| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Pediatric Endocrinology Department and Centre of Reference for Pathologies in Hormonal Receptivity (R.C., N.B.-N.), and Departments of Genetics (A.G.), Pediatric Surgery (L.C.), and Pathology (A.C.), University Hospital, 49033 Angers, France; Molecular Biology and Hormonology Department (D.M., Y.M.), Debrousse Hospital, 69000 Lyon, France; and Department of Pediatric Hormonology and Metabolic Diseases (N.L.), Saint Vincent de Paul Hospital, 75014 Paris, France
Address all correspondence and requests for reprints to: Régis Coutant, Department of Pediatric Endocrinology, University Hospital, 4 rue Larrey, 49033 Angers CEDEX 01, France. E-mail: recoutant{at}chu-angers.fr.
| Abstract |
|---|
|
|
|---|
Objective: We report a SF-1 mutation that caused ambiguous genitalia associated with strikingly different hormonal phenotypes in two affected 46,XY children from the same family.
Methods: Hormonal evaluation included testosterone (T), anti-Mullerian hormone (AMH), inhibin B, FSH, and LH measurements during the first weeks of life, a period when physiological activation of the gonadotropin-gonadal system occurs. Direct DNA sequencing of the coding sequence of the SF-1 and the androgen receptor (AR) genes was performed.
Results: Both 46,XY children had ambiguous genitalia with no Mullerian structures and no adrenal insufficiency. The older child showed normal elevation of T (up to 7.6 nmol/liter, 2.2 ng/ml), AMH (504 pmol/liter, 70.6 ng/ml), inhibin B (245 pg/ml), FSH, and LH during the first weeks, which led to a presumptive diagnosis of partial androgen insensitivity syndrome. The AR sequence was, however, normal. In the second child, T, AMH, and inhibin B were low, suggesting gonadal dysgenesis. In both children and their mother, a c.536delC frameshift mutation in the SF-1 gene was found. This mutation terminates translation at position 295, removing the ligand-binding domain and the activation function 2 (AF-2) domain, a critical domain for SF-1 transactivating activity.
Conclusions: The usual markers of testis dysgenesis may be normal in 46,XY individuals with SF-1 mutation. Screening for SF-1 mutation should be performed in subjects with apparent partial androgen insensitivity syndrome and no mutation in the AR gene.
| Introduction |
|---|
|
|
|---|
Steroidogenic factor-1 (SF-1; Ad4BP/NR5A1) belongs to the NR5A subfamily of nuclear receptors (3). SF-1 is expressed in the adrenal gland, testis, ovary, pituitary, hypothalamus, spleen, and skin and regulates genes involved in adrenal and gonadal development, steroidogenesis, and reproduction (4, 5, 6). In mice, homozygous targeted disruption of the SF-1 gene causes adrenal and gonadal agenesis and impaired gonadotropin expression, resulting in postnatal death due to severe adrenal insufficiency (5), whereas heterozygous animals have a milder phenotype (7). SF-1 knockout also causes abnormalities of the ventromedial hypothalamic nucleus, the control center for satiety and feeding, which suggests that SF-1 may have broader roles in the control of metabolism and obesity (8). In humans, two SF-1 mutations have been described that resulted in adrenal insufficiency and 46,XY sex reversal (female phenotype), no palpable gonads, and presence of Mullerian structures, thus suggesting a severe defect in adrenal and gonadal development (9, 10, 11). Other reported 46,XY patients with heterozygous SF-1 mutations had normal adrenal function and ambiguous genitalia due to dysgenetic testis (12, 13, 14, 15), with (14, 15) or without Mullerian structures (12, 13, 15). However, in all cases described to date, a severe defect in testosterone production has consistently been found owing to dysgenetic gonads (9, 10, 11, 12, 13, 14, 15).
We here describe two previously unreported 46,XY siblings with heterozygous frameshift mutations in the SF-1 gene; they had ambiguous genitalia with no adrenal insufficiency, one with normal testosterone, AMH, and inhibin B production up to the age of 4 months [which initially led to a presumptive diagnosis of partial androgen insensitivity syndrome (PAIS)], the second with evident gonadal dysgenesis.
| Subjects and Methods |
|---|
|
|
|---|
Inhibin B was measured by means of a solid-phase sandwich assay (Serotec, Oxford, UK) as previously described (16). Inhibin A exhibited a 1% cross-reactivity in the inhibin B assay. Intraassay precision was 7.4 and 4.2% at levels of 44 and 225 pg/ml, respectively. AMH was measured by means of a solid-phase sandwich assay using reagents provided by Immunotech/Beckman Coulter (Villepinte, France) (16, 17). There was no cross-reactivity of related proteins including TGFß. Intraassay precision at the levels of 245 and 1106 pmol/liter were 5.1 and 4.9%, respectively. The sensitivity was 0.7 pmol/liter. RIAs were used to measure plasma concentrations of testosterone (CIS Biointernational, Gif sur Yvette, France; nonautomated assay), cortisol, and aldosterone (Immunotech/Beckman Coulter; Diagnostic Systems Laboratories, Webster, TX). LH and FSH were measured by immunoradiometric assays (Immunotech/Beckman Coulter; Coat-a-Count, Diagnostic Products Corp., Los Angeles, CA). The reference values for testosterone, inhibin B, AMH, FSH, and LH in boys aged 0–1 yr have been published (14, 17, 18, 19).
Amplification and sequencing of the androgen receptor (AR) gene and the SF-1 gene
Peripheral blood samples were collected from patient and parents after informed consent had been obtained, according to our institutional guidelines. Genomic DNA was isolated from whole-blood leukocytes using a DNA extraction kit (Nucleon BACC3; Amersham Biosciences, Arlington Heights, IL). Primers were designed to amplify all exons and exon-intron boundaries of the AR gene, according to the published sequence (20, 21). Linkage analysis was performed by determination of the number of CAG repeats within the first exon of the AR gene. Primers were also designed to amplify all exons, exon-intron boundaries, and part of the promoter region of the SF-1 gene according to the published sequences (GenBank accession numbers NC_000009.10 and AB009577), and PCR conditions were determined for each fragment as previously described (14). Direct sequencing was performed using the ABI Prism BigDye Terminator sequencing kit on a 3130 automated sequencer (Applied Biosystems, Foster City, CA).
| Results |
|---|
|
|
|---|
The index case was a baby with ambiguous genitalia (birth weight, 2.9 kg, full term), with a perineal urogenital sinus, a genital tubercle measuring 12 mm in length, palpable gonads (12 x 5 mm) in the bilateral inguinal regions, and a hypotrophic scrotum or corrugation of major labia. Karyotype was 46,XY. The diagnosis of PAIS was initially thought to be plausible because testosterone, AMH, inhibin B, FSH, and LH levels were in the normal range from birth up to 4 months (Table 1
), no uterus was seen on pelvic ultrasound, and genitography revealed a vaginal pouch (Fig. 1
). However, sequencing of the coding regions of the AR gene showed no mutation. Because the decision was made to rear this newborn as a girl, no testosterone treatment was implemented, and feminizing genitoplasty and gonadectomy were performed at 6 months. Histology of the gonad showed normal infantile testicular structure; seminiferous tubules lined with Sertoli cells contained some spermatogonia, and interstitium showed small clusters of Leydig cells (Fig. 1
).
|
|
Molecular study of the SF-1 gene
Because the clinical and biological presentation of the second child suggested gonadal dysgenesis, sequencing of the SF-1 gene was performed in both children and the parents. A heterozygous deletion of the cytosine at position 536 in exon 4 was found in the two children and their mother (c.536delC) (Fig. 2
, A and B). This mutation replaces the proline at position 179 by a histidine, causes a frameshift, and prematurely terminates translation at position 295, 116 codons after the codon encoding for proline 179 (p.P179HfsX116 mutation). This mutation should produce a protein with normal DNA-binding domain (DBD) and A box but lacking the normal C-terminal half of the hinge region, the ligand-binding domain (LBD) and the activation function 2 (AF-2) domain (Fig. 2C
). The hormone measurements performed in the mother when it was found that she carried the SF-1 mutation were indicated in Table 1
.
