| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Section of Reproductive Endocrinology, Infertility, and Genetics, Department of Obstetrics and Gynecology (L.C.L.), and Neurobiology Program (L.C.L.), The Institute of Molecular Medicine and Genetics, The Medical College of Georgia, Augusta, Georgia 30912; Section of Endocrinology (A.L.A.P., L.A.C.R.d.M.), Department of Medicine, University of Brasilia, Brasilia 71525, DF, Brazil; Abbott Pharmaceutical Company (J.X.), Abbott Park, Illinois 60664; Section of Reproductive Endocrinology (L.D.C.d.M.), Department of Obstetrics and Gynecology, University of Brasilia, Brasilia, DF, Brazil; Serono (W.W.), Rockland, Massachusetts 02370; and Reproductive Endocrine Laboratory Unit (P.M.S.), Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Lawrence C. Layman, M.D., Section of Reproductive Endocrinology, Infertility, and Genetics, Department of Obstetrics and Gynecology, The Medical College of Georgia, 1120 15th Street, Augusta, Georgia 30912-3360. E-mail: . Llayman{at}mail.mcg.edu
Abstract
FSH is a dimeric pituitary glycoprotein hormone that regulates gonadal function. Human mutations in the FSHß gene have been shown to produce complete deficiency states in which pubertal development and reproductive capacity are inhibited. To date, no patients with partial or complete pubertal development due to FSHß mutations have been documented in humans. We describe and characterize affected siblings, a male and a female, with evidence of pubertal development due to homozygosity for a Tyr76X nonsense mutation in the FSHß gene. In vitro analysis of this mutant demonstrates unmeasurable FSH by immunoassay and by two different bioassays, using either cAMP (homologous FSH bioassay) or estradiol (rat granulosa cell assay) as the endpoints. In additional in vitro analyses, mutants previously found in patients with a phenotype of complete FSH deficiency (Cys51Gly and Val61X) and the Tyr76X were compared in the same immuno- and bioassays. All mutations failed to produce measurable FSH by all assays. Unexpectedly, these siblings with isolated FSH deficiency due to a nonsense FSHß mutation had some evidence of puberty, suggesting that other factors might preserve gonadal steroidogenesis in the absence of FSH or that current bioassays cannot discriminate among very low FSH levels.
THE PITUITARY GLYCOPROTEIN hormone FSH has important tropic roles in gonadal steroid and gamete production in males and females. FSH is a dimeric glycoprotein hormone composed of an 
-subunit common to LH, human chorionic gonadotropin, and TSH and a unique ß-subunit providing hormone specificity (1, 2). The functions of FSH include follicular development and sex steroid production necessary for fertility in females (2). In males, FSH stimulates Sertoli function and may be involved in androgen production, both components of which are necessary for fertility (2). These functions are supported by studies involving knockout of the FSHß gene ligand (3, 4) and the FSH receptor (5), as well as by human gene mutations.
Several mutations in FSHß gene have been described in humans, but all of those studied by functional analysis demonstrated complete FSH deficiency with absent puberty and sterility (6, 7, 8). Three females described had absent breast development, primary amenorrhea, low levels of estrogen, undetectable serum immunoreactive FSH, and elevated immunoreactive LH (6, 7, 9). The FSHß gene mutations identified in these patients failed to yield any measurable immunoreactive or bioactive FSH in vitro (7). Only two males with FSHß mutations have been described, with one having normal puberty and azoospermia (10), and the other having absent puberty and azoospermia (8). Neither of these males had their mutant FSHß genes characterized in vitro, although the male with absent puberty had the same mutation in which we demonstrated undetectable immunoreactive and bioactive FSH levels in vitro (7).
We report and characterize in vitro the first FSHß gene mutation in two siblings with partial (female) or complete (male) puberty. The FSHß mutation was identified in a consanguineous Brazilian family containing two siblings with isolated FSH deficiency. In contrast to several previously described patients with FSHß mutations, affected individuals in this family manifested pubertal development, but were infertile as others with FSHß mutations have been. This is also the first family reported that contains both an affected male and affected female with isolated FSH deficiency in the same pedigree. We demonstrate that the mutant FSHß gene displays impaired immunoreactive and bioactive FSH in vitro and analyze this mutation in vitro in the same assay with other previously described FSHß mutations. These studies confirm that the newly described Tyr76X FSHß mutation causes the phenotype of isolated FSH deficiency.
Patients and Methods
Patients
This Brazilian family included a male and female, both of whom had evidence of pubertal development (Fig. 1
). The proband was a 32-yr-old, 156-cm-tall female who presented with symptoms of partial (Tanner stages II-III) breast development, primary amenorrhea, and a low estradiol (44 pmol/liter; normal, 92–1468 pmol/liter). Her brother, a 30-yr-old, 170-cm-tall male, presented with infertility. He had normal puberty, normal erections and ejaculation. He had a normal adult male testosterone level (26 nmol/liter; normal, 6.3–63.7 nmol/liter) and no gynecomastia, but had small testes (12 cc) and azoospermia. A testicular biopsy revealed Leydig cell hyperplasia and sparse, small seminiferous tubules with germinal cell aplasia, peritubular fibrosis, and few Sertoli cells (Fig. 2). Both affected individuals had low serum FSH and elevated serum LH levels, both basally and after GnRH stimulation, suggesting the possibility of isolated FSH deficiency. All other hormones studied, computed tomography scans of the brain, and karyotypes were normal (Table 1). This study was approved by the Institutional Review Boards from the University of Chicago and the Medical College of Georgia.
|
|
|
DNA was extracted (11) from 11 family members, indicated in Fig. 1
. Thirty cycles of PCR were performed as described previously on exons 2 and 3 of the FSHß gene, which encode for the translated FSHß protein (7). For each exon, 30 cycles of PCR were performed at 95 C for 1 min, 55 C for 30 sec, and 72 C for 30 sec. The MgCl2 concentration was 2.0 mM for exon 2 and 2.5 mM for exon 3. Primer sequences include: exon 2 sense (AGT TTC TAG TGG GCT TCA TTG TTT G) exon 2 antisense (TGG CTA AAG GAC TCA TGG CTG); exon 3 sense (GCT AAA TAG GAA CTT CCA C) and exon 3 antisense (TAT GTG GCC TGA AAT GTC C) (7). A negative control, containing all reagents except DNA, was included in each PCR. The PCR products were then electrophoresed on agarose gels, ethanol precipitated, and subjected to dideoxy DNA sequencing using Big Dye Terminator Cycle (ABI Prism, PE Applied Biosystems, Foster City, CA).
Sequencing reactions were done using the ABI Big Dye Terminator kit (PE Applied Biosystems). Briefly, duplicate reactions of 20 ng template of the PCR products were amplified in a total reaction volume of 20 µl using either a forward or reverse primer. After the sequencing amplification, unincorporated nucleotides and primers were removed by spin column chromatography (Princeton Separations, Trenton, NJ) and dried in a vacuum centrifuge. Template suppression reagent was added to each of the samples, which were then vortexed briefly and denatured at 95 C for 2 min. Samples were immediately placed on ice for 2 min and vortexed again to mix. Then, the samples were analyzed on the ABI 310 Automated DNA Sequencer (ABI Prism, PE Applied Biosystems). Sequencing was confirmed in forward and reverse directions three times each for both probands and the parents.
To analyze additional family members for the mutation, a mutant primer was synthesized, which introduced a restriction cut site for the enzyme HpaI (Y76* CCA TGC AGA TTC CTT GTA TAG TTA). PCR products were then subjected to HpaI digestion at 37 C overnight, electrophoresed on agarose gels, stained with ethidium bromide, and photographed. Only mutant alleles will be digested with the HpaI enzyme, whereas wild-type alleles will not.
In vitro analysis of FSHß mutation
Two plasmids, pGEM7z (Promega Corp., Madison, WI) and pM2FSHß (a generous gift from J. Larry Jameson, Northwestern University), containing the entire coding region of human FSHß (
3.4 kb) were used for in vitro analysis (7). Site-directed mutagenesis of the mutant fragment was performed by PCR on the pGEM7z FSHß fragment and confirmed by DNA sequencing. The mutant was then ligated into the PstI site of a pGEM7z FSHß construct. Digestion of pGEM7z and pM2FSHß with BamHI was then performed, and the mutant fragment from pGEM7z was ligated into pM2FSHß, then grown in Escherichia coli (7). Proper orientation was confirmed with restriction enzyme digestion. Chinese hamster ovary (CHO; American Type Culture Collection, Manassas, VA) cells were then stably transfected with the mutant or wild-type pM2FSHß construct. Resistant clones were selected with Geneticin (Life Technologies, Inc., Grand Island, NY) as described previously (7). RNA was extracted and subjected to RT-PCR (Superscript Preamplification System, Life Technologies, Inc.) in both cell types for FSHß-subunit mRNA, using appropriate negative controls (tubes without reverse transcriptase and without RNA). Cells from stable clones were then grown to near confluence and after 48 h, cellular media was collected and stored at -80 C for assay.
In addition, a comparative in vitro analysis was performed of the present Tyr76X mutation and two previously described FSHß mutations. In a separate set of experiments, stably transfected CHO cells containing either the wild-type FSHß gene, Cys51Gly, Val61X, or the Tyr76X mutants were grown in media for 48 h, then collected and stored at -80 C for assay for both immuno- and bioactive FSH.
Cellular medium was assayed for immunoassay and two different FSH bioassays, one a homologous FSH bioassay (7, 12) and the other, the rat granulosa cell assay (13).
Immunometric FSH assay
Cell-conditioned medium was assayed for immunoactive FSH using a two-site sandwich assay (Abbott Diagnostics, Abbott Park, IL), which has been described and validated previously (14). The FSH immunoassay is specific for dimeric FSH, was calibrated with a hMG reference standard (WHO 71/223), and has a lower limit of detection (1 mIU/ml) than the homologous FSH bioassay. The intra- and interassay coefficients of variation across the working range of the assay were less than 5% and 8%, respectively.
Homologous FSH bioassay
The homologous FSH bioassay uses CHO cells stably transfected with the human FSH receptor (CHO/FSH-R cells) and a cAMP-responsive luciferase reporter gene (12, 15). Cells were grown in culture in MEM with 10% fetal calf serum and 100 µg/ml concentration of Geneticin. FSH standards (WHO 71/223) or samples (100 µl of cell-conditioned medium from the wild-type or FSHß mutant cell lines) were diluted in assay medium containing 0.25 mmol/liter methylisobutylxanthine and 2% dimethylsulfoxide. After 4.5 h of incubation at 37 C, medium was removed, and cells were placed in lysis solution as described previously (12, 15). One hundred microliters of assay buffer were then added to the cell lysate, and luciferase activity was measured using a luminometer (ML3000, Dynatech Corp. Laboratories, Chantilly, VA) after injection of 50 µl 250 µmol/liter luciferin (Sigma Chemical Co., St. Louis, MO) containing 10 mmol/liter dithiothreitol at room temperature (12, 15). Luciferase activity was measured as relative light units over a 10-sec time period. The lower limit of detection is 2 mIU/ml for bioactive FSH levels. The intra- and interassay coefficients of variation across the working range of the assay were less than 10% and 15%, respectively, at a level of 30 IU/liter.
Rat granulosa cell assay
The rat granulosa cell FSH bioassay was performed on the cell-conditioned medium. The protocol, modified from Erickson and Hsueh (13), used granulosa cells from diethylstibestrol (DES)-exposed immature Sprague Dawley rats and estradiol as the endpoint. Briefly, immature (21 d old) Sprague Dawley rats with sc implanted DES pellets (DES 5 mg pellets/rat) were used. After 3 d of DES exposure, follicles were punctured repetitively using two 27-gauge needles mounted on 1-cc syringes to release granulosa cells. After washing, cells at 100,000 per well were seeded in microtiter wells and incubated at 37 C in McCoy’s 5A medium supplemented with 0.1% BSA, 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin sulfate, 2 µM androstendione, and 0.5 mM 3-isobutyl-1-methyl-xanthine. Ten 2-fold serial dilutions (from 200 pM to 1.25 pM) of recombinant human FSH (rhFSH; Serono SA, Aubonne, Switzerland) or samples at desired concentrations were added in triplicate to cells. After 72 h of incubation, culture supernatants were removed. Production of estradiol was analyzed by estradiol enzyme immunoassay (EIA; DSL-10-4300, Diagnostic Systems Laboratories, Inc., Webster, TX). All samples were quantitated using the internal standard provided by the manufacturer of the estradiol EIA kits. The lower limit of quantitation of estradiol according to the manufacturer is 20 pg/ml. The intra- and interassay precision of this assay is more than 90% and 88%, respectively. The EC50 of rhFSH as measured by induction of estradiol production using this EIA is usually between 450 and 750 pM, and the lowest concentration of rhFSH that showed consistent induction of measurable estradiol production is at least 0.2 pM.
Results
Agarose gel electrophoresis of PCR products from exons 2 and 3 was normal in both probands and all other family members. DNA sequencing of the PCR products of exon 2 was normal (data not shown), but exon 3 revealed a homozygous nonsense mutation at codon 76 (Tyr76X) in which a TAC (Tyr) was changed to a TAA (Stop; Fig. 3). This mutation was identified in both affected individuals, and both parents were heterozygous (Figs. 1 and 4). Digestion with HpaI of exon 3 PCR products from other family members (Figs. 1 and 4) demonstrated heterozygosity in two siblings (II3 and II5) and three of the siblings’ children (III2, III5, and III6). In addition to the parents of the affected individuals, the heterozygotes of reproductive maturity (II3 and II5) had normal puberty and fertility. DNA sequencing of one of the presumed heterozygotes by HpaI digestion confirmed heterozygosity for the nonsense mutation (data not shown).
|
|
FSHß vector (containing both the -subunit and the FSHß-subunit genes), and stably transfected into CHO cells (7). Expression of both the -subunit and FSHß (wild-type, Y76X, Val61X, Cys51Gly) genes was confirmed by RT-PCR (data not shown) from RNA extracted from cell lines containing either the wild-type or mutant FSHß genes. Immunometric assay of cellular medium demonstrated measurable levels of FSH for the wild-type FSHß cell line, but was undetectable for the Tyr76X mutant (Fig. 5). FSH bioassay of media from wild-type FSH cells revealed measurable levels of FSH for both the homologous FSH bioassay (Fig. 5) and the rat granulosa cell assay (data not shown). However, the Tyr76X FSHß mutant yielded unmeasurable FSH levels in both the homologous FSH bioassay and the rat granulosa cell assay (Fig. 5). In addition, cellular media from the Val61X, Cys51Gly, and Tyr76X mutants all demonstrated unmeasurable levels of immunoactive FSH and undetectable bioactive FSH, as determined by both the homologous FSH bioassay and the rat granulosa cell assay (Fig. 5).
|
A phenotype of absent puberty and infertility has been observed in homozygous FSHß knockout mice (3, 4) and in several humans with FSHß mutations (6, 7, 8, 9, 10). Female FSHß knockout mice demonstrate low levels of FSH and sterility due to defects in follicular development, whereby follicles do not mature beyond the antral stage (3). Three previously described human females with FSHß mutations (two homozygotes for Val61X and one compound heterozygote Val61X/Cys51Gly) manifest profound hypoestrogenism, with a complete lack of breast development, absent menses, and infertility (6, 7, 9). When both mutants were stably transfected into CHO cells, cellular media assayed for immuno- and bioactive FSH levels (using the homologous FSH bioassay) revealed undetectable levels (7). Complete FSH deficiency in male FSHß knockout mice results in normal testosterone levels, small testes, and oligospermia, but fertility does occur (3). In contrast, only one human male with the severe Val61X mutation has been characterized (8). He presented with absent pubertal development, low testosterone levels, small testes, and azoospermia.
To date, no well characterized patients with partial or complete pubertal development due to FSHß mutations have been described in humans. Although one male with normal puberty and azoospermia had a homozygous Cys82Arg missense FSHß mutation (10), no in vitro analysis was performed to determine the effect upon the FSHß protein. In the present study, we describe a family containing both a male and a female with evidence of pubertal development, both of whom demonstrated homozygosity for a newly described Tyr76X nonsense FSHß mutation. Of the two sibs, the female had a more severe phenotype with incomplete breast development, primary amenorrhea, and sterility. Although FSH deficient, this female had some evidence of breast development, unlike the previously described FSHß-deficient females. The male in this family had slightly reduced testicular size and azoospermia, but a normal serum testosterone. Testicular histology demonstrated absent germ cells, diminished Sertoli cells, and Leydig cell hyperplasia in the presence of FSH deficiency and LH hypersecretion.
In vitro analysis of the Tyr76X FSHß mutation demonstrated both unmeasurable immuno- and bioactive FSH, similar to the previously described severe mutations (7). This finding is somewhat unexpected because both patients had evidence of secondary sex characteristics. This absent FSH bioactivity was observed for both bioassays used. The homologous FSH bioassay uses second messenger cAMP as the endpoint (12), whereas the rat granulosa cell bioassay uses estradiol as the endpoint (13).
Despite the similarity in vitro, the Tyr76X produces a milder phenotype than either Val61X or Cys51Gly mutants. The Cys51Gly disrupts a conserved Cys, important in disulfide bonds, that probably interferes with dimer formation and subsequent secretion (2, 6). The Val61X is predicted to produce a frameshift, altering amino acids 61–86, and then introduces a stop codon at position 86, truncating amino acids 87–111 of the mature FSHß-subunit (6). In contrast, the Tyr76X predicts a translated protein missing amino acids 76–111. Because our female with the Tyr76X mutation had a more severe phenotypic deficiency (partial breast development, primary amenorrhea, and hypoestrogenism) than the male (small testes, normal pubic hair, and normal testosterone), it suggests that amino acids 76–111 are important for FSH action in males and, perhaps more importantly, in females. It also appears from the available human FSHß mutations that those which abolish and/or alter amino acids 61–111 (as in the Val61X) produce a more severe phenotype than if amino acids 76–111 altered (as in the Tyr76X). These findings indicate that perhaps amino acids 61–76 might be important determinants of the severity of phenotype. According to the crystal structure of FSH, all three of these mutants should interfere with the cysteine knot of FSHß, important in dimer formation and intracellular stability (16). In addition, both the Tyr76X and Val61X mutants would be lacking the noose (determinant loop)–amino acids 87–94-involved in FSH receptor specificity and the seat belt region contained in amino acids 84–104 that enables the ß-subunit to wrap around the -subunit (16).
Our in vitro analysis of known FSHß mutations, involving stable transfection into CHO cells, along with one immunometric assay and two FSH bioassays, was not able to differentiate between mutations causing absent puberty and sterility (severe mutations Val61X and Cys51Gly causing complete FSH deficiency) vs. those with partial pubertal development and sterility (mild mutation Tyr76X causing partial deficiency). Despite the use of different endpoints [estradiol in the rat granulosa cell assay (13) and cAMP in the homologous FSH bioassay (12, 15)], bioactive FSH was unmeasurable for all mutations studied. These findings suggest that neither of these available bioassays are sensitive enough to differentiate between FSH deficiency and very low levels of FSH (<1–2 IU/liter) seen in states of partial FSH deficiency. Our observations could also suggest that other modifying genes and/or environmental factors play a role in determining the phenotype of patients with isolated FSH deficiency.
We have characterized the first human FSHß mutation in patients with evidence of pubertal development. Although severe FSHß mutations result in absent puberty and sterility, this Tyr76X mutation permits a variable degree of pubertal development, but still causes infertility without treatment. This is also the first complete characterization of testicular histology in a male with isolated FSH deficiency, and it differs from that described in the male with the Cys82Arg mutation who had normal puberty and azoospermia (10). Photographs of the testes were not included in the report by these authors, who described the presence of Leydig cells and spermatogenic arrest (Sertoli cells were not mentioned). In our affected male with evidence of puberty, histological features included small (both in size and number) seminiferous tubules with reduced numbers of Sertoli cells and absent germ cells, as well as prominent Leydig cell hyperplasia. It is interesting that spermatogenic arrest, and now absent germ cells have been described in FSH-deficient men. Further studies of these interesting patients will be necessary to understand the true role of FSH in spermatogenesis.
Although FSHß gene mutations are considered to be relatively rare, their identification increases our understanding of normal reproductive physiology. FSH is thought to be essential for both steroid production necessary for puberty and gametogenesis in human males and females. Although there is great phenotypic similarity between the FSHß knockout mice and human FSHß mutants, fertility appears to be more severely affected in human mutants. Ovarian follicles in females do not progress beyond the antral stage (even if breast development occurs), whereas all of the human males described are azoospermic, whether or not puberty develops. It is quite possible that the full spectrum of FSHß gene mutations has not yet been realized, because women with ovulation disorders or men with oligospermia or azoospermia might possess still milder FSHß mutations. Future studies will be important to ascertain the true prevalence of FSHß mutations in humans.
Acknowledgments
We thank J. Larry Jameson and Eun Jig Lee (Northwestern University, Chicago, IL) for providing the pM2FSHß construct and helpful comments for mutagenesis.
Footnotes
This work was supported by Grant HD33004 from the U.S. Public Health Service (USPHS)-National Institute of Child Health and Human Development (to L.C.L.). Partial support for some of the analytical work was provided by USPHS Grant HD29164 (to P.M.S.).
Present address for W.W.: Millennium Pharmaceuticals, Inc., Cambridge, Massachusetts.
Abbreviations: CHO, Chinese hamster ovary; DES, diethylstibestrol; EIA, enzyme immunoassay; rhFSH, recombinant human FSH.
Received January 3, 2002.
Accepted April 23, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  A. Lofrano-Porto, G. B. Barra, L. A. Giacomini, P. P. Nascimento, A. C. Latronico, L. A. Casulari, and F. d. A. da Rocha Neves Luteinizing Hormone Beta Mutation and Hypogonadism in Men and Women N. Engl. J. Med., August 30, 2007; 357(9): 897 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Allan, A. Garcia, J. Spaliviero, M. Jimenez, and D. J. Handelsman Maintenance of Spermatogenesis by the Activated Human (Asp567Gly) FSH Receptor During Testicular Regression Due to Hormonal Withdrawal Biol Reprod, May 1, 2006; 74(5): 938 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P N Themmen An update of the pathophysiology of human gonadotrophin subunit and receptor gene mutations and polymorphisms Reproduction, September 1, 2005; 130(3): 263 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T R. Kumar What have we learned about gonadotropin function from gonadotropin subunit and receptor knockout mice? Reproduction, September 1, 2005; 130(3): 293 - 302. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Lamminen, P. Jokinen, M. Jiang, P. Pakarinen, H. Simonsen, and I. Huhtaniemi Human FSH{beta} subunit gene is highly conserved Mol. Hum. Reprod., August 1, 2005; 11(8): 601 - 605. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.-P. Zhang, T. Pakarainen, M. Poutanen, J. Toppari, and I. Huhtaniemi The low gonadotropin-independent constitutive production of testicular testosterone is sufficient to maintain spermatogenesis PNAS, November 11, 2003; 100(23): 13692 - 13697. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
