The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 349:1614-1627 October 23, 2003 Number 17
NextNext

The GPR54 Gene as a Regulator of Puberty
Stephanie B. Seminara, M.D., Sophie Messager, Ph.D., Emmanouella E. Chatzidaki, B.Sc., Rosemary R. Thresher, Ph.D., James S. Acierno, Jr., B.S., Jenna K. Shagoury, B.S., Yousef Bo-Abbas, M.D., Wendy Kuohung, M.D., Kristine M. Schwinof, M.A., Alan G. Hendrick, Ph.D., Dirk Zahn, Ph.D., John Dixon, B.A., Ursula B. Kaiser, M.D., Susan A. Slaugenhaupt, Ph.D., James F. Gusella, Ph.D., Stephen O'Rahilly, M.D., Mark B.L. Carlton, Ph.D., William F. Crowley, Jr., M.D., Samuel A.J.R. Aparicio, B.M., B.Ch., Ph.D., and William H. Colledge, Ph.D.

 

This Article
-Abstract
- PDF
-PDA Full Text
-PowerPoint Slide Set
-Supplementary Material

Commentary
-Perspective
by Beier, D. R.

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited
-E-mail When Letters Appear

More Information
-PubMed Citation
ABSTRACT

Background Puberty, a complex biologic process involving sexual development, accelerated linear growth, and adrenal maturation, is initiated when gonadotropin-releasing hormone begins to be secreted by the hypothalamus. We conducted studies in humans and mice to identify the genetic factors that determine the onset of puberty.

Methods We used complementary genetic approaches in humans and in mice. A consanguineous family with members who lacked pubertal development (idiopathic hypogonadotropic hypogonadism) was examined for mutations in a candidate gene, GPR54, which encodes a G protein–coupled receptor. Functional differences between wild-type and mutant GPR54 were examined in vitro. In parallel, a Gpr54-deficient mouse model was created and phenotyped. Responsiveness to exogenous gonadotropin-releasing hormone was assessed in both the humans and the mice.

Results Affected patients in the index pedigree were homozygous for an L148S mutation in GPR54, and an unrelated proband with idiopathic hypogonadotropic hypogonadism was determined to have two separate mutations, R331X and X399R. The in vitro transfection of COS-7 cells with mutant constructs demonstrated a significantly decreased accumulation of inositol phosphate. The patient carrying the compound heterozygous mutations (R331X and X399R) had attenuated secretion of endogenous gonadotropin-releasing hormone and a left-shifted dose–response curve for gonadotropin-releasing hormone as compared with six patients who had idiopathic hypogonadotropic hypogonadism without GPR54 mutations. The Gpr54–deficient mice had isolated hypogonadotropic hypogonadism (small testes in male mice and a delay in vaginal opening and an absence of follicular maturation in female mice), but they showed responsiveness to both exogenous gonadotropins and gonadotropin-releasing hormone and had normal levels of gonadotropin-releasing hormone in the hypothalamus.

Conclusions Mutations in GPR54, a G protein–coupled receptor gene, cause autosomal recessive idiopathic hypogonadotropic hypogonadism in humans and mice, suggesting that this receptor is essential for normal gonadotropin-releasing hormone physiology and for puberty.

The neuroendocrine and genetic control of sexual maturation at puberty remains one of the great mysteries of human biology. The secretion of gonadotropin-releasing hormone by the hypothalamus represents the first known step in the reproductive cascade — initiating pulsatile release of gonadotropins, gonadal secretion of sex steroids, pubertal development, and gametogenesis. Although the central role of gonadotropin-releasing hormone in the reproductive hierarchy of all mammals is undisputed, little is understood of the genetic factors that modulate its secretion. The identification of such factors is critical for advancing the understanding of normal reproduction and for providing insight into disorders of the pubertal process.

We used complementary genetic approaches in humans and mice to study a gene involved in the onset of puberty. Idiopathic hypogonadotropic hypogonadism in humans is characterized by the absence of spontaneous sexual maturation in the face of concentrations of gonadotropins in the low-normal range. Affected patients have a complete or partial absence of luteinizing hormone pulsations induced by endogenous gonadotropin-releasing hormone1 and have normal responsiveness to physiological replacement with exogenous gonadotropin-releasing hormone. These observations localize the defect to an abnormality of gonadotropin-releasing hormone synthesis, secretion, or activity.2,3,4

Using linkage analysis and candidate-gene screening, we identified mutations in a G protein–coupled receptor gene, GPR54, in a large, consanguineous Saudi Arabian family with idiopathic hypogonadotropic hypogonadism and in one unrelated black male proband in the United States. Through a complementary approach, we generated Gpr54-deficient mice with a phenotype that demonstrated a lack of adult sexual development and low circulating gonadotropin concentrations — features that closely resemble their human counterparts.

Methods

Family History

A large Saudi Arabian family in which there had been three marriages between first cousins sought medical attention for infertility (Figure 1). Six of the 19 offspring (4 men and 2 women) met the standard diagnostic criteria for idiopathic hypogonadotropic hypogonadism (inappropriately low gonadotropin concentrations in the presence of prepubertal concentrations of sex steroids, normal anterior pituitary function, and normal findings on imaging of the brain) and had responsiveness to exogenous, pulsatile gonadotropin-releasing hormone, as previously reported.5 The collection of blood samples for genetic studies and the clinical protocols were approved by the Subcommittee on Human Studies of the Massachusetts General Hospital, and all participants provided written informed consent.


View larger version (13K):
[in this window]
[in a new window]
  		Figure 1. Pedigree of a Family with Idiopathic Hypogonadotropic Hypogonadism.

Squares represent male family members, circles female family members, diamonds multiple family members with the number given within, and solid symbols affected family members; the proband is indicated with an arrow. L denotes leucine, and S serine at position 148 of the GPR54 gene. Adapted with permission from Bo-Abbas et al.5 and Acierno et al.6

 
Mutation Analysis

Linkage to a 1.06-Mb interval on chromosome 19p13.3 was previously demonstrated.6 Mutation analysis of candidate genes, beginning with GPR54, was initiated with the use of DNA extracted from whole blood. The sequence of GPR54 complementary DNA (cDNA) (GenBank accession number AY253981 [GenBank] ) was aligned with the published genomic sequence7 to identify the genomic structure. Details of polymerase-chain-reaction (PCR) amplifications, sequencing, control populations, transient transfections, the targeting construct, and biochemical assays are provided in Supplementary Appendix 1 (available with the full text of this article at http://www.nejm.org).

Additional Populations

To determine whether the observed base-pair changes in GPR54 were normal variants, control populations of 80 North American persons (primarily anonymous blood donors), 50 Middle Eastern persons, and 50 black persons from North America were also screened. An additional 63 patients with normosmic idiopathic hypogonadotropic hypogonadism and 20 patients with anosmic idiopathic hypogonadotropic hypogonadism (Kallmann's syndrome) were also screened for coding-sequence mutations in GPR54. Six patients with normosmic idiopathic hypogonadotropic hypogonadism who had participated in dose–response studies of exogenous, pulsatile gonadotropin-releasing hormone and who were negative for GPR54 mutations on genomic screening were selected for comparisons of genotypes and phenotypes.

Allele-Specific Cloning

To demonstrate that there were base-pair changes on separate alleles, specific PCR products were cloned into a pCRII-TOPO plasmid vector (Invitrogen). Colonies were grown, and their DNA sequenced.

Reverse-Transcriptase PCR

Subjects in whom coding sequence changes were identified in GPR54 were further screened by means of reverse-transcriptase (RT)–PCR to rule out cryptic splicing events. Total RNA was extracted from lymphoblastoid cell lines, and GPR54 cDNA was amplified and sequenced.

Generation of Mutant Constructs

The sequence of the mammalian expression vector pCMVsport 6 containing full-length wild-type GPR54 (clone CS0DE005YC17, Invitrogen) was confirmed by direct sequencing and found to contain a polyA tail. Site-directed mutagenesis was performed to introduce the three mutations (L148S, R331X, and X399R) into this vector. In addition, a stop codon was introduced immediately after the polyA tail (the construct was called "X399R polyA stop").

Studies of GPR54 Signaling

A natural ligand for GPR54, kisspeptin-1 (encoded by the gene KISS1), has been identified by three separate groups.8,9,10 Its C-terminal decapeptide kisspeptin-1 112–1219 was demonstrated to be the minimal-length peptide required for the full stimulation of GPR54 (the Gq class of G proteins, coupled to phospholipase C). Stimulation of GPR54 by kisspeptin-1 112–121 has been shown to increase phosphatidylinositol turnover.10

Kidney (COS-7) cells from African green monkeys were transiently transfected with 1.5 µg of each GPR54 construct or empty vector (pCMVsport6) per well, were stimulated with varying doses of kisspeptin-1 112–121 for 45 minutes, and were subsequently extracted with 20 mM of formic acid. Supernatants were loaded onto anion-exchange columns, and inositol phosphates were extracted.11 Assays were performed in triplicate.

Quantitative RT-PCR

Quantitative RT-PCR was performed on RNA isolated from immortalized lymphoblasts obtained from patients (TaqMan One-Step RT-PCR Master Mix, Applied Biosystems). Different primers and probes capable of specifically amplifying the R331X and X399R alleles were used. Samples were run in quadruplicate in a minimum of two independent experiments. The {beta}-actin gene was used as an endogenous control to standardize the assays in terms of expression levels.

Genotype–Phenotype Correlations

The patient carrying mutations R331X and X399R was admitted to the General Clinical Research Center of the Massachusetts General Hospital. Blood sampling was performed every 10 minutes for 12 hours. The patient then received gonadotropin-releasing hormone subcutaneously every two hours, and his dose was titrated while he was an outpatient until his pituitary–gonadal axis had normalized. After 11 months of treatment, the patient underwent a dose–response study in which four doses of gonadotropin-releasing hormone spanning 1.5 logarithmic orders (7.5 to 250 ng per kilogram of body weight per bolus) were administered intravenously in random order, and luteinizing hormone was sampled frequently.12 Pulsatile hormone secretion was assessed with the use of the modified version of the method of Santen and Bardin.13,14

Studies in Mice

The transgenic mice (Gpr54tm1PTL) were maintained as an inbred stock on a 129S6/SvEv genetic background. The gene-targeting strategy engineered a germ-line deletion of transmembrane loops 1 and 2 and the encompassing domains (Figure 2A). Correct targeting was verified for the 3' and 5' arms by Southern blotting and PCR (Figure 2B). The generation of a null Gpr54 allele was confirmed by RT-PCR (Figure 2C). Genotyping was performed by PCR (see Supplementary Appendix 2, available with the full text of this article at http://www.nejm.org). All experiments were performed under the authority of a U.K. Home Office Project License and were approved by a local ethics panel.


View larger version (47K):
[in this window]
[in a new window]
  		Figure 2. Targeted Deletion in the Gpr54 Locus in Mice.

Panel A shows a schematic representation of the allele of Gpr54 that was targeted for deletion (Panel A). Blue boxes represent exons, and purple boxes resistance cassettes and markers. Key restriction sites, primers, and probes are shown above the loci. Panel B shows the correct targeting of the 5' and 3' arms as demonstrated by Southern blotting. The arrows indicate the mutant and wild-type bands detected by diagnostic restriction digestion and probing. The sizes are given in numbers of base pairs. Southern blots show the expected pattern in heterozygous, wild-type, and mutant mice. In mice that are homozygous for the Gpr54 deletion (Panel C), transcription 3' of the locus has been ablated. RT-PCR analysis of the segment spanning exons 4 and 5 shows that a detectable transcript is absent. IRES denotes internal ribosome entry site, B-Gal beta-galactosidase gene, MC MC promoter, Neo neomycin resistance gene, and Ex exon.

 
            Gonadotropin-Releasing Hormone Injection

Wild-type female mice were staged with the use of vaginal smears. Wild-type female mice at diestrus and Gpr54 –/– female mice received four intraperitoneal injections of 25 ng of gonadotropin-releasing hormone (Sigma) at 30-minute intervals.15 The mice were killed 30 minutes after the last injection. Blood samples and pituitary specimens were treated as previously described,15 except that the pituitary specimens were homogenized in 0.3 ml of phosphate-buffered saline.

            Hormone Assays

The sensitivity of the immunoradiometric assay for luteinizing hormone was 0.07 ng per milliliter (intraassay variation, 6.0 percent; interassay variation, 12.5 percent), and the sensitivity of the radioimmunoassay for follicle-stimulating hormone was 2 ng per milliliter (intraassay variation, 10 percent; interassay variation, 18 percent). Gonadotropin-releasing hormone was measured by radioimmunoassay,16 with a detection limit of 0.2 pg per tube (0.83 pg per milliliter) and an intraassay variation of 13 percent. Testosterone was measured by radioimmunoassay, with a sensitivity of 0.2 nmol per liter (intraassay variation, 6.0 percent; interassay variation, 18 percent). 17{beta}-estradiol was measured by enzyme-linked immunosorbent assay (ELISA), with a sensitivity of 10 pg per milliliter (intraassay variation, 3.9 percent; interassay variation, 10 percent).

            Histologic Studies

Mouse tissues were dissected and fixed for four hours in 4 percent formaldehyde and were then washed three times in 0.01 percent phosphate-buffered saline. Ovaries, testes, and adrenal glands were wax-embedded and sectioned at 3 to 4 µm. Tissue sections were stained with hematoxylin and eosin. Mammary glands were dissected and fixed for 2 to 4 hours at 23°C in fixative (six parts absolute ethanol to three parts chloroform to one part glacial acetic acid), washed in 70 percent ethanol for 15 minutes, rehydrated, and stained overnight in carmine alum stain that has been boiled for 20 minutes in 500 ml of distilled water. Slides were washed in increasing concentrations of ethanol (70 percent, 95 percent, and 100 percent) for 15 minutes each, cleared in xylene for 30 minutes, and mounted.

Results

Mutation Analysis

Linkage in a consanguineous Saudi Arabian family was previously demonstrated on chromosome 19p13.3 with a maximal two-point lod score of 5.17.6 The candidate region on chromosome 19 contained 23 known genes,7 including GPR54, which is expressed in the human brain, pituitary gland, and placenta, as assessed with the use of RT-PCR.9,10 GPR54 has five exons and contains an open reading frame of 1197 bp that encodes a 398-amino-acid protein.

A homozygous single-nucleotide variant (443T>C) in exon 3, which substitutes a serine for the normal leucine at position 148 (L148S) in the second intracellular loop, was found in all six affected persons in the Saudi pedigree and did not occur in a homozygous state in any unaffected family members (all references to base-pair positions are reported according to standard numbering and nomenclature17) (see Supplementary Appendix 3, available with the full text of this article at http://www.nejm.org). This variant does not appear to be a polymorphism, since it occurs only in affected family members; is absent in 160 chromosomes from unrelated, unaffected controls from the United States and 100 chromosomes from controls from the Middle East; is present in an amino acid residue that is conserved among species including mouse, rat, amphioxus, and pufferfish (GenBank accession numbers AF343726 [GenBank] , BAB55447 [GenBank] , and AAM18884 [GenBank] , and Fugu Genome Server18 accession number SINFRUP00000071513,19 respectively); and changes the polarity of the encoded amino acid from hydrophobic to neutral.

Of the 63 unrelated patients with normosmic idiopathic hypogonadotropic hypogonadism and the 20 patients with Kallmann's syndrome, one black man with idiopathic hypogonadotropic hypogonadism was discovered to have a heterozygous C-to-T transition at nucleotide 991 in exon 5, in which an arginine at residue 331 was replaced with a premature stop codon (991C>T [R331X]). In addition, a heterozygous T-to-A transversion was identified at nucleotide 1195 in exon 5, which replaced the stop codon at residue 399 with an arginine (1195T>A [X399R]) (Supplementary Appendix 3). This nonstop mutation results in the continuation of the open reading frame to the polyA signal, with no intervening stop codon. Neither change was identified in the 160 chromosomes from North American controls or the 100 chromosomes from black controls.

To confirm that the nonsense and nonstop mutations are found on separate chromosomes, allele-specific cloning was performed. Seventeen clones contained either R331X or X399R, and no clones contained both, confirming that the two variants occur on separate alleles (making the patient a compound heterozygote) (data not shown).

RT-PCR

RT-PCR products generated from the Saudi Arabian proband and the compound heterozygote were of the expected size for all segments. Sequence analysis of these products revealed that the newly identified mutations did not result in cryptic splicing (data not shown).

Functional Assays

To determine whether the identified changes in GPR54 affect the function of the receptor, inositol phosphate production was measured in COS-7 cells in response to kisspeptin-1 112–121. The maximal inositol phosphate response of the cells that were transfected with the mutant L148S and R331X constructs was decreased by 65 percent and 67 percent, respectively, as compared with the cells that were transfected with the wild-type gene (Figure 3A and Figure 3B). RT-PCR of COS-7 cells transfected with the X399R construct revealed a transcript that contained 3' untranslated region, polyA tail, and expression-vector sequence (data not shown). In the absence of the physiologic stop codon at position 399 (but with a stop codon in the vector sequence), the in vitro transcript resulted in an elongated receptor protein. Because this in vitro construct did not accurately mimic in vivo physiology, it was not used in the functional studies. The X399R polyA stop construct, which makes a protein identical to that encoded by the nonstop transcript, stimulates inositol phosphate production that is 61 percent of that of wild-type GPR54 (Figure 3C). No inositol phosphate stimulation was observed with pCMVsport 6.


View larger version (23K):
[in this window]
[in a new window]
  		Figure 3. Dose–Response Curves for the Ligand-Stimulated Production of Inositol Phosphate in Mutant Constructs, Corrected for Protein Content.

The data points represent the means of multiple replicates, each measured in triplicate. Cotransfection with beta-galactosidase revealed no differences in the efficiency of transfection; quantitative RT-PCR and Western blotting revealed equivalent transcript and protein expression (data not shown). Panel A shows the curve for the L148S mutation (three independent experiments, each performed in triplicate), Panel B the curve for the R331X mutation (two independent experiments, each performed in triplicate), and Panel C the curve for the X399R polyA stop mutation (two independent experiments, each performed in triplicate); the percentages on the y axis represent the percentages of the maximal stimulation for each GPR54 construct. Panel D shows the relative quantification of the wild-type and mutant GPR54 allele expression in lymphoblastoid cell lines as measured by quantitative RT-PCR. I bars represent standard errors.

 
Quantitative RT-PCR

Expression analysis of the GPR54 alleles was performed by means of real-time PCR, with the use of lymphoblastic messenger RNA (mRNA) as a template. The mutant alleles were expressed at concentrations correlated with concentrations of control lymphoblasts. When compared with the standardized control mRNA, the mean (±SE) total concentration of GPR54 mRNA in the compound-heterozygous patient was 17.6±1.6 percent of the normal concentration (P<0.001 by Student's t-test); the expression concentration of the R331X allele was 17.9±1.9 percent of the normal concentration, and the expression concentration of the X399R allele was 2.5±0.3 percent of the normal concentration (Figure 3D).

Endocrinologic Phenotyping

A base-line profile of luteinizing hormone in the proband carrying the heterozygous mutations R331X and X399R is shown in Figure 4A. The patient had low concentrations of luteinizing hormone and low testosterone concentrations, particularly as compared with the mean (±2 SD) luteinizing hormone level measured in 20 normal men.20,21 Nonetheless, nine low-amplitude pulses of luteinizing hormone are present, as determined by formal pulse analysis. The responses to four doses of intravenous gonadotropin-releasing hormone are compared with the mean amplitude of luteinizing hormone pulses in six other patients with idiopathic hypogonadotropic hypogonadism who were treated with the same regimen (Figure 4B and Figure 4C). The dose–response curve for the proband is shifted leftward as compared with the 95 percent confidence intervals around the mean responses in men with idiopathic hypogonadotropic hypogonadism who did not have any GPR54 mutations.


View larger version (21K):
[in this window]
[in a new window]
  		Figure 4. Biochemical Phenotyping in a Patient Carrying the GPR54 Mutations R331X and X399R.

Panel A shows the base-line pattern of luteinizing hormone secretion, with measurements of the gonadotropin at 10-minute intervals over a 24-hour period. The pooled luteinizing hormone concentration was 2.0 IU per liter, the pooled follicle-stimulating hormone concentration 3.9 IU per liter, and the pooled testosterone concentration 1.22 nmol per liter (normal range, 11.42 to 33.6); the testicular volume was 1 ml for each testis. The shaded areas in Panels A and B represent the mean concentration (±2 SD) of secretion of luteinizing hormone among 20 normal men. Arrowheads represent pulses. Panel B shows the luteinizing hormone response to exogenous pulsatile gonadotropin-releasing hormone administered intravenously in a range of doses. The doses were administered in random order. The pooled testosterone concentrations ranged from 20.90 to 23.65 nmol per liter; the testicular volume was 6 ml for the right testis and 8 ml for the left testis. The patient received exogenous pulsatile gonadotropin-releasing hormone subcutaneously every two hours on an outpatient basis for 11 months before the study began. Panel C shows the dose–response curve for the patient, drawn with a regression line. The data for the patient fall to the left of the 95 percent confidence intervals around the mean amplitudes of the luteinizing hormone response in six other men with idiopathic hypogonadotropic hypogonadism who were receiving the same regimen.

 
Pathophysiology, Anatomy, and Behavior of Homozygous Gpr54-Deficient Mice

Homozygous mutant mice (Gpr54 –/–) (Figure 2A and Figure 2B) were viable and obtained at the expected mendelian frequency from heterozygous breeding pairs (Figure 2B and Supplementary Appendix 2). RT-PCR analysis of transcripts showed no detectable transcription in the 3' end of the homozygous Gpr54tm1PTL allele (Figure 2C). Gpr54 +/– mice were phenotypically normal and were fertile. Gpr54 –/– mice did not display any of the physiologic changes associated with sexual maturation. The testes of male Gpr54 –/– mice were significantly smaller than those of age-matched controls (Figure 5A) (mean weight in nine Gpr54 –/– mice, 0.05±0.00 g; mean weight in eight age-matched controls, 0.18±0.01 g; P<0.001 by the unpaired Mann–Whitney U test) and did not contain spermatozoa in the lumen of the seminiferous tubules or the epididymis (Figure 5C, Figure 5D, Figure 5E, and Figure 5F). Primary spermatocytes were present, but there were very few haploid spermatids, which suggests that spermatogenesis had been initiated but had stopped before the meiotic-division stage.


View larger version (43K):
[in this window]
[in a new window]
  		Figure 5. Gonadal Anatomy and Secondary Sexual Characteristics of Gpr54 –/– Mice.

Panel A shows the reduction in the size of the testes (wild-type as compared with mutant male mice), and Panel B shows the small ovaries and uteri found in female Gpr54 –/– mice; the scale bars represent 0.5 cm. In Panels C through N, the wild-type mouse is represented by the left-hand column and the mutant mouse is represented by the right-hand column. Panel D shows the reduction in the number of spermatozoa in the seminiferous tubules, as compared with Panel C; the scale bars represent 50 µm. Panels E and F show the presence and absence, respectively, of sperm in the epididymis; the scale bars represent 100 µm. Panel H shows reduced development of the preputial gland, as compared with Panel G; the scale bars represent 1 cm. Panels I and J show the absence and presence, respectively, of the prepubescent zone X in the adrenal gland; the scale bars represent 20 µm. Panel L shows reduced mammary-duct formation, as compared with Panel K (the dark mass is lymph node); the scale bars represent 0.5 cm. Panels M and N show the presence and absence, respectively, of graafian follicles and corpora lutea; CL denotes corpus luteum; the scale bars represent 300 µm.

 
Male mice also lacked development of secondary sex glands, including the preputial gland (Figure 5G and Figure 5H), the seminal vesicles, and the prostate (not shown). In the adrenal glands of the mutant animals, the innermost region of the cortex, which normally regresses at puberty, was still present (Figure 5I and Figure 5J). Sexual mounting behavior was also not observed among the male mice. No gross morphologic abnormalities were found in the central nervous system of Gpr54 –/– mice, and the mutant mice thrived, apart from the reproductive defect.

Female mutant mice also had defective sexual development; they had small vaginal openings and did not become pregnant after appropriate mating exposure. Vaginal smears consisted of nonkeratinized epithelia and mucus strands similar to those observed in immature female mice, indicating the lack of an estrus cycle. The uterine horns in female Gpr54 –/– mice were threadlike, and the ovaries were significantly smaller than normal (mean weight in nine wild-type mice, 5.7±0.7 mg; mean weight in eight mutant mice, 1.0±0.1 mg; P<0.001 by the unpaired Mann–Whitney U test) (Figure 5B). Mammary tissue showed no postpubertal maturation of branched epithelial ducts (Figure 5K and Figure 5L). The ovaries contained primary and secondary follicles and occasionally an early antral follicle but no large graafian follicles or corpora lutea (Figure 5M and Figure 5N).

Endocrinologic Phenotypes in Gpr54-Deficient Mice

Male Gpr54 –/– mice had significantly lower blood testosterone concentrations than age-matched +/+ controls (mean among 12 mutant mice, 0.1±0.02 pg per milliliter; mean among 11 wild-type mice, 4.6±1.6 pg per milliliter; P<0.001 by the unpaired Mann–Whitney U test). The testosterone concentrations in male Gpr54 –/– mice were similar to those observed in female Gpr54 +/+ mice (mean among eight female mice, 0.2±0.02 pg per milliliter) (Figure 6A). The 17{beta}-estradiol concentrations in female Gpr54 –/– mice were similar to those in Gpr54 +/+ females at nonestrus stages of the reproductive cycle (Figure 6B) and to the base-line serum estradiol concentrations in male Gpr54 +/+ mice (data not shown). No Gpr54 –/– females were identified that had a 17{beta}-estradiol concentration similar to that found at estrus (mean concentration among five wild-type females, 96.5±16.3 pg per milliliter) (Figure 6B).


View larger version (24K):
[in this window]
[in a new window]
  		Figure 6. Hormone Levels in Gpr54 –/– Mice.

Five to eight mice were used in each assay group. P values are provided for statistically significant differences; all P values were calculated with nonparametric Mann–Whitney tests.

 
The lack of an estrus cycle in female mice was not caused by an inability of gonadal tissue to respond to gonadotropins. Female Gpr54 –/– mice could be induced to ovulate after sequential injection of the gonadotropins pregnant mares serum and human chorionic gonadotropin (data not shown). The lack of an estrus cycle and the failure to produce sperm in Gpr54 –/– mice were caused by a significant reduction in the serum follicle-stimulating hormone concentration (P=0.009) and a more moderate decrease in the luteinizing hormone concentration (Figure 6C and Figure 6D). Possible explanations of the reduced concentrations of circulating gonadotropins include an absence of pituitary gonadotropes, an inability of existing gonadotropes to respond to stimulation by gonadotropin-releasing hormone, and a lack of gonadotropin-releasing hormone production. This last possibility was ruled out because there was no significant difference between normal and mutant mice in the concentration of gonadotropin-releasing hormone in hypothalamic extracts (Figure 6E).

In addition, measurements of pituitary luteinizing hormone and follicle-stimulating hormone showed that although the total amount of each hormone was lower in Gpr54 –/– mice than in wild-type mice, significant quantities of each hormone were found, indicating that the pituitary gonadotropes are present in Gpr54 –/– mice and are capable of synthesizing luteinizing hormone and follicle-stimulating hormone. Furthermore, in adult female Gpr54 –/– mice, luteinizing hormone was secreted into the bloodstream in response to the injection of gonadotropin-releasing hormone (Figure 6F), and there was a corresponding depletion in pituitary luteinizing hormone (Figure 6G). Studies of the secretion of follicle-stimulating hormone in response to the injection of gonadotropin-releasing hormone had similar results (data not shown). Although the absolute concentration of serum luteinizing hormone after the injection of gonadotropin-releasing hormone was lower in Gpr54 –/– mice than in +/+ mice, the proportional increase in the luteinizing hormone concentration was similar (an increase by a factor of five from base line). These responses are consistent with a first exposure to gonadotropin-releasing hormone.22,23

Discussion

In primates, the hypothalamic–pituitary–gonadal axis is fully active during neonatal life, followed by a mysterious period of dormancy during childhood. The triggers of the onset of gonadotropin-releasing hormone secretion at puberty are as unclear as those that halt its secretion at the end of the neonatal period. Insight into this process has been gained through the study of various diseases in humans and animal models in which genetic defects cause abnormalities of sexual maturation. Mutations in both GPR54 in humans and Gpr54 in mice cause hypogonadotropic hypogonadism, pubertal delay, and sexual infantilism that can be corrected by the administration of exogenous gonadotropin-releasing hormone. Taken together, these observations establish that the effect of GPR54 on gonadotropin-releasing hormone secretion is conserved in multiple mammalian species and is a genetic determinant of sexual maturation.

GPR54 is a member of the rhodopsin family of G protein–coupled receptors whose sequences are most similar to those of members of the galanin-receptor family (35 to 40 percent identity).9 Although galanin and galanin-like peptide appear not to bind to GPR54,9,24 endogenous peptides derived from a precursor protein, kisspeptin-1, have recently been identified that do display agonist activity.8,9,10 The longest of these peptides is kisspeptin-1 68–121, or metastin, so called because of its ability to suppress metastatic potential in melanoma and breast-cancer cell lines.25,26,27 Metastin is secreted into the circulation by the placenta in relatively large quantities throughout gestation, although its physiologic role in pregnancy remains unknown.28

In this study, a variety of mutations in GPR54 were identified in patients with idiopathic hypogonadotropic hypogonadism. The index family was found to carry a homozygous L148S substitution. When expressed in cell lines, the L148S mutant construct markedly reduced inositol phosphate production as compared with the wild-type construct. A male patient with sporadic idiopathic hypogonadotropic hypogonadism was also found to be a compound heterozygote for the mutations R331X and X399R, which gave rise to nonsense and nonstop transcripts. It has been shown that mRNAs with premature termination codons can be targeted to nonsense-mediated decay,29 whereas mRNAs without an in-frame termination codon can be subject to nonstop decay, a recently identified degradation pathway that is initiated when the ribosome reaches the 3' terminal of the mRNA.30,31

We hypothesized that this combination of nonstop and nonsense mutations in GPR54 in the patient with sporadic idiopathic hypogonadotropic hypogonadism would result in the absence of a functional receptor. Quantitative RT-PCR confirmed the presence of dramatically reduced concentrations of GPR54 mRNA in the patient with the R331X and X399R alleles. Moreover, this total concentration appeared to be composed almost exclusively of the nonsense transcript. These results indicate that the contribution of the nonstop transcript was almost negligible, supporting the hypothesis of nonstop decay. In the unlikely event that a protein were produced by the X399R transcript, our findings regarding the X399R polyA stop construct suggest that it would function poorly.

The clinical phenotype of the patient carrying the R331X and X399R mutations was associated with a neuroendocrine profile involving low-amplitude pulses of luteinizing hormone, suggesting reduced secretion of gonadotropin-releasing hormone. This notion is supported by the leftward-shifted dose–response curve as compared with those for other patients with idiopathic hypogonadotropic hypogonadism who were undergoing the same therapy, suggesting that this patient was more sensitive to exogenous gonadotropin-releasing hormone. Since these studies were performed, deletions in GPR54 have been described in a separate family with idiopathic hypogonadotropic hypogonadism, although the phenotypic features of this family were not detailed.32 Although the investigators in that case agree with our conclusion that the frequency of GPR54 mutations as a cause of idiopathic hypogonadotropic hypogonadism is low (4 of 113 total cases), GPR54 reveals a new direction for the exploration of other genes that are essential for the secretion of gonadotropin-releasing hormone.

The Gpr54-deficient mice had striking physiological similarities to the patients with idiopathic hypogonadotropic hypogonadism, including a lack of sexual maturation associated with low concentrations of gonadotropins. In addition, their gonads remained sensitive to exogenous gonadotropins, and their pituitary gonadotropes remained responsive to stimulation by gonadotropin-releasing hormone. This strong similarity between the findings in the patients and those in the mouse model establishes a central role for GPR54 in gonadotropin-releasing hormone secretion and the onset of sexual maturation among mammalian species. Moreover, the use of Gpr54-deficient mice permitted the quantitation of their hypothalamic gonadotropin-releasing hormone concentrations, which were normal in the face of their hypogonadotropism. The presence of normal concentrations of gonadotropin-releasing hormone in the hypothalamus of Gpr54-deficient, sexually immature mice is reminiscent of prepubertal rats and monkeys who have normal numbers of gonadotropin-releasing hormone–containing neurons, normal mRNA concentrations, and normal concentrations of gonadotropin-releasing hormone in the hypothalamus.33,34 Extrapolation from the Gpr54-deficient mice to nonhuman primates and humans suggests that GPR54 may have a substantial effect on the processing or secretion of gonadotropin-releasing hormone.

There are three possible mechanisms that may allow abnormalities in GPR54 to cause pubertal delay. The first possibility is that defects in the metastin–GPR54 system perturb gonadotropin-releasing hormone neuronal migration that is analogous to the abnormal axonal targeting that occurs in the X-linked form of Kallmann's syndrome (idiopathic hypogonadotropic hypogonadism with anosmia).35,36,37 In vitro, the metastin–GPR54 system induces an "adhesive phenotype" with inhibition of chemotaxis, focal adhesions and stress fibers, and phosphorylation of focal adhesion kinase and paxillin.8 However, the normal content of gonadotropin-releasing hormone in the hypothalamus of Gpr54-deficient mice argues that there has been an appropriate migration of the neurons containing gonadotropin-releasing hormone from their origin in the olfactory placode to their destination in the hypothalamus. The possibility remains, however, that there is a subtle defect in the terminal migration or differentiation of these neurons within the hypothalamus.

The second possibility is that GPR54 modulates the activity of gonadotropin-releasing hormone at the level of the pituitary. The presence of small but detectable pulses of luteinizing hormone induced by gonadotropin-releasing hormone in the patient with the R331X and X399R mutations and his leftward-shifted dose–response curve suggest that pituitary responsiveness in the GPR54-deficient patient is, if anything, enhanced, suggesting that loss-of-function mutations in GPR54 do not diminish the sensitivity of gonadotropes to gonadotropin-releasing hormone.

The third possibility is that GPR54 regulates the release of gonadotropin-releasing hormone at the level of the hypothalamus. This hypothesis is supported by three observations: the low-amplitude pulses of luteinizing hormone in the patient carrying the R331X and X399R mutations, his leftward-shifted dose–response curve, and the normal content of gonadotropin-releasing hormone in the hypothalamus of Gpr54-deficient mice. Further studies will be required to determine the precise mechanisms of action within the hypothalamus as well as the physiological functions of the peptide ligands for GPR54.

Currently, it appears that the frequency of GPR54 mutations as a cause of idiopathic hypogonadotropic hypogonadism is not high. However, the elucidation of the role of GPR54 as a regulator of gonadotropin-releasing hormone–related physiology and puberty may well provide a seminal clue, offering a new perspective on other candidate genes that may control sexual maturation and puberty. These include the genes affecting the biosynthesis, processing, and secretion of the putative ligands, kisspeptin–metastin; the transcriptional regulation of GPR54 itself; and the signaling pathways downstream of the gene. Our data from patients with idiopathic hypogonadotropic hypogonadism and a mouse model provide strong evidence that GPR54 is a key regulator of the biology of puberty.

Supported by grants (U54 HD28138-13, 5R01 HD15788-17, 3M01 RR01066-22S2, GM61354, and T32 HD40135) from the National Institutes of Health.

We are indebted to the staff of the General Clinical Research Center and the Reproductive Endocrine Unit of the Massachusetts General Hospital for their clinical care and investigation of patients with hypogonadotropic hypogonadism; to Alan Schneyer, Israel Sidis, Gregoy Bedecarrats, and Sandra Ryeom for their insights and expertise in molecular biology; to Cricket and Jon Seidman, David Altshuler, and Eric Lander for their advice regarding genetic analysis; to the Genomics and Peptide Core Facilities of Massachusetts General Hospital; to Victoria Petkova of the TaqMan Real-Time PCR Core Laboratory of the Beth Israel Deaconess Medical Center; to Astrid Meysing for assistance with sequencing; to Dr. A. Caraty from the Unité Mixte de Recherche Physiologie de la Reproduction et des Comportements, Institut National de la Recherche Agronomique, Nouzilly, France, for assistance with the gonadotropin-releasing hormone radioimmunoassay; to the Ligand Core Laboratory, supported by grant U54 HD28934; and to all the patients with hypogonadotropic hypogonadism who have donated their time, energy, and blood samples and have been our coinvestigators.


Source Information

From the Reproductive Endocrine Unit (S.B.S., J.S.A., J.K.S., K.M.S., W.F.C.) and the Molecular Neurogenetics Unit, Center for Human Genetic Research (S.A.S., J.F.G.), Massachusetts General Hospital; the Division of Endocrinology, Diabetes, and Hypertension, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School (W.K., U.B.K.); and the Harvard Institute of Human Genetics, Harvard Medical School (S.A.S., J.F.G.) — all in Boston; Paradigm Therapeutics (S.M., R.R.T., A.G.H., D.Z., J.D., M.B.L.C., S.A.J.R.A, W.H.C.); the Departments of Physiology (E.E.C., W.H.C.), Oncology (S.A.J.R.A.), and Clinical Biochemistry (S.O.) and the Cambridge Institute for Medical Research, Addenbrooke's Hospital (S.O.), University of Cambridge, Cambridge, United Kingdom; and the Faculty of Medicine, Kuwait University, Al-Jabriyah (Y.B.-A.).

Drs. Seminara, Messager, Chatzidaki, and Thresher contributed equally to the article. Drs. Crowley, Aparicio, and Colledge were the senior authors.

Address reprint requests to Dr. Colledge at whc23{at}cam.ac.uk; to Dr. Aparicio at Paradigm Therapeutics, 214 Cambridge Science Park, Milton Rd., Cambridge CB4 0WA, United Kingdom, or at saparicio{at}paradigm-therapeutics.com; or to Dr. Crowley at the Reproductive Endocrine Unit, BHX 505, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114, or at crowley.william{at}mgh.harvard.edu.

References

  1. Seminara SB, Hayes FJ, Crowley WF Jr. Gonadotropin-releasing hormone deficiency in the human (idiopathic hypogonadotropic hypogonadism and Kallmann's syndrome): pathophysiological and genetic considerations. Endocr Rev 1998;19:521-539. [Free Full Text]
  2. Crowley WF Jr, McArthur JW. Simulation of the normal menstrual cycle in Kallmann's syndrome by pulsatile administration of luteinizing hormone-releasing hormone (LHRH). J Clin Endocrinol Metab 1980;51:173-175. [Free Full Text]
  3. Hoffman AR, Crowley WF Jr. Induction of puberty in men by long-term pulsatile administration of low-dose gonadotropin-releasing hormone. N Engl J Med 1982;307:1237-1241. [Abstract]
  4. Santoro N, Filicori M, Crowley WF Jr. Hypogonadotropic disorders in men and women: diagnosis and therapy with pulsatile gonadotropin-releasing hormone. Endocr Rev 1986;7:11-23. [Free Full Text]
  5. Bo-Abbas Y, Acierno JS Jr, Shagoury JK, Crowley WF Jr, Seminara SB. Autosomal recessive idiopathic hypogonadotropic hypogonadism: genetic analysis excludes mutations in the gonadotropin-releasing hormone (GnRH) and GnRH receptor genes. J Clin Endocrinol Metab 2003;88:2730-2737. [Free Full Text]
  6. Acierno JS Jr, Shagoury JK, Bo-Abbas Y, Crowley WF Jr, Seminara SB. A locus for autosomal recessive idiopathic hypogonadotropic hypogonadism on chromosomal 19p13.3. J Clin Endocrinol Metab 2003;88:2947-2950. [Abstract]
  7. UCSC genome browser. November 2002 Assembly. (Accessed September 26, 2003, at http://www.genome.ucsc.edu.)
  8. Ohtaki T, Shintani Y, Honda S, et al. Metastasis suppressor gene KiSS-1 encodes peptide ligand of a G-protein-coupled receptor. Nature 2001;411:613-617. [CrossRef][Medline]
  9. Muir AI, Chamberlain L, Elshourbagy NA, et al. AXOR12, a novel human G protein-coupled receptor, activated by the peptide KiSS-1. J Biol Chem 2001;276:28969-28975. [Free Full Text]
  10. Kotani M, Detheux M, Vandenbogaerde A, et al. The metastasis suppressor gene KiSS-1 encodes kisspeptins, the natural ligands of the orphan G protein-coupled receptor GPR54. J Biol Chem 2001;276:34631-34636. [Free Full Text]
  11. Bedecarrats GY, Linher KD, Kaiser UB. Two common naturally occurring mutations in the human gonadotropin-releasing hormone (GnRH) receptor have differential effects on gonadotropin gene expression and on GnRH-mediated signal transduction. J Clin Endocrinol Metab 2003;88:834-843. [Free Full Text]
  12. Spratt DI, Carr DB, Merriam GR, Scully RE, Rao PN, Crowley WF Jr. The spectrum of abnormal patterns of gonadotropin-releasing hormone secretion in men with idiopathic hypogonadotropic hypogonadism: clinical and laboratory correlations. J Clin Endocrinol Metab 1987;64:283-291. [Free Full Text]
  13. Filicori M, Santoro N, Merriam GR, Crowley WF Jr. Characterization of the physiological pattern of episodic gonadotropin secretion throughout the human menstrual cycle. J Clin Endocrinol Metab 1986;62:1136-1144. [Free Full Text]
  14. Santen RJ, Bardin CW. Episodic luteinizing hormone secretion in man: pulse analysis, clinical interpretation, physiologic mechanisms. J Clin Invest 1973;52:2617-2628. [Web of Science][Medline]
  15. Collins TJ, Parkening TA, Smith EK. Plasma and pituitary concentrations of LH, FSH, and prolactin after injection of GnRH in aged female C57BL/6 mice. Neurobiol Aging 1981;2:125-131. [CrossRef][Web of Science][Medline]
  16. Caraty A, Antoine C, Delaleu B, et al. Nature and bioactivity of gonadotropin-releasing hormone (GnRH) secreted during the GnRH surge. Endocrinology 1995;136:3452-3460. [Abstract]
  17. Antonarakis SE. Recommendations for a nomenclature system for human gene mutations. Hum Mutat 1998;11:1-3. [CrossRef][Web of Science][Medline]
  18. Fugu genome server. (Accessed September 26, 2003, at http://ensembl.fugu-sg.org/Fugu_rubripes/.)
  19. Aparicio S, Chapman J, Stupka E, et al. Whole-genome shotgun assembly and analysis of the genome of Fugu rubripes. Science 2002;297:1301-1310. [Free Full Text]
  20. Spratt DI, O'Dea LSL, Schoenfeld D, Butler J, Rao PN, Crowley WF Jr. Neuroendocrine-gonadal axis in men: frequent sampling of LH, FSH, and testosterone. Am J Physiol 1988;254:E658-E666. [Web of Science][Medline]
  21. Whitcomb RW, Crowley WF Jr. Diagnosis and treatment of isolated gonadotropin-releasing hormone deficiency in men. J Clin Endocrinol Metab 1990;70:3-7. [Free Full Text]
  22. McDowell IF, Morris JF, Charlton HM. Characterization of the pituitary gonadotroph cells of hypogonadal (hpg) male mice: comparison with normal mice. J Endocrinol 1982;95:321-330. [Free Full Text]
  23. McDowell IF, Morris JF, Charlton HM, Fink G. Effects of luteinizing hormone releasing hormone on the gonadotrophs of hypogonadal (hpg) mice. J Endocrinol 1982;95:331-340. [Free Full Text]
  24. Lee DK, Nguyen T, O'Neill GP, et al. Discovery of a receptor related to galanin receptors. FEBS Lett 1999;446:103-107. [CrossRef][Web of Science][Medline]
  25. Lee JH, Miele ME, Hicks DJ, et al. KiSS-1, a novel human malignant melanoma metastasis-suppressor gene. J Natl Cancer Inst 1996;88:1731-1737. [Erratum, J Natl Cancer Inst 1997;89:1549.] [Free Full Text]
  26. Lee JH, Welch DR. Suppression of metastasis in human breast carcinoma MDA-MB-435 cells after transfection with the metastasis suppressor gene, KiSS-1. Cancer Res 1997;57:2384-2387. [Free Full Text]
  27. Lee JH, Welch DR. Identification of highly expressed genes in metastasis-suppressed chromosome 6/human malignant melanoma hybrid cells using subtractive hybridization and differential display. Int J Cancer 1997;71:1035-1044. [CrossRef][Web of Science][Medline]
  28. Horikoshi Y, Matsumoto H, Takatsu Y, et al. Dramatic elevation of plasma metastin concentration in human pregnancy: metastin as a novel placenta-derived hormone in humans. J Clin Endocrinol Metab 2003;88:914-919. [Free Full Text]
  29. Maquat LE, Carmichael GG. Quality control of mRNA function. Cell 2001;104:173-176. [CrossRef][Web of Science][Medline]
  30. Frischmeyer PA, van Hoof A, O'Donnell K, Guerrerio AL, Parker R, Dietz HC. An mRNA surveillance mechanism that eliminates transcripts lacking termination codons. Science 2002;295:2258-2261. [Free Full Text]
  31. van Hoof A, Frischmeyer PA, Dietz HC, Parker R. Exosome-mediated recognition and degradation of mRNAs lacking a termination codon. Science 2002;295:2262-2264. [Free Full Text]
  32. De Roux N, Genin E, Carel J-C, Matsuda F, Chaussain J-L, Milgrom E. Hypogonadotropic hypogonadism due to loss of function of the KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci U S A 2003;100:10972-10976. [Free Full Text]
  33. Wiemann JN, Clifton DK, Steiner RA. Pubertal changes in gonadotropin-releasing hormone and proopiomelanocortin gene expression in the brain of the male rat. Endocrinology 1989;124:1760-1767. [Free Full Text]
  34. Cameron JL, McNeill TH, Fraser HM, Bremner WJ, Clifton DK, Steiner RA. The role of endogenous gonadotropin-releasing hormone in the control of luteinizing hormone and testosterone secretion in the juvenile male monkey, Macaca fascicularis. Biol Reprod 1985;33:147-156. [Abstract]
  35. Franco B, Guioli S, Pragliola A, et al. A gene deleted in Kallmann's syndrome shares homology with neural cell adhesion and axonal path-finding molecules. Nature 1991;353:529-536. [CrossRef][Medline]
  36. Legouis R, Hardelin JP, Levilliers J, et al. The candidate gene for the X-linked Kallmann syndrome encodes a protein related to adhesion molecules. Cell 1991;67:423-435. [CrossRef][Web of Science][Medline]
  37. Legouis R, Lievre CA, Leibovici M, Lapointe F, Petit C. Expression of the KAL gene in multiple neuronal sites during chicken development. Proc Natl Acad Sci U S A 1993;90:2461-2465. [Free Full Text]

 

This Article
-Abstract
- PDF
-PDA Full Text
-PowerPoint Slide Set
-Supplementary Material

Commentary
-Perspective
by Beier, D. R.

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited
-E-mail When Letters Appear

More Information
-PubMed Citation

This article has been cited by other articles:


HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved.