www.elsevier.com/locate/ygcen

General and Comparative Endocrinology 153 (2007) 132–140

Review

KiSS-1 system and reproduction: Comparative aspects and rolesin the control of female gonadotropic axis in mammals

Juan Roa, Manuel Tena-Sempere *

Physiology Section, Department of Cell Biology, Physiology and Immunology, University of Cordoba, 14004 Cordoba, Spain

Received 13 November 2006; revised 19 January 2007; accepted 21 January 2007Available online 30 January 2007

Abstract

In late 2003, inactivating mutations of the G protein-coupled receptor GPR54 were found in patients suffering hypogonadotropichypogonadism. This observation led to the proposal that this receptor and its putative ligands (kisspeptins, encoded by the KiSS-1 gene)are essential in the control of reproduction; a contention that has been now substantiated by an ever growing number of experimentalstudies. However, most (if not all) of this work has been carried out in mammals (human, sheep and laboratory rodents). Moreover,characterization of gonadotropin responses to kisspeptin was conducted in males, whereas its actions on the female gonadotropic axisinitially received much less attention. Notwithstanding, recent experimental data have unveiled very prominent roles of the KiSS-1 sys-tem in the control of key aspects of female reproduction, which include not only the timing puberty onset and its modulation by met-abolic factors, but also the dynamic regulation of the gonadotropic axis in adulthood. On the latter, the KiSS-1 neuron has beenproposed as key intermediary element for the negative and positive feedback effects of sex steroids on gonadotropin secretion. Moreover,expression of KiSS-1 (mRNA and peptide) and its receptor have been recently reported in the ovary, adding further complexity to thepotential actions of this system in the female. In sum, compelling experimental evidence, obtained in mammals, has recently defined thepivotal role of the KiSS-1/GPR54 system in the control of essential aspects of female reproduction, from puberty to ovulation. Whilecharacterization of its role in non-mammalian species remains largely unexplored, the presence of GPR54 in GnRH neurons and thechanges in its expression during pubertal development, reported recently in fish species, are suggestive of a conserved function of theKiSS-1/GPR54 system in the control of reproduction during evolution.� 2007 Elsevier Inc. All rights reserved.

Keywords: Kisspeptin; KiSS-1; GPR54; Gonadotropins; Gonadotropin-releasing hormone (GnRH); Ovary; Ovarian cycle; Ovulation; Hypogonadism

1. Introduction

Reproductive function, defined by the capacity to gener-ate viable gametes, to promote fertilization and to supportpregnancy, is an essential function for the perpetuation ofthe species, and thus it is under the control of a precise net-work of regulatory signals, which conforms the so-calledgonadotropic axis (Fink, 2000; Tena-Sempere and Huhtan-iemi, 2003). In mammals and other vertebrates, this neuro-hormonal system is arranged into three major levels oforganization: the hypothalamus, the pituitary and thegonads. Within this axis, the hypothalamic decapeptide

0016-6480/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.ygcen.2007.01.026

* Corresponding author. Fax: +34 957 218288.E-mail address: fi1tesem@uco.es (M. Tena-Sempere).

gonadotropin-releasing hormone (GnRH) is the majordriving signal, with ability to induce self-priming at thepituitary and to stimulate the pulsatile release of bothgonadotropins, luteinizing hormone (LH) and follicle-stim-ulating hormone (FSH). These, in turn, operate on specificreceptors located in the gonads (LH receptors in Leydigcells of the testis as well as theca and luteal cells of the ova-ry; FSH receptors in Sertoli cells of the testis and granulosacells of the ovary), where they conduct different trophicactions, including not only the promotion of spermatogen-esis, folliculogenesis and ovulation, but also the stimulationof gonadal hormone secretion. These hormonal productsof the gonads participate, through negative and eventuallypositive feedback loops, in the dynamic regulation of thegonadotropic axis (Tena-Sempere and Huhtaniemi, 2003).

J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140 133

In addition, multiple regulators, of central and peripheralorigin, impinge on the above elements of the hypothalam-ic-pituitary-gonadal (HPG) axis, thus participating in thetuning of this system and its functional coupling todifferent endogenous conditions (e.g. energy status) andenvironmental factors (e.g. dark–light cycles).

Although the major components of the HPG axisappeared to be well defined for decades (see above), ourknowledge of the neuroendocrine networks controllingreproduction was recently revolutionized by the identifica-tion of the essential roles in this system of kisspeptins (thepeptide products of the KiSS-1 gene) and their receptor,GPR54. Detailed reviews on the disclosure and initial char-acterization of the reproductive dimension of the KiSS-1/GPR54 system have been recently published elsewhere(Iovane et al., 2004; Seminara, 2005; Seminara, 2006;Tena-Sempere, 2006). In this work, we will focus our atten-tion on specific aspects of this system that originallyreceived less attention (but will be likely the subject of sub-stantial research efforts in the coming years), such as theroles of kisspeptin in the neuroendocrine control of thefemale reproductive axis. Nonetheless, for sake of termi-nology, we will first describe the components of theKiSS-1/GPR54 system, and will provide a brief overviewof its most salient roles in the control of reproduction. Ofnote, although the experimental work reviewed herein hasbeen conducted in mammals, some comments pertainingthe structure and function of the elements of the KiSS-1/GPR54 system in non-mammalian species are alsoincluded.

2. The KiSS-1/GPR54 system: identification and structural

features

The KiSS-1/GPR54 is a ligand-receptor system whosecomponents were characterized between 1996 and 2001.The first element to be identified was KiSS-1, which in1996 was cataloged as a metastasis-suppressor gene givenits differential profiles of expression in melanoma cells withdifferent metastatic capacity (Lee et al., 1996). In 1999,GPR54 was independently cloned in the rat as orphanreceptor with a partial sequence similarity (>40%) withthe transmembrane regions of galanin receptors (Leeet al., 1999). Finally, in 2001, the peptide products of theKiSS-1 gene were fully characterized. These turned out toderive from the differential proteolytic processing a com-mon precursor, and were globally termed kisspeptins(Kotani et al., 2001). Among kisspeptins (Kp), the majorproduct is a 54 amino acid peptide, expressed mainly inthe placenta, which was termed metastin by virtue of itsability to inhibit tumor metastasis (Ohtaki et al., 2001).In addition, other peptide fragments of the KiSS-1 precur-sor, such as Kp-14, Kp-13, and Kp-10, were also identified(Kotani et al., 2001). All kisspeptins share the C-terminalregion of the metastin molecule, where they have a canon-ical Arg-Phe-NH2 motif distinctive of the RF-amide pep-tide family, and are able to bind and activate GPR54,

now catalogued as the cognate receptor of KiSS-1 peptides(Kotani et al., 2001; Muir et al., 2001; Ohtaki et al., 2001).

To date, the cDNA sequences of KiSS-1 have beencloned in a number of mammalian species, including thehuman (Homo sapiens), chimpanzee (Pan troglodytes),macaque (Macaca mulatta), bull (Bos taurus), rat (Rattusnorvegicus) and mouse (Mus musculus) (GenBank Acces-sion Nos. NM_002256, XM_514123, AY_823262,XM_867473, NM_181692, and NM_178260, respectively).In addition, the cDNA sequence of ovine (Ovis aries) KiSS-1 has been recently submitted to GenBank (Accession No.DQ_059506). Comparative analysis of the predicted aminoacid sequences reveals a high degree of homology amongprimate species (>85%), whereas sequence identity betweenthe human, bovine and rodent kisspeptins ranges between45 and 50% (Fig. 1). To our knowledge, no cDNA or pro-tein sequences of KiSS-1/kisspeptins from non-mammalianspecies have been annotated in gene databases.

In addition, the cDNA sequences of GPR54 from differ-ent species have been reported. These include not onlymammals, such the human, macaque, rat, mouse and boar(Sus scrofa) (GenBank Accession Nos. NM_032551, AY833261, NM_023992, NM_053244, and DQ_459345/46,respectively), but also several non-mammalian species,such as the zebra fish (Danio rerio), tilapia (Oreochromis

niloticus) and purple see urchin (Strongylocentrotus

purpuratus) (see GenBank Accession Nos. XM_685300,AB_162143, and XM_001188545, respectively). In addi-tion, the cDNA sequences of GPR54 from the bullfrog(Rana catesbeiana; Cho et al., 2006), the grey mullet(Mugil cephalus; ABG76790.1) and the Atlantic croaker(Micropogonias undulatus; ABC75101.1) have been veryrecently reported. Alignment of the predicted amino acidsequences of GPR54 from representative mammalian andnon-mammalian species is presented in Fig. 2. It is interest-ing to note that the percentage of homology between pri-mates and rodent sequences is much higher for GPR54(>80%) than for KiSS-1 peptide. Moreover, sequence iden-tity between human, fish and frog GPR54 is >45%, whilethe homology between the human and urchin receptors is>20%. These data suggest a higher degree of sequence con-servation for the receptor than for ligand in the KiSS-1/GPR54 system across evolution.

3. Kisspeptin and GPR54 in reproduction: overview

The reproductive dimension of the KiSS-1/GPR54 sys-tem was disclosed in late 2003, when inactivating mutationsof GPR54 gene were first reported in humans and mice suf-fering hypogonadotropic hypogonadism (de Roux et al.,2003; Seminara et al., 2003). From that moment onwards,an ever growing number of studies worldwide have helpedto substantiate the essential functions of kisspeptin inthe control of different aspects of reproduction. Althoughdetailed description of the contents of these studiesis beyond the scope of this work, some of the most impor-tant roles recently assigned to kisspeptin in the area of

Fig. 1. Alignment of amino acid sequences of KiSS-1 peptides from different mammalian species. The sources of sequences are: Homo sapiens (GenBank,NP_002247.2), Pan troglodytes (GenBank, XP_514123.1), Macaca mulatta (GenBank, XP_001098284.1), Bos taurus (GenBank, XP_872566.1), Rattus

norvegicus (GenBank, NP_859043.1), and Mus musculus (GenBank, NP_839991.1). GeneDoc software was used for homology shading. Three shadinglevels are set: black for 100% similarity groups, deep grey for 80% similarity groups and light grey for 60% similarity groups. Metastin (upper lined) andkisspeptin-10 (underlined) amino acid sequences are shown.

134 J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140

Reproduction are summarized in the following points (seeTena-Sempere, 2006; and references therein):

• Kisspeptin is a very potent stimulator of the GnRH/gonadotropin axis. This effect is detected after its intra-cerebral and systemic administration, in a diversity ofmammalian species.

• The KiSS-1 system plays an essential role in the control(timing) of puberty onset. This action likely involves anenhancement of the endogenous kisspeptin tone as wellas an increase in the sensitivity to kisspeptin at the timeof puberty.

• The stimulatory actions of kisspeptin on the gonado-tropic axis are primarily conducted at the hypothala-mus, where it can activate GnRH neurons and elicitGnRH secretion.

• Two major populations of KiSS-1 expressing neuronshave been detected at the hypothalamus in rodents,which are located at the arcuate nucleus (ARC) andthe anteroventral periventricular nucleus (AVPV).

• Hypothalamic expression of KiSS-1 gene is under thecontrol of sex steroids, and KiSS-1 neurons are strongcandidates for mediating the negative and positive feed-back effects of estradiol on gonadotropin secretion (seefollowing sections).

• The hypothalamic KiSS-1 system operates also asmolecular conduit for the control of the gonadotropicaxis by additional relevant regulators, such as metabolicsignals and, likely, environmental cues.

It is to be stressed that virtually all the experimentalstudies published to date addressing the biological rolesof the KiSS-1/GPR54 system in the control of reproduc-tion have been conducted in mammalian species, includingthe rat, mouse, sheep and primates (see Tena-Sempere,2006). Nonetheless, some initial evidence suggests thatGPR54 signaling is also related with reproduction in

non-mammalian species; a phenomenon whose functionalcharacterization is still pending.

4. The female gonadotropic axis in mammals: functional

features

As indicated in previous sections, most of the experi-mental studies aiming at the characterization of the rolesof kisspeptin in the control of gonadotropin secretionwere originally focused in the male. Thus, pharmacolog-ical analyses conducted in different mammalianspecies, from rodents to humans, set the contention thatkisspeptin is likely the most potent elicitor of the GnRH/gonadotropic axis known so far (see Navarro et al.,2005a,b; Tena-Sempere, 2006). Interestingly, however,little attention was initially paid to characterization ofthe gonadotropin-releasing effects of kisspeptin in theadult female, neither was it evaluated whether kisspeptinconducts similar gonadotropin-stimulating actions innon-mammalian species. While the latter remains, toour knowledge, unexplored, evaluation of the specificroles of KiSS-1 system in the female gonadotropic axishas significantly progressed recently.

Although the basic regulatory elements of the HPG axisin mammals are essentially similar between males andfemales, there are specific functional features of the femalegonadotropic system that evidence a higher degree ofsophistication. Thus, a hallmark of reproductive function(and thus, gonadotropin secretion) in the adult female isits cyclic nature, defined by periodic episodes of ovulationthat are ultimately triggered by the pre-ovulatory surge ofgonadotropins (Schwartz, 2000). This is a timely neuroen-docrine phenomenon that involves a complex series oforchestrated hormonal events, such as the pre-ovulatoryrise in estradiol secretion by dominant follicles of the ovary(positive feedback), as well as the increase in hypothalamicGnRH secretion and GnRH self-priming at the pituitary

Fig. 2. Alignment of amino acid sequences of G protein-coupled receptor GPR54 from different mammalian and non-mammalian species. The sources ofsequences are: Homo sapiens (GenBank, AAK83235.1), Mus musculus (GenBank, AAK83236.1), Rattus norvegicus (GenBank, NP_076482.1), Sus scrofa

(GenBank, ABE73453.1), Danio rerio (GenBank, XP_690392.1), Oreochromis niloticus (GenBank, BAD34454.1), Micropogonias undulatus (GenBank,ABC75101.1) and Mugil cephalus (GenBank, ABG76790.1). GeneDoc software was used for homology shading. Three shading levels are set: black for100% similarity groups, deep grey for 80% similarity groups, and light grey for 60% similarity groups.

J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140 135

(Fink, 2000; Schwartz, 2000). Moreover, the pre-ovulatorypeak of gonadotropins is dependent on different environ-mental cues and endogenous factors. As additional levelof complexity, the female gonadotropic axis undergoes dra-matic functional changes during the periods of pregnancyand lactation when, among others, important modifica-tions in serum gonadotropin levels are detected (Linkieand Niswender, 1972; Smith and Grove, 2002). In this

work, we will summarize our current knowledge on thecontribution of kisspeptin to the dynamic regulation offemale gonadotropic axis, with special emphasis on itspotential role in the generation of the pre-ovulatory surgeof LH and its effects on gonadotropin secretion in differentfunctional states of the female. In addition, recent evidenceon the expression of KiSS-1 and GPR54 (mRNAs andproteins) in the ovary will be briefly reviewed.

136 J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140

5. Kisspeptin and gonadotropin secretion in the female along

the ovarian cycle

In an attempt to define the potential physiological rolesof kisspeptin in the control of female gonadotropic axis,LH secretory responses to kisspeptin-10 were recently eval-uated in our laboratory at different phases of the ovariancycle, using the rat as animal model (Roa et al., 2006).These analyses revealed that LH secretion is maximallyactivated by intracerebral injection of kisspeptin duringthe transition of proestrus to estrus. In fact, in the morningof estrus, Kp-10 was able to evoke maximal LH responseseven though the pre-ovulatory peak had taken place lessthan 12 h before. Yet, albeit of lower magnitude, significantLH secretory responses to kisspeptin were also detected atdiestrus. Collectively considered, these data suggest that,despite persistent responsiveness to kisspeptin along thecycle, there is a window of maximal effectiveness for theLH-releasing effects of kisspeptin from proestrus to estrus.Given the proposed role of the KiSS-1 system in the timelygeneration of the pre-ovulatory surge of LH (see below), itis clear that a tight temporal regulation of kisspeptin inputson GnRH neurons are required for the generation of thesurge in the evening of proestrus and its cessation thereaf-ter; a contention that has been experimentally confirmedvery recently (Smith et al., 2006).

The hormonal mechanisms for the observed fluctuationsin LH responsiveness to kisspeptin likely involve changes inthe sex steroid milieu along the cycle. Indeed, our function-al tests in ovariectomized (OVX) rats, with or withouthormone replacement, evidenced that combined supple-mentation with estradiol and progesterone was maximallyeffective in promoting LH responsiveness to kisspeptin(Roa et al., 2006). This observation suggests that the stateof hyper-responsiveness to kisspeptin detected at the proes-trus-to-estrus transition might be induced, at least partial-ly, by the pre-ovulatory rise of estradiol in the presence ofactivated progesterone receptors. On the latter, activationof progesterone receptors is well-known to be needed forfull expression of the pre-ovulatory surge (Levine et al.,2001), and the enhancement of LH responses to kisspeptinin the presence of progesterone might operate as contribut-ing factor to this phenomenon. Of note, the above changesin the magnitude of LH responses to kisspeptin along theestrous cycle and after manipulation of ovarian sex steroidlevels might be due to changes in the sensitivity of GnRHneurons to kisspeptin and/or modifications in pituitary sen-sitivity to GnRH. In this sense, an increase in the pituitarysensitivity due to GnRH self-priming has been reported atproestrus.

Finally, besides fluctuations in maximal LH responses,changes in the sensitivity to Kp-10 were also monitoredalong the estrous cycle. Intriguingly, despite absolute LHresponses to kisspeptin were significantly lower at diestrus,the sensitivity to low doses of Kp-10 was not only pre-served, but even enhanced, at this phase (Roa et al.,2006). This observation suggests that the sensitivity and

maximal responsiveness to kisspeptin are distinctly regulat-ed along the estrous cycle in the rat.

6. Kisspeptin and the pre-ovulatory surge of gonadotropins:

positive feedback

One of the first evidences for the critical role of kisspep-tin in the physiological control of gonadotropin secretionin the female came from expression analyses showing thatovariectomy not only evokes the expected rise in circulat-ing LH but induces also a significant elevation of KiSS-1(and GPR54) mRNA levels at the hypothalamus (Navarroet al., 2004). Moreover, the fact that estradiol supplemen-tation of OVX rats prevented both hormonal (LH) andexpression (KiSS-1) responses to gonadectomy further sup-ported a major role of hypothalamic KiSS-1 as centralmediator of the negative feedback effects of ovariansteroids on gonadotropin secretion.

However, subsequent localization analyses aiming at thedissection of the changes in KiSS-1 gene expression withinthe hypothalamus following gonadectomy, by means ofin situ hybridization, evidenced a higher degree of complex-ity of the organization of the KiSS-1 system at the hypothal-amus and its involvement in the control of gonadotropinsecretion (Smith et al., 2005a,b). As indicated in previoussections, these analyses revealed two major populations ofKiSS-1 neurons, one located at the arcuate nucleus(ARC) and the other at the anteroventral periventricularnucleus (AVPV). Interestingly, while negative regulationof hypothalamic KiSS-1 mRNA by sex steroids was demon-strated at the ARC, in the AVPV, KiSS-1 mRNA levelsdecreased after gonadectomy and increased after estrogenreplacement. Considering that the AVPV had beenpreviously identified as hypothalamic area involved in thepositive feedback control of gonadotropins, these observa-tions were suggestive of a major role of KiSS-1 neurons atthis site in conveying the positive feedback effects of estro-gen upon gonadotropin secretion selectively during thepre-ovulatory phase (Smith et al., 2005b).

The above findings were extremely appealing, as theyprovided compelling evidence for a discernible neuronalnetwork involved in the generation of the positive feedbackeffects of estradiol; a contention that had remained elusivefor decades. Indeed, the hypothesis of a major role ofAVPV KiSS-1 neurons in the generation of the pre-ovula-tory surge of LH has been recently supported by the dem-onstration that KiSS-1 mRNA levels at this hypothalamicsite (but not at the ARC) dramatically increase in the ratduring the evening of proestrus (Smith et al., 2006). Theserecent data are in line with previous evidence showing thatimmunoneutralization of central metastin abolishes thepre-ovulatory surge of gonadotropins (Kinoshita et al.,2005). Moreover, the population of KiSS-1 neurons atthe AVPV is strikingly sexually dimorphic (much greaterin the female rat; i.e., where positive feedback takes place),and its development appears to be sensitive to the organiz-ing effects of sex steroids during the critical period of sexual

J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140 137

brain differentiation, as neonatal androgenization perma-nently decreased (masculinized) the expression of KiSS-1gene at the AVPV in adult female rats (Kauffman et al.,2007). Taken as a whole, these observations demonstratethat the hypothalamic KiSS-1 system is an essential ele-ment for relaying not only negative, but also positive feed-back inputs of sex steroids to GnRH neurons, thus playinga pivotal function in the generation of the hormonal signaltriggering ovulation.

7. Kisspeptin and gonadotropin secretion in pregnancy and

lactation

To further define the contribution of kisspeptin to theregulation of gonadotropin secretion at other functionalstates of the female reproductive axis, expression studiesand pharmacological assays, testing the effects of Kp-10on LH and FSH levels, were conducted by our group inthe rat at pregnancy and lactation (Roa et al., 2006). Ofnote, significant changes in basal secretion of gonadotro-pins take place in pregnant and lactating dams, with amarked suppression of the gonadotropic axis especially atlactation (Smith and Grove, 2002). Moreover, in humans,a dramatic increase in circulating levels of metastin hasbeen described (Horikoshi et al., 2003); an observation thatseems to be at odds with the mild decrease of LH levelsreported at mid-pregnancy. The latter might be suggestiveof potential desensitization events, as >7000-fold rise inmetastin concentrations has been detected in pregnantwomen.

However, our analyses in pregnant rats, at mid- andlate-gestation, revealed that the absolute responses to effec-tive (high) doses of Kp-10 were not only conserved, butactually enhanced, when compared with the effectsobserved at diestrus. Moreover, the sensitivity to low dosesof Kp-10 was fully preserved at mid-gestation, when dosesas low as 0.1 pmol were able to elicit LH release (Roa et al.,2006). This state of hyper-responsiveness to kisspeptinmight be related with the exposure to elevated levels ofestradiol and progesterone during gestation, as suggestedby our previous analyses in gonadectomized rats (seeabove). As further proof for the lack of desensitization ofkisspeptin signaling, the expression levels of GPR54mRNA at the hypothalamus did not significantly changein pregnant rats. Intriguingly, hypothalamic expression ofKiSS-1 gene significantly increased along pregnancy; a phe-nomenon whose functional relevance is yet to be defined.In any event, our studies in the rat strongly suggest thatno significant desensitization to the stimulatory effects ofkisspeptin on gonadotropin secretion takes place duringgestation, at least in this species.

Concerning lactation, our studies evidenced that, despitethe dramatic suppression of basal gonadotropin levels,intracerebral injection of high doses of Kp-10 was able toelicit LH secretory responses in lactating dams. Yet, themagnitude of these responses was significantly attenuatedvs. non-lactating controls. Moreover, a significant impact

of lactation on the sensitivity to kisspeptin in terms of gon-adotropin secretion was demonstrated. Thus, while dosesof 0.1 pmol Kp-10 were sufficient to elicit LH secretionin cyclic (and pregnant) rats, doses of 10 pmol were com-pletely ineffective in lactating rats (Roa et al., 2006). Thissuggest a dramatic suppression in the sensitivity of thegonadotropic system to kisspeptin stimulation, with an ele-vation of the threshold effective doses of al least two ordersof magnitude. This phenomenon might be mechanisticallyrelevant for the observed inhibition of the HPG axis at lac-tation. On the contrary, the possibility of a primary defectof endogenous kisspeptin tone seems less likely, as ourexpression analyses failed to demonstrate significantchanges in KiSS-1 mRNA expression at the hypothalamusat early lactation (Roa et al., 2006).

8. Expression of KiSS-1 and GPR54 in the ovary: functional

relevance?

While clinical and experimental evidence gathered inrecent years undisputedly demonstrates an indispensablerole of hypothalamic KiSS-1 system in the neuroendocrinecontrol of reproduction in mammals, the possibility ofadditional modulatory actions of kisspeptin at other levelsof the female reproductive axis cannot be excluded. In fact,the prominent roles of kisspeptin and GPR54 at the hypo-thalamus make it possible that such peripheral functionsmight have been overlooked, even at sites of the gonadalaxis where evidence for the expression of KiSS-1 and/orGPR54 genes had been reported. In this context, our grouphas got recently interested in the analysis of the potentialexpression of KiSS-1 and GPR54 in the ovary, as the pres-ence of both mRNAs had been preliminarily described inthis tissue in the rat (Terao et al., 2004). Our immunohisto-chemical and gene expression analyses have now demon-strated that the elements of the KiSS-1 system areactually expressed in the rat ovary, with detectable mRNAlevels of both targets across the estrous cycle, and clear-cutimmunoreactive signals for kisspeptin in the theca layer ofgrowing follicles, corpora lutea and interstitial gland; com-partments where modest GPR54 immunoreactivity wasalso observed (Castellano et al., 2006).

One of the most salient aspects of KiSS-1 expression atthe ovary revealed by our studies was that, contrary toGPR54, its mRNA levels fluctuate in a cyclic-dependentmanner, with a robust increase in the afternoon of proes-trus, i.e., preceding ovulation. This observation promptedus to evaluate in detail the hormonal regulation of ovarianexpression of KiSS-1 gene. Notably, the rise in ovarianKiSS-1 mRNA at the afternoon of proestrus was fully pre-vented by blockade of the pre-ovulatory surge of gonado-tropins by means of pre-treatment with an antagonist ofGnRH. Conversely, in females treated with a GnRHantagonist, replacement with a super-agonist of LH, suchas human choriogonadotropin, restored KiSS-1 over-ex-pression at the ovary. Interestingly, immature rat ovariesshowed low to negligible levels of KiSS-1 mRNA, but

AVPV

LH

FSH+

-

+

Kp+

Kp

GnRH

GnRH

+ +

Arc

HYPOTHALAMUS

PITUITARY

OVARY

KiSS-1?

E2 PE2 P

Fig. 3. Tentative model for the putative roles of KiSS-1 system in thecontrol of the reproductive axis in the female in mammals. The primarysite of action of this system is located at the hypothalamus, wherekisspeptin is known to act directly on GnRH neurons, which expressGPR54, to activate the GnRH/gonadotropin axis. Two different popula-tions of KiSS-1 (Kp) neurons have been identified within the hypothal-amus, at the arcuate nucleus (ARC) and the anteroventral periventriculararea (AVPV). While Kp neurons at the ARC have been implicated inrelaying negative feedback actions of estradiol (E2) and, eventually,progesterone (P), Kp neurons at the AVPV are likely involved inmediating the positive feedback actions of E2, selectively in the female.In addition, expression of the elements of KiSS-1 system has been recentlydescribed in the rat ovary. However, the potential relevance of directgonadal actions of kisspeptin is yet to be defined. For more details, see thetext.

138 J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140

expression of KiSS-1 was significantly enhanced by a stan-dard protocol of gonadotropin priming (Castellano et al.,2006). Overall, the above observations evidence that theKiSS-1 gene is expressed in the rat ovary in a developmen-tal- and hormonally-regulated manner. Although the func-tional relevance of this phenomenon is yet to be defined,the ability of the LH surge to timely induce ovarian expres-sion of KiSS-1 gene at the pre-ovulatory period is sugges-tive of a potential role of locally produced kisspeptin inthe control of ovulation. Of note, expression ofthe elements of the KiSS-1/GPR54 system appears to beconserved in the ovary, as we have recently obtained evi-dence for the presence of kisspeptin in the primate ovary(our unpublished data), while the ovarian expression ofGPR54 mRNA has been reported very recently in theteleost fish (Nocillado et al., 2007).

9. Conclusions and future perspectives: kisspeptin in

comparative endocrinology

As indicated in previous sections, during the last threeyears, we have witnessed the emergence of the hypothalam-ic KiSS-1 system as fundamental element in the central cir-cuits controlling key aspects of reproductive developmentand function, such as puberty onset, gonadotropin secre-tion and the metabolic control of fertility. Although initialefforts in the characterization of the roles of kisspeptin inthe control of the gonadotropic axis were mostly focusedin the male, recently developments in this area have clearlydemonstrated that the KiSS-1 system is essential in the reg-ulation of key facets of female reproduction (as schemati-cally depicted in Fig. 3). These include the generation ofthe pre-ovulatory surge of gonadotropins, via mediationof the positive feedback effects of estradiol, whereby kiss-peptin plays a central role in the hormonal control of ovu-lation. In addition, expression of KiSS-1 (mRNA andpeptide) and its receptor has been recently demonstratedin rat ovary, where kisspeptin (whose ovarian expressionappears tightly regulated, with peak levels during the pre-ovulatory period) might play additional modulatory rolesin the local control of ovulation. Finally, the patterns ofgonadotropin responses to kisspeptin at different function-al states of the female reproductive axis have been recentlycharacterized, thus allowing to set the basis for the rationaluse of kisspeptin analogs in the therapeutic manipulationof the gonadotropic axis of the female.

In spite of the rapid progress in this field, several aspectsof kisspeptin physiology as major regulator of the HPG axisare yet to be fully elucidated. Among those, there is a con-spicuous lack of knowledge on the central (neural) circuitsresponsible for the control of the KiSS-1 system, whosecharacterization will likely concentrate considerable atten-tion in the near future. More relevant for the present review,it is also striking that the available information regardingthe structure and function of the elements of the KiSS-1/GPR54 system in non-mammalian species remains aston-ishingly limited. In terms of structure, the cDNA sequences

of GPR54 in fish (zebra fish, tilapia, grey mullet and Atlan-tic croaker), bullfrog and purple sea urchin have beenreported, with an overall degree of homology of 45–50%between human, fish and amphibian amino acid sequences.In contrast, to our knowledge, no KiSS-1 cDNA sequencesfrom non-mammalian species have been annotated to date.This might suggest a higher degree of conservation for thereceptor than for the ligand across evolution. In terms offunction, analyses of the effects of kisspeptin upon thegonadotropic axis in non-mammalian species have not beenso far reported, which might be due to the lack of cloning ofKiSS-1 gene in such phyla. Nonetheless, it is noticeable thatrecent transfection studies demonstrated the ability ofrodent Kp-10 to activate bullfrog GPR54 (Cho et al.,2006); a phenomenon which could allow the conductionof heterologous hormonal tests to identify potential rolesof this system in the control of the reproductive axis in

J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140 139

non-mammals. These studies also evidenced that the sensi-tivity of bullfrog GPR54 to rodent Kp-10 is �10-fold lowerthan that of the rat receptor, suggesting the existence of anspecific form of kisspeptin in the bullfrog, which is yet to beisolated. In any event, despite functional tests are yet to beconducted, some fragmentary evidence strongly suggeststhat GPR54 and its ligand(s) are actually involved in thecontrol of reproduction also in non-mammalian species.These data have been mainly gathered in fish, where expres-sion of GPR54 has been described in GnRH neurons in tila-pia (Parhar et al., 2004), and dynamic fluctuations ofGPR54 mRNA levels have been demonstrated alongpubertal maturation in the brain and ovary of grey mullet(Nocillado et al., 2007).

In sum, we have reviewed herein recent molecular andphysiological data that have substantiated the importantroles of kisspeptin and GPR54 in the control of reproduc-tion in general, and in particular of the female gonadotrop-ic axis. While these data have been gathered in mammals(from humans to laboratory rodents), some initial evidenc-es allow us to speculate that GPR54 and its ligand(s) mightserve similar functions in non-mammalian species. Consid-ering the paradigmatic example of GnRH, and how identi-fication of its variants in avian and fish species wasessential for our proper understanding of the evolutionand functions of this family of peptides, it is anticipatedthat the KiSS-1/GPR54 system will be the subject of inten-sive investigation by comparative endocrinologists in theyears to come.

Acknowledgments

The authors are indebted with C. Dieguez, L. Pinilla, E.Aguilar, and other members of their research team, forcontinuous collaboration in studies on neuroendocrine as-pects of kisspeptin physiology. The experimental workfrom the authors’ laboratory summarized in this reviewhas been supported by Grants BFI 2002-00176 and BFU2005-07446 from Ministerio de Educacion y Ciencia,Spain, funds from Instituto de Salud Carlos III (ProjectPI042082 and CIBER Physiopathology of Obesity andNutrition; Ministerio de Sanidad, Spain), and EU researchcontract EDEN QLK4-CT-2002-00603.

References

Castellano, J.M., Gaytan, M., Roa, J., Vigo, E., Navarro, V.M., Bellido,C., Dieguez, C., Aguilar, E., Sanchez-Criado, J.E., Pellicer, A., Pinilla,L., Gaytan, F., Tena-Sempere, M., 2006. Expression of KiSS-1 in ratovary: putative local regulator of ovulation? Endocrinology 147,4852–4862.

Cho, H.J., Moon, J.S., Yang, J.S., Kim, J.I., Seong, J.Y., 2006. Molecularcloning of the bullfrog metastin receptor GPR54 (bfGPR54) and afunctional analysis of bfGPR54 in comparison with rat GPR54. 23rdCECE, p. 099.

de Roux, N., Genin, E., Carel, J.C., Matsuda, F., Chaussain, J.L.,Milgrom, E., 2003. Hypogonadotropic hypogonadism due to loss offunction of the KiSS1-derived peptide receptor GPR54. Proc. Natl.Acad. Sci. USA 100, 10972–10976.

Fink, G., 2000. Neuroendocrine regulation of pituitary function: generalprinciples. In: Conn, P.M., Freeman, M.E. (Eds.), Neuroendocrinol-ogy in Physiology and Medicine. Humana Press, Totowa, New Jersey,pp. 107–134.

Horikoshi, Y., Matsumoto, H., Takatsu, Y., Ohtaki, T., Kitada, C.,Usuki, S., Fujino, M., 2003. Dramatic elevation of plasma metastinconcentrations in human pregnancy: metastin as a novel placenta-derived hormone in humans. J. Clin. Endocrinol. Metab. 88, 914–919.

Iovane, A., Aumas, C., de Roux, N., 2004. New insights in the genetics ofisolated hypogonadotropic hypogonadism. Eur. J. Endocrinol. 151,U83–U88.

Kauffman, A.S., Gottsch, M.L., Roa, J., Byquist, A.C., Crown, A.,Clifton, D.K., Hoffman, G.E., Steiner, R.A., Tena-Sempere, M., 2007.Sexual differentiation of Kiss1 gene expression in the brain of the rat.Endocrinology. doi:10.1210/en.2006-1540.

Kinoshita, M., Tsukamura, H., Adachi, S., Matsui, H., Uenoyama, Y.,Iwata, K., Yamada, S., Inoue, K., Ohtaki, T., Matsumoto, H., Maeda,K., 2005. Involvement of central metastin in the regulation ofpreovulatory luteinizing hormone surge and estrous cyclicity in femalerats. Endocrinology 146, 4431–4436.

Kotani, M., Detheux, M., Vandenbogaerde, A., Communi, D., Vanderw-inden, J.M., Le Poul, E., Brezillon, S., Tyldesley, R., Suarez-Huerta,N., Vandeput, F., Blanpain, C., Schiffmann, S.N., Vassart, G.,Parmentier, M., 2001. The metastasis suppressor gene KiSS-1 encodeskisspeptins, the natural ligands of the orphan G protein-coupledreceptor GPR54. J. Biol. Chem. 276, 34631–34636.

Lee, J.H., Miele, M.E., Hicks, D.J., Phillips, K.K., Trent, J.M., Weissman,B.E., Welch, D.R., 1996. KiSS-1, a novel human malignant melanomametastasis-suppressor gene. J. Natl. Cancer Inst. 88, 1731–1737.

Lee, D.K., Nguyen, T., O’Neill, G.P., Chang, R., Liu, Y., Howard, A.D.,Coulombe, N., Tan, C.P., Tang-Nguyen, A.T., George, S.R., O’Dowd,B.F., 1999. Discovery of a receptor related to the galanin receptors.FEBS Lett. 446, 103–107.

Levine, J.E., Chappell, P.E., Schneider, J.S., Sleiter, N.C., Szabo, M.,2001. Progesterone receptors as neuroendocrine integrators. Front.Neuroendocrinol. 22, 69–106.

Linkie, D.M., Niswender, G.D., 1972. Serum levels of prolactin, lutein-izing hormone and follicle-stimulating hormone during pregnancy inthe rat. Endocrinology 90, 632–637.

Muir, A.I., Chamberlain, L., Elshourbagy, N.A., Michalovich, D.,Moore, D.J., Calamari, A., Szekeres, P.G., Sarau, H.M., Chambers,J.K., Murdock, P., Steplewski, K., Shabon, U., Miller, J.E., Middle-ton, S.E., Darker, J.G., Larminie, C.G.C., Wilson, S., Bergsma, D.J.,Emson, P., Faull, R., Philpott, K.L., Harrison, D.C., 2001. AXOR12,a novel human G protein-coupled receptor, activated by the peptideKiSS-1. J. Biol. Chem. 276, 28969–28975.

Navarro, V.M., Castellano, J.M., Fernandez-Fernandez, R., Barreiro,M.L., Roa, J., Sanchez-Criado, J.E., Aguilar, E., Dieguez, C., Pinilla,L., Tena-Sempere, M., 2004. Developmental and hormonally regulatedmessenger ribonucleic acid expression of KiSS-1 and its putativereceptor, GPR54, in rat hypothalamus and potent luteinizinghormone-releasing activity of KiSS-1 peptide. Endocrinology 145,4565–4574.

Navarro, V.M., Castellano, J.M., Fernandez-Fernandez, R., Tovar, S.,Roa, J., Mayen, A., Nogueiras, R., Vazquez, M.J., Barreiro, M.L.,Magni, P., Aguilar, E., Dieguez, C., Pinilla, L., Tena-Sempere, M.,2005a. Characterization of the potent LH releasing activity of KiSS-1peptide, the natural ligand of GPR54. Endocrinology 146, 156–163.

Navarro, V.M., Castellano, J.M., Fernandez-Fernandez, R., Tovar, S.,Roa, J., Mayen, A., Barreiro, M.L., Casanueva, F.F., Aguilar, E.,Dieguez, C., Pinilla, L., Tena-Sempere, M., 2005b. Effects of KiSS-1peptide, the natural ligand of GPR54, on follicle-stimulating hormonesecretion in the rat. Endocrinology 146, 1689–1697.

Nocillado, J.N., Levavi-Sivan, B., Carrick, F., Elizur, A., 2007. Temporalexpression of G-protein-coupled receptor 54 (GPR54), gonadotropin-releasing hormones (GnRH), and dopamine receptor D2 (drd2) inpubertal female grey mullet, Mugil cephalus. Gen. Comp. Endocrinol.150, 278–287.

140 J. Roa, M. Tena-Sempere / General and Comparative Endocrinology 153 (2007) 132–140

Ohtaki, T., Shintani, Y., Honda, S., Matsumoto, H., Hori, A., Kanehashi,K., Terao, Y., Kumano, S., Takatsu, Y., Masuda, Y., Ishibashi, Y.,Watanabe, T., Asada, M., Yamada, T., Suenaga, M., Fujino, C.,Usuki, S., Kurokawa, T., Onda, H., Nishimura, O., Fujino, M., 2001.Metastasis suppressor gene KiSS-1 encodes peptide ligand of a G-protein-coupled receptor. Nature 411, 613–617.

Parhar, I.S., Ogawa, S., Sakuma, Y., 2004. Laser captured singledigoxigenin-labeled neurons of gonadotropin-releasing hormone typesreveal a novel g protein-coupled receptor (GPR54) during maturationin cichlid fish. Endocrinology 145, 3613–3618.

Roa, J., Vigo, E., Castellano, J.M., Navarro, V.M., Fernandez-Fernandez,R., Casanueva, F.F., Dieguez, C., Aguilar, E., Pinilla, L., Tena-Sempere, M., 2006. Hypothalamic expression of KiSS-1 system andgonadotropin-releasing effects of kisspeptin in different reproductivestates of the female rat. Endocrinology 147, 2864–2878.

Schwartz, N.B., 2000. Neuroendocrine regulation of reproductive cyclic-ity. In: Conn, P.M., Freeman, M.E. (Eds.), Neuroendocrinology inPhysiology and Medicine. Humana Press, Totowa, New Jersey, pp.135–146.

Seminara, S.B., 2005. Metastin and its G protein-coupled receptor,GPR54: critical pathway modulating GnRH secretion. Front. Neuro-endocrinol. 26, 131–138.

Seminara, S.B., 2006. Mechanisms of disease: the first kiss-a crucial rolefor kisspeptin-1 and its receptor, G-protein-coupled receptor 54, inpuberty and reproduction. Nat. Clin. Pract. Endocrinol. Metab. 2,328–334.

Seminara, S.B., Messager, S., Chatzidaki, E.E., Thresher, R.R., Acierno,J.S., Shagoury, J.K., Bo-Abbas, Y., Kuohung, W., Schwinof, K.M.,

Hendrick, A.G., Zahn, D., Dixon, J., Kaiser, U.B., Slaugenhaupt,S.A., Gusella, J.F., O’Rahilly, S., Carlton, M.B., Crowley, W.F.,Aparicio, S.A., Colledge, W.H., 2003. The GPR54 gene as a regulatorof puberty. N. Engl. J. Med. 349, 1614–1627.

Smith, M.S., Grove, K.L., 2002. Integration of the regulation ofreproductive function and energy balance: lactation as a model.Front. Neuroendocrinol. 23, 225–256.

Smith, J.T., Dungan, H.M., Stoll, E.A., Gottsch, M.L., Braun, R.E.,Eacker, S.M., Clifton, D.K., Steiner, R.A., 2005a. Differential regu-lation of KiSS-1 mRNA expression by sex steroids in the brain of themale mouse. Endocrinology 146, 2976–2984.

Smith, J.T., Cunningham, M.J., Rissman, E.F., Clifton, D.K., Steiner,R.A., 2005b. Regulation of Kiss1 gene expression in the brain of thefemale mouse. Endocrinology 146, 3686–3692.

Smith, J.T., Popa, S.M., Clifton, D.K., Hoffman, G.E., Steiner, R.A.,2006. Kiss1 neurons in the forebrain as central processors forgenerating the preovulatory luteinizing hormone surge. J. Neurosci.26, 6687–6694.

Tena-Sempere, M., 2006. GPR54 and kisspeptin in reproduction. Hum.Reprod. Update 12, 631–639.

Tena-Sempere, M., Huhtaniemi, I., 2003. Gonadotropins and gonadotro-pin receptors. In: Fauser, B.C.J.M. (Ed.), Reproductive Medicine –Molecular, Cellular and Genetic Fundamentals. Parthenon Publishing,New York, pp. 225–244.

Terao, Y., Kumano, S., Takatsu, Y., Hattori, M., Nishimura, A., Ohtaki,T., Shintani, Y., 2004. Expression of KiSS-1, a metastasis suppressorgene, in trophoblast giant cells of the rat placenta. Biochim. Biophys.Acta 1678, 102–110.