|
| Discussion |
|---|
|
|
|---|
The c536delC frameshift mutation found here alters the SF-1 protein sequence from the 179th amino acid residue in the hinge region and terminates translation at position 295, removing the LBD and the AF-2 domain, whereas the DBD and the A box are preserved. In humans, seven SF-1 mutations in 46,XY patients with disorders of sex development with no adrenal insufficiency have been previously published, which were frameshift, nonsense, or missense mutations (12, 13, 14, 15). The most distal mutation was a heterozygous deletion of 8 bp in exon 6 that caused a frameshift that altered the protein sequence from the 352nd amino acid and prematurely terminated translation at position 378, upstream of the AF-2 domain (12). The mutant protein, which was less truncated than in our case, showed a decrease in DNA binding (although the DBD was intact) and a dominant-negative effect in three different cell lines (mouse Leydig tumor cells, human embryonic kidney cells, and mouse adrenocortical tumor cells) but had a positive effect in human adrenocortical tumor cells. The AF-2 domain is necessary for SF-1 transactivating activity (23) and interactions with numerous coactivators (23, 24), many of which are involved in adrenal and gonadal steroidogenesis. In addition to removing the AF-2 domain, the frameshift mutation in our observation replaced the serine residue at 203, the phosphorylation of which is essential for SF-1 activity, both steroidogenic enzymes and AMH gene transcription activation (23, 25). These observations indicate that if the p.P179HfsX116 mutant protein is expressed, a severe impairment of its function should be expected.
Although the clinical and biological features in the second child of the family described here were consistent with the biological diagnosis of gonadal dysgenesis, this was not evident in the index case; serum testosterone, LH, FSH, inhibin B, and AMH levels showed a physiological increase between birth and 4 months (17, 18, 19), a window of opportunity in which the diagnosis of hypogonadism and/or hypogonadotropism can usually be established (26). The clinical, hormonal, echographic, radiological, and histological features of this child suggested a diagnosis of partial PAIS. Testosterone levels, albeit not elevated, were in the range of those measured in PAIS subjects (27), as were AMH levels (17), in accord with the lack of uterus seen on pelvic ultrasound. A vaginal pouch was present, as in 90% of subjects with PAIS (21). Because sequencing of the coding regions of the AR gene showed no mutation, this child had initially received no diagnosis, as in the case of about 50% of male ambiguous subjects (1, 21). One should note that FSH levels were slightly higher than those measured in normal male or subjects with PAIS of similar age (18, 19, 27); this could be considered as an indicator of subtle Sertoli cell dysfunction. However, SF-1 mutation was searched for only after the birth of the second child.
Ambiguous genitalia in our cases, as in the nine reported 46,XY subjects with SF-1 mutation, indicate that testosterone production from Leydig cells was defective at least during the critical period of formation of the external genitalia (28). This defect was likely variable, because the patients differed in the extent of virilization, from sex reversal (9, 10) to ambiguous genitalia (12, 13, 14, 15). Our observation showed that normal testosterone production during the first weeks of life is possible, whereas the individuals previously observed shortly after birth had low testosterone levels (10, 14, 15). Because gonadectomy was performed at 6 months of age in our cases, we cannot exclude the possibility that Leydig cell function would have later declined. We did not perform human chorionic gonadotropin tests in our infants, because we assumed that the basal hormone measurements during the first weeks of life provided an adequate characterization of gonad function (26); however, human chorionic gonadotropin tests would have helped to describe the functional consequences of SF-1 mutation more exactly.
Variability in Sertoli cell function owing to SF-1 mutation has already been suggested from several observations. In affected 46,XY subjects, Mullerian structures were either present (9, 10, 14, 15) or absent (13, 14, 15) (our children), thus indicating, respectively, defective and efficient AMH production during the first trimester of gestation. Serum AMH levels, measured in three cases, were low (14, 15). In our observation, the index case had no Mullerian structures, in agreement with the normal AMH levels, whereas the second child had undetectable AMH levels, in apparent contrast with the lack of Mullerian structures. Although this should be interpreted with some caution, because Mullerian structures may not be seen on pelvic ultrasound but found at laparotomy (14), it underlines the variability in the degree of gonadal dysgenesis in affected 46,XY subjects. Consistently, gonad histology, when performed, has shown variable alterations, from normal histology (15) (our cases) to subnormal testicular structure (14, 15) and severe gonadal dysgenesis (9, 10, 12, 13, 15).
In mice, SF-1 has been shown to be essential for pituitary gonadotrope function (5, 29). In humans, variability in gonadotrope function in the presence of defective testosterone production has been shown in several observations of SF-1 mutation. LH and/or FSH levels were either elevated (9, 12, 13, 14, 15), suggesting appropriate gonadotrope response to gonadal dysgenesis, or normal/low (10, 15) (our cases), in accord with the gonadotrope defect found in the murine model of SF-1 inactivation (5, 29).
The mechanisms underlying the variable effect of SF-1 gene mutations are largely unknown. Heterozygous mutