The relaxin family peptide receptors and their ligands: New developments and paradigms in the...

General and Comparative Endocrinology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

General and Comparative Endocrinology

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

The relaxin family peptide receptors and their ligands: Newdevelopments and paradigms in the evolution from jawlessfish to mammals

http://dx.doi.org/10.1016/j.ygcen.2014.07.0140016-6480/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding authors. Current addresses: Department of Immunology,University of Toronto, 1 King’s Circle, Toronto, ON, Canada (S. Yegorov); Departmentof Biology, University of Winnipeg, 599 Portage Ave., Winnipeg, MB, Canada(S.V. Good).

E-mail address: [email protected] (S.V. Good).

Please cite this article in press as: Yegorov, S., et al. The relaxin family peptide receptors and their ligands: New developments and paradigms in tlution from jawless fish to mammals. Gen. Comp. Endocrinol. (2014), http://dx.doi.org/10.1016/j.ygcen.2014.07.014

Sergey Yegorov a,⇑, Jan Bogerd b, Sara V. Good a,⇑a Department of Biology, University of Winnipeg, 599 Portage Ave., Winnipeg, MB, Canadab Department of Biology, Faculty of Science, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history:Available online xxxx

Keywords:Relaxin family peptide receptors (RXFP)Relaxin and insulin-like peptidesInsulin superfamilyPhylogeneticsGene nomenclatureGPCR evolutionWhole genome duplication

Relaxin family peptide receptors (Rxfps) and their ligands, relaxin (Rln) and insulin-like (Insl) peptides,are broadly implicated in the regulation of reproductive and neuroendocrine processes in mammals.Most placental mammals harbour genes for four receptors, namely rxfp1, rxfp2, rxfp3 and rxfp4. The num-ber and identity of rxfps in other vertebrates are immensely variable, which is probably attributable tointraspecific variation in reproductive and neuroendocrine regulation. Here, we highlight several inter-esting, but greatly overlooked, aspects of the rln/insl-rxfp evolutionary history: the ancient origin, recruit-ment of novel receptors, diverse roles of selection, differential retention and lineage-specific loss of genesover evolutionary time. The tremendous diversity of rln/insl and rxfp genes appears to have arisen fromtwo divergent receptors and one ligand that were duplicated by whole genome duplications (WGD) inearly vertebrate evolution, although several genes, notably relaxin in mammals, were also duplicatedvia small scale duplications. Duplication and loss of genes have varied across lineages: teleosts retainedmore WGD-derived genes, dominated by those thought to be involved in neuroendocrine regulation(rln3, insl5 and rxfp 3/4 genes), while eutherian mammals witnessed the diversification and rapid evolu-tion of genes involved in reproduction (rln/insl3). Several genes that arose early in evolutionary historywere lost in most mammals, but retained in teleosts and, to a lesser extent, in early diverging tetrapods.To elaborate on their evolutionary history, we provide updated phylogenies of the Rxfp1/2 and Rxfp3/4receptors and their ligands, including new sequences from early diverging vertebrate taxa such as coela-canth, skate, spotted gar, and lamprey. We also summarize the recent progress made towards under-standing the functional biology of Rxfps in non-mammalian taxa, providing a new conceptualframework for research on Rxfp signaling across vertebrates.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

Relaxin family peptide receptors (Rxfps) and their ligands,relaxin (Rln) and insulin-like (Insl) peptides, are broadly implicatedin the regulation of reproductive and neuroendocrine processes inmammals. Despite playing important roles in animal physiology, ithas been little more than a decade since Rxfps have been deorph-anized (Halls et al., 2006; Hsu et al., 2002) and the comparativeendocrinology of the Rln/Insl-Rxfp signaling systems in vertebratesoutside mammals remains for the most part enigmatic. Humans

and the majority of other placental mammals harbour genes cod-ing for four RXFP proteins, namely RXFP1, RXFP2, RXFP3 andRXFP4, which bind and are preferentially stimulated by the pep-tides relaxin (RLN, or RLN1 and RLN2 in human), insulin-like pep-tide 3 (INSL3), relaxin-3 (RLN3), and insulin-like peptide 5 (INSL5),respectively (Bathgate et al., 2013). The number and identity ofRxfps in other vertebrates, as elucidated by several recent studies,are immensely variable and can be attributed to the differences inreproductive and neuroendocrine regulation that exist among dif-ferent vertebrate taxa, varying from only three genes in chicken tosix in opossum to eleven genes in zebrafish. In line with this, thephylogeny of vertebrate Rxfps closely reflects their taxonomic ori-gin and shows strong evidence of the influence of whole genomeduplications (WGDs) and lineage-specific gene loss on the extantvertebrates’ gene repertoires (Yegorov and Good, 2012).

he evo-

http://dx.doi.org/10.1016/j.ygcen.2014.07.014

mailto:[email protected]


http://www.sciencedirect.com/science/journal/00166480

http://www.elsevier.com/locate/ygcen


Table 1Proposed revised nomenclature for the rln/insl (ligands) and rxfp (receptors) genesbased on their evolutionary relationship and history in vertebrates.

Predicted post-2R andcoelacanth genes

Humanorthologues

Teleost post-3Rgenes

Ligands rln RLN* rlninsl3 INSL3 insl3rln3 RLN3 rln3a

rln3binsl5 INSL5 insl5a

insl5b

Receptors rxfp1 RXFP1 rxfp1rxfp2 RXFP2 rxfp2a or rxfp2

rxfp2b^

rxfp2-like � rxfp2-like^

rxfp3-1 RXFP3 rxfp3-1rxfp3-2 � rxfp3-2a

rxfp3-2brxfp3-3 � rxfp3-3a1 and

rxfp3-3a2 **

rxfp3-3brxfp3-4 RXFP4 rxfp4**

* The RLN locus has undergone extensive duplication in placental mammals; insl4and insl6 (not shown in the Table) duplicated in the ancestor to placental mammals(Arroyo et al., 2012a; Hoffmann and Opazo, 2011), while rln1 duplicated in the lastcommon ancestor of catarrhine primates (Arroyo et al., 2014).

^ Present in zebrafish and cavefish, but a pseudogene in other teleosts.** rxfp3-3a1 and rxfp3-3a2 are the products of small scale duplication (SSD) in

teleosts. Zebrafish and cavefish additionally harbour another SSD gene, 3-3a3, butlack rxfp4.� Pseudogene or complete loss of gene.

2 S. Yegorov et al. / General and Comparative Endocrinology xxx (2014) xxx–xxx

The nomenclature currently used for relaxin family peptidereceptors was proposed by Bathgate et al. (2006) in an effort tostandardize and simplify the naming of the receptors in humanand mice (Bathgate et al., 2006). However, recent progress in char-acterizing the Rxfps in multiple non-human or non-mouse verte-brates (Donizetti et al., 2010; Fiengo et al., 2013; Yegorov andGood, 2012) has necessitated the need for a modification of theexisting receptor gene nomenclature to take into account thegreater diversity of receptors in non-mammalian taxa and derivea nomenclature based on their evolutionary relationships. There-fore the goal of this review is to summarize these advances andpromote the use of a gene and protein nomenclature that supportsthe true evolutionary relationship among the receptors and theirligands across vertebrate lineages (Table 1). We also aim to high-light several of the most interesting, but greatly overlooked,aspects of the rln/insl-rxfp diversification: their ancient origin,recruitment of novel receptors over evolutionary time, and thediverse roles of selection, differential retention and lineage-specificloss of genes. We conclude by highlighting the recent progress andparadigms in relaxin research in non-mammals.

2. Structure and function – a mammalian perspective

The functional and structural biology of RXFP receptors andtheir ligands in human and several mammalian models haverecently been extensively reviewed elsewhere (Anand-Ivell andIvell, 2014; Bathgate et al., 2013; Kong et al., 2010). This sectionaims to equip the reader with a basic understanding of the physi-ological role of Rxfp receptors, prior to addressing the diversifica-tion of the Rln/Insl-Rxfp signaling systems across vertebrates.

Although relaxin (RLN) has been known since 1926 (Hisaw,1926), the majority of relaxin family peptides were characterizedmore recently, primarily in the 1990s/2000s, through a combina-tion of data mining of sequenced genomes and experimental work(Adam et al., 1993; Conklin et al., 1999; Hsu, 1999; Hsu et al., 2002;

Please cite this article in press as: Yegorov, S., et al. The relaxin family peptidelution from jawless fish to mammals. Gen. Comp. Endocrinol. (2014), http://dx

Liu et al., 2003). The ligation and signaling of relaxin peptidesbegan to be defined in 2002 following studies in mice (Hsu et al.,2002), and prior to this Rln/Insl peptides were anticipated to func-tion via receptor tyrosine kinases (RTKs), the cognate receptors forinsulin and insulin-like growth factors (IGF) (Halls et al., 2006).However, subsequent experimental studies showed that Rln/Inslpeptides bound and activated two distinct subclasses of G pro-tein-coupled receptors (GPCRs): Rxfp1/Rxfp2 and Rxfp3/Rxfp4receptors, which were previously coined orphan receptors Lgr7/Lgr8 and Gpr135/Gpr142, respectively (Bathgate et al., 2006; Liuet al., 2003, 2005).

2.1. Structure and function of Rxfp1/Rxfp2 class proteins

The Rxfp1 and Rxfp2 proteins are classified in subgroup d ofrhodopsin GPCRs and further classified as subtype C leucine-richrepeats containing GPCRs (LGRs) (Fredriksson et al., 2003). Thecommon signature of all LGRs is the leucine-rich repeat (LRR)domain at the N-terminus of the molecule, which solely inRxfp1/2 receptors, terminates with a low density lipoprotein(LDL) domain (Van Loy et al., 2008). The LRR domain is a crucialplayer in ligand binding and subsequent signal transfer (Bathgateet al., 2013; Halls et al., 2006).

In human and several other studied mammals, both RXFP1 andRXFP2 with their respective ligands, RLN and INSL3, are primarilyassociated with reproductive functions, such as the relaxation ofuterine musculature and pubic symphysis during labor (in the caseof Rln-Rxfp1) or germ cell survival (Anand-Ivell and Ivell, 2014;Kawamura et al., 2004), and trans-abdominal testicular descent(in the case of Insl3-Rxfp2) (Feng et al., 2009; Nef and Parada,1999; Zimmermann et al., 1999). For simplicity, we refer to thecognate ligand of RXFP1 in humans as RLN, however, the rln locusduplicated in the ancestor to catarrhini giving rise to tandem par-alogs rln1 and rln2 (Arroyo et al., 2014); by convention humanRLN2 is considered the homologue of Rln in other mammals.Despite its important role during mammalian parturition, theRLN-RXFP1 signaling system also plays diverse non-reproductivefunctions such as tissue remodeling during wound healing, angio-genesis, and tumor formation (Ho et al., 2007; McGuane and Parry,2005) and is expressed in non-reproductive organs (Bathgate et al.,2013; Gunnersen et al., 1995; Osheroff and Ho, 1993). On the otherhand, Insl3 has been hypothesized to be a biomarker for malereproductive status (Adam et al., 1993; Ivell et al., 2005, 2013),and the Insl3-Rxfp2 signaling system has been implicated in somenon-reproductive functions such as bone morphogenesis and thy-roid tumour growth and angiogenesis (Ferlin et al., 2013;Hombach-Klonisch et al., 2010).

