GERMLINE PROKINETICIN RECEPTOR 2 (PROKR2)VARIANTS ASSOCIATED WITH CENTRALHYPOGONADISM CAUSE DIFFERENTAL MODULATIONOF DISTINCT INTRACELLULAR PATHWAYS

Domenico Vladimiro Libri1, Gunnar Kleinau2, Valeria Vezzoli3, Marta Busnelli4,Fabiana Guizzardi1, Antonio Agostino Sinisi5, Angela Ida Pincelli6,Antonio Mancini7, Gianni Russo8, Paolo Beck-Peccoz3,9, Sandro Loche10,Claudio Crivellaro11, Mohamad Maghnie12, Csilla Krausz13, Luca Persani1,3, andMarco Bonomi1

on behalf of the Italian study group on Idiopathic Central Hypogonadism (ICH)*

1Unità di Medicina Generale ad indirizzo Endocrino-Metabolico e Laboratorio di Ricerche Endocrino-Metaboliche, Istituto Auxologico Italiano, Milano, Italy; 2 Institute of Experimental PediatricEndocrinology, Charité-Universitätsmedizin, Berlin, Germany; 3 Dipartimento di Scienze Cliniche e diComunità, Università degli Studi di Milano, Milano, Italy; 4 Consiglio Nazionale delle Ricerche, Istituto diNeuroscienze, Milano, Italy; 5 Dipartimento di Scienze Cardiotoraciche e Respiratorie, Seconda Universitàdi Napoli, Napoli, Italy; 6 Clinica Medica, AO “San Gerardo dei Tintori”, Università Milano-Bicocca,Monza, Italy; 7 Divisione di Endocrinologia, Dipartimento di Medicina Interna, Università del Sacro Cuore,Roma, Italy; 8 Dipartimento di Pediatria, Ospedale San Raffaele, Milano, Italy; 9 Unità di Endocrinologia eDiabetologia, Fondazione Policlinico “Cà Granda”, Milano, Italy; 10 Servizio di Endocrinologia Pediatrica,Ospedale Microcitemico, ASL Cagliari, Cagliari, Italy; 11 Ospedale Regionale di Bolzano, Clinica Medica,Bolzano, Italy; 12Dipartimento di Pediatria, G. Gaslini, Università di Genova, Genova, Italy; 13

Dipartimento di Fisiopatologia Clinica, Unità di Andrologia, Università di Firenze, Firenze, Italy

Introduction: Defects of prokineticin pathway affect the neuroendocrine control of reproduction,but their role in the pathogenesis of central hypogonadism remains undefined and the functionalimpact of missense PROKR2 variants has been incompletely characterized. Material and Methods:In a series of 246 idiopathic central hypogonadism patients we found three novel (p.V158I,p.V334M, p.N15TfsX30) and six already known (p.L173R, p.T260M, p.R268C, p.V274D, p.V331M,p.H20MfsX23) germline variants in PROKR2 gene. We evaluated effects of seven missense alter-ations on two different PROKR2-dependent pathways: IP-Ca2� (Gq-coupling) and cAMP (Gs-cou-pling). Results: PROKR2 variants were found in 16 patients (6.5%). Expression levels of variantsp.V158I, p.V331M were moderately reduced whereas they were markedly impaired in the remain-ing cases, except p.V334M that was significantly overexpressed. The variants p.T260M, p.R268C andp.V331M showed no remarkable changes in cAMP response (EC50) while the IP signalling appearedmore profoundly affected. In contrast, cAMP accumulation cannot be stimulated through thep.L173R and p.V274D, but IP EC50 was similar to wt inp.L173R and increased by 10-fold in p.V274D.The variant p.V334M lead to a 3-fold increase of EC50 for both cAMP and IP. Conclusion: Our studyshows that single PROKR2 missense allelic variants can either affect both signaling pathwaysdifferently or selectively. Thus, the integrity of both PROKR2-dependent cAMP and IP signalsshould be evaluated for a complete functional testing of novel identified allelic variants.

Research on the prokineticin system has revealed itsinvolvement in several physiological mechanisms

and pathological conditions. The knockout models for

both ligand (Prok2) and the G protein-coupled receptor(GPCR), Prokr2, revealed a role in olfactory bulb mor-phogenesis and sexual maturation, indicating PROK2

ISSN Print 0021-972X ISSN Online 1945-7197Printed in U.S.A.Copyright © 2013 by The Endocrine SocietyReceived June 5, 2013. Accepted November 13, 2013.

Abbreviations: PROK2, prokineticin 2; PROKR2, prokineticin receptor 2; GnRH, Gonado-tropin Releasing Hormone; GPCRs, G-protein coupled receptors; mAb, monoclonal anti-body; CHg, Central Hypogonadism; KS, Kallmann’s syndrome; nIHH, normosmic isolated

doi: 10.1210/jc.2013-2431 J Clin Endocrinol Metab jcem.endojournals.org 1

J Clin Endocrin Metab. First published ahead of print November 25, 2013 as doi:10.1210/jc.2013-2431

Copyright (C) 2013 by The Endocrine Society

and PROKR2 as strong candidate genes for human GnRHdeficiency (1, 2). The involvement of PROKR2 in thepathogenesis of Hypogonadotropic Hypogonadism (HH)was first described in 2006 (3), and subsequently reportedin several other patients series (4). Idiopathic HH is a raredisease with a complex pathogenesis but a strong geneticcomponent. It is characterized by delayed or absent pu-berty secondary to gonadotropin deficiency and associ-ated or not with olfactory defects in the Kallmann’s syn-drome (KS) or in the normosmic Isolated HH (nIHH),respectively (4). Very recently, PROKR2 variants havebeen also described in patients with idiopathic combinedpituitary hormone deficits (CPHDs), including gonado-tropin deficiency (5–7). Importantly, all previous in vitrostudies on the identified allelic variants were showing vari-able impairments of PROKR2 function (5, 6, 8–11) eitherinvolving the expression level of the variant receptors ortheir ability to stimulate the Gq-dependent pathway. Nodirect data are so far available about the effect of thesevariants on the Gs-dependent cAMP signaling, despite ev-idence of its activation by nM concentrations of PROK2(12). Since, a variable involvement of diverse transductionpathways by a single genetic variant has been reported forother GPCRs (13), we decided to characterize thePROKR2 genetic variations identified in a large Italianseries of idiopathic CHg patients by evaluating both eithertheir direct expression on living-cell membrane and theirability to activate the two signal transduction pathwayslinked to the PROKR2.

Materials and Methods

PatientsA large cohort of Italian patients with CHg was collected (n �

246, 197 males and 49 females) thanks to the Italian network forIdiopathic CHg supported by the Italian Societies of Endocri-nology, and of Pediatric Endocrinology and Diabetes. The CHgcohort includes KS (41%), nIHH (54%) or idiopathic CPHDs(5%) patients. The institutional Ethic Committee approved thestudy and all patients gave their informed consent for the geneticinvestigations. Patients’ entire ORF of the human PROKR2 genewas amplified by PCR (primers available upon request) and se-quenced according to standard protocols.

Expression and bioassays of PROKR2 variantsPROKR2 wild-type (wt) plasmid (Missouri S&T cDNA Re-

source Center, USA) was used as template to engineer the testedmutants (see Supplemental material). PROKR2 variant expres-sion was checked by FACS, immunocytochemistry and westernblot, while cAMP and IP1 bioassays were performed in transienttransfection on HEK293 or COS7 cells following previously de-

scribed protocols (14) with specific modifications (see Supple-mental material).

StatisticsThe data presented were from representative experiments.

Differences between means were compared using unpaired t-Student’s test and analysis of variance (ANOVA) with Dunnettpost hoc test.

Results

Genetic analysisThe analysis of PROKR2 gene led to the identification

of 16 patients carrying exonic variations out of 246 CHgpatients (6.5%). All the variants were present in theheterozygous state, except in one case (Table 1). Nine dif-ferent variants were identified: seven missense substitu-tions (p.V158I, p.L173R, p.T260M, p.R268C, p.V274D,p.V331M and p.V334M) and two frameshift deletionsgenerating premature stop-codons (p.N15TfsX30;p.H20MfsX23). Six variants had been previously identi-fied (p.L173R; p.T260M, p.R268C, p.V274D, p.V331M,p.H20MfsX23) (15, 16). All these variations affect wellconserved residues, and none of these variations was ob-served in 300 control alleles from the adult general pop-ulation. The analysis of other candidate genes (KAL-1,FGFR1, FGF8, GnRHR, GnRH1, TAC3, TAC3R,KISS1R, CHD7, PROK2, SEMA3A) allowed the identi-fication of a second heterozygous variant in 4/16 patients(Table 1).

Clinical characteristicsThe main clinical features of the 16 patients with

PROKR2 gene variations are reported in Table 1. No sta-tistically significant correlations were found between thepresence/absenceof thegenetic variationsandclinical phe-notypes and/or parameters such as LH/FSH and sex ste-roids levels (data not shown). All patients, but one, expe-rienced a pubertal delay and the reversal of CHg wasdocumented in three. Variable CPHDs were detected inthree cases. Familial history revealed the recurrence ofanosmia in an aunt of KS-1 carrying the wt-PROKR2allele and a pubertal delay in the heterozygous mother ofKS-6. The PROKR2 gene investigations in the familiesfailed to show cosegregation with phenotype, as previ-ously reported (5, 7, 11, 17). In KS-2, nIHH-1, nIHH-7 thephenotype appeared in the family in association with thebiallelic defect.hypogonadotropic hypogonadism; nM, nanomolar; ORF, open reading frame; SPRT, signalpeptide rhodopsin tag; h, human; hr, human recombinant; TMH, transmembrane helix;ICL, intracellular loop; ECL, extracellular loop; ADRB2; beta-2 adrenergic receptor.

2 PROKR2 variants and intracellular pathways J Clin Endocrinol Metab

Functional and molecular studiesThe two frameshift variants, p.N15TfsX30 and

p.H20MfsX23, lead to the generation of a protein lackingall the transmembrane and intracellular receptor domains,thus justifying the classification as mutations with a severeloss-of-function (LOF).

All missense mutants showed an impaired targeting onthe cell surface, exception made for p.V334M which ismore efficiently expressed on the HEK293 membranecompared to the wt (Figure 1A,E and Supplemental Figure1). Both FACS analysis after cell permeabilization andwestern blotting showed a similar levels of receptor pro-tein expression (Supplemental Figures 2A, 2B), consistentwith reproducible transfection efficiencies and stability oftranslated proteins.

Two different bioassays were performed. Basal IP1 lev-els were similar for all PROKR2 variants. Differently, con-centration-effect curves showed a variable right-shiftmodification of the EC50 in four tested variants (p.R268C,p.V274D, p.V331M, p.V334M), while p.T260M had aborderline shift of the response curve (Figure 1B,E) and anormal response was observed for the remaining two vari-ants (p.V158I, p.L173R). The testing of the Gs-cAMPpathway did not reveal difference at baseline in any case,but showed normal or statistically significant increase ofthe EC50 for two (p.V158I and p.V331M) or three vari-ants (p.T260M, p.R268C, p.V334M), respectively. Theremaining two variants, p.L173R and p.V274D displayedthe complete lack of intracellular cAMP response, dem-onstrating a severe derangement of the Gs coupling (Fig-ure 1C,E).

Since the variants behave differently in the three bio-logical parameters examined, we considered the genera-tion of a functional score integrating all these results. Thisscore generates the ranking of PROKR2 variants (Supple-mental Figure 3), with p.V274D, p.T260M, p.R268C andp.L173R showing a complete or severe LOF; the remain-ing missense mutants (p.V331M, p.V334M and p.V158I)conserve 50%–70% of wild-type activities.

Discussion

We report the frequency of PROKR2 gene variants inabout 6.5% of CHg patients belonging to the largest Ital-ian cohort so far investigated. This frequency appearscomparable to that of the French population (3), butslightly higher in comparison with the US population (9).In order to understand their pathogenic role, we exten-sively characterized all the missense variants by testing thedual activation of Gq- and Gs-dependent pathways and

we identified variations with a predominant impact on oneor the other downstream signal.

At variance with previous studies (8, 9), we used a directmethod to test the expression of the mutated constructs atthe cell surface. A hampered membrane targeting was de-tected in all cases, except the variant p.V334M that ex-hibited an enhanced level of expression on the cell mem-brane. We therefore verified either the capacity for Gq orthe Gs signal transduction pathway of wt and mutants(Figure 1B,C,E). While the ability to stimulate the Gqpathway was affected in most of the cases, this functionwas well conserved for the p.V158I. The mutantsp.R268C and p.V274D are both strongly lacking the abil-ity to activate the Gq protein, while no activation of cAMPaccumulation was detected in cells expressing thep.L173R and p.V274D constructs. Importantly, the se-vere LOF of the p.V274D mutant on both of the trans-duction pathways cannot be explained only on the basis ofa poor membrane targeting because similar low surfaceexpression levels were seen for both p.V274D andp.L173R mutants, but p.L173R was still able to normallyactivate the Gq-dependent pathway. The functional eval-uation of the p.L173R mutant represents another cleardemonstration of the importance to evaluate the dualPROKR2 signaling. While the membrane expression andthe mechanisms leading to the Gq activation appear onlypartially compromised, the complete loss of the ability tostimulate the Gs-dependent pathway amplify the patho-genic impact of this particular variant. In addition, struc-tural receptor models (Figure 1D) suggest two main mes-sages on the molecular level for the tested substitutions(see the figure legend for details): 1) Mutations likep.V274D (TMH6) or p.V334M and p.L173R modify theintrinsic signaling capacity of PROKR2 by modifying spe-cific biophysical properties in structural parts participat-ing significantly in receptor activation (eg, TMH6 andTMH7 interfaces). 2) Especially amino acids T260, R268selectively influence Gq coupling and they are located atthe intracellular transitions between ICL3 and TMH5/TMH6, which is a potential direct G-protein contact re-gion. In light of our findings, the pathogenic role of mu-tations found in CHg patients can be misclassified if asingle pathway is analyzed, since the dual signal ability ofPROKR2 appears to be differentially affected by the vari-ations detected in this large CHg cohort. The extensivefunctional analysis here reported (FACS, IP1 and cAMPassays) were therefore integrated in a formula that gener-ates a score giving a comprehensive characterization ofeach variant (see Supplemental Figure 3). Then, thep.V158I receptor represents a rather benign variation, in-stead p.L173R and p.V331M generate a severe damage ofonly one intracellular pathway (either Gq or Gs), while

doi: 10.1210/jc.2013-2431 jcem.endojournals.org 3

Figure 1. A) Flow citofluorometry analysis, FACS, of intact HEK293 cells transiently transfected with wt PROKR2 or the identified variants. Resultsare expressed as fold of wt. The Prism computer program (GraphPad Software, Inc.) was used for statistical analysis: Student’s t-Test in comparisonwith the wt: ***, P � .0001; **, P � .001; Analysis of variance (ANOVA) with Dunnett post hoc test: °°, P � .01; °, P � .05. The figure illustratesthe integrated results (mean�SE) of 3 independent experiments; B-C) Concentration-effect curves for the PROKR2 variants under stimulation ofincreasing concentrations of hrPROK2. HEK293 cells transiently transfected with the various constructs were stimulated by the concentration ofhrPROK2 indicated, and intracellular IP1 (B) or cAMP (C) values were determined. The results are expressed as fold of wt. The Prism computerprogram (GraphPad Software, Inc.) was used for curve fitting and for EC50 determination. The figure shows the integrated results (mean�SE) of 6independent experiments; D) Homology models of the inactive (boxed window) and active PROKR2 conformations. The receptors (helices andloops) are visualized by cartoon representation (white) and the three subunits of Gs are shown with different colors. On these structural PROKR2models, wt positions (sticks) of investigated single side-chain substitutions are highlighted. The colors encode specific functional effects of thesevariants: green � Gs and Gq signaling were both decreased, cyan � slightly effected expression level, magenta – selectively impaired Gq signaling.During receptor activation - transition between inactive and active signaling state - the transmembrane helices (TMHs) 5, 6 and 7 are predicted toundergo significant structural rearrangements (indicated by red arrows) with participation of specific amino acids. Helix 5 is extended in the activeconformation compared to the inactive state and TMH6 is shorter at the intracellular site. The here described mutations are located at differentspatial receptor regions: I) Variant p.V158I (slightly decreased cell-surface expression level) is located at the junction between the transmembranehelix 3 (TMH3) and intra-cellular loop 2 (ICL2) and sticks into the cytoplasm without interor intramolecular contacts. II) The side-chain ofsubstitution p.L173R (positively charged) in TMH4 points towards TMH3 (to A148). This mutant leads to decreased levels of cell-surface expression,decreased maximal IP and impaired cAMP signaling. This supports that the arginine substitution (large, branched, positively charged) is nottolerated in a hydrophobic cage at TMH3 instead of a large and branched hydrophobic wt leucine side-chain. III) The selectively inactive (Gq)p.T260M variant is located in the ICL3, but is part of the junction between TMH5 and ICL3 in the active state conformation with an extended helix

4 PROKR2 variants and intracellular pathways J Clin Endocrinol Metab

both of the transduction pathways were simultaneouslyaffected to variable degrees in the remaining cases, beingp.V274D the variant provided with an almost completeLOF.

Since the inactive p.V274D variant was found in thehomozygous state in KS-6 case with spontaneous reversalof GnRH neuron function (18), our data indicate that asevere PROKR2 LOF can only cause a partial impairmentof GnRH neuron function, which may be overcome by theplasticity of the neuronal network. In addition, the caseCPHD-3 is carrying a mutation predicted to generate atruncated protein lacking all the transmembrane domains,but came to the physician attention because of typical hy-pogonadal manifestations at 58 years. He reported to havespontaneously fathered 2 daughters, and transmitted thesame heterozygous PROKR2 mutation to one of them,who spontaneously conceived and very recently deliveredone baby. These data are consistent with a report describ-ing the complex inheritance with highly variable pen-etrance among carriers of the ancient founder mutationp.L173R (17). The whole of these data indicate thatPROKR2 variants may indeed be found as rare polymor-phisms in the general population and that the presence ofPROKR2 variations can partially weaken (likely to vari-able extents, depending upon the severity of the functionalalteration) the GnRH function and predispose to CHg andreproductive defects, but other genetic or acquired defectsmay be required to obtain the full blown IHH phenotype.Consistently, the digenic defects were found here only inKS or nIHH patients.

In conclusion, routine assessment of the impact ofPROKR2 variants on both the cAMP and IP pathways iswarranted to avoid absolving mutations that selectivelycompromise one signaling branch while leaving the otherintact, and thereby to avoid erroneously predicting a re-duced risk of disease in the variant-carrying progeny ofsuch patients.

Acknowledgments

The Authors acknowledge the contribution of all participants tothe Italian ICH Study Group belonging to the Italian Societies of

Endocrinology and of Pediatric Endocrinology and Diabetes:Gianluca Aimaretti (Novara), Monica Altobelli (Bergamo),Giorgio Arnaldi (Ancona), Maurizia Baldi (Genoa), Luigi Bar-talena (Varese), Luciano Beccaria (Lecco), Giuseppe Bellastella(Naples), Maria Bellizzi (Trento); Gianni Bona (Novara); Gior-gio Borretta (Cuneo), Fabio Buzi (Brescia), Salvatore Cannavò(Messina), Marco Cappa (Rome), Anna Cariboni (Milan), Tom-maso Ciampani (Varese), Alessandro Cicognani (Bologna),Mariangela Cisternino (Pavia), Sabrina Corbetta (Milan), Ni-cola Corciulo (Lecce), Giovanni Corona (Bologna), RenatoCozzi (Milan), Angela Valentina D’Elia (Udine), Ettore DegliUberti (Ferrara), Mario De Marchi (Turin), Gianni Forti (Flor-ence), Natascia di Iorgi (Genoa), Andrea Isidori (Rome), AndreaFabbri (Rome), Alberto Ferlin (Padua), Carlo Foresta (Padua),Roberto Franceschi (Trento), Andrea Garolla (Padua), RossellaGaudino (Verona), Vito Giagulli (Bari), Enrico Grosso (Turin),Emmanuele Jannini (L’Aquila), Fabio Lanfranco (Turin), LidiaLarizza (Milan), Andrea Lenzi (Rome), Francesco Lombardo(Rome), Paolo Limone (Turin), Mario Maggi (Florence), Ro-berto Maggi (Milan), Maria Cristina Maggio (Palermo), GiorgiaMandrile (Turin), Marco Marino (Modena), Maria AntoniettaMencarelli (Siena), Nicola Migone (Turin), Giovanni Neri(Rome), Lucia Perroni (Genova), Elisa Pignatti (Modena), AlbaPilotta (Brescia), Alessandro Pizzocaro (Milan), Alfredo Pon-tecorvi (Rome), Gabriella Pozzobon (Milan), Flavia Prodam(Novara); Giorgio Radetti (Bolzano), Paola Razzore (Turin),Maria Carolina Salerno (Naples), Alessandro Salvatoni (Var-ese), Filippo Salvini (Milan), Andrea Secco (Genoa), Maria Segni(Rome), Manuela Simoni (Modena), Riccardo Vigneri (Cata-nia), Giovanna Weber (Milan).

Address all correspondence and requests for reprints to: LucaPersani, MD PhD, UO di Medicina Generale ad Indirizzo En-docrino-Metabolico, Piazzale Brescia 20, 20149 Milano, Phone:�39–02619112738, Fax: �39–02619112777, email:luca.persani@unimi.it.

* Members of the Italian ICH study group are listed in theacknowledgment.

Disclosure Summary: All the Authors have nothing to discloserelated to this manuscript.

This work was supported by Fundings: This work was sup-ported by funds for Young Investigators from the Italian Min-istry of Health (Grant No. GR2008–1137632) and IRCCS Is-tituto Auxologico Italiano (Ricerca Corrente Funds: 05C701).

Legend to Figure 1 Continued. . .5 that is of importance for receptor/G-protein contact and activation. IV) The mutation p.R268C is located at the transition between TMH6-ICL3and, therefore, might have a more direct impact on the receptor/Gq contact. V) According to the functional characterization of the p.V274Dsubstitution, V274 is a key-player for signal transduction. The drastic change from hydrophobic to hydrophilic side-chain properties likely impedesthe helix-movement between TMH5 and TMH6 that is a general prerequisite for G-protein activation. VI) Mutants p.V331M and p.V334M werecharacterized to inhibit Gq signaling and they are located at the TMH7, close to the intracellular site. In the inactive state side-chains of V331 andV334 point towards the membrane, whereby V334 is predicted to rotate slightly towards TMH1 during receptor activation (Gs). It cannot beexcluded, that they might participate in intermolecular contacts, as previously suggested (19). Further experiments based on the cotransfection ofwt and variant PROKR2 will be worthy to confirm such possibility. Structure images were produced using PyMOL software (The PyMOL MolecularGraphics System, Version 1.3 Schrödinger, LLC); E) Summarized data indicating the functional results of the PROKR2 allelic variants. Data areexpressed as mean�SE of 3 FACS or 6 independent signaling experiments, respectively (***P � .0001 vs WT; ** P � .001 vs WT; *P � .01 vs WTusing Student’s t-Test; °P � .05 vs WT, °°P � .01 vs WT using ANOVA with Dunnett post hoc test.

doi: 10.1210/jc.2013-2431 jcem.endojournals.org 5

References

1. Ng KL, Li JD, Cheng MY, Leslie FM, Lee AG, Zhou QY. Depen-dence of olfactory bulb neurogenesis on prokineticin 2 signaling.Science. 2005;308:1923–1927.

2. Matsumoto S, Yamazaki C, Masumoto KH, Nagano M, Naito M,Soga T, Hiyama H, Matsumoto M, Takasaki J, Kamohara M, Mat-suo A, Ishii H, Kobori M, Katoh M, Matsushime H, Furuichi K,Shigeyoshi Y. Abnormal development of the olfactory bulb and re-productive system in mice lacking prokineticin receptor PKR2. ProcNatl Acad Sci U S A. 2006;103:4140–4145.

3. Dode C, Teixeira L, Levilliers J, Fouveaut C, Bouchard P, KottlerML, Lespinasse J, Lienhardt-Roussie A, Mathieu M, Moerman A,Morgan G, Murat A, Toublanc JE, Wolczynski S, Delpech M, PetitC, Young J, Hardelin JP. Kallmann syndrome: mutations in thegenes encoding prokineticin-2 and prokineticin receptor-2. PLoSGenet. 2006;2:e175.

4. Bonomi M, Libri DV, Guizzardi F, Guarducci E, Maiolo E, PignattiE, Asci R, Persani L. New understandings of the genetic basis ofisolated idiopathic central hypogonadism. Asian J Androl. 2012;14:49–56.

5. Reynaud R, Jayakody SA, Monnier C, Saveanu A, Bouligand J,Guedj AM, Simonin G, Lecomte P, Barlier A, Rondard P, Martinez-Barbera JP, Guiochon-Mantel A, Brue T. PROKR2 variants in mul-tiple hypopituitarism with pituitary stalk interruption. J Clin En-docrinol Metab. 2012;97:E1068–E1073.

6. Raivio T, Avbelj M, McCabe MJ, Romero CJ, Dwyer AA, Tom-miska J, Sykiotis GP, Gregory LC, Diaczok D, Tziaferi V, EltingMW, Padidela R, Plummer L, Martin C, Feng B, Zhang C, Zhou QY,Chen H, Mohammadi M, Quinton R, Sidis Y, Radovick S, DattaniMT, Pitteloud N. Genetic overlap in Kallmann syndrome, combinedpituitary hormone deficiency, and septo-optic dysplasia. J Clin En-docrinol Metab. 2012;97:E694–E699.

7. McCabe MJ, Gaston-Massuet C, Gregory LC, Alatzoglou KS, Tzi-aferi V, Sbai O, Rondard P, Masumoto KH, Nagano M, ShigeyoshiY, Pfeifer M, Hulse T, Buchanan CR, Pitteloud N, Martinez-Bar-bera JP, Dattani MT. Variations in PROKR2, But Not PROK2, AreAssociated With Hypopituitarism and Septo-optic Dysplasia. J ClinEndocrinol Metab. 2013

8. Monnier C, Dode C, Fabre L, Teixeira L, Labesse G, Pin JP, HardelinJP, Rondard P. PROKR2 missense mutations associated with Kall-mann syndrome impair receptor signalling activity. Hum MolGenet. 2009;18:75–81.

9. Cole LW, Sidis Y, Zhang C, Quinton R, Plummer L, Pignatelli D,Hughes VA, Dwyer AA, Raivio T, Hayes FJ, Seminara SB, Huot C,Alos N, Speiser P, Takeshita A, Van VG, Pearce S, Crowley WF, Jr.,Zhou QY, Pitteloud N. Mutations in prokineticin 2 and prokineticinreceptor 2 genes in human gonadotrophin-releasing hormone defi-

ciency: molecular genetics and clinical spectrum. J Clin EndocrinolMetab. 2008;93:3551–3559.

10. Abreu AP, Noel SD, Xu S, Carroll RS, Latronico AC, Kaiser UB.Evidence of the importance of the first intracellular loop of proki-neticin receptor 2 in receptor function. Mol Endocrinol. 2012;26:1417–1427.

11. Caronia LM, Martin C, Welt CK, Sykiotis GP, Quinton R, Tham-bundit A, Avbelj M, Dhruvakumar S, Plummer L, Hughes VA, Semi-nara SB, Boepple PA, Sidis Y, Crowley WF, Jr., Martin KA, Hall JE,Pitteloud N. A genetic basis for functional hypothalamic amenor-rhea. N Engl J Med. 2011;364:215–225.

12. Chen J, Kuei C, Sutton S, Wilson S, Yu J, Kamme F, Mazur C,Lovenberg T, Liu C. Identification and pharmacological character-ization of prokineticin 2 beta as a selective ligand for prokineticinreceptor 1. Mol Pharmacol. 2005;67:2070–2076.

13. Calebiro D, Gelmini G, Cordella D, Bonomi M, Winkler F, Bieber-mann H, de MA, Marelli F, Libri DV, Antonica F, Vigone MC,Cappa M, Mian C, Sartorio A, Beck-Peccoz P, Radetti G, Weber G,Persani L. Frequent TSH receptor genetic alterations with variablesignaling impairment in a large series of children with nonautoim-mune isolated hyperthyrotropinemia. J Clin Endocrinol Metab.2012;97:E156–E160.

14. Bonomi M, Busnelli M, Persani L, Vassart G, Costagliola S. Struc-tural differences in the hinge region of the glycoprotein hormonereceptors: evidence from the sulfated tyrosine residues. Mol Endo-crinol. 2006;20:3351–3363.

15. Dode C, Rondard P. PROK2/PROKR2 Signaling and KallmannSyndrome. Front Endocrinol (Lausanne). 2013;4:19.

16. Sykiotis GP, Plummer L, Hughes VA, Au M, Durrani S, Nayak-Young S, Dwyer AA, Quinton R, Hall JE, Gusella JF, Seminara SB,Crowley WF, Jr., Pitteloud N. Oligogenic basis of isolated gonad-otropin-releasing hormone deficiency. Proc Natl Acad Sci U S A.2010;107:15140–15144.

17. Avbelj SM, Jeanpierre M, Sykiotis GP, Young J, Quinton R, AbreuAP, Plummer L, Au MG, Balasubramanian R, Dwyer AA, Florez JC,Cheetham T, Pearce SH, Purushothaman R, Schinzel A, Pugeat M,Jacobson-Dickman EE, Ten S, Latronico AC, Gusella JF, Dode C,Crowley WF, Jr., Pitteloud N. An ancient founder mutation inPROKR2 impairs human reproduction. Hum Mol Genet. 2012;21:4314–4324.

18. Sinisi AA, Asci R, Bellastella G, Maione L, Esposito D, Elefante A,De BA, Bellastella A, Iolascon A. Homozygous mutation in the pro-kineticin-receptor2 gene (Val274Asp) presenting as reversible Kall-mann syndrome and persistent oligozoospermia: case report. HumReprod. 2008;23:2380–2384.

19. Marsango S, Bonaccorsi di Patti MC, Barra D, Miele R. Evidencethat prokineticin receptor 2 exists as a dimer in vivo. Cell Mol LifeSci. 2011;68:2919–2929.

Table 1. Clinical and genetic parameters of the CHg patients with PROKR2 allelic variants

Patients Sex and age at diagnosiscDNA

mutationProtein

mutation Zygosity Associated gene variants§MRI Olfactory

structureSense of

smell Familial manifesta-tions Phenotypic details

KS-1 M, 14 c.472 G�A p.V158I hetero none Hypoplastic H Anosmia in aunt with wt PROKR2 —

KS-2 M, 15 c.518 T�G p.L173R hetero FGFR1: c.IVS17 � 6 C�T Hypoplastic A — —

KS-3 M, 16 c.518 T�G p.L173R hetero none Hypoplastic A — ---

KS-4 M, 20 c.779 C�T p.T260 M hetero none Absent A — ---

KS-5 M, 18 c.802 C�T p.R268C hetero none Absent A — ---

KS-6 M, 19 c.820 T�A p.V274D homo none Absent A Pubertal delay in hetero mother CHg reversal

nIHH-1 M, 14 c.518 T�G p.L173R hetero FGFR1: p.Q89R Normal N — Partial Empty Sella

nIHH-2 M, 20 c.518 T�G p.L173R hetero none Normal N — —

nIHH-3 M, 18 c.802 C�T p.R268C hetero none Normal N — —

nIHH-4 M, 12 c.1000 G�A p.V334 M hetero PROK2: p.C383FfsX1 Normal N — —

nIHH-5 F, 16 c.1000 G�A p.V334 M hetero none Normal N — ---

nIHH-6 M, 21 c.56delC p.H20MfsX23 hetero none Normal N — CHg reversal

nIHH-7 M, 18 c.43insACTTT p.N15TfsX30 hetero GNRHR: p.Q106R nd N — CHg reversal

CPHD-1 M, 22 c.779 C�T p.T260 M hetero none Normal N — Adult GHD; OB

CPHD-2 M, 37 c.991 G�A p.V331 M hetero none Normal N — CHa;T2DM;OB

CPHD-3 M, 58 c.56delC p. H20MfsX23 hetero none Normal N — Adult onset CHt; T2DM; OB

KS � Kallmann’s syndrome; nIHH � normosmic IHH; CPHD � combined pituitary hormone deficiencies; A � anosmia; H � hyposmia; N �normal; T2DM � Type 2 Diabetes Mellitus; OB � obesity; GHD � GH deficit; CHa � central hypoadrenalism; CHt � central hypothyroidism; §

associated genetic variants found in heterozygosity

6 PROKR2 variants and intracellular pathways J Clin Endocrinol Metab