor subfamily, and the metabo- or TSH, follitropin or FSH, and human choriogonado- tropic glutamate receptor subfamily (see ref. 2 for review). tropin or hCG) are hetero-dimeric, large peptide mole- A common feature of the three superfamilies is the disul- cules that signal through G protein-coupled receptors (see ref. 8 for review). The receptors of the glycoprotein hormones belong to the rhodopsin subfamily and are 1 To whom correspondence should be addressed. Fax: (215)652- 2075. Ç45% identical at the amino acid level among each 0006-291X/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved. 266

Transcript of Identification and Cloning of an Orphan G Protein-Coupled Receptor of the Glycoprotein Hormone...

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 247, 266–270 (1998)ARTICLE NO. RC988774

Identification and Cloning of an Orphan GProtein-Coupled Receptor of the GlycoproteinHormone Receptor Subfamily

Terry McDonald,* Ruiping Wang,* Wendy Bailey,† Guochun Xie,† Fang Chen,*C. Thomas Caskey,* and Qingyun Liu*,1

*Department of Human Genetics, Merck Research Laboratories, WP26A-3000, West Point, Pennsylvania 19486; and†Department of Bioinformatics, Merck Research Laboratories, WP42-300, West Point, Pennsylvania 19486

Received May 4, 1998

fide bridge two cysteine residues located in the first andMining of the EST database identified a human EST second extracellular loop (3). Within each subfamily, dis-

that was predicted to encode a novel member of the tinct, highly conserved motifs have been identified . Theseglycoprotein hormone receptor subfamily. Based on motifs have been suggested to be critical for the struc-the sequence information, the full-length coding re- tural integrity of the receptor as well as for coupling togion of this gene was isolated and sequenced. This

G proteins. These GPCR-specific motifs, furthermore, cangene, designated HG38, is predicted to encode a poly-be utilized to predict whether a gene encoding a 7-TMpeptide of 907 amino acid residues with a putative sig-protein is likely to be a receptor coupled to G protein ornal peptide sequence at its very N-terminus. HG38 isnot (4).most closely related to members of the glycoprotein

Historically, GPCR genes have been isolated either byhormone receptor subfamily with Ç35% overall iden-expression cloning, i.e., the use of high affinity ligands totity at the protein sequence level. As with the glycopro-screen expression cDNA libraries, or by the synthesis oftein hormone receptors, HG38 contains a long extra-specific oligonucleotides based on amino acid sequencecellular domain with a total of 16 leucine-rich repeats.

Northern blot analysis showed that HG38 was ex- from purified receptor protein. Both approaches have beenpressed in skeletal muscle, placenta, spinal cord, and highly successful in cloning receptors for a large varietyvarious regions of the brain. Radiation hybrid map- of ligands, including the cloning of the firstb-adrenoceptorping placed HG38 into human chromosome 12q22-23. and the first muscarinic acetylcholine receptor (5, 6).HG38 is most likely to be a receptor for a novel class Low stringency cross-hybridization and degenerate PCRof glycoprotein ligands. q 1998 Academic Press guided by cloned receptor sequences resulted in the clon-

ing of many novel receptor genes that were closely related,as exemplified by the rapid expansion of the seven trans-membrane G protein-coupled receptor family. While these

G protein-coupled receptors (GPCRs) are seven trans- approaches are still employed to clone receptors for specificmembrane (7-TM) domain proteins that mediate signal ligands, the recent explosion in the number of ESTs (ex-transduction of a large, diverse number of ligands pressed sequence tags) by both public and private effortsthrough heterotrimeric G proteins (see ref. 1 for review). provides a rich source for the finding and cloning of novelApproximately 140 different subtypes GPCR genes have receptor genes (7). GPCR-encoding ESTs are found by us-been cloned even without counting those genes encoding ing computer search programs with known GPCR asolfactory and gustatory receptors. The GPCRs isolated so query sequences and converted to full-length genes by afar can be classified into three subfamilies based on their variety of experiments approaches.sequence similarity among each other: the rhodopsin sub- Glycoprotein hormones (lutropin or LH, thyrotropinfamily, the secretin receptor subfamily, and the metabo- or TSH, follitropin or FSH, and human choriogonado-tropic glutamate receptor subfamily (see ref. 2 for review). tropin or hCG) are hetero-dimeric, large peptide mole-A common feature of the three superfamilies is the disul- cules that signal through G protein-coupled receptors

(see ref. 8 for review). The receptors of the glycoproteinhormones belong to the rhodopsin subfamily and are1 To whom correspondence should be addressed. Fax: (215)652-

2075. Ç45% identical at the amino acid level among each

0006-291X/98 $25.00Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

266

AID BBRC 8774 / 6955$$$361 06-01-98 20:48:22 bbrcgs AP: BBRC

Vol. 247, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 1. cDNA and predicted peptide sequence of HG38. The predicted signal peptide sequence is double underlined. The seven transmem-brane domains were single underlined. The Genbank accession number of HG38 is AF062006.