2.2. Structure and function of Rxfp3/Rxfp4 class proteins

The other subclass of Rxfp receptors includes Rxfp3 and Rxfp4proteins that belong to subgroup c of the rhodopsin class GPCRsand phylogenetically cluster with the receptors for small peptidessuch as bradykinin, angiotensin and somatostatin (Fredrikssonet al., 2003). Similar to other small peptide receptors, Rxfp3/4 pro-teins have a short N-terminus, and the transmembrane domain isthe primary site of interaction with ligands (Gloriam et al., 2009).

The distinct evolutionary origins of Rxfp3/4s from Rxfp1/2s mir-ror their disparate biological functions: Rxfp3 and Rxfp4, with theirrespective ligands, Rln3 and Insl5, have primarily been associatedwith neuroendocrine, not reproductive, regulation. Both Rln3 andits receptor Rxfp3 are predominantly expressed in the brain, par-ticularly the nucleus incertus (Bathgate et al., 2002; Liu et al.,2003; Matsumoto et al., 2000; Tanaka et al., 2005) and play rolesin the modulation of feeding activities, body weight regulationand in stress coordination, learning and memory (McGowan

receptors and their ligands: New developments and paradigms in the evo-.doi.org/10.1016/j.ygcen.2014.07.014


S. Yegorov et al. / General and Comparative Endocrinology xxx (2014) xxx–xxx 3

et al., 2008, 2005; Tanaka et al., 2005). Broadly, the Rln3-Rxfp3 sig-naling system influences the hypothalamo–pituitary–adrenal axis(HPA) via its influence on luteinizing hormone (LH) and corticotro-pin releasing hormone, and thereby appears to link nutritional andreproductive statuses (McGowan et al., 2014, 2008).

The function of Insl5-Rxfp4 remains somewhat elusive. BothInsl5 and Rxfp4 were found to have wide and overlapping expres-sion in many peripheral tissues (Boels and Schaller, 2003; Liu et al.,2003) and to have similar patterns of evolution including the find-ing that both are pseudogenes in rat. Recent research has found arole for Insl5-Rxfp4 signaling in an autocrine/paracrine fashion inthe colorectal epithelium, rectal neuroendocrine tumors(Mashima et al., 2013), and glucose metabolism and fertility(Burnicka-Turek et al., 2012).

Although RLN2-RXFP1, INSL3-RXFP2, RLN3-RXFP3 and INSL5-RXFP4 are well-established ligand-receptor pairs in human, therelaxin and insulin-like peptides overlap in their abilities to bindthe receptors in vitro. For instance, in humans RLN2 binds RXFP1and RXFP2, and RLN3 binds RXFP1, RXFP3 and RXFP4. The promis-cuous interaction of Rln/Insl peptides with two such evolutionarilydistant classes of receptors, suggests that there are multiple modesof ligand-receptor binding, and eludes to an unusual history ofligand-receptor co-evolution (Gloriam et al., 2009).

3. Evolutionary origins and diversification of Rln/Insl and Rxfpgenes in vertebrates

3.1. Origins of Rxfp signaling: clues from vertebrate ancestors

That relaxin family peptides exist in invertebrate organismswas first proposed in 1999, when Georges and Schwabe used poly-merase chain reaction to characterize a relaxin-like gene in Cionaintestinalis and determined it to be nearly identical to porcineand whale rln (Georges and Schwabe, 1999). This finding has sincebeen intensively scrutinized and the discovery of Ciona relaxin hasbeen attributed to contamination with mammalian tissues (Hafnerand Korthof, 2006). Since then, some research has been done on theinvertebrate origins of relaxin-insulin superfamily (Holland et al.,2008; Olinski et al., 2006a,b), but it has only become possible tostudy the origin of the receptors since the emergence of genomicsand functional data from Echinoderms and Cephalochordates.

3.1.1. EchinodermsIn a variety of invertebrates, rxfp1/2-like genes are known to

exist (Van Loy et al., 2008). However, the first evidence of arelaxin-like system in invertebrate deuterostomes was providedby the biochemical characterization of gonad-stimulating sub-stance (GSS) in the starfish, Asterina pectinifera (Mita et al., 2011,2009). Starfish GSS is secreted by the echinoderm’s radial nervesand effects follicular cells to produce 1-methyladenine, the echino-derm counterpart of vertebrate progestin, which promotes oocytematuration. While the receptor of GSS remains unknown, its signaltransduction is mediated by G proteins and results in a dose-dependent increase in intracellular cyclic AMP concentration(Mita, 2013a,b). Interestingly, the signaling role of GSS in starfishis reminiscent of vertebrate LH, which is produced in the pituitaryand signals via a glycoprotein hormone receptor located on follic-ular cells (Mita, 2013a).

The publicly available genome database of another echinoderm,the sea urchin Strongylocentrotus purpuratus, reveals the presenceof at least twenty-seven (27) rxfp1/2-like genes (Yegorov andGood, 2012), but our efforts to find rxfp3/4- or GSS-like geneswithin this database have not been successful. Similarly, data min-ing the genome of the hemichordate acorn worm Saccoglossuskowalevskii identified multiple rxfp1/2-like homologues but no


rxfp3/4-like genes; identification of GSS-like molecules in the acornworm will require closer examination and synteny analyses due tothe small size of the ligand (unpublished data). Nevertheless,cumulative evidence from echinoderms for the presence of Rxfp1/2- and Rln/Insl-related genes suggests that Rxfp1/2-mediated sig-naling played a key role in the regulation of reproductive functionsas early as 840 million years ago (mya) in the common ancestor ofdeuterostomes.

3.1.2. Cephalochordates and tunicatesCephalochordates (amphioxus, lancelets) and tunicates (e.g.

Ciona), which diverged from a common ancestor shared with ver-tebrates approximately 797 mya, show genomic evidence for thepresence of multiple rxfp1/2-like and insulin-like genes. However,due to a lack of functional characterization it is unclear whethersome of these insulin-like genes (which phylogenetically groupwith vertebrate relaxin family peptides) could be functionallyrelated to genes coding for Rln/Insl family peptides. Previously,using ancestral genome reconstruction data from Putnam et al.(2008), we determined that three out of five amphioxus rxfp1/2-like genes are found in the same genomic linkage group as humanRXFP1 and RXFP2, i.e. they appear to be orthologous (Yegorov andGood, 2012). Interestingly, the number of rln/insl-like genes inamphioxus matches that of rxfp1/2-type genes. In addition, datamining in the tunicate C. intestinalis (but not in the amphioxusBranchiostoma floridae) database determined the presence of twosequences with rxfp3/4-like features (Yegorov and Good, 2012).Taking into account that Ciona is the closest living invertebrateto vertebrates, we hypothesize that the first receptor for the ances-tral Rln/Insl peptide was an Rxfp1/2-like receptor, and that Rxfp3/4-like receptors were recruited prior to the divergence of tunicates.This hypothesis is supported by the finding that starfish GSS bindsto a GPCR receptor expressed in ovary causing an up-regulation ofintracellular cAMP (Mita et al., 2011); this signaling cascade isshared with the vertebrate Rxfp1/2 receptors, while activation ofRxfp3/4-type receptors leads to cAMP down-regulation. To testthe pre-tunicate origins of Rln/Insl-Rxfp signaling, it will be impor-tant to perform extensive analysis of other pre-vertebrate chordategenomes such as another tunicate, Oikopleura dioica. Given thatinvertebrate deuterostomes harbour many rxfp1/2-like genes, andthat Ciona has two rxfp3/4 genes (Yegorov and Good, 2012), itcan be hypothesized that the majority of pre-2R rln/insl-like andrxfp genes were lost as a result of a bottle neck in the early verte-brate evolution to yield a tripartite two receptor-one ligand net-work that was further duplicated by 2R (Good et al., 2012).

3.2. The role of whole genome and small scale duplications in theformation of vertebrate rxfp repertoires

The evolution of vertebrates from a common ancestor sharedwith tunicates and cephalochordates was accompanied by tworounds of whole genome duplication (abbreviated as 2R WGD)and massive genomic rearrangements (Nakatani et al., 2007;Putnam et al., 2008). It is currently thought that 2R WGD occurredaround 550–530 mya before the divergence of jawless fish (lam-preys and myxines). However, there is some genomic evidence thatagnathans may belong to an intermediate post-1R group of verte-brates that did not undergo the 2nd round of WGD (Kuraku et al.,2009). Recently, we confirmed previous suggestions regarding therole of 2R in the diversification of the rln/insl genes (Hoffmann andOpazo, 2011) and demonstrated that both 2R and large-scale gen-ome rearrangements in the vertebrate ancestor played a criticalrole in shaping the extant vertebrate rln/insl and rxfp gene reper-toires. Specifically, we used ancestral genome reconstructions(AGR) of the vertebrate and chordate ancestor based on the workof Nakatani et al. (2007) and Putnam et al. (2008) to trace the




origin, duplication, loss and changing linkage relationships ofligand and receptor genes from the pre-1R to post 2R genomesand within the teleost and tetrapod lineages (Yegorov and Good,2012). This analysis determined that the ancestral rxfp1/2 andrxfp3/4 genes were derived from loci located in distinct linkagegroups (chromosomes) in the pre-2R vertebrate ancestor, althoughinterestingly the single pre-2R ancestral rln/insl gene was linked(on the same vertebrate ancestral chromosome) to the ancestralrxfp3/4 gene, a linkage relationship that has been preserved insome lineages (Yegorov and Good, 2012). We also show that mostof the extant rln/insl-rxfp genes in modern vertebrates arose viaWGD events and two genes underwent local, tandem, duplication(Fig. 1).

Vertebrates harbour three members of the rxfp1/2 subclass:rxfp1, rxfp2 and rxfp2-like. While the first two genes are well knownand widely distributed, rxfp2-like is a novelty found in early diverg-ing tetrapods (e.g. marsupials, reptiles, birds and amphibians) andin early diverging teleosts (zebrafish and cavefish). Interestingly,previous studies concluded that chicken lacks insl3 and rxfp2(Hsu et al., 2000; Semyonov et al., 2008; Wilkinson et al., 2005b).Our research confirms this, yet identifies that chicken has anotherancestral gene, rxfp2-like. Using synteny and AGRs, we confirmedthat rxfp1 and rxfp2 are the products of 2R, i.e. ohnologues, but itcould not be resolved if the rxfp2-like gene was present also a prod-uct of 2R (discussed below).

Additionally, we previously demonstrated that the single ances-tral rxfp3/4-like gene generated four ohnologues following 2R. Tokeep reference to the established terms, we named the ohnologuesrxfp3-1, rxfp3-2, rxfp3-3 and rxfp3-4, whereby RXFP3 and RXFP4 inhumans are equivalent to rxfp3-1 and rxfp3-4 across vertebrates.The two other ohnologues, rxfp3-2 and rxfp3-3, were lost in mostmammals, including human and mouse. However, several tetrapodlineages, including cow (Bos taurus) and pig (Sus scrofa), harbour arxfp3-3 gene alongside with rxfp3-1 and rxfp3-4 (aka rxfp4). On theother hand, teleosts retained all four of these 2R-derived rxfp3/4genes and two of them, rxfp3-3 and rxfp3-4, were further

Jawless fish

Car�laginfish

1R

Agnathan ancestor

Gnathostoancesto

Vertebrate ancestor

AncRln-likeAncRxfp1/2 AncRxfp3/4

AncRln-II

AncRln-I

AncRxfp3-II

AncRxfp3-I

Rxfp1

Rxfp2-like

Rxfp2

2R

AncRxfp2

AncRxfp1

535

526

Insl3

Insl5

Rln

Rln3

Fig. 1. The number and identity of rln/insl and rxfp genes in the major vertebrate clades (jAncestral hypothetical genes are indicated with prefix ‘‘Anc’’. 1R and 2R: the first andcrossed circles indicate gene loss. Numbers at nodes indicate times of divergence (mya)


duplicated and retained during the third round (3R) of WGD(Yegorov and Good, 2012) that occurred in teleosts between 250and 320 mya (Hurley et al., 2007; Santini et al., 2009). One of theseduplicates, rxfp3-3a also underwent local duplication in the teleostancestor, rendering the total number of rxfp receptor genesbetween 9 and 11 in teleosts (Good et al., 2012) (Table 1, Fig. 1).

In the following section we present an extensive phylogeneticre-analysis of the rxfp and rln/insl genes across vertebrates. Wedo so by including several taxa from early diverging vertebrate lin-eages, such as skate, coelacanth, spotted gar, and also members ofthe reptile/bird and marsupial clades.

3.3. Phylogenetic reconstruction of Rxfp1/2, Rxfp3/4 and their ligands

3.3.1. Phylogenetic methods and data collectionFor the purpose of this review, we updated the sequence data-

base of 22 taxa from a previous analysis (Yegorov and Good, 2012)to include sequences from 8 recently annotated genomes (tasma-nian devil (Sarcophilus harrisii), chinese painted turtle (Pelodiscussinensis), duck (Anas platyrhynchos), cavefish (Astyanax mexicanus),lamprey (Petromyzon marinus), coelacanth (Latimeria chalumnae)and spotted gar (Lepisosteus oculatus)) of early diverging taxa totest hypotheses regarding the role of 2R and 3R in the diversifica-tion of rln and rxfp genes. Additionally, to include a representativeof cartilaginous fish, we searched the skate genome for rln, insl andrxfp genes. By including skate, coelacanth and spotted gar, wesought to confirm the presence of the genes coding for the fourligands, three Rxfp1/2 subclass and four Rxfp3/4 subclass receptorspredicted to be encoded in post-2R non-mammalian and non-tele-ostean genomes (Table 1, Fig. 1). We also included more represen-tatives of reptiles and birds since they appeared to have a uniquepattern of gain and loss of genes in previous research, as well ascavefish because its sister species, zebrafish, harbours a uniquesuite of genes compared with other later diverging teleosts(Yegorov and Good, 2012). Several of these taxa have divergencedates closer to 2R based on the expert molecular result available

ous

Teleost fish

Tetrapods

3Rme r

Osteichtyan ancestor

Ac�nopterygian ancestor

Teleost ancestor

Rxfp3-2

Rxfp3-1

Rxfp4

Rxfp3-3

Gar

454

343

430Coelacanth

300

rxfp1

rxfp2-like

rxfp2arxfp3-2a

rxfp3-1

rxfp4

rxfp3-3a

insl3

insl5a

rln3a

rxfp2b

insl5b

rln3b

rln

rxfp3-2b

rxfp3-3b

awless fish, cartilaginous fish, teleosts), gar and the hypothetical vertebrate ancestor.second rounds of whole genome duplication. Stars indicate local gene duplication,.




at Timetree (http://www.timetree.org) (Fig. 1). Lastly, it should benoted that the divergence of lamprey from the Agnathan ancestoris estimated to be 535 mya, which places it either between 1R and2R or after 2R, since both rounds of WGD are thought to haveoccurred within 10–20 mya during that time period.

3.3.1.1. Sequence collection and annotation. To construct phyloge-netic trees, annotated Rln/Insl- and Rxfp-coding sequences wereretrieved from Ensembl v. 60 (for those sequences obtained fromYegorov and Good (2012) and v. 74 (newly retrieved sequences)databases (http://www.ensembl.org) for 14 mammals (11 placen-tals, two marsupials (opossum and tasmanian devil) and platypus),5 representatives of the reptile/bird clade (anole lizard, chinesepainted turtle, duck, chicken and zebra finch), 1 amphibian (clawedfrog) and 6 teleosts. Some of the sequences reported here remainunannotated; their identity was confirmed using synteny and reci-procal blast as described in Yegorov and Good (2012). Annotatedligand sequences for rhesus monkey and Rana esculenta wereobtained from NCBI (http://www.ncbi.nlm.nih.gov/gene).Sequences for skate were obtained by data mining Build2 of theskate genome (http://www.skatebase.org) using coelacanth andspotted gar sequences as queries. The accession numbers for allof the sequences are given in Supplementary Table S1.

3.3.1.2. Sequence alignments and phylogenetic reconstruction. Align-ment of the rln/insl and rxfp DNA sequences was performed usingMuscle (Edgar, 2004), as implemented in version 5.2 of MEGA(Tamura et al., 2011) using codon sequences as a guide and thenin Tcoffee (Notredame et al., 2000) to obtain quality scores fromwhich alignments could be fine-tuned manually. Additionally, pro-tein sequences were aligned using a suite of MAFFT alignmentalgorithms (L-INS-i, E-INS-I and G-INS-i) as implemented in Jal-view (alignments provided in Supplementary Figs. S1–S3) (Katohand Standley, 2013). Subsequently, the best models of DNA andprotein sequence evolution were estimated using Maximum Like-lihood (ML) and Akaike Information Criterion (AIC) as imple-mented in the program Treefinder (Jobb et al., 2004). The bestmodels for the rxfp1/2 genes were TVM (DNA) and MIX + G (pro-tein), for rxfp3/4 were TVM + G (DNA) and JTT + G (protein) andfor the ligands were witHIV (DNA) based on the first two codonposition only or GTR + G (protein). Optimization of the phyloge-netic relationship among sequences was then performed usingML and Minimum Evolution (ME) algorithms using the appropriatemodel of DNA/protein evolution and 1000 rounds of bootstrap rep-lication to resolve nodal support as implemented in the programsTreeFinder (Jobb et al., 2004) and Mega 6.0 (Tamura et al., 2011)respectively.

3.3.2. Phylogeny reveals that members of almost all taxonomic groupsexcept placental mammals harbour rxfp2-like

The phylogenetic reconstructions of Rxfp 1/2-like based onamino acid sequences and using ML and ME algorithms andamphioxus as the outgroup, both place rxfp2-like as sister to rxfp1,while rxfp2 clustered outside this clade (Fig. 2 (ML) and Supple-mentary Fig. S4 (ME)). In contrast, in the reconstructions basedon DNA sequence data, rxfp1 and rxfp2 are sister groups, andrxfp2-like is basal to this clade (Supplementary Fig. S5 (ML) andS6 (ME)), making it difficult to resolve the origin of Rxfp2-likebased on phylogenetic data alone.

Most of the early diverging lineages harbour an rxfp2-like ortho-logue (Fig. 2, Supplementary Table S1): coelacanth, the oldest liv-ing relative of tetrapods possesses rxfp2-like as do all of theamphibians, reptile/birds, and marsupials, such that rxfp2-like onlyappears to have been lost in the eutherian ancestor. Additionally,we find rxfp2-like in spotted gar, and confirm its presence in zebra-fish and cavefish. Interestingly, many of the taxa that harbour


rxfp2-like, such as coelacanth and spotted gar, also have both rxfp1and rxfp2, such that they have three rxfp1/2-like genes but only sin-gle copies of rln and insl3. Interestingly, we did not find an ortho-logue of rxfp2-like in lamprey, and the one copy of an rxfp1/2-likegene in lamprey is intermediate in divergence between rxfp1 andrxfp2, suggesting that if lamprey did diverge after 1R (not 2R),rxfp2-like may have originated during 2R. Finally, although wefound several contigs with rxfp1/2-like genes in the skate genome,we could not piece together specific genes due to small size of thecontigs. Given that phylogenetic reconstruction is not definitiveregarding the origin of rxfp2-like, it is interesting that rxfp2-like alsolacks synteny with rxfp1 or rxfp2 in vertebrates, and with the rxfp1/2 subclass of genes in pre-2R deuterostomes (Yegorov and Good,2012). Thus it appears that rxfp2-like was either present in theearly chordate genome (in a unique synteny group) or that it is aproduct of 2R, and was subsequently translocated out of its ances-tral syntenic cluster. However, given the overall similarity of rxfp2-like to both rxfp1 and rxfp2, we speculate that rxfp2-like is an ohno-logue of rxfp1 and rxfp2.

3.3.3. The phylogeny of the rxfp3/4 genes strongly confirms theimportant role of WGDs in their diversification

The phylogenetic relationships among rxfp3/4 sequences estab-lished using both ML and ME algorithms and either amino acid ornucleotide data exhibit highly similar topologies and lend over-whelming support to the WGD-driven diversification of rxfp3/4genes. They also shows that two of these receptor genes, rxfp3-2and rxfp3-3, were lost in most mammalian species but retainedin many other lineages (the ML tree based on amino acidsequences is shown in Fig. 3, the remaining trees are not shownas they were highly similar). The inclusion of more sequences fromearly diverging lineages confirms that most post-2R taxa have allfour 2R-derived ohnologues, although the absence of some genesin some species (e.g. skate) may be attributed to difficulties in datamining. It is notable that spotted gar, a pre-3R ray-finned fish, har-bours all four rxfp3-4 ohnologues, and that coelacanth has three ofthe four genes. Coelacanth lacks rxfp3-2, which was also not foundin any tetrapod, suggesting that rxfp3-2 may have been lost in theeutherian ancestor. Based on phylogenetic reconstruction, weidentified an rxfp3-2-like gene in turtle, but lack of assembly ofthe turtle genome prevented us from obtaining synteny supportfor this grouping (Fig. 3).

