General and Comparative Endocrinology 170 (2011) 68–78

Contents lists available at ScienceDirect

General and Comparative Endocrinology

journal homepage: www.elsevier .com/locate /ygcen

Review

Revisiting the evolution of gonadotropin-releasing hormones and their receptorsin vertebrates: Secrets hidden in genomes

Dong-Kyu Kim a,1, Eun Bee Cho a,1, Mi Jin Moon a, Sumi Park a, Jong-Ik Hwang a, Olivier Kah b,Stacia A. Sower c, Hubert Vaudry d, Jae Young Seong a,⇑a Graduate School of Medicine, Korea University, Seoul 136-705, Republic of Koreab Neurogenesis and Oestrogens, Université de Rennes 1, UMR CNRS 6026, IFR140, Campus de Beaulieu, 35042 Rennes cedex, Francec Center for Molecular and Comparative Endocrinology, University of New Hampshire, 46 College Road, Durham, NH 3824, USAd INSERM U413, Neuronal and Neuroendocrine Differentiation and Communication, European Institute for Peptide Research (IFRMP 23), University of Rouen, 76821Mont-Saint-Aignan, France

a r t i c l e i n f o

Article history:Received 4 August 2010Revised 19 October 2010Accepted 23 October 2010Available online 29 October 2010

Keywords:Gonadotropin-releasing hormoneG protein-coupled receptorsComparative genomicsEvolution

0016-6480/$ - see front matter � 2010 Elsevier Inc. Adoi:10.1016/j.ygcen.2010.10.018

Abbreviations: ADAM12, ADAM metallopeptidaseANKRD34A, ankyrin repeat domain 34A; ANKRD34B,alanyl (membrane) aminopeptidase; AP3S1, adaptor-protein fragment; ARRDC2, arrestin domain containinbridging integrator 3; C3AR, C3a anaphylatoxin chuncharacterized protein C20orf141; C20orf194, unchaCGNL1, cingulin-like 1; CILP, cartilage intermediatepeptidase X homolog; CPXM1, carboxypeptidase X mefamily 2, subfamily R, polypeptide 1; DDRGK1, DDRDOCK2, dedicator of cytokinesis 2; DOCK5, dedicatorfactor 4; ELL, elongation factor RNA polymerase II; ELA5; EPRS, glutamyl-prolyl-tRNA synthetase; FAM11FAM196B, family with sequence similarity 196, membinding protein (G protein), beta 5; GOLGA7, golgin Ahomeobox containing 1; HOMER1, homer homolog 1;IDH3B; isocitrate dehydrogenase 3 (NAD+) beta; IGDmember 4; INA, internexin neuronal intermediate filactivating protein 2; IQGAP3, IQ motif containing GTPpotassium channel tetramerisation domain containin⇑ Corresponding author. Fax: +82 2 921 4355.

E-mail address: jyseong@korea.ac.kr (J.Y. Seong).1 These two authors equally contributed to this wor

a b s t r a c t

Gonadotropin-releasing hormone (GnRH) and its G protein-coupled receptor, GnRHR, play a pivotal rolein the control of reproduction in vertebrates. To date, many GnRH and GnRHR genes have been identifiedin a large variety of vertebrate species using conventional biochemical and molecular biological tools incombination with bioinformatic tools. Phylogenetic approaches, primarily based on amino acid sequenceidentity, make it possible to classify these multiple GnRHs and GnRHRs into several lineages. Four verte-brate GnRH lineages GnRH1, GnRH2, GnRH3, and GnRH4 (for lamprey) are well established. Four verte-brate GnRHR lineages have also been proposed—three for nonmammalian GnRHRs and mammalianGnRHR2 as well as one for mammalian GnRHR1. However, these phylogenetic analyses cannot fullyexplain the evolutionary origins of each lineage and the relationships among the lineages. Rapid and vastaccumulation of genome sequence information for many vertebrate species, together with advances inbioinformatic tools, has allowed large-scale genome comparison to explore the origin and relationshipof gene families of interest. The present review discusses the evolutionary mechanism of vertebrateGnRHs and GnRHRs based on extensive genome comparison. In this article, we focus only on vertebrategenomes because of the difficulty in comparing invertebrate and vertebrate genomes due to their markeddivergence.

� 2010 Elsevier Inc. All rights reserved.

ll rights reserved.

domain 12; ADAM19, ADAM metaankyrin repeat domain 34B; ANKrelated protein complex 3, sigmag 2; ARRDC3, arrestin domain conemotactic receptor; C9orf25, uncracterized protein C20orf194; CDClayer protein, nucleotide pyrophomber 1; CPXM2, carboxypeptidase

GK domain containing 1; DHX32,of cytokinesis 5; EBF1, early B-cel

L2, elongation factor RNA polymer3A, family with sequence similar

ber B; GDNF, Glial cell line-deri7; GRINL1A, glutamate receptor, iHOMER2, homer homolog 2; HOMCC3, immunoglobulin superfamil

ament protein, alpha; IQGAP1, IQase activating protein 3; ITGA10, i

g 9; KIAA1161, uncharacterized fam

k.

llopeptidase domain 19; ADD3, adducin 3; ADRA2A, adrenergic alpha 2A receptor;RD34C, ankyrin repeat domain 34C; ANKRD35, ankyrin repeat domain 35; ANPEP,1 subunit; AP3S2, adaptor-related protein complex 3, sigma 2 subunit; APC, apc

taining 3; ARRDC4, arrestin domain containing 4; AVP, arginine vasopressin; BIN3,haracterized protein C9orf25; C15orf44, UPF0464 protein C15orf44; C20orf141,A2, cell division cycle associated 2; CENPC1, centromere protein C 1; CGN, cingulin;sphohydrolase; CILP2, cartilage intermediate layer protein 2; CLPX, caseinolyticX (M14 family), member 2; CRYBB3, crystallin beta B3; CYP2R1, cytochrome P450,DEAH (Asp-Glu-Ala-His) box polypeptide 32; DOCK1, dedicator of cytokinesis 1;

l factor 1; EBF2, early B-cell factor 2; EBF3, early B-cell factor 3; EBF4, early B-cellase II-like 2; ELL3, elongation factor RNA polymerase II-like 3; EPHA5, EPH receptority 113 member A; FAM196A, family with sequence similarity 196, member A;ved neurotrophic factor precursor; GLO1, glyoxalase I; GNB5, guanine nucleotideonotropic, N-methyl D-aspartate-like 1B; HFE2, hemochromatosis type 2; HMBOX1,

ER3, homer homolog 3; IDH2, isocitrate dehydrogenase 2 (NADP+), mitochondrial;y, DCC subclass, member 3; IGDCC4, immunoglobulin superfamily, DCC subclass,motif containing GTPase activating protein 1; IQGAP2, IQ motif containing GTPasentegrin, alpha 10; ITGA11, integrin, alpha 11; ITPA, inosine triphosphatase; KCTD9,

ily 31 glucosidase KIAA1161; KIF7, kinesin family member 7; LIX1, LIX1 homolog.

D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78 69

1. Introduction

Neuropeptides and their receptors originating from a commonancestral gene have become diversified through evolutionary pro-cesses, such as gene/chromosome duplication and gene modifica-tion, generating families of related yet distinct peptides andreceptors [3,9,20]. The hypothesis for two rounds of large-scalegenome duplication in an early vertebrate ancestor has been pro-posed: however the timing of these duplications is controversial[6,8,14,17,18,31,48]. These events have resulted in the formationof up to four copies of each gene followed by gene modificationor loss. In fish, a teleost fish-specific third-round genome duplica-tion occurred after the divergence of fish and land vertebrates, the-oretically producing eight copies of a gene family [19]. Genomeduplication events have produced paralogous chromosomal re-gions, also called paralogons, providing a basis for tracing generelationships and origins [23]. Currently, the large accumulationof genome sequence information for various vertebrate species,combined with recent advances in bioinformatic tools, has allowedlarge-scale genome comparisons [47].

The neuropeptide gonadotropin-releasing hormone (GnRH) andits G protein-coupled receptor GnRHR play a pivotal role in verte-brate reproduction. Since the initial isolation of mammalian GnRHfrom the porcine and ovine hypothalamus [2,24], a number ofGnRH isoforms and their genes have been characterized in verte-brates and invertebrates [16,26,34,42,51–53]. To date, 25 uniquechordate GnRH forms have been identified, all with a similarmolecular architecture [15,16]. Most vertebrate species have twoor three forms of GnRH in the brain: GnRH1, GnRH2, and GnRH3[4] with lampreys having three GnRHs, one GnRH2 and twoGnRH4. Despite variations in the amino acid sequences of theGnRH1 decapeptide, primarily at positions 5 and 8, vertebrateGnRH1s are orthologous to one another because of their conservedexpression patterns and functions [15,32,37,52]. The amino acidsequence of the GnRH2 decapeptide (also called cGnRH2) is fullyconserved in almost all vertebrates with the exception of the lam-prey GnRH2 that has one amino acid difference to gnathostomeGnRHs [4,16,51]. GnRH3, also called salmon GnRH, is a teleostfish-specific form that is generated by gene/genome duplicationof either the GnRH1 or GnRH2 gene [15].

GnRHR is a member of the rhodopsin-like G-protein-coupledreceptor family. It took five years after the first isolation of a cDNA

LIX1L, Lix1 homolog (mouse)-like; LNPEP, leucyl/cystinyl aminopeptidase; MCTP1, multipMEOX2, mesenchyme homeobox 2; MGMT, O-6-methylguanine-DNA methyltransferaseribosomal protein S26; MTFMT, mitochondrial methionyl-tRNA formyltransferase; MYneurofilament light polypeptide; NEFM, neurofilament medium polypeptide; NKX2-3, NK5; NKX2-6, NK2 transcription factor related locus 6; NKX3-1, NK3 homeobox 1; NOP5member 1; NR2F2, nuclear receptor subfamily 2, group F, member 2; NR2F6, nuclear recepmoiety X)-type motif 17; NUP155, nucleoporin 155 kDa; OEP, one-eyed pinhead; OXT, oxribose) polymerase family, member 8; PARP16, poly (ADP-ribose) polymerase family, mbiogenesis factor 11 beta; PGPEP1, pyroglutamyl-peptidase I; PGPEP1L, pyroglutamyl-pepPMVK, phosphomevalonate kinase; PNOC, prepronociceptin; POLR3G, polymerase (RNAdirected) polypeptide G (32kDa)-like; PPP2R2B, protein phosphatase 2, regulatory subuPRUNE, prune homolog; PTPRA, protein tyrosine phosphatase receptor type A; PTPRE, prRFESD, Rieske (Fe-S) domain containing; RGMA, RGM domain family, member A; RGMB, Rglycoprotein; RHOBTB2, Rho-related BTB domain containing 2; RNF115, ring finger protein(TM) and short cytoplasmic domain, (semaphorin) 4A; SEMA4B, sema domain, immuno(semaphorin) 4B; SGSM1, small G protein signaling modulator 1; SLC25A28, solute carrisolute carrier organic anion transporter family, member 3A1; SLCO4C1, solute carrier orgaSNAP-associated protein; SNRPB, small nuclear ribonucleoprotein polypeptides B and B1ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 4; STAP1, signal transducintranscription factor 12; TMEM106B, transmembrane protein 106B; TMPRSS11A, transmTMPRSS11E, transmembrane protease, serine 11E; TUFT1, tuftelin 1; TXNIP, thioredoxinrepeats; UBA6, ubiquitin-like modifier activating enzyme 6; UBAP2, ubiquitin associated p2 family, polypeptide A3; UNC119B, unc-119 homolog B; UNC13A, unc-13 homolog A; UNrepeat domain 70; WDR93, WD repeat domain 93; YTHDC1, YTH domain containing 1; Z

encoding the mammalian GnRHR [35,46], before nonmammalianGnRHR cDNAs were identified [13,44,45,50], likely because of theoccurrence of large structural differences among mammalian andnonmammalian GnRHRs. In particular, nonmammalian GnRHRscontain a C-terminal cytoplasmic tail, unlike mammalian GnRHRs.Nonmammalian GnRHRs have conserved Asp2.50/Asp7.49 residues,whereas mammalian GnRHRs contain Asn2.50/Asp7.49 residues atthese positions [13,50]. With regard to function and ligand affinity,most nonmammalian GnRHRs respond much better to GnRH2 thanGnRH1, but mammalian GnRHRs have a higher affinity for GnRH1than GnRH2 [13,49,50]. After the discovery of nonmammalianGnRHRs, a second form of mammalian GnRHR was identified in amonkey species, the marmoset [25,30]. This mammalian type-IIGnRHR closely resembles nonmammalian GnRHRs in structureand function. Single vertebrate species are known to express multi-ple forms of GnRHR [25,43,50]. Classification of multiple forms ofGnRHR is more complicated than that of GnRH because most ver-tebrate species express more GnRHRs than GnRHs. For example,teleost fish express four or five isoforms of GnRHRs [11,21,27],while amphibian and reptilian species have three forms of GnRHR[10,50]. Phylogenetic analyses of GnRHRs have generally revealedthat four lineages of GnRHR exist, though the tree patterns slightlydiffer and the relationships among lineages have been explaineddifferently [5,11,29,43].

In spite of the extensive phylogenetic analyses of GnRHs andGnRHRs largely based on amino acid sequence identity, the evolu-tionary origin of each lineage and/or the relationship among lin-eages remains unclear. For example, the lineages from whichteleost fish-specific GnRH3 have originated are still unclear. Theevolutionary relationship among the four lineages of GnRHR needsto be further clarified. Availability of genome information fromvarious vertebrate species has allowed large-scale genome com-parisons to explore the origin and relationship of gene families ofinterest [47]. As the function and molecular interaction of GnRH–GnRHR pairs have been largely discussed elsewhere [3,22,32,36],the present article focuses on the evolutionary history of GnRHsand GnRHRs in vertebrates, primarily based on extensive genomecomparison. This article does not cover the evolution of inverte-brate GnRHs and GnRHRs, since invertebrate and vertebrate gen-omes are markedly divergent and cannot be compared.

le C2 domains, transmembrane 1; MCTP2, multiple C2 domains, transmembrane 2;; MKI67, antigen identified by monoclonal antibody Ki-67; MRPS26, mitochondrialL2, myosin light chain 2; MYO5A, myosin VA (heavy chain 12, myoxin); NEFL,

2 transcription factor related, locus 3; NKX2-5, NK2 transcription factor related, locus6, small nucleolar RNA, C/D box 86; NR2F1, nuclear receptor subfamily 2, group F,tor subfamily 2, group F, member 6; NUDT17, nudix (nucleoside diphosphate linkedytocin; PARP6, poly (ADP-ribose) polymerase family, member 6; PARP8, poly (ADP-ember 16; PEX11A, peroxisomal biogenesis factor 11 alpha; PEX11B, peroxisomaltidase I-like; PI4KB, phosphatidylinositol 4-kinase, catalytic, beta; PLIN1, perilipin 1;) III (DNA directed) polypeptide G (32kDa); POLR3GL, polymerase (RNA) III (DNAnit B, beta; PPP2R2D, protein phosphatase 2, regulatory subunit B, delta isoform;

otein tyrosine phosphatase receptor type E; RBM8A, RNA binding motif protein 8A;GM domain family, member B; RHBG, Rh family, B glycoprotein; RHCG, Rh family, C115; SEMA4A, sema domain, immunoglobulin domain (Ig), transmembrane domain

globulin domain (Ig), transmembrane domain (TM) and short cytoplasmic domain,er family 25, member 28; SLC25A37, solute carrier family 25 member 37; SLCO3A1,nic anion transporter family, member 4C1; SMAD4, SMAD family member 4; SNAPIN,; ST8SIA2, ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 2; ST8SIA4,g adaptor family member 1; STC1, stanniocalcin 1; STC2, stanniocalcin 2; TCF12,embrane protease, serine 11A; TMPRSS11B, transmembrane protease, serine 11B;interacting protein; UACA, uveal autoantigen with coiled-coil domains and ankyrinrotein 2; UBOX5, U-box domain containing 5; UGT2A3, UDP glucuronosyltransferaseC13C, unc-13 homolog C; VPS16, vacuolar protein sorting 16 homolog; WDR70, WDNF366, zinc finger protein 366; ZNF710, zinc finger protein 710.

70 D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78

2. Evolution of GnRH peptides

The cDNA sequences corresponding to the GnRH precursorpolypeptide have been isolated in a limited range of vertebratespecies, enabling the construction of a phylogenetic tree in 1999[4]. Subsequently, as more GnRH sequences were identified, thephylogenetic analyses supported this hypothesis that these cDNAsequences represent the division of the GnRH peptide into threedistinct branches, GnRH1, GnRH2, and GnRH3 in addition to anew lineage, GnRH4 [39,41]. GnRH4 represents two GnRHs, lam-prey GnRH-I and -III that are only found in the most basal verte-brate, the sea lamprey [39]. A novel form of GnRH, GnRH2, wasdiscovered in lampreys [16]. In vertebrates and likely due to a gen-ome/gene duplication event, an ancestral gene gave rise to two lin-eages of GnRHs—the gnathostome GnRH and lamprey GnRH-II[16,41]. The gene duplication events that generated the differentfish and tetrapod paralogous groups likely took place within thegnathostome lineage, after its divergence from the ancestral agna-thans. Lamprey GnRH-I and -III (type 4) can be identified now asparalogous homologs of gnathostome GnRH and lamprey GnRH-IIgroup, resulting from a duplication within the lamprey lineage.This implies that there probably was a genome/gene duplicationevent that gave rise to all forms of vertebrate GnRH affecting thecommon ancestor of lamprey and gnathostome isoforms. The gna-thostome branch of the lamprey GnRH-I and -III orthologous groupwas probably lost during evolution.

The teleost fish and tetrapod GnRH1 genes are positioned ingenomic regions containing common loci including BIN3, RHOBTB2,SLC25A37, NKX3-1, NKX2-6, STC1, NEFM, NEFL, DOCK5, KCTD9,CDCA2, EBF2, and GOLGA7, exhibiting well conserved synteny(Fig. 1a and b). Thus, all GnRH1 genes considered in the presentstudy are orthologous to one another. Synteny analysis data com-bined with BLAST search results suggest that zebrafish does notcontain the GnRH1 gene, though its neighboring genes are presentin the corresponding genome region (Fig. 1b). A BLAST search indi-cates the presence of a GnRH1 gene in stickleback that might pro-duce a novel GnRH decapeptide form with Asn at position eight(unpublished data).

For GnRH2 genes, SNRPB, NOP56, IDH3B, and PTPRA are commonGnRH2 neighbors across vertebrate species (Fig. 1c and d). In tetra-pods, GnRH2 genes are located in the genomic region comprisingcommon loci such as EBF4, CPXM1, C20ORF141, FAM113A, VPS16,PTPRA, MRPS26, OXT, AVP, UBOX5, DDRGK1, ITPA, and C20ORF194(Fig. 1c). It is noteworthy that EBF4, a paralog of EBF2 that is foundnear GnRH1, resides near GnRH2, exhibiting a relationship betweenGnRH1- and GnRH2- containing genome fragments. In teleost fish,GnRH2 genes are surrounded by RFESD, UBAP2, KIAA1161,C9ORF25, UNC119B, CRYBB3, MYL2, and GOLGA7. Notably, these fishGnRH2-neighboring genes are also located near the fish GnRH1gene (Fig. 1b and d). Treefam (http://www.treefam.org/) andENSEMBL (http://www.ensembl.org) database analyses reveal thatthese teleost GnRH1/GnRH2-neighboring genes were likely gener-ated by the teleost fish-specific third round genome duplicationwhile the precise mechanism how these genes located near GnRH1and GnRH2 needs to be further investigated.

Fish GnRH3 genes reside within a gene cluster that commonlyincludes INA, DHX32, ADAM12, FAM196A, MGMT, EBF3, PPP2R2D,ADRA2A, ADD3, SLC25A28, and NKX2-3 (Fig. 1f). The similar genearrangement is found in the tetrapod genomes while fish GnRH3orthologs are absent in these genomes (Fig. 1e), indicating that tet-rapod GnRH3 has been lost after divergence of fish and tetrapodlineages. Interestingly, fish and tetrapods have fourth paralogonsrelated to three GnRH-containing paralogons. These paralogonsfound in the tetrapod and fish genomes contain PPP2R2B, ADAM19,EBF1, DOCK2, FAM196B, NKX2-5, and STC2 that are paralogous to

genes found in GnRH-containing genome fragments (Fig. 1g). Thisresult suggests two rounds of GnRH-containing genome duplica-tion in early vertebrates, resulting in four copies of paralogonwhere the fourth GnRH gene has been completely lost either in fishor tetrapod lineages.

Comparison of gene arrangement of these four paralogonsfound in tetrapods reveals that EBF1/2/3/4 are common for all par-alogons. Three paralogons, GnRH1- and GnRH3- containing paralo-gons and the fourth paralogon share paralogous genes one another,while the GnRH2-containing paralogon shares paralogous genessuch as EBF3/4, CPXM1/2, and PTPRA/E only with the GnRH3-con-taining paralogon, indicating closer relationship of the GnRH2-con-taining paralogon with the GnRH3-containing paralogon than withother paralogons. It is also noteworthy that GnRH2 is closely linkedto PTPRA (5–6 kb upstream) such that these two genes have movedtogether during the massive chromosome rearrangement. Like-wise, PTPRE is also located �2 kb upstream of GnRH3 in medaka,tetraodon, stickleback and fugu genomes except for zebrafish gen-ome (Fig. 1e). This observation may revive an old question as towhere the GnRH3 gene has evolved. Genome synteny and closelocalization of PTPRA/E with GnRH2/3 may suggest that GnRH3 ar-ose by a duplication of a GnRH2/GnHR3 common ancestor. How-ever, we cannot exclude the possibility of generation of GnRH3from a GnRH1/GnRH3 common ancestor, as GnRH1 and GnRH3 alsoshare EBF2/3, SLC25A28/37, NKX2-3/6, NEFL/INA, DOCK1/5, andMKI67/CDCA2 as neighbors. Further, phylogenetic analysis showsthe closest relationship between EBF2 (GnRH1-linked) and EBF3(GnRH3-linked) among the EBF family. In addition, because of thesimilarity in distribution of GnRH1 and GnRH3 neurons and theirembryonic origin in teleost fish, the GnRH1 and GnRH3 genes arebelieved to have a common ancestor [15]. Indeed, GnRH3 likely re-places function of GnRH1 when the latter is lost (e.g. Cypriniformesand most Salmoniformes) and vice versa when GnRH3 is lost (e.g.catfish and eel) [15,32]. These genome surveys of teleost fish andtetrapods suggest that two rounds of genome duplication froman ancestral GnRH gene produced three GnRH genes, GnRH1,GnRH2, and GnRH3, while the precise evolutionary mechanism isnot yet conclusive (Fig. 1h). Further, until the lamprey genomehas been annotated and until GnRH genes are identified in chon-drichthyes and hagfish, the origins of the GnRH family in verte-brates will remain unresolved.

3. Evolution of GnRH receptors

Like GnRHs, multiple forms of GnRHR exist in a single verte-brate species. According to published articles and bioinformaticsearch data, bony fish likely have four or five isoforms of GnRHRs[27]. Lampreys have three functional pituitary GnRHRs, one thathas been published [40]; the other two have not (unpublished,Sower). The amphibian species bullfrog (S. catesbeina) and Dybow-ski’s frog (R. dybowskii) have three forms of GnRHR: frog GnRHR1,GnRHR2, and GnRHR3 [36,50]. On the other hand, the Africanclawed frog (X. laevis) is known to have two forms correspondingto frog GnRHR2 and GnRHR3 [45]. Recently, updated genome se-quence information for the Western clawed frog (X. tropicalis orS. tropicalis) revealed the presence of a X. tropicalis GnRHR form(NP_001107547) corresponding to frog GnRHR1. Thus, amphibianspecies likely harbor three forms of GnRHRs. The reptilian leopardgecko has three forms of GnRHR: GnRHR1/III, GnRHR2, andGnRHR2/II which are orthologous to frog GnRHR1, GnRHR2, andGnRHR3, respectively [10,11]. A TBLASTN genome search of the an-ole lizard genome, however, revealed the presence of only oneform of GnRHR corresponding to leopard gecko GnRHR2. The lackof other forms of GnRHR may be due to insufficient genome infor-

Fig. 1. Synteny for the genomic regions comprising GnRH1 (a), fish GnRH1 (b), GnRH2 (c), fish GnRH2 (d), GnRH3 (e), fish GnRH3 (f) and the lost GnRH4 (g) as well as theproposed evolutionary history of these genes (h). Gene loci organization in the genomic region containing the GnRH gene was obtained from the Ensembl Genome Browser(http://www.ensembl.org). The direction of each chromosome (Ch) and scaffold (Sf) is indicated with line arrows. The direction of each gene is represented by a pointed box.Numerics in the pointed box are given when genes in the individual chromosomes are largely reshuffled. Numerics below the gene indicate the location in the chromosome.An absence of the GnRH gene is indicated by an ‘‘X’’ in an open rectangle. Reversed triangles (.) indicate the common paralogous genes across paralogons found in otherchromosomes. Name of paralogous genes are presented in parentheses. The numbers in open circles indicate common neighboring genes found across vertebrate genomefragments containing the GnRH form of a given type. �, Common loci in fish GnRH1- and GnRH2-containing genome fragments. h, Genome duplication.

D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78 71

72 D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78

mation for anole lizard species. Avian species possess two forms ofGnRHR, namely GnRHR1/III and cGnRHR [38]. For some mammalssuch as monkeys and platypuses, two forms of GnRHR exist: mam-malian GnRHR1 and GnRHR2. Because mammalian GnRHR2 exhib-its a close similarity to nonmammalian GnRHRs in structure,function, tissue expression, and ligand selectivity, mammalian typeII and nonmammalian GnRHRs have been grouped together[22,25]. However, in many mammals (e.g. human, chimpanzee,sheep, cow, rat, and mouse), the GnRHR2 gene has been inactivatedor deleted [7].

Phylogenetic analyses based on amino acid sequence identitycategorize known GnRHRs into four lineages of GnRHR or two majorgroups that branch into several lineages [5,11,43]. However, suchclassification of proteins does not fully explain the evolutionaryrelationship of each GnRHR lineage. Thus, additional data from gen-ome synteny analysis are necessary to fully explore the evolutionaryhistory of GnRHR lineages. We have classified vertebrate GnRHRsinto the following four groups based on genome synteny: nonmam-malian type I (GnRHRn1), nonmammalian type II (GnRHRn2),nonmammalian type III/ mammalian type II (GnRHRn3/m2), andmammalian type I (GnRHRm1). We used three X. tropicalis GnRHRsfor comparison of the genome structure across vertebrate speciesfor two reasons. (1) This species has three distinct forms of GnRHR,each belonging to the GnRHRn1, GnRHRn2, or GnRHRn3/m2 group,while other tetrapod species have one or two forms or genome infor-mation is not sufficient. (2) X. tropicalis is a diploid species, avoidinggenome complexity caused by teleost fish-specific genome duplica-tion. The GnRHRm2 lineage belongs to the amphibian GnRHR3group based on a high degree of sequence similarity. GnRHRm1 isa group independent of the three nonmammalian GnRHR groups.

Table 1Classification of vertebrate GnRHRs based on genome synteny combined with phylogenet

Vertebrate species Receptor Classification*

GnRHRn1 GnRHRn1b GnRHRn2

Zebrafish(D. renio)

GnRHR4NP_001091663

GnRHR2NP_001138

Medaka(O. latipes)

GnRHR3NP_001098393

GnRHR1NP_001098352

GnRHR (UENSORLP00

Green pufferfish(T. nigroviridis)

GnRHR1/III-3BAE45698

GnRHR1/III-1BAE45694

GnRHR1/IIBAE45696

Fugu(T. rubripes)

GnRHR (UA)ENSTRUP00000019152

GnRHR (UA)ENSTRUP00000018665

GnRHR (UENSTRUP0

Stickleback(G. aculeatus)

GnRHR (UA)ENSGACP00000014249

GnRHR (UA)ENSGACP00000019583

GnRHR (UENSGACP0

European seabass(D. labrax)

GnRHR2BCAE54807

GnRHR2CCAE54805

GnRHRCAD11992

Bullfrog(R. catesbeina)

GnRHR1AAG42575

GnRHR2AAG42949

Western clawed frog(X. tropicalis)

GnRHR1/IIINP_001107547

GnRHR2/nNP_001107

African clawed frog(X. laevis)

GnRHR1NP_001079

Chicken(G. gallus)

GnRHR1/IIINP_001012627

cGnRH-RNP_989984

Zebra finch(T. guttata)

GnRHR1/IIINP_001012627

GnRHR (UNP_001116

Anole lizard(A. carolinensis)

GnRHR2ENSACAP0

Leopard gecko(E. macularius)

GnRHR1/IIIABB89900

GnRHR2BAD11150

Platypus(O. anatinus)

Mouse(M. musculus)

Marmoset(C. jacchus)

Human(H. sapiens)

* Accession numbers for NCBI GenBank or ENSEMBL are given below the names of the

3.1. Evolution of GnRHRn1 and GnRHRn2 lineages

Genome synteny analysis, combined with phylogenetic analysis,indicates that the GnRHRn1 group includes X. tropicalis GnRHR1,chicken GnRHR1/III, and several GnRHR forms identified in fishspecies such as zebrafish GnRHR4, medaka GnRHR3, tetraodonGnRHR1/III-3, a fugu GnRHR found in scaffold_217 (ENSTRUP00000019152), and a stickleback GnRHR found in group_XIX (ENS-GACP00000014249) (Table 1, Fig. 2a and b). X. tropicalis GnRHR1 isarranged among the RHCG, KIF7, PLIN1, PEX11A, WDR93, ANPEP,AP3S2, ZNF710, IDH2, and SEMA4B loci in scaffold_22. Although mostof these neighboring genes are also clustered in human, marmoset,and mouse chromosomes, X. tropicalis GnRHR1 orthologs are notfound in these species (Fig. 2a). In the avian species chicken and ze-bra finch, X. tropicalis GnRHR1 orthologous genes are found in chro-mosome 10 and chromosome 10_random, respectively. We couldnot find the X. tropicalis GnRHR1 orthologous gene in a reptilian spe-cies, the anole lizard. However, because the reptilian leopard geckoharbors a GnRHR gene, GnRHR1/III [12] exhibiting a high degree ofsequence similarity to X. tropicalis GnRHR1, we believe that the X.tropicalis GnRHR1 ortholog in the anole lizard will be found aftergenome sequencing of this species has been completed. In teleostfish, two forms of GnRHR orthologous to the X. tropicalis GnRHR1are found. They are referred to as fish GnRHRn1 (Fig. 2b) andGnRHRn1b (Fig. 2c) here. The fish GnRHRn1-containing genome frag-ments have a large array of common neighboring genes includingIDH2, ZNF710, SEMA4B, PEX11A, SLCO3A1, RGMA, ST8SIA2, NR2F2,MCTP2, RHCG, and ANPEP which are orthologous to genes found intetrapod GnRHR1-containing genome fragments (Fig. 2b). Thesegenome fragments also include PARP16, CGNL1, TCF12, MYO5A,

ic analysis.

GnRHRn3/m2 GnRHRn3b GnRHRm1

451GnRHR3XP_693111

GnRHR1NP_001138452

A)000015859

GnRHR2NP_001098392

I-2 GnRHR2/nmI-2BAE45702

GnRHR2/nmI-1BAE45700

A)0000014430

GnRHR (UA)ENSTRUP00000014399

GnRHR (UA)ENSTRUP00000005457

A)0000021774

GnRHR (UA)ENSGACP00000004101

GnRHR (UA)ENSGACP00000000651

GnRHRIIAAS49921

GnRHR1ACAE54804

GnRHR3AAG42574

ml548

GnRHR3/IINP_001107549

176GnRHR2NP_001079211

A)830

0000004352GnRHR3/IIABB89901GnRHR (UA)ENSOANP00000020534

GnRHRNP_001116830GnRHRNP_034453

GnRHR2AAK60927

GnRHRXP_002745835

GnRHR2NR_002328

GnRHRAAI13547

receptors. UA, unannotated.

Fig. 2. Synteny for the genomic regions comprising GnRHRn1 (a), fish GnRHRn1 (b), fish GnRHRn1b (c), GnRHRn2 (d), and fish GnRHRn2 (e). Gene loci organization in thegenomic region containing the GnRHR gene was retrieved from the Ensembl Genome Browser (http://www.ensembl.org). The direction of each chromosome (Ch), contig (Ct),group (Gp), and scaffold (Sf) is indicated with line arrows. The direction of each gene is represented by a pointed box. Numerics below the gene indicate the location in thechromosome. An absence of the GnRHR gene is indicated by an ‘‘X’’ in an open rectangle. Genes located in different genome fragments are indicated with diagonals.Pseudogenes are depicted as dashed boxes. Reversed triangles indicated the common paralogous genes across paralogons. Open circles containing numbers indicate commonneighboring genes found across vertebrate genome fragments containing GnRHR forms of any given classification. �, Common loci between fish GnRHRn1- and GnRHRn1b- orbetween fish GnRHRn1- and GnRHRn2-containing genome fragments.

D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78 73

GNB5, CLPX, UACA, ELL3, ITGA11, and PARP6 that are present in tetra-pod GnRHRn2-containing genome fragments, which will be furtherdiscussed. The fish GnRHRn1b are surrounded by ST8SIA2, RGMA,AP3S2, ANPEP, SLCO3A1, NR2F2 and MCTP2 which are also shown in

tetrapod GnRHRn1-containing genomes (Fig. 2c). As these genesare found in both GnRHRn1- and GnRHRn1b-containing fish gen-omes, they including GnRHRn1 and GnRHRn1b have been likely gen-erated by the teleost-specific third round genome duplication.

Fig. 3. Local duplication of GnRHRn1 and GnRHRn2. Location of GnRHRn1 and GnRHRn2 in the same chromosomes of human, chicken, and stickleback (a). The representativeteleost-specific third round duplication of stickleback GnRHRn1- and GnRHRn2-containing genome fragments is shown (b). A proposed model showing local duplication ofGnRHRn1 and GnRHRn2 in vertebrate species (c). The local GnRHRn1 gene duplication is depicted by ‘‘X’’ in the open circle.

74 D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78

The GnRHRn2 group comprises X. tropicalis GnRHR2, chickencGnRHR, anole lizard GnRHR2 (ENSACAP00000004352), and sev-eral fish GnRHRs (Table 1, Fig. 2d and e). The X. tropicalis GnRHR2gene is found in the genomic region containing the MTFMT, CLPX,CILP, PARP16, IGDCC3, IGDCC4, and C15ORF44 loci. The X. tropicalisGnRHR2 ortholog in mammalian species such as human, marmoset,mouse, and platypus has not been found in the correspondinggenomic region. In the chicken and anole lizard, genome fragmentscontaining X. tropicalis GnRHR2 orthologs exhibited good syntenywith that of X. tropicalis. In fish species, MTFMT, ITGA11, UACA,IGDCC3, CILP, GRINL1A, CGNL1, TCF12, ELL3, GNB5, MYO5A, IGDCC4,and PARP6 genes are common in tetrapod genomic regions con-taining GnRHRn2.

It is of interest to note that the fish GnRHRn2-containing frag-ments also share RHCG, IDH2, ZNF710, SEMA4B, ANPEP, AP3S2,SLCO3A1, NR2F2 and MCPT2 that are found either in GnRHRn1- orGnRHRn1b-containing genome fragments (Fig. 2b, c and e). How-

ever, careful comparison of gene arrangement of individual fishspecies reveals that there are no overlapping genes betweenGnRHRn1b- and GnRHRn2- containing fragments, and that neigh-boring genes in GnRHRn1b- plus GnRHRn2-containing fragmentswell pair with those in GnRHRn1-containing fragments (Fig. 2b, cand e). There would be one possible scenario accounting for thisphenomenon. GnRHRn1 and GnRHRn2 were closely localized withinthe same genome before teleost-specific third round genomeduplication and GnRHRn2b has been then deleted after third roundduplication of this genome. The close localization of GnRHRn1 andGnRHRn2 are supported by finding that GnRHRn1- and GnRHRn2-containing genome fragments locate in close vicinity within thesame chromosome with a 25 mb distance on human chromosome15, with a 3 mb distance on chicken chromosome 10, and with a10 mb distance on stickleback group_II (Fig. 3a). Thus, it is likelythat GnRHRn1 and GnRHRn2 were generated by a local gene dupli-cation within the paralogon before divergence of tetrapods and tel-

D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78 75

eost fish. The neighboring genes in this paralogon have split intotwo fragments either within the same chromosome (for human,chicken and stickleback) or on different chromosomes (for otherspecies) by chromosome rearrangement. This may explain the rea-son why tetrapod GnRHRn1- and GnRHRn2-containing genomefragments do not share any common neighboring genes. The splitof neighboring genes in the paralogon is further evidenced by com-parison of gene arrangement of human genome fragments(Fig. 3b). Human chromosome 15 has GnRHRn1 neighbors in90–96 mb region and GnRHRn2 in 44–73 mb region. Paralogousgenes for these GnRHRn1 and GnRHRn2 neighbors are reshuffledin 145–158 mb region of chromosome 1 where the GnRHRm2pseudogene is localized, in 50–115 mb region of chromosome 5,and in 17–20 mb region of chromosome 19 (Fig. 3b). Thus, it islikely that GnRHRn1 and GnRHRn2 have been generated by a local

Fig. 4. Synteny for the genomic regions comprising GnRHRn3/m2 (a), fish GnRHRn3 (b), fisof GnRHRs (f). Open circles containing numbers or ‘‘m’’ indicate common neighboring genclassification. +, Common loci in fish GnRHRn3- and GnRHRn3b-containing genome fragm

duplication before divergence of tetrapods and fish. For fish, thirdround genome duplication produced a copy of GnRHRn1 andGnRHRn2 followed by gene loss of one copy of GnRHRn2 (Fig. 3c).

3.2. Evolution of GnRHRn3/m2 and GnRHRm1 lineages

The GnRHRn3/m2 group includes human GnRHR2 (which is apseudogene because of the early introduction of a stop codon[28]), marmoset GnRHR2, platypus GnRHR2, and X. tropicalisGnRHR3 (Table 1). The marmoset GnRHR2 gene is located in thegenome fragment of chromosome 18 containing the common lociHFE2, TXNIP, POLR3GL, ANKRD34A, LIX1L, RBM8A, PEX11B, ITGA10,ANKRD35, NUDT17, RNF115, PRUNE, PI4KB, CGN, TUFT1, SNAPIN,PMVK, SEMA4B, RHBG, and IQGAP3. While a similar gene arrange-ment exists in human chromosome 1 and mouse chromosome 3,

h GnRHRn3b (c), GnRHRm1 (d), fish GnRHRn3b (e) and proposed evolutionary historyes found across vertebrate genome fragments containing GnRHR forms of any givenents.

76 D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78

these chromosomes have a pseudo GnRHR2 gene for human or lackthe GnRHR2 gene for mouse, respectively. The platypus and X. trop-icalis genomes harbor a marmoset GnRHR2 orthologous gene, asthey demonstrated a higher degree of amino acid sequence iden-tity with marmoset GnRHR2 than other GnRHR types examined.Synteny data is missing because of insufficient genome informa-tion for these species. We could not find marmoset GnRHR2 ortho-logs in avian and reptilian species using BLAST searches (Fig. 4aand Table 1). In fish such as zebrafish, stickleback, fugu, and tetra-odon, two copies of the marmoset GnRHR2-like gene have beenidentified and are referred to as GnRHRn3 (Fig. 4b) and GnRHRn3b(Fig. 4c) lineages here. In the fish genome containing GnRHRn3,marmoset GnRHR2-neigboring genes such as PMVK, ANKRD34A, IQ-GAP3, IGDCC3L, POLR3GL, PRUNE, PI4KB, HFE2, and NUDT17 are clus-tered. Fish genome fragments containing GnRHRn3b share IGDCC3L,TXNIP, SNAPIN and PEX11B with those of human and marmoset.These genome fragments also share common TMEM106B, MEOX2and SMAD4 genes with those containing GnRHRn3 in fish (Fig. 4band c), indicating that GnRHRn3 and GnRHRn3b genes have beenderived from a common origin and were likely generated duringfish-specific genome duplication.

Mammalian GnRHR1 genes are surrounded by EPHA5, CENPC1,STAP1, UBA6, TMPRSS11A, TMPRSS11B, YTHDC1, TMPRSS11E2, andUGT2A3 on chromosome 4 in humans and on chromosome 5 inmice. Similar gene locations are observed for marmoset chromo-some 3 and platypus contig_214. Although many GnRHRm1-neigh-boring genes are also clustered in the genome fragment of chicken(chromosome 4), zebra finch (chromosome 4), anole lizard (scaf-fold_316), and X. tropicalis (scaffold_333), GnRHRm1 orthologousgenes are missing in these species (Fig. 4d). This gene cluster hasdiverged in fish species such as fugu and zebrafish. The BLASTsearch also revealed that the GnRHRm1 orthologous genes were ab-sent in all fish species examined. Thus, the structural and func-tional origin of GnRHRm1 is uncertain. In mammals, GnRHR1-containing genome fragments have common YTHDC1 and UGT2A3genes, while no fish GnRHR genes are present in the genome frag-ment containing these two genes. The YTHDC1 and UGT2A3 geneslocalize with APC near the C3AR, GDNF, WDR70, SGSM1, and/or

Fig. 5. Proposed evolutionary history of GnRHRs. Official gene names providedwww.ensembl.org) are presented in each box of the lineage. Accession numbers for thGnRHR-containing genome fragments have resulted in the generation of four GnRHR linelikely generated by a local gene duplication just before divergence of fish and tetrapodcommon ancestor. The third and fourth genome fragments lost the GnRHR gene. GnRHRchromosomes. Teleost-specific genome duplication has produced the fish-specific GnRHRis shown by an ‘‘X’’ in a white box. The pseudogene is marked with a dashed box. The ‘‘?’lineage. This is because the gene could not be identified due to insufficient genome info

NUP155 genes within a �6 mb distance. Very interestingly, in zeb-rafish chromosome 10, a GnRHRn3 form exists within this clusterthat lacks the YTHDC1 and UGT2A3 genes (Fig. 4e), raising the pos-sible relationship between GnRHRm1 and GnRHRn3.

We propose a possible evolutionary history for GnRHR as shownin Fig. 5. The presence of third and fourth paralogons sharing para-logous genes in the GnRHR-containing genome fragments (Fig. 3b)suggests two rounds of large-scale genome duplication occurredduring early vertebrate evolution. GnRHRn1 and GnRHRn2 are likelygenerated by local gene duplication within a paralogon beforedivergence of fish and tetrapod lineages. Later then, neighboringgenes within this paralogon have split into two fragments on dif-ferent chromosomes in some species. The GnRHRn3/m2-contain-ing paralogon shares paralogous genes with either GnRHRn1- orGnRHRn2-containing paralogon, indicating GnRHRn3/m2 arose byfirst or second round genome duplication from a common ancestor.The third and fourth paralogons lost the GnRHR gene. Origin ofGnRHRm1 is currently uncertain. GnRHRm1 might translocate fromeither the third or fourth paralogons to other chromosomes beforeor after divergence of tetrapod and fish lineages. GnRHRm1 thensurvived only in mammalian species. In addition, fish have twoadditional GnRHR lineages known as GnRHRn1b and GnRHRn3bthat might have been produced by fish-specific genome duplica-tion of the GnRHRn1 and GnRHRn3 lineages, respectively.

4. Concluding remarks

Genome synteny, combined with phylogenetic analyses of tele-ost fish and tetrapods, revealed that two rounds of genome dupli-cation events may have generated three vertebrate lineages of theGnRH peptides GnRH1, GnRH2, and GnRH3. The GnRH3 gene hasbeen lost in the tetrapod. The presence of lamprey-specificGnRH-I and -III (GnRH4 group), however, reveals more compli-cated evolutionary scenario of the GnRH gene during early verte-brate evolution. This issue will be further disclosed whengenome data of agnathan and chondrichtyan species are available.For GnRHRs, four GnRHR lineages have most likely been generated

by NCBI (http://www.ncbi.nlm.nih.gov/) or Ensembl Genome Browser (http://e indicated genes are shown in Table 1. Two rounds of duplication of an ancestralages: GnRHRn1, GnRHRn2, GnRHRn3/m2, and GnRHRm1. GnRHRn1 and GnRHRn2 arelineages. GnRHRn3/m2 arose by first or second round genome duplication from am1 likely translocated from either the third or fourth genome fragments to other

lineages, GnRHRn1b and GnRHRn3b. An absence of the GnRHR gene in a given lineage’ mark in the box indicates the possible presence of the GnRHR gene in the indicatedrmation. , Genome/gene duplication.

D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78 77

by two rounds of genome duplication as four copies of paralogonssharing paralogous genes one another exist in tetrapod genomeswhile two paralogons lost GnRHR after or before divergence of fishand tetrapod. Currently, most known GnRHRs have been namednumerically in the order of their discovery or based on phyloge-netic relationships [1,33]. Alternatively, ligand affinity toward thethree different forms of GnRH, expression patterns such as brainand pituitary distribution, or a specific domain (SD/EP, PEY, andPPS) at the extracellular loop 3 has been used for classification[5,49,50]. However, the nomenclature for GnRHRs provided by pre-vious annotation strategies do not correlate well with those givenby the combination of genome synteny and phylogenetic analysesperformed here, underscoring the need for intensive reannotationof vertebrate GnRHR genes. Together, GnRH and GnRHR providean excellent model for understanding the molecular evolution ofpeptide ligands and their receptor pairs at the genome level.

Acknowledgments

This work was supported by a grant (M103KV010005-08K2201-00510) to J.Y.S. from the Brain Research Center of the 21st CenturyFrontier Research Program and a STAR exchange program to J.Y.S.and H.V.

References

[1] J. Bogerd, W.B. Diepenbroek, E. Hund, F. van Oosterhout, A.C. Teves, R. Leurs, M.Blomenrohr, Two gonadotropin-releasing hormone receptors in the Africancatfish: no differences in ligand selectivity, but differences in tissuedistribution, Endocrinology 143 (2002) 4673–4682.

[2] R. Burgus, M. Butcher, M. Amoss, N. Ling, M. Monahan, J. Rivier, R. Fellows, R.Blackwell, W. Vale, R. Guillemin, Primary structure of the ovine hypothalamicluteinizing hormone-releasing factor (LRF) (LH-hypothalamus-LRF-gaschromatography-mass spectrometry-decapeptide-Edman degradation), Proc.Natl. Acad. Sci. U. S. A. 69 (1972) 278–282.

[3] H.J. Cho, S. Acharjee, M.J. Moon, D.Y. Oh, H. Vaudry, H.B. Kwon, J.Y. Seong,Molecular evolution of neuropeptide receptors with regard to maintaininghigh affinity to their authentic ligands, Gen. Comp. Endocrinol. 153 (2007) 98–107.

[4] R.D. Fernald, R.B. White, Gonadotropin-releasing hormone genes: phylogeny,structure, and functions, Front. Neuroendocrinol. 20 (1999) 224–240.

[5] C.A. Flanagan, C.C. Chen, M. Coetsee, S. Mamputha, K.E. Whitlock, N.Bredenkamp, L. Grosenick, R.D. Fernald, N. Illing, Expression, structure,function, and evolution of gonadotropin-releasing hormone (GnRH) receptorsGnRH-R1SHS and GnRH-R2PEY in the teleost, Astatotilapia burtoni,Endocrinology 148 (2007) 5060–5071.

[6] C. Fried, S.J. Prohaska, P.F. Stadler, Independent Hox-cluster duplications inlampreys, J. Exp. Zool. B Mol. Dev. Evol. 299 (2003) 18–25.

[7] P.M. Gault, K. Morgan, A.J. Pawson, R.P. Millar, G.A. Lincoln, Sheep exhibit novelvariations in the organization of the mammalian type II gonadotropin-releasing hormone receptor gene, Endocrinology 145 (2004) 2362–2374.

[8] P.W. Holland, J. Garcia-Fernandez, N.A. Williams, A. Sidow, Gene duplicationsand the origins of vertebrate development, Development 33 (1994) 125–133.Suppl.

[9] C.H. Hoyle, Neuropeptide families and their receptors: evolutionaryperspectives, Brain Res. 848 (1999) 1–25.

[10] T. Ikemoto, M. Enomoto, M.K. Park, Identification and characterization of areptilian GnRH receptor from the leopard gecko, Mol. Cell. Endocrinol. 214(2004) 137–147.

[11] T. Ikemoto, M.K. Park, Identification and molecular characterization of threeGnRH ligands and five GnRH receptors in the spotted green pufferfish, Mol.Cell. Endocrinol. 242 (2005) 67–79.

[12] T. Ikemoto, M.K. Park, Comparative analysis of the pituitary and ovarian GnRHsystems in the leopard gecko: signaling crosstalk between multiple receptorsubtypes in ovarian follicles, J. Mol. Endocrinol. 38 (2007) 289–304.

[13] N. Illing, B.E. Troskie, C.S. Nahorniak, J.P. Hapgood, R.E. Peter, R.P. Millar, Twogonadotropin-releasing hormone receptor subtypes with distinct ligandselectivity and differential distribution in brain and pituitary in the goldfish(Carassius auratus), Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 2526–2531.

[14] S.Q. Irvine, J.L. Carr, W.J. Bailey, K. Kawasaki, N. Shimizu, C.T. Amemiya, F.H.Ruddle, Genomic analysis of Hox clusters in the sea lamprey Petromyzonmarinus, J. Exp. Zool. 294 (2002) 47–62.

[15] O. Kah, C. Lethimonier, G. Somoza, L.G. Guilgur, C. Vaillant, J.J. Lareyre, GnRHand GnRH receptors in metazoa: a historical, comparative, and evolutiveperspective, Gen. Comp. Endocrinol. 153 (2007) 346–364.

[16] S.I. Kavanaugh, M. Nozaki, S.A. Sower, Origins of gonadotropin-releasinghormone(GnRH) in vertebrates: identification of a novel GnRH in a basalvertebrate, the sea lamprey, Endocrinology 149 (2008) 3860–3869.

[17] S. Kuraku, A. Meyer, S. Kuratani, Timing of genome duplications relative to theorigin of the vertebrates: did cyclostomes diverge before, or after?, Mol Biol.Evol. 26 (2009) 47–49.

[18] D. Larhammar, L.G. Lundin, F. Hallbook, The human Hox-bearing chromosomeregions did arise by block or chromosome (or even genome) duplications,Genome Res. 12 (2002) 1910–1920.

[19] T.A. Larsson, F. Olsson, G. Sundstrom, L.G. Lundin, S. Brenner, B. Venkatesh,D. Larhammar, Early vertebrate chromosome duplications and theevolution of the neuropeptide Y receptor gene regions, BMC. Evol. Biol. 8(2008) 184.

[20] Y.R. Lee, K. Tsunekawa, M.J. Moon, H.N. Um, J.I. Hwang, T. Osugi, N. Otaki, Y.Sunakawa, K. Kim, H. Vaudry, H.B. Kwon, J.Y. Seong, K. Tsutsui, Molecularevolution of multiple forms of kisspeptins and GPR54 receptors in vertebrates,Endocrinology 150 (2009) 2837–2846.

[21] C. Lethimonier, T. Madigou, J.A. Muñoz-Cueto, J.J. Lareyre, O. Kah, Evolutionaryaspects of GnRHs, GnRH neuronal systems and GnRH receptors in teleost fish,Gen. Comp. Endocrinol. 135 (2004) 1–16.

[22] J.H. Li, H. Choe, A.F. Wang, K. Maiti, C. Wang, A. Salam, S.Y. Chun, W.K. Lee, K.Kim, H.B. Kwon, J.Y. Seong, Extracellular loop 3 (EL3) and EL3-proximaltransmembrane helix 7 of the mammalian type I and type II gonadotropin-releasing hormone (GnRH) receptors determine differential ligand selectivityto GnRH-I and GnRH-II, Mol. Pharmacol. 67 (2005) 1099–1110.

[23] L.G. Lundin, Evolution of the vertebrate genome as reflected inparalogous chromosomal regions in man and the house mouse, Genomics 16(1993) 1–19.

[24] H. Matsuo, Y. Baba, R.M. Nair, A. Arimura, A.V. Schally, Structure of the porcineLH- and FSH-releasing hormone. I. The proposed amino acid sequence,Biochem. Biophys. Res. Commun. 43 (1971) 1334–1339.

[25] R. Millar, S. Lowe, D. Conklin, A. Pawson, S. Maudsley, B. Troskie, T. Ott, M.Millar, G. Lincoln, R. Sellar, B. Faurholm, G. Scobie, R. Kuestner, E. Terasawa, A.Katz, A novel mammalian receptor for the evolutionarily conserved type IIGnRH, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9636–9641.

[26] K. Miyamoto, Y. Hasegawa, M. Nomura, M. Igarashi, K. Kangawa, H. Matsuo,Identification of the second gonadotropin-releasing hormone in chickenhypothalamus: evidence that gonadotropin secretion is probably controlledby two distinct gonadotropin-releasing hormones in avian species, Proc. Natl.Acad. Sci. U. S. A. 81 (1984) 3874–3878.

[27] N. Moncaut, G. Somoza, D.M. Power, A.V. Canario, Five gonadotrophin-releasing hormone receptors in a teleost fish: isolation, tissue distributionand phylogenetic relationships, J. Mol. Endocrinol. 34 (2005) 767–779.

[28] K. Morgan, D. Conklin, A.J. Pawson, R. Sellar, T.R. Ott, R.P. Millar, Atranscriptionally active human type II gonadotropin-releasing hormonereceptor gene homolog overlaps two genes in the antisense orientation onchromosome 1q.12, Endocrinology 144 (2003) 423–436.

[29] K. Morgan, R.P. Millar, Evolution of GnRH ligand precursors and GnRHreceptors in protochordate and vertebrate species, Gen. Comp. Endocrinol.139 (2004) 191–197.

[30] J.D. Neill, L.W. Duck, J.C. Sellers, L.C. Musgrove, A gonadotropin-releasinghormone (GnRH) receptor specific for GnRH II in primates, Biochem. Biophys.Res. Commun. 282 (2001) 1012–1018.

[31] S. Ohno, Evolution by Gene Duplication, Allen & Unwin Springer-Verlag, NewYork, 1970.

[32] K. Okubo, Y. Nagahama, Structural and functional evolution of gonadotropin-releasing hormone in vertebrates, Acta Physiol. (Oxf). 193 (2008) 3–15.

[33] K. Okubo, S. Nagata, R. Ko, H. Kataoka, Y. Yoshiura, H. Mitani, M. Kondo, K.Naruse, A. Shima, K. Aida, Identification and characterization of two distinctGnRH receptor subtypes in a teleost, the medaka Oryzias latipes,Endocrinology 142 (2001) 4729–4739.

[34] J.F. Powell, Y. Zohar, A. Elizur, M. Park, W.H. Fischer, A.G. Craig, J.E. Rivier, D.A.Lovejoy, N.M. Sherwood, Three forms of gonadotropin-releasing hormonecharacterized from brains of one species, Proc. Natl. Acad. Sci. U. S. A. 91 (1994)12081–12085.

[35] J. Reinhart, L.M. Mertz, K.J. Catt, Molecular cloning and expression of cDNAencoding the murine gonadotropin-releasing hormone receptor, J. Biol. Chem.267 (1992) 21281–21284.

[36] J.Y. Seong, L. Wang, D.Y. Oh, O. Yun, K. Maiti, J.H. Li, J.M. Soh, H.S. Choi, K. Kim,H. Vaudry, H.B. Kwon, Ala/Thr(201) in extracellular loop 2 and Leu/Phe(290) intransmembrane domain 6 of type 1 frog gonadotropin-releasing hormonereceptor confer differential ligand sensitivity and signal transduction,Endocrinology 144 (2003) 454–466.

[37] N.M. Sherwood, D.A. Lovejoy, I.R. Coe, Origin of mammalian gonadotropin-releasing hormone, Endocrinol. Rev. 14 (1993) 241–254.

[38] M. Shimizu, G.Y. Bedecarrats, Identification of a novel pituitary-specificchicken gonadotropin-releasing hormone receptor and its splice variants,Biol. Reprod. 75 (2006) 800–808.

[39] M.R. Silver, H. Kawauchi, M. Nozaki, S.A. Sower, Cloning and analysis of thelamprey GnRH-III cDNA from eight species of lamprey representing the threefamilies of Petromyzoniformes, Gen. Comp. Endocrinol. 139 (2004) 85–94.

[40] M.R. Silver, N.V. Nucci, A.R. Root, K.L. Reed, S.A. Sower, Cloning andcharacterization of a functional type II gonadotropin-releasing hormonereceptor with a lengthy carboxy-terminal tail from an ancestral vertebrate,the sea lamprey, Endocrinology 146 (2005) 3351–3361.

[41] S.A. Sower, M. Freamat, S.I. Kavanaugh, The origins of the vertebratehypothalamic–pituitary–gonadal (HPG) & hypothalamic–pituitary thyroid(HPT) endocrine systems: new insights from lampreys, Gen. Comp.Endocrinol. 161 (2009) 16120–16129.

78 D.-K. Kim et al. / General and Comparative Endocrinology 170 (2011) 68–78

[42] J.A. Tello, J.E. Rivier, N.M. Sherwood, Tunicate gonadotropin-releasing hormone(GnRH) peptides selectively activate Ciona intestinalis GnRH receptors and thegreen monkey type II GnRH receptor, Endocrinology 146 (2005) 4061–4073.

[43] J.A. Tello, N.M. Sherwood, Amphioxus: beginning of vertebrate and end ofinvertebrate type GnRH receptor lineage, Endocrinology 150 (2009) 2847–2856.

[44] C. Tensen, K. Okuzawa, M. Blomenrohr, F. Rebers, R. Leurs, J. Bogerd, R. Schulz,H. Goos, Distinct efficacies for two endogenous ligands on a single cognategonadoliberin receptor, Eur. J. Biochem. 243 (1997) 134–140.

[45] B.E. Troskie, J.P. Hapgood, R.P. Millar, N. Illing, Complementary deoxyribo-nucleic acid cloning, gene expression, and ligand selectivity of a novelgonadotropin-releasing hormone receptor expressed in the pituitary andmidbrain of Xenopus laevis, Endocrinology 141 (2000) 1764–1771.

[46] M. Tsutsumi, W. Zhou, R.P. Millar, P.L. Mellon, J.L. Roberts, C.A. Flanagan, K. Dong,B. Gillo, S.C. Sealfon, Cloning and functional expression of a mouse gonadotropin-releasing hormone receptor, Mol. Endocrinol. 6 (1992) 1163–1169.

[47] H.N. Um, J.M. Han, J.I. Hwang, S.I. Hong, H. Vaudry, J.Y. Seong, Molecularcoevolution of kisspeptins and their Receptors from fish to mammals, Ann. N.Y. Acad. Sci. 1200 (2010) 67–74.

[48] K. Vandepoele, W. De Vos, J.S. Taylor, A. Meyer, Y. Van de Peer, Major events inthe genome evolution of vertebrates: paranome age and size differconsiderably between ray-finned fishes and land vertebrates, Proc. Natl.Acad. Sci. U. S. A. 101 (2004) 1638–1643.

[49] C. Wang, O. Yun, K. Maiti, D.Y. Oh, K.K. Kim, C.H. Chae, C.J. Lee, J.Y. Seong,H.B. Kwon, Position of Pro and Ser near Glu7.32 in the extracellular loop 3of mammalian and non-mammalian gonadotropin-releasing hormone(GnRH) receptors is a critical determinant for differential ligand selectivityfor mammalian GnRH and chicken GnRH2, Mol. Endocrinol. 18 (2004) 105–116.

[50] L. Wang, J. Bogerd, H.S. Choi, J.Y. Seong, J.M. Soh, S.Y. Chun, M. Blomenrohr, B.E.Troskie, R.P. Millar, W.H. Yu, S.M. McCann, H.B. Kwon, Three distinct types ofGnRH receptor characterized in the bullfrog, Proc. Natl. Acad. Sci. U. S. A. 98(2001) 361–366.

[51] R.B. White, J.A. Eisen, T.L. Kasten, R.D. Fernald, Second gene for gonadotropin-releasing hormone in humans, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 305–309.

[52] M.S. Yoo, H.M. Kang, H.S. Choi, J.W. Kim, B.E. Troskie, R.P. Millar, H.B. Kwon,Molecular cloning, distribution and pharmacological characterization of anovel gonadotropin-releasing hormone ([Trp8] GnRH) in frog brain, Mol. Cell.Endocrinol. 164 (2000) 197–204.

[53] L. Zhang, J.A. Tello, W. Zhang, P.S. Tsai, Molecular cloning, expression pattern,and immunocytochemical localization of a gonadotropin-releasing hormone-like molecule in the gastropod mollusk, Aplysia californica, Gen. Comp.Endocrinol. 156 (2008) 201–209.