267

AID BBRC 8774 / 6955$$8774 06-01-98 20:48:22 bbrcgs AP: BBRC

Vol. 247, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 2. Alignment of HG38 with the FSH, TSH, and LH receptors. Protein sequences of human FSH, TSH, LH receptors are translatedfrom GenBank cDNA sequences (Accession No. M65085 for the FSH receptor, M31774 for the TSH receptor, M63108 for the LH receptor.The seven transmembranes are as marked.

was amplified with two gene-specific primers from a cDNA pool thatother (9-11) . An interesting feature of these receptorscontained the entire sequence, and cloned into the expression vectorare their large extracellular domain with leucine-richpcDNA3.1. The identity of the clones was then verified by completerepeats which are short sequence motifs found in a sequencing.

variety of proteins involved in protein-protein interac- Multi-tissue Northern analysis. Multi-tissue Northern blots weretion (see ref. 12 for review). Here we report the identi- purchased from Clontech (Clontech, Inc.). Hybridization probes were

generated by labeling the entire coding fragment of each gene withfication and cloning of a putative novel member of thea-32P-dCTP using the Redy Prime labeling kit (Amersham Life Sci-glycoprotein hormone receptor subfamily.ence). Hybridizations were carried out at 687C for two hours usingthe ExpressHyb buffer supplied by Clontech. Final washes of the

EXPERIMENTAL PROCEDURES blots were carried out at 607C for 30 min in 0.21 SSC, 0.1% SDS.The membranes were then analyzed by a phosphorimager (Molecu-

EST database mining. Searching of the EST database (dbEST) lar Dynamics).was carried out with the program tblastn using polypeptide se- Chromosomal localization. HG38 was mapped by radiation hy-quences of known G protein coupled receptors (13). Sequences of the brid mapping using the G4 radiation hybrid panel (16). One STSinitial ‘‘hits’’ were then analyzed for their similarity with known (sequenced tagged site) was developed for HG38 and used to amplifysequences and for their capacity to encode critical motifs of GPCRs. from the G4 radiation hybrid panel. The PCR results were scoredA human EST with the Genbank accession # aa 424098 was found and submitted to the Whitehead Institute Genome Center for linkageand its corresponding gene was designated HG38. analysis.

Cloning of full-length genes. The full-length coding sequences ofRESULTS AND DISCUSSIONHG38 was isolated by a modified cDNA RACE(rapid amplification

of cDNA ends) method called RCCA (Reduced Complexity cDNAUsing peptide sequences of known GPCRs as queries,Analysis) from a placenta cDNA library (14,15). After its full-length

sequence was determined, the entire coding sequence of each gene we identified a human EST that potentially encode a

268

AID BBRC 8774 / 6955$$$361 06-01-98 20:48:22 bbrcgs AP: BBRC

Vol. 247, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Ç5.0 kb transcript (Fig. 3). Further analysis indicatedthat HG38 was expressed in various sub-regions of thebrain and also in the spinal cord (data not shown). Thetissue distribution of HG38 is distinct from the othermembers of the glycoprotein hormone receptor family,suggesting unique functions of this novel receptor.

The chromosomal location of HG38 was determinedby radiation hybrid mapping. HG38, was found to betightly linked to the marker WI-5980, indicating that itis located within chromosome 12q22-23. Interestingly,this region contains several other genes involved in

FIG. 3. Multi-tissue Northern analysis of HG38. muscle growth and development. Since HG38 wasfound to be highly expressed in skeletal muscle, it mayplay a role in muscle growth or function.

We identified and cloned an orphan GPCR that isnovel member of the glycoprotein hormone receptorsubfamily. The EST sequence was then used to isolate predicted to be a novel member of the glycoprotein hor-

mone receptor subfamily. The three known members ofits full-length coding sequence with a modified race(rapid amplification of cDNA ends) method RCCA, and the glycoprotein receptors play pivotal roles in growth,

development, and reproduction. This new receptor isthe gene is designated HG38. The DNA and predictedpeptide sequence of HG38 is shown in Fig. 1. HG38 is also likely to play important roles in certain functions

of the body. There are no ‘‘orphan’’ ligands of the glyco-predicted to encode a polypeptide of 907 amino acidresidues. Examination of the protein sequence of HG38 protein hormone family, indicating the existence of an-

other class of glycoprotein hormones that functionsreveals that it contains a putative signal peptide at thevery N-terminus and seven transmembrane domains through G protein-coupled receptors. Elucidation of the

functions of HG38 in vivo may lead to the discovery ofat the C-terminus (Fig. 1). The seven transmembranedomains of HG38 contain some of the high conserved a novel neuroendocrine axis.residues of G protein-coupled receptors of the rhodop-sin family, indicating the HG38 is a member of the G ACKNOWLEDGMENTSprotein coupled receptor family.

The authors are grateful to the staff of the Human Genetics Se-When its sequence was compared to those of otherquencing Core for DNA sequencing. The authors also thank theG protein-coupled receptors, HG38 is most similar toWhitehead Institute for Biomedical Research/MIT Center for Ge-members of the glycoprotein hormone receptor subfam- nome Research for the radiation hybrid mapping results. The authors

ily. It shares an overall identify ofÇ35% with the FSH, thank other members of the Human Genetics Department for valu-TSH, LH/hCG receptors. Alignment of the four peptide able discussions.sequences shows that the transmembrane domains aremuch more conserved (Fig. 2). All the four proteins REFERENCEShave a Glu instead of Asp residue upstream of the Arg

1. Strader, C. D., Fong, T. M., Tota, M. R., Underwood, D., andresidue at the end of the third transmembrane domainDixon, R. A. (1994) Annu. Rev. Biochem. 63, 101–132.(Fig. 2). A signal peptide is also common to all the four

2. Fraser, C. M. (1995) J. Nucl. Med. 36, 17S–21S.receptors. Preliminary results suggested that HG383. Beck-Sickinger, A. G. (1996) Drug Discovery Today 1, 502–513.was not activated by FSH, TSH, and LH (data not4. Michiel van Rhee, A., and Jacobson, K. A. (1996) Drug Dev. Res.shown).

37, 1–38.As with the glycoprotein hormone receptors, HG385. Dixon, R. A., Kobilka, B. K., Strader, D. J., Benovic, J. L., Dohl-is predicted to have a large extracellular domain with man, H. G., Frielle, T., Bolanowski, M. A., Bennett, C. D., Rands,

leucine-rich repeats (LRR). Interestingly, HG38 con- E., Diehl, R. E., et al. (1986) Nature 321, 75–79.tains a total of 16 LRR while the FSH receptor contains 6. Kubo, T., Fukuda, K., Mikami, A., Maeda, A., Takahashi, H.,seven, the LH receptor contains five, the TSH receptor Mishina, M., Haga, T., Haga, K., Ichiyama, A., Kangawa, K., et

al. (1986) Nature 323, 411–416.contains six repeats of this kind. The presence of LRR7. Venter, J. C. (1993) J. Pharm. Pharmacol. 45(Suppl. 1), 355–in HG38 suggest that the ligand(s) for this receptor is

360.likely to be large protein ligands similar to the a and8. Ryan, R. J., Charlesworth, M. C., McCormick, D. J., Milius, R. P.,b subunits of the FSH, TSH, and LH. The increased

and Keutmann, H. T. (1988) FASEB J. 2, 2661–2669.number of LRR in HG38 also suggests that its ligand(s)9. McFarland, K. C., Sprengel, R., Phillips, H. S., Kohler, M., Ro-may be of higher molecular weight. semblit, N., Nikolics, K., Segaloff, D. L., and Seeburg, P. H.

Expression of HG38 in various human tissues was (1989) Science 245, 494–499.examined by multi-tissue Northern blot analysis. 10. Sprengel, R., Braun, T., Nikolics, K., Segaloff, D. L., and See-

burg, P. H. (1990) Mol. Endocrinol. 4, 525–530.HG38 was expressed skeletal muscle and brain as a

269

AID BBRC 8774 / 6955$$$361 06-01-98 20:48:22 bbrcgs AP: BBRC

Vol. 247, No. 2, 1998 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

11. Parmentier, M., Libert, F., Maenhaut, C., Lefort, A., Gerard, C., Zhang, Q., Morsy, M. A., and Caskey, C. T. (1998) Gene 207,1–7.Perret, J., Van Sande, J., Dumont, J. E., and Vassart, G. (1989)

Science 246, 1620–1622. 15. Meyerson, M., Counter, C. M., and Eaton, E. N., Ellisen, L. W.,12. Kobe, B., and Deisenhofer, J. (1994) Trends Biochem. Sci. 19, Steiner, P., Caddle, S. D., Ziaugra, L., Beijersbergen, R. L., Da-

415–421. vidoff, M. J., Liu, Q., Baccehetti, S., Haber, D. A., and Weinberg,R. (1997) Cell 90, 785–795.13. Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman,

D. J. (1990) J. Mol. Biol. 215, 403–410. 16. Cox, D. R., Burmeister, M., Price, E. R., Kim, S., and Myers,R. M. (1990) Science 250, 245–250.14. Liu, Q., Chang, B., Chen, F., Wang, R., MacDonald, T., Gu, M.,

270

AID BBRC 8774 / 6955$$$361 06-01-98 20:48:22 bbrcgs AP: BBRC