Another important observation is that while most of the rxfp3/4genes show similar and quite low levels of genetic diversity amongclades, rxfp3-4 is highly divergent within and between mammalianand teleostean taxa. Lastly, as shown by the sheer number of rxfp3/4 genes in teleosts, there has been an impressive expansion ofthese genes in teleosts, largely driven by 3R but also by the localduplication of rxfp3-3a (see Table 1, Fig. 1).

3.3.3.1. Gene number and sequence identity of early divergingvertebrates. Interestingly, we find only two rxfp3/4 subclass genesin lamprey. The inclusion of both skate and lamprey in the recon-struction allows us to speculate on the origin of the lamprey genes.Lamprey is estimated to have diverged 535 mya and skate at526 mya and yet the genes in skate cluster with other post-2Rdiverging taxa while those of lamprey do not. We identifiedrxfp3-1, rxfp3-2 and rxfp3-3 in skate. Skate rxfp3-2 and rxfp3-3genes group basal, as predicted, to the clade containing all othervertebrate rxfp3-2 and rxfp3-3 genes, while skate rxfp3-1 groupscloser to rxfp3-1 genes in fish and early tetrapods than to mamma-lian rxfp3-1 (Fig. 3). Collectively, this suggests two possibilities:either lamprey has not been through 2R or the rxfp3/4-like geneswere strongly selected in the short period of time between thedivergence of jawless and cartilaginous fish.


http://www.timetree.org

http://www.ensembl.org

http://www.ncbi.nlm.nih.gov/gene

http://www.skatebase.org



3.3.4. Phylogenetic reconstruction of rln/insl peptides clustersorthologous genes together despite highly variable rates in evolutionamong taxa and genes

Several previous studies have examined the molecular evolu-tion of relaxin family peptides (Arroyo et al., 2012b,c; Good-Avilaet al., 2009; Good et al., 2012; Wilkinson et al., 2005a). Interpreta-tion of the phylogenetic relationship among rln/insl ligands hasbeen difficult due to the small size of the molecule (consisting of�60 amino acids) and the strong and differential role of selectionacross taxonomic groups and genes (Good-Avila et al., 2009; Parket al., 2008). However, using the first two nucleotides in a codonand ML methods, the four 2R-derived rln/insl ohnologs were cor-rectly grouped into four monophyletic clades based on the identi-fication of genes achieved through syntenic analyses performedelsewhere (Good-Avila et al., 2009; Yegorov and Good, 2012).Using the first two bases in the codon, starfish GSS as an outgroupand the ME algorithm produced a similar, but not completelymonophyletic reconstruction among sequences (SupplementaryFig. S7), while phylogenetic reconstructions based on amino acidsequences generate non-monophyletic topologies (not shown).Nevertheless, simple patterns regarding rln/insl gene evolutioncan be observed from the branch lengths and grouping ofsequences (Fig. 4). (1) The rln3 gene is highly conserved acrossall taxa as noted in many previous studies (Good-Avila et al.,2009; Park et al., 2008; Wilkinson and Bathgate, 2007; Wilkinsonet al., 2005a,b). (2) For the remaining 2R-generated ligands, rln,insl3 and insl5, there is an overall pattern of high conservationamong reptiles/birds or amphibians and fish but then a markeddivergence from marsupial to placental mammals.

3.4. Differential gain and loss of rxfp-rln/insl genes across vertebrategroups

Vertebrates have experienced differential patterns of gene gainand loss, diagrammatically summarized for early tetrapods, mam-mals and teleosts (Figs. 5–7). Among early diverging tetrapods, allreptiles/birds included here lack insl3 and rxfp2 genes, with theexception of duck which has rxfp2, and most of the species, exclud-ing turtle and frog, have also lost rxfp4, but not insl5. Finally, alltaxa harbour the rxfp2-like gene (Fig. 5). The evolution of the sys-tem in mammals is strongly marked by duplication of the rln locus(Fig. 6). Various studies have proposed that insl6 duplicated fromthe rln locus in the ancestor of placental mammals (Arroyo et al.,2012d; Wilkinson et al., 2005a), but it was recently shown thatinsl4 also arose via duplication of the rln locus at approximatelythe same time (Arroyo et al., 2012a), rather than during primateevolution as previously proposed (Wilkinson and Bathgate,2007). Notably the receptors for Insl4 and Insl6 remain unknown(Bathgate et al., 2013). Additionally, several mammalian lineagesexhibit variable numbers of tandem duplications at the rln locus;in rabbit there are six tandem copies (Arroyo et al., 2012b), andthe duplication of the rln locus that gave rise to tandemly dupli-cated paralogues rln1 and rln2 in higher primates, includinghumans, has recently been proposed to have occurred in the ances-tor of catarrhini (Arroyo et al., 2014), not in the ancestor to greatapes. Apart from this marked evolutionary pattern, other remark-able trends in mammalian rln/insl-rxfp evolution are the wide-spread loss of rxfp3-2, retention of rxfp3-3 in only a few lineages,and the loss of insl5-rxfp4 in some lineages (Fig. 6). In teleosts, 3Rgenerated paralogous copies of those genes involved with neuro-endocrine regulation (rln3/insl5 and rxfp3/4), one of which, rxfp3-3a, was additionally further duplicated by one or two tandemduplication events. Lastly, only early diverging teleosts retainedrxfp2-like, such that there are 6 rln/insl genes in most teleosts and9-11 rxfp genes (Fig. 7).


3.4.1. Unique evolutionary history of the relaxin locus in vertebrateevolution: evidence for neo-functionalization

Sequence comparisons show that rln exhibited a slower rate ofevolution than rln3 in early vertebrate evolution, but acquired afew lineage specific mutations, predominantly in the A-chain inthe ancestors of teleosts and most tetrapods. Then, exceptionally,in placental mammals a burst of mutations in the B- and A-chainsoccurs and a remarkable 23 amino acids show evidence of codon-specific positive selection including sites in the B-chain prohor-mone cleavage site (unpublished data, our laboratory). This placen-tal-specific sudden leap in the slow paced evolution of vertebraterln is coincident with: (1) the diversification of placentals as agroup and (2) the massive local duplications of the rln locus, inwhich two of its duplicates insl4 and insl6, also exhibit evidenceof positive Darwinian selection in mammals (Wilkinson et al.,2005a). The ancestral relaxin-like peptide appears to have resem-bled today’s vertebrate rln3 and non-eutherian rln genes. However,despite this fantastic example of positive Darwinian selection onrln, which was presumably recruited to play a widespread and pre-dominant role of softening the pubic symphysis during mamma-lian birth, the rxfp1 gene shows a very similar and conservedmode of evolution across all vertebrates (Fig. 2) (Good et al.,2012). This puzzling observation deserves further investigationand provocatively suggests that the strong selection on the mam-malian relaxin locus may have occurred because relaxin co-evolved with other receptors, such as the glucocorticoid receptor(Duarte et al., 2014; Stolte et al., 2006).

4. Emerging perspectives in the comparative endocrinology ofRxfp-Rln/Insl systems

Outside of mammals, the first studies on the relaxin family ofsignaling molecules were conducted in cartilaginous fish(Gelsleichter et al., 2003; Reinig et al., 1981; Steinetz et al.,1998), chicken (Gallus domesticus) (Brackett et al., 1997) and frog(Rana esculenta) (De Rienzo et al., 2006; de Rienzo et al., 2001).In these experiments, researchers isolated relaxin-like peptidesfrom reproductive tissues, typically testes or ovary, and oftentested the physiological effect of these substances in mammalianmodel species. Studies of relaxin family peptides in non-mamma-lian taxa gained further momentum following the first molecularevolutionary studies that made use of whole genome sequencedata from teleosts and other taxa (Good-Avila et al., 2009; Parket al., 2008; Wilkinson and Bathgate, 2007; Wilkinson et al.,2005a). This has since given rise to experimental work on therelaxin family peptides in zebrafish (Donizetti et al., 2010, 2009;Good-Avila et al., 2009; Good et al., 2012; Wilson andSummerlee, 1994), killifish (Amamoo and Wilson, 2005), eel (Huet al., 2011), medaka (unpublished data, our laboratory) and, morerecently stickleback (Kusakabe et al., 2014). Now that the evolu-tionary history of the Rln/Insl-Rxfp genes is clearer, it is timelyfor more research to be performed on the functions of the Rln/Insl-Rxfp signaling systems in non-human taxa. For example, therole of Rxfp2-like in reptiles/birds and many early diverging lin-eages awaits investigation, as does the role of Rxfp3-3 in somemammalian lineages (e.g. cow). Lastly, the extensive diversificationand apparent sub-functionalization of Rxfps in teleosts will pro-vide an exciting addition to our understanding of Rxfp evolutionin vertebrates.

4.1. Teleost Rxfp and Rln/Insl systems

The fish orthologues of rxfp1/rxfp2 and rln/insl3 have beenrecently studied by different research groups and compared totheir counterparts in mammals as summarized below.



Fig. 2. Maximum Likelihood (ML) optimization of the phylogenetic relationship among rxfp1, rxfp2 and rxfp2-like genes across vertebrates. An rxfp1/2-like sequence fromamphioxus was used to root the tree. The ML tree is based on protein sequences employing the JTT + G model of sequence evolution. Support for each node is given in percent,based on 1000 bootstrapped samples of the data. Colored boxes around taxonomic groups are used to aid in visualizing the tree.


Please cite this article in press as: Yegorov, S., et al. The relaxin family peptide receptors and their ligands: New developments and paradigms in the evo-lution from jawless fish to mammals. Gen. Comp. Endocrinol. (2014), http://dx.doi.org/10.1016/j.ygcen.2014.07.014


Fig. 3. Maximum Likelihood (ML) optimization of the phylogenetic relationshipamong rxfp3/4 subclass of genes across vertebrates. The two lamprey rxfp3/4-likesequences were used to root the tree. The ML tree is based on protein sequence andthe JTT + G model of sequence evolution. Support for each node is given as apercentage based on 1000 bootstrapped samples of the data.



4.1.1. Rxfp1-Rln

Donizetti et al. (2010) confirmed that rxfp1 in zebrafish has ahighly similar exon–intron structure as the human gene, similarrelaxin peptide-binding motif and a similar spatial expression inzebrafish and mammalian brain. However, as mentioned, giventhe very different evolutionary history of rln in teleosts versusmammals, and the high amino acid similarity of teleost Rln toRln3, it is possible that teleost Rln has a functional role more akinto that of the ancestral vertebrate Rln3 peptide.

4.1.2. Rxfp2-Insl3

Real-time quantitative PCR experiments in zebrafish (Goodet al., 2012) and medaka (our group, unpublished) show that insl3and rxfp2 are both highly expressed in testis of fish and mammals,and in situ hybridization experiments in zebrafish localize insl3mRNA expression to Leydig cells (García-López et al., 2009).Despite the similarity of these patterns of expression, molecularevolutionary and functional analyses indicate that different aminoacids have been the target of positive Darwinian selection in bothteleosts and mammals (Good et al., 2012). Park et al. (2008)showed that replacements at several key residues in the codingregion of insl3 were responsible for Insl3 acquiring specificity forRxfp2, leading to its specific involvement in testicular descent inplacental mammalian species. This hypothesis has been used todescribe Insl3 as a ‘‘neohormone’’ in mammalian evolution, andit has since been advertised as a molecular marker for male repro-ductive status (Ivell et al., 2013). Our selection analyses suggestthat both insl3 and rxfp2 have also undergone positive Darwinianselection in teleosts (Good-Avila et al., 2009; Good et al., 2012).Thus, it appears that the role of the Insl3-Rxfp2 signaling systemin spermatogenesis is ancestral, but that specialized neo-function-alization of the hormone in mammals and teleosts has occurred.Furthermore in zebrafish there are a surprising four members ofthe Rxfp1/2 subclass (Rxfp1, 2a, 2b, and 2-like, Table 1), but onlysingle Rln and Insl3 ligands. Thus it is timely to perform in situhybridization of the diverse receptors to discriminate the cell typesexpressing these genes and conduct ligand-binding assays toexamine the binding affinities of Insl3 for each receptor inzebrafish.

4.1.1. Massive expansion of Rxfp3-4 subclass of receptors in teleostsRecent research demonstrated that duplicated genes putatively

involved in neuroendocrine regulation, namely the rxfp3/4 subclassof receptors and their candidate ligands rln3 and insl5, were prefer-entially retained in teleosts. Using bioinformatics approaches, wepreviously proposed a model for the pairing of the teleost rxfp3/4 receptors with candidate peptides (rln3a/b and insl5a/b), inwhich each ligand signals via one to three receptors (Good et al.,2012), and functional studies to confirm these hypotheses are attheir inception.

There is evidence for differential expression and sub-functional-ization of the rln3a/rln3b and insl5a/insl5b paralogues. Not surpris-ingly, given the highly conserved nature of rln3 in all vertebrates,there is evidence that teleost and mammalian rln3 have broadoverlapping spatiotemporal expression profiles, but there is alsosub-functionalization of rln3a versus rln3b, and rln3a is morehighly localized to the nucleus incertus (as in mammals), whilerln3b has broader expression including gonad (Donizetti et al.,2009; Good-Avila et al., 2009; Hu et al., 2011). On the other hand,only two studies have examined differential expression of thereceptor genes: Fiengo et al. (2013) found that rxfp3-2a/rxfp3-2bexhibit differential developmental and spatiotemporal expressionthat supported sub-functionalization and signaling via rln3a.Good et al. (2012), examining the expression of all ligand and



Fig. 4. Maximum Likelihood (ML) optimization of the phylogenetic relationshipamong rln/insl genes across vertebrates, with midpoint rooting. The ML tree is basedon DNA sequences including only the first two nucleotides in each codon andemploying the miHIV + G model of sequence evolution. Support for each node isgiven as a percentage based on 1000 bootstrapped samples of the data.



receptor genes using 7 tissues in zebrafish, found that most recep-tors were expressed in brain, but some were also expressed inintestine (insl5 and rxfp3-3 genes) or gonad. Interestingly, a tran-scriptomic analysis of the response of three-spine stickleback(Gasterosteus aculeatus) to predation, identified a ‘‘relaxin signal-ing’’ pathway as one of the four most up-regulated networks(Sanogo et al., 2011). Although the authors do not specify whichrelaxin signaling pathway they refer to, the ability of Rln3 to actas a neuropeptide and be implicated in the acute stress response(McGowan et al., 2014) suggests that the role of Rxfp3-Rln3 genesin the response to predation warrants further investigation.

4.2. Roles of Rxfp and Rln/Insl systems in osmoregulation: directionstill worth pursuing?

In mammals, the relaxin peptides, Rln and Rln3, acting viaRxfp1 and Rxfp3, up-regulate components of the forebrain renin-angiotensin system resulting in increased drinking, release of vaso-pressin and retention of water in the kidney (Hornsby et al., 2001;Hsu, 1999; Sherwood, 2004). For this reason, it has been hypothe-sized that relaxin peptides may facilitate acclimation of fish fromsalt water (SW) to fresh water (FW) by inhibiting renal signalingleading to decreased water consumption and ion loss in hypotonicenvironment (Amamoo and Wilson, 2005; Hu et al., 2011). Addi-tionally, Rln3 is an important peptide for modulating appetiteespecially in response to reproductive status (Ganella et al.,2013). Given the trade-offs between growth and reproduction inSW vs FW for euryhaline species (Schluter, 1995), relaxin peptideshave been hypothesized to influence SW adaptation directlythrough their impact on osmoregulation or indirectly via physio-logical changes associated with growth or reproduction in FW hab-itats. In a study of euryhaline populations of Japanese eel, Hu et al.(2011), did not find differences in the expression of rln3 or rln tran-scripts using qPCR between FW and SW acclimated fish, but didfind evidence of differential expression of rln3 and rln transcriptsin brain, and expression of rln in pituitary and gut. In line with this,they concluded Rln3 or Rln peptides may play regulatory roles inthe brain of euryhaline fish. Intriguingly, a recent study in theeuryhaline species three-spine stickleback (G. aculeatus), found dif-ferential expression of rln3a, rln3b and rln in SW or FW adaptedindividuals presented with a salinity challenge (Kusakabe et al.,2014).

5. New and outdated paradigms in the evolution of the Rxfpsignaling systems in vertebrates

We conclude this review by summarizing the major paradigms,both new and outdated, in the evolution of Rxfp and Rln/Insl sig-naling systems in vertebrates. Unfortunately and despite the recentresearch, the first two paradigms (outlined below) have persistedin the literature, while the new ones have been largelyunacknowledged.

5.1. ‘‘Non-mammals only harbour Rxfp3-Rln3 receptor-ligandsystems’’: outdated

The first hypotheses regarding the origin and diversification ofthe Rln/Insl-Rxfp signaling systems in vertebrates proposed thatthree of the peptides (Rln3, Insl5 and Rln) were present prior tothe diversification of teleosts, and that the fourth peptide (Insl3)arose in tetrapods (Wilkinson and Bathgate, 2007; Wilkinsonet al., 2005a). With the identification of multiple Rln3- andRxfp3-like molecules in teleosts (Wilkinson et al., 2005a), it wasproposed that the signaling of Rln/Insl peptides in teleosts wasmediated solely by Rxfp3-like receptors (Wilkinson and Bathgate,



Rxfp3-2Rxfp4

Rxfp2-like

Rxfp3-2

Rxfp3-3Rxfp3-2

Rln3Rxfp3-2Rxfp4

Insl5 Rxfp2-like

Rxfp3-3 Rxfp4

Insl3 Rxfp2

Rxfp3-3

Frogs

Lizard Placentals (Eutherians)

Tetrapod ancestor

Birds

Platypus

OpossumTurtle

361324

220

176

Rxfp1

Rxfp2-like

Rxfp2

Rxfp3-2

Rxfp3-1

Rxfp4

Rxfp3-3

Insl3

Insl5

Rln

Rln3

Fig. 5. Ancestral suite of genes hypothesized to be present in the tetrapod ancestor, followed by the patterns in lineage specific gene loss in amphibians, reptiles/birds,monotremes, marsupials and the eutherian mammal ancestor. Crossed circles indicate gene loss. Numbers at nodes indicate times of divergence (mya).

Rxfp3-3Rxfp4Insl5

Rxfp3-3

Rxfp3-1

Rxfp3-3

Rln

Rxfp4Insl5

Rxfp3-3

Rxfp3-3

Armadillo

Eutherian ancestor

Elephant

Shrew

Dog

CowPig

Rabbit

Guinea pig

Mouse

Rat

Human

Rhesus macaque

Laurasiatheria

Cetar�odactyla

Glires

Euarchontoglires

Roden�a

Catarrhini

Atlantogenata

Horse

Rxfp3-3Rxfp4Insl5

Rxfp2-like

Insl6

Insl4

6

3

2

2

Rln1

105

97

91

30

Rxfp1

Rxfp2-like

Rxfp2

Rxfp3-1

Rxfp4

Rxfp3-3

Insl3

Insl5

Rln

Rln3

Fig. 6. Ancestral suite of genes hypothesized to be present in the eutherian ancestor, followed by the patterns in lineage specific gene gain and loss in Atlantogenata,Laurasiatheria, Euarchontoglires and Catarrhini lineages. Stars indicate local gene duplication at the rln locus, crossed circles indicate gene loss. Note that insl4 and insl6 wererecently shown to have duplicated from the rln locus in the ancestor to all placental mammals, the duplication giving rise to rln1 and rln2 occurred in the ancestor tocatarrhini (see text for details). Numbers at nodes indicate times of divergence (mya).


2007). Although database searches identified orthologues of rxfp1and rxfp2 in teleosts (Wilkinson and Bathgate, 2007), it washypothesized that the recruitment of the Rxfp1/2 subclass ofreceptors into Rln/Insl signaling pathways occurred only in the tet-rapod lineage, in part due to a belief that Insl3 was a tetrapod-(perhaps even mammal-) specific gene. This paradigm still persistsin the literature [see for example: (Anand-Ivell and Ivell, 2014)],however as shown in this review, it is outdated and needs to bereplaced.

5.2. ‘‘The insl3 gene is a product of small scale duplication’’: outdated

The hypothesis that insl3 was specific to the tetrapod lineagewas postulated, in part, because in humans RLN3 and INSL3 areco-localized on chromosome 19 (�3.8 MB apart), which ledresearchers to propose that insl3 arose from a local duplication ofrln3 in the ancestor of tetrapods (Hoffmann and Opazo, 2011;Park et al., 2008). However, large scale synteny analysis has since


uncovered that eutherian rln3 and insl3 belong to two linkagegroups, which were separate in the gnathostome ancestor,remained separated in teleosts, but which fused in the ancestorof mammals (Yegorov and Good, 2012). Furthermore, using areconstruction of the ancestral vertebrate genome it has beenshown that insl3 originated from a duplication of a gene ancestralto rln at the onset of vertebrate evolution (Yegorov and Good,2012), as had been hypothesized based on the functional similari-ties between rln and insl3 (Kong et al., 2010).

5.3. Pre-WGD relaxin signaling: tripartite model

All known vertebrate relaxin and insulin-like loci originated froma single gene in the pre-2R ancestor, thus implying that the ancestralrelaxin/insulin-like peptide signaling system may have consisted ofone hormone and two receptors [one of each subclass, see (Goodet al., 2012; Yegorov and Good, 2012)]. Interestingly, the hypothesisthat relaxin family peptides ancestrally functioned with both Rxfp1/



rxfp2b

rxfp2-like

rxfp3-1

rln3b

rxfp4

Zebrafish

Teleost ancestor

Medaka

Tetraodon

Fugu

S�ckleback

rxfp3-3a3 rxfp3-3a1

rxfp3-3a2

307

250

125

75

Cavefish

rxfp1

rxfp2-like

rxfp2a rxfp3-2a

rxfp3-1

rxfp4

rxfp3-3a

insl3

insl5a

rln3a

rxfp2b

insl5b

rln3b

rln

rxfp3-2b

rxfp3-3b

Fig. 7. Ancestral suite of genes hypothesized to be present in the teleost ancestor, followed by patterns of lineage specific gene gain and loss in six teleosts. Stars indicate localgene duplication at the rxfp3-3a locus, crossed circles indicate gene loss. Numbers at nodes indicate times of divergence (mya).


2 and Rxfp3/4 receptor subclasses was also proposed by Park et al.(2008) to explain the promiscuous binding of Rln3 to three differentreceptors: Rxfp1, Rxfp2 and Rxfp3. We hypothesize that the ances-tral tripartite signaling system had a dual function, which was par-titioned after the first round of WGD such that two sets of ligand-receptor pairs sub-functionalized into predominantly neuroendo-crine- or reproductive-focused functions, with an Rln3-like ligandinteracting with Rxfp3/4 subclass receptors and an Rln-like ligandfunctioning via the Rxfp1/2 subclass receptors (Good et al., 2012).This contrasts with the hypothesis that the promiscuous bindingof relaxin family peptides with multiple receptors arose by a processof convergent evolution (Bathgate et al., 2013).

5.4. Rxfp2-like: a novelty of unclear origin in the Rxfp1/2 subclass

Until recently, vertebrates were thought to harbor only twogenes coding for Rxfp1/2-like receptors (Rxfp1 and Rxfp2). However,there is a third gene, which we refer to as rxfp2-like, which is found inmany early diverging vertebrates, including coelacanth, spotted gar,all sequenced reptiles/birds/amphibians, marsupials/monotremes,and zebrafish/cavefish (see Section 3.3.2.). Unlike rxfp1 and rxfp2,which are clearly the products of 2R, it is unclear whether rxfp2-likehas ancient pre-2R origins or is an ohnologue of rxfp1 and rxfp2. Theidentification of rxfp2-like has important implications for studies inbirds, and other taxa. Previous bioinformatic analyses showed thatchicken lack rxfp2 (Hsu et al., 2000; Semyonov et al., 2008;Wilkinson et al., 2005b), but did not identify that chicken harboursa different gene, namely rxfp2-like. Interestingly, we identified rxfp1and rxfp2-like in chicken and rxfp2 and rxfp2-like in duck, suggestinga complex pattern of gene retention and loss in birds. The possibilitythat rxfp2-like was lost relatively recently in mammals is suggestedby the fact that in chicken, rxfp2-like is located on the W sex chromo-some and a pseudogene for RXFP2-like was identified on humanchromosome X, syntenic to STARD8 in both species (Yegorov andGood, 2012).This highlights another important aspect of utilizing anomenclature based on evolutionary origins, because it also facili-tates the naming and identification of pseudogenes.

5.5. The Rxfp 3/4 subclass includes more than just Rxfp3 and Rxfp4

Although previous data mining and phylogenetic reconstructionof vertebrate rxfps uncovered multiple rxfp3-like genes in teleosts(Wilkinson and Bathgate, 2007), the origin and orthologous/paralo-


gous relationship among these genes was only recently clarified(Yegorov and Good, 2012). This new knowledge will be valuablefor researchers working on various tetrapod taxa, such as cow,pig, opossum and turtle and teleosts, which harbour either or bothrxfp3-2 and rxfp3-3 orthologues.

6. Conclusions

In spite of their relatively recent discovery, characterization andde-orphanization, RXFPs have rapidly become a focus of intensivebiomedical research (Bathgate et al., 2013; Feng et al., 2009;Ganella et al., 2013; Ivell et al., 2011; Kong et al., 2010; McGuaneand Parry, 2005; Sasser, 2013; Smith et al., 2010; Summers et al.,2009; van der Westhuizen et al., 2008; Vodstrcil et al., 2012). Withmore data accumulating on the physiological functions of RXFP-mediated signaling, it is sensible to understand their evolutionarybasis and to consider broadening the range of model organismsavailable for studying these signaling systems.

Research may be substantially hindered by incorrect classifica-tion or annotation of genes and proteins in databases. This hasbeen the case with Rln/Insl genes in teleosts, which were anno-tated as rln3a-3f in databases such as NCBI, Ensembl and Zfin.These and other mistakes persist today for many rxfp and rln/inslgenes that we describe in this work [e.g. Takifugu rubripes rln geneis named ‘‘rln3c’’ in NCBI (DQ462200), chicken rxfp2-like and cave-fish rxfp3-3a1 are annotated as ‘‘rxfp2’’ and ‘‘rxfp4’’, respectively, inEnsembl]. This is partly due to the absence of experimentallyobtained data required for gene/protein annotation in public dat-abases. We are hopeful that the novel findings summarized in thisreview will be incorporated into the literature and into appropriatedatabases and that the revised nomenclature will provide a unifiedvision for comparative evolutionary research on relaxin signalingsystems in vertebrates.

Acknowledgments

We thank Dr. Juan Opazo for valuable comments on the manu-script. This research was funded by Manitoba Graduate Studentand University of Winnipeg Fellowships to SY and an NSERC dis-covery grant (# 261616) awarded to SVG.




Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.ygcen.2014.07.014.

References

Adam, I.M., Burkhardt, E., Benahmed, M., et al., 1993. Cloning of a cDNA for a novelinsulin-like peptide of the testicular Leydig cells. J. Biol. Chem. 268, 26668–26672.

Amamoo, A., Wilson, B.C., 2005. Relaxin inhibits central angiotensin II expression inkillifish: a central osmoregulatory role for relaxin and angiotensin II in thekillifish Fundulus heteroclitus. Endocrinology 1041, 229–232.

Anand-Ivell, R., Ivell, R., 2014. Regulation of the reproductive cycle and earlypregnancy by relaxin family peptides. Mol. Cell. Endocrinol. 382, 472–479.

Arroyo, J., Hoffmann, F., Good, S., Opazo, J., 2012a. INSL4 pseudogenes help definethe relaxin family repertoire in the common ancestor of placental mammals. J.Mol. Evol., 1–6.

Arroyo, J., Hoffmann, F., Opazo, J., 2012b. Gene duplication and positive selectionexplains unusual physiological roles of the relaxin gene in the European rabbit.J. Mol. Evol. 74, 52–60.

Arroyo, J.I., Hoffmann, F.G., Opazo, J.C., 2012c. Gene duplication and positiveselection explains unusual physiological roles of the relaxin gene in theEuropean rabbit. J. Mol. Evol. 74, 52–60.

Arroyo, J.I., Hoffmann, F.G., Opazo, J.C., 2012d. Gene turnover and differentialretention in the relaxin/insulin-like gene family in primates. Mol. Phylogenet.Evol. 63, 768–776.

Arroyo, J.I., Hoffmann, F.G., Opazo, J.C., 2014. Evolution of the relaxin/insulin-likegene family in anthropoid primates. Genome Biol. Evol. 6, 491–499.

Bathgate, R.A.D., Samuel, C.S., Burazin, T.C.D., Layfield, S., Claasz, A.A., Reytomas,I.G.T., Dawson, N.F., Zhao, C., Bond, C., Summers, R.J., Parry, L.J., Wade, J.D.,Tregear, G.W., 2002. Human relaxin gene 3 (H3) and the equivalent mouserelaxin (M3) gene: novel members of the relaxin peptide family. J. Biol. Chem.277, 1148–1157.

Bathgate, R.A., Ivell, R., Sanborn, B.M., Sherwood, O.D., Summers, R.J., 2006.International union of pharmacology LVII: recommendations for thenomenclature of receptors for relaxin family peptides. Pharmacol. Rev. 58, 7–31.

Bathgate, R.A.D., Halls, M.L., van der Westhuizen, E.T., Callander, G.E., Kocan, M.,Summers, R.J., 2013. Relaxin family peptides and their receptors. Physiol. Rev.93, 405–480.

Boels, K., Schaller, H.C., 2003. Identification and characterisation of GPR100 as anovel human G-protein-coupled bradykinin receptor. Br. J. Pharmacol. 140,932–938.

Brackett, K.H., Fields, P.A., Dubois, W., Chang, S.-M.T., Mather, F.B., Fields, M.J., 1997.Relaxin: an ovarian hormone in an avian species (Gallus domesticus). Gen. Comp.Endocrinol. 105, 155–163.

Burnicka-Turek, O., Mohamed, B.A., Shirneshan, K., Thanasupawat, T., Hombach-Klonisch, S., Klonisch, T., Adham, I.M., 2012. INSL5-deficient mice display analteration in glucose homeostasis and an impaired fertility. Endocrinology 153,4655–4665.

Conklin, D., Lofton-Day, C.E., Haldeman, B.A., Ching, A., Whitmore, T.E., Lok, S.,Jaspers, S., 1999. Identification of INSL5, a new member of the insulinsuperfamily. Genomics 60, 50–56.

de Rienzo, G., Aniello, F., Branno, M., Minucci, S., 2001. Isolation andcharacterization of a novel member of the relaxin/insulin family from thetestis of the frog Rana esculenta. Endocrinology 142, 3231–3238.

De Rienzo, G., Aniello, F., Branno, M., Izzo, G., Minucci, S., 2006. The expression levelof frog relaxin mRNA (fRLX), in the testis of Rana esculenta, is influenced bytestosterone. J. Exp. Biol. 209, 3806–3811.

Donizetti, A., Fiengo, M., Minucci, S., Aniello, F., 2009. Duplicated zebrafish relaxin-3gene shows a different expression pattern from that of the co-orthologue gene.Dev. Growth Differ. 51, 715–722.

Donizetti, A., Fiengo, M., del Gaudio, R., Di Giaimo, R., Minucci, S., Aniello, F., 2010.Characterization and developmental expression pattern of the relaxin receptorrxfp1 gene in zebrafish. Dev. Growth Differ. 52, 799–806.

Duarte, C., Kobayashi, Y., Kawamoto, T., Moriyama, K., 2014. Relaxin receptors 1 and2 and nuclear receptor subfamily 3, group C, member 1 (glucocorticoidreceptor) mRNAs are expressed in oral components of developing mice. Arch.Oral Biol. 59, 111–118.

Edgar, R., 2004. MUSCLE: a multiple sequence alignment method with reduced timeand space complexity. BMC Bioinformatics 5, 113.

Feng, S., Ferlin, A., Truong, A., Bathgate, R., Wade, J.D., Corbett, S., Han, S., Tannour-Louet, M., Lamb, D.J., Foresta, C., Agoulnik, A.I., 2009. INSL3/RXFP2 signaling intesticular descent. Endocrinology 1160, 197–204.

Ferlin, A., Selice, R., Carraro, U., Foresta, C., 2013. Testicular function and bonemetabolism-beyond testosterone. Nat. Rev. Endocrinol. 9, 548–554.

Fiengo, M., del Gaudio, R., Iazzetti, G., Di Giaimo, R., Minucci, S., Aniello, F., Donizetti,A., 2013. Developmental expression pattern of two zebrafish rxfp3 paraloguegenes. Dev. Growth Differ. 55, 766–775.

Fredriksson, R., Lagerström, M.C., Lundin, L.-G., Schiöth, H.B., 2003. The G-protein-coupled receptors in the human genome form five main families. Phylogeneticanalysis, paralogon groups, and fingerprints. Mol. Pharmacol. 63, 1256–1272.


Ganella, D.E., Ma, S., Gundlach, A.L., 2013. Relaxin-3/RXFP3 signaling andneuroendocrine function – A perspective on extrinsic hypothalamic control.Front. Endocrinol. 4.

García-López, Á., Bogerd, J., Granneman, J.C.M., van Dijk, W., Trant, J.M., Taranger,G.L., Schulz, R.W., 2009. Leydig cells express follicle-stimulating hormonereceptors in African catfish. Endocrinology 150, 357–365.

Gelsleichter, J., Steinetz, B.G., Manire, C.A., Ange, C., 2003. Serum relaxinconcentrations and reproduction in male bonnethead sharks, Sphyrna tiburo.Gen. Comp. Endocrinol. 132, 27–34.

Georges, D., Schwabe, C., 1999. Porcine relaxin, a 500 million-year-old hormone?The tunicate Ciona intestinalis has porcine relaxin. FASEB J. 13, 1269–1275.

Gloriam, D.E., Foord, S.M., Blaney, F.E., Garland, S.L., 2009. Definition of the Gprotein-coupled receptor transmembrane bundle binding pocket andcalculation of receptor similarities for drug design. J. Med. Chem. 52, 4429–4442.

Good, S., Yegorov, S., Martijn, J., Franck, J., Bogerd, J., 2012. New insights into ligand-receptor pairing and coevolution of relaxin family peptides and their receptorsin teleosts. Int. J. Evol. Biol. 2012, 14.

Good-Avila, S.V., Yegorov, S., Harron, S., Bogerd, J., Glen, P., Ozon, J., Wilson, B.C.,2009. Relaxin gene family in teleosts: phylogeny, syntenic mapping, selectiveconstraint, and expression analysis. BMC Evol. Biol. 9, 293.

Gunnersen, J.M., Crawford, R.J., Tregear, G.W., 1995. Expression of the relaxin genein rat tissues. Mol. Cell. Endocrinol. 110, 55–64.

Hafner, M., Korthof, G., 2006. Does a ‘‘500 million-year-old hormone’’ disproveDarwin? FASEB J. 20, 1290–1292.

Halls, M.L., Bathgate, R.A., Summers, R.J., 2006. Relaxin family peptide receptorsRXFP1 and RXFP2 modulate cAMP signaling by distinct mechanisms. Mol.Pharmacol. 70, 214–226.

Hisaw, F., 1926. Experimental relaxation of the pubic ligament of the guinea pig.Proc. Soc. Exp. Biol. Med. 23, 661–663.

Ho, T.-Y., Yan, W., Bagnell, C.A., 2007. Relaxin-induced matrix metalloproteinase-9expression is associated with activation of the NF-jB pathway in human THP-1cells. J. Leukoc. Biol. 81, 1303–1310.

Hoffmann, F.G., Opazo, J.C., 2011. Evolution of the relaxin/insulin-like gene family inplacental mammals: implications for its early evolution. J. Mol. Evol. 72, 72–79.

Holland, L.Z., Albalat, R., Azumi, K., Benito-Gutierrez, E., Blow, M.J., Bronner-Fraser,M., Brunet, F., Butts, T., Candiani, S., Dishaw, L.J., Ferrier, D.E., Garcia-Fernandez,J., Gibson-Brown, J.J., Gissi, C., Godzik, A., Hallbook, F., Hirose, D., Hosomichi, K.,Ikuta, T., Inoko, H., Kasahara, M., Kasamatsu, J., Kawashima, T., Kimura, A.,Kobayashi, M., Kozmik, Z., Kubokawa, K., Laudet, V., Litman, G.W., McHardy,A.C., Meulemans, D., Nonaka, M., Olinski, R.P., Pancer, Z., Pennacchio, L.A.,Pestarino, M., Rast, J.P., Rigoutsos, I., Robinson-Rechavi, M., Roch, G., Saiga, H.,Sasakura, Y., Satake, M., Satou, Y., Schubert, M., Sherwood, N., Shiina, T.,Takatori, N., Tello, J., Vopalensky, P., Wada, S., Xu, A., Ye, Y., Yoshida, K.,Yoshizaki, F., Yu, J.K., Zhang, Q., Zmasek, C.M., de Jong, P.J., Osoegawa, K.,Putnam, N.H., Rokhsar, D.S., Satoh, N., Holland, P.W., 2008. The amphioxusgenome illuminates vertebrate origins and cephalochordate biology. Genome.Res. 18, 1100–1111.

Hombach-Klonisch, S., Bialek, J., Radestock, Y., Truong, A., Agoulnik, A.I., Fiebig, B.,Willing, C., Weber, E., Hoang-Vu, C., Klonisch, T., 2010. INSL3 has tumor-promoting activity in thyroid cancer. Int. J. Cancer 127, 521–531.

Hornsby, D.J., Wilson, B.C., Summerlee, A.J., 2001. Relaxin and drinking in pregnantrats. Prog. Brain Res. 133, 229–240.

Hsu, S.Y., 1999. Cloning of two novel mammalian paralogs of relain/insulin familyproteins and their expression in testis and kidney. Mol. Endocrinol. 13, 2163–2174.

Hsu, S.Y., Kudo, M., Chen, T., Nakabayashi, K., Bhalla, A., van der Spek, P.J., van Duin,M., Hsueh, A.J.W., 2000. The three subfamilies of leucine-rich repeat-containingG protein-coupled receptors (LGR): identification of LGR6 and LGR7 and thesignaling mechanism for LGR7. Mol. Endocrinol. 14, 1257–1271.

Hsu, S.Y., Nakabayashi, K., Nishi, S., Kumagai, J., Kudo, M., Sherwood, O.D., Hsueh,A.J.W., 2002. Activation of orphan receptors by the hormone relaxin. Science295, 671–674.

Hu, G.B., Kusakabe, M., Takei, Y., 2011. Localization of diversified relaxin genetranscripts in the brain of eels. Gen. Comp. Endocrinol. 172, 430–439.

Hurley, I.A., Mueller, R.L., Dunn, K.A., Schmidt, E.J., Friedman, M., Ho, R.K., Prince,V.E., Yang, Z., Thomas, M.G., Coates, M.I., 2007. A new time-scale for ray-finnedfish evolution. Proc. R. Soc. Edinburgh Biol. 274, 489–498.

Ivell, R., Hartung, S., Anand-Ivell, R., 2005. Insulin-like factor 3: Where are we now?Endocrinology 1041, 486–496.

Ivell, R., Kotula-Balak, M., Glynn, D., Heng, K., Anand-Ivell, R., 2011. Relaxin familypeptides in the male reproductive system—a critical appraisal. Mol. Hum.Reprod. 17, 71–84.

Ivell, R., Wade, J.D., Anand-Ivell, R., 2013. INSL3 as a biomarker of leydig cellfunctionality. Biol. Reprod. 88 (147), 141–148.

Jobb, G., von Haeseler, A., Strimmer, K., 2004. TREEFINDER: a powerful graphicalanalysis environment for molecular phylogenetics. BMC Evol. Biol. 4, 18.

Katoh, K., Standley, D.M., 2013. MAFFT multiple sequence alignment software version7: improvements in performance and usability. Mol. Biol. Evol. 30, 772–780.

Kawamura, K., Kumagai, J., Sudo, S., Chun, S.Y., Pisarska, M., Morita, H., Toppari, J.,Fu, P., Wade, J.D., Bathgate, R.A., Hsueh, A.J., 2004. Paracrine regulation ofmammalian oocyte maturation and male germ cell survival. Proc. Natl. Acad.Sci. USA 101, 7323–7328.

Kong, R.C., Shilling, P.J., Lobb, D.K., Gooley, P.R., Bathgate, R.A., 2010. Membranereceptors: structure and function of the relaxin family peptide receptors. Mol.Cell. Endocrinol. 320 (1–2), 1–15.




http://refhub.elsevier.com/S0016-6480(14)00285-8/h0505




































































































































































Kuraku, S., Meyer, A., Kuratani, S., 2009. Timing of genome duplications relative tothe origin of the vertebrates: did cyclostomes diverge before or after? Mol. Biol.Evol. 26, 47–59.

Kusakabe, M., Ishikawa, A., Kitano, J., 2014. Relaxin-related gene expression differsbetween anadromous and stream-resident stickleback (Gasterosteus aculeatus)following seawater transfer. Gen. Comp. Endocrinol. http://dx.doi.org/10.1016/j.ygcen.2014.06.017.

Liu, C., Chen, J., Sutton, S., Roland, B., Kuei, C., Farmer, N., Sillard, R., Lovenberg, T.W.,2003. Identification of relaxin-3/INSL7 as a ligand for GPCR142. J. Biol. Chem.278, 50765–50770.

Liu, C., Kuei, C., Sutton, S., Chen, J., Bonaventure, P., Wu, J., Nepomuceno, D., Kamme,F., Tran, D.T., Zhu, J., Wilkinson, T., Bathgate, R., Eriste, E., Sillard, R., Lovenberg,T.W., 2005. INSL5 is a high affinity specific agonist for GPCR142 (GPR100). J.Biol. Chem. 280, 292–300.

Mashima, H., Ohno, H., Yamada, Y., Sakai, T., Ohnishi, H., 2013. INSL5 may be aunique marker of colorectal endocrine cells and neuroendocrine tumors.Biochem. Biophys. Res. Commun. 432, 586–592.

Matsumoto, M., Kamohara, M., Sugimoto, T., Hidaka, K., Takasaki, J., Saito, T., Okada,M., Yamaguchi, T., Furuichi, K., 2000. The novel G-protein coupled receptorSALPR shares sequence similarity with somatostatin and angiotensin receptors.Gene 248, 183–189.

McGowan, B.M.C., Stanley, S.A., Smith, K.L., White, N.E., Connolly, M.M., Thompson,E.L., Gardiner, J.V., Murphy, K.G., Ghatei, M.A., Bloom, S.R., 2005. Central relaxin-3 administration causes hyperphagia in male wistar rats. Endocrinology 146,3295–3300.

McGowan, B.M., Stanley, S.A., Donovan, J., Thompson, E.L., Patterson, M., Semjonous,N.M., Gardiner, J.V., Murphy, K.G., Ghatei, M.A., Bloom, S.R., 2008. Relaxin-3stimulates the hypothalamic–pituitary–gonadal axis. Am. J. Physiol. Endocrinol.Metab. 295, E278–286.

McGowan, B., Minnion, J., Murphy, K.G., Roy, D., Stanley, S., Dhillo, W.S., Gardiner, J.,Ghatei, M., Bloom, S.R., 2014. Relaxin-3 stimulates the neuroendocrine stressaxis via corticotrophin-releasing hormone. J. Endocrinol. 221, 337–346.

McGuane, J.T., Parry, L.J., 2005. Relaxin and the extracellular matrix: molecularmechanisms of action and implications for cardiovascular disease. Expert Rev.Mol. Med. 7, 1–18.

Mita, M., 2013a. Relaxin-like gonad-stimulating substance in an echinoderm, thestarfish: A novel relaxin system in reproduction of invertebrates. Gen. Comp.Endocrinol. 181, 241–245.

Mita, M., 2013b. Release of relaxin-like gonad-stimulating substance from starfishradial nerves by lonomycin. Zool. Sci. 30, 602–606.

Mita, M., Yoshikuni, M., Ohno, K., Shibata, Y., Paul-Prasanth, B., Pitchayawasin, S.,Isobe, M., Nagahama, Y., 2009. A relaxin-like peptide purified from radial nervesinduces oocyte maturation and ovulation in the starfish, Asterina pectinifera.Proc. Natl. Acad. Sci. USA 106, 9507–9512.

Mita, M., Yamamoto, K., Nagahama, Y., 2011. Interaction of relaxin-like gonad-stimulating substance with ovarian follicle cells of the starfish Asterinapectinifera. Zoolog. Sci. 28, 764–769.

Nakatani, Y., Takeda, H., Kohara, Y., Morishita, S., 2007. Reconstruction of thevertebrate ancestral genome reveals dynamic genome reorganization in earlyvertebrates. Genome Res. 17, 1254–1265.

Nef, S., Parada, L.F., 1999. Cryptorchidism in mice mutant for Insl3. Nat. Genet. 22,295–299.

Notredame, C., Higgins, D.G., Heringa, J., 2000. T-coffee: a novel method for fast andaccurate multiple sequence alignment. J. Mol. Biol. 302, 205–217.

Olinski, R.P., Dahlberg, C., Thorndyke, M., Hallbook, F., 2006a. Three insulin-relaxin-like genes in Ciona intestinalis. Peptides 27, 2535–2546.

Olinski, R.P., Lundin, L.G., Hallbook, F., 2006b. Conserved synteny between the Cionagenome and human paralogons identifies large duplication events in themolecular evolution of the insulin-relaxin gene family. Mol. Biol. Evol. 23, 10–22.

Osheroff, P.L., Ho, W.H., 1993. Expression of relaxin mRNA and relaxin receptors inpostnatal and adult rat brains and hearts. Localization and developmentalpatterns. J. Biol. Chem. 268, 15193–15199.

Park, J.I., Semyonov, J., Chang, C.L., Yi, W., Warren, W., Hsu, S.Y., 2008. Origin ofINSL3-mediated testicular descent in therian mammals. Genome Res. 18, 974–985.

Putnam, N.H., Butts, T., Ferrier, D.E., Furlong, R.F., Hellsten, U., Kawashima, T.,Robinson-Rechavi, M., Shoguchi, E., Terry, A., Yu, J.K., Benito-Gutiérrez, E.L.,Dubchak, I., Garcia-Fernàndez, J., Gibson-Brown, J.J., Grigoriev, I.V., Horton, A.C.,de Jong, P.J., Jurka, J., Kapitonov, V.V., Kohara, Y., Kuroki, Y., Lindquist, E., Lucas,


S., Osoegawa, K., Pennacchio, L.A., Salamov, A.A., Satou, Y., Sauka-Spengler, T.,Schmutz, J., Shin, I.T., Toyoda, A., Bronner-Fraser, M., Fujiyama, A., Holland, L.Z.,Holland, P.W., Satoh, N., Rokhsar, D.S., 2008. The amphioxus genome and theevolution of the chordate karyotype. Nature 453 (7198), 1064–1071.

Reinig, J.W., Daniel, L.N., Schwabe, C., Gowan, L.K., Steinetz, B.G., O’Byrne, E.M.,1981. Isolation and characterization of relaxin from the sand tiger shark(Odontaspis taurus). Endocrinology 109, 537–543.

Sanogo, Y.O., Hankison, S., Band, M., Obregon, A., Bell, A.M., 2011. Braintranscriptomic response of threespine sticklebacks to cues of a predator.Brain Behav. Evol. 77, 270–285.

Santini, F., Harmon, L., Carnevale, G., Alfaro, M., 2009. Did genome duplication drivethe origin of teleosts? A comparative study of diversification in ray-finnedfishes 9, 194.

Sasser, J.M., 2013. WEH Section New Investigator Award Lecture. The emerging roleof relaxin as a novel therapeutic pathway in the treatment of chronic kidneydisease. Am. J. Physiol. Regul. Integr. Comp. Physiol. 305 (6), R559–R565. http://dx.doi.org/10.1152/ajpregu.0052.

Schluter, D., 1995. Adaptive radiation in sticklebacks: trade-offs in feedingperformance and growth 76, 82–90.

Semyonov, J., Park, J.-I., Chang, C.L., Hsu, S.Y.T., 2008. GPCR genes are preferentiallyretained after whole genome duplication. PLoS One 3, e1903.

Sherwood, O.D., 2004. Relaxin’s physiological roles and other diverse actions.Endocr. Rev. 25, 205–234.

Smith, C.M., Shen, P.J., Banerjee, A., Bonaventure, P., Ma, S., Bathgate, R., Sutton,S.W., Gundlach, A.L., 2010. Distribution of relaxin-3 and RXFP3 withinarousal, stress, affective, and cognitive circuits of mouse brain. J. Comp.Neurol. 518, 4014–4045.

Steinetz, B.G., Schwabe, C., Callard, I.P., Goldsmith, L.T., 1998. Dogfish shark (Squalusacanthias) testes contain a relaxin. J. Androl. 19, 110–115.

Stolte, E.H., van Kemenade, B.M.L.V., Savelkoul, H.F.J., Flik, G., 2006. Evolution ofglucocorticoid receptors with different glucocorticoid sensitivity. J. Endocrinol.190, 17–28.

Summers, R.J., Bathgate, R.A., Wade, J.D., van der Westhuizen, E.T., Halls, M.L., 2009.Roles of the receptor, the ligand, and the cell in the signal transductionpathways utilized by the relaxin family peptide receptors 1–3. Endocrinology1160, 99–104.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:molecular evolutionary genetics analysis using maximum likelihood,evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28,2731–2739.

Tanaka, M., Iijima, N., Miyamoto, Y., Fukusumi, S., Itoh, Y., Ozawa, H., Ibata, Y., 2005.Neurons expressing relaxin 3/INSL 7 in the nucleus incertus respond to stress.Eur. J. Neurol. 21, 1659–1670.

van der Westhuizen, E.T., Halls, M.L., Samuel, C.S., Bathgate, R.A., Unemori, E.N.,Sutton, S.W., Summers, R.J., 2008. Relaxin family peptide receptors–fromorphans to therapeutic targets. Drug Discov. Today 13, 640–651.

Van Loy, T., Vandersmissen, H.P., Van Hiel, M.B., Poels, J., Verlinden, H., Badisco, L.,Vassart, G., Vanden Broeck, J., 2008. Comparative genomics of leucine-richrepeats containing G protein-coupled receptors and their ligands. Gen. Comp.Endocrinol. 155, 14–21.

Vodstrcil, L.A., Tare, M., Novak, J., Dragomir, N., Ramirez, R.J., Wlodek, M.E., Conrad,K.P., Parry, L.J., 2012. Relaxin mediates uterine artery compliance duringpregnancy and increases uterine blood flow. FASEB J. 26, 4035–4044.

Wilkinson, T.N., Bathgate, R.A., 2007. The evolution of the relaxin peptide family andtheir receptors. Adv. Exp. Med. Biol. 612, 1–13.

Wilkinson, T.N., Speed, T.P., Tregear, G.W., Bathgate, R.A., 2005a. Evolution of therelaxin-like peptide family. BMC Evol. Biol. 5, 14.

Wilkinson, T.N., Speed, T.P., Tregear, G.W., Bathgate, R.A., 2005b. Evolution of therelaxin-like peptide family: from neuropeptide to reproduction. Endocrinology1041, 530–533.

Wilson, B.C., Summerlee, A.J., 1994. Effects of exogenous relaxin on oxytocin andvasopressin release and the intramammary pressure response to centralhyperosmotic challenge. J. Endocrinol. 141, 75–80.

Yegorov, S., Good, S., 2012. Using paleogenomics to study the evolution of genefamilies: origin and duplication history of the relaxin family hormones andtheir receptors. PLoS One 7, e32923.

Zimmermann, S., Steding, G., Emmen, J.M.A., Brinkmann, A.O., Nayernia, K., Holstein,A.F., Engel, W., Adham, I.M., 1999. Targeted disruption of the Insl3 gene causesbilateral cryptorchidism 13, 681–691.
















































































http://dx.doi.org/10.1152/ajpregu.0052

http://dx.doi.org/10.1152/ajpregu.0052



















































The relaxin family peptide receptors and their ligands: New developments and paradigms in the...

Documents

Transcript of The relaxin family peptide receptors and their ligands: New developments and paradigms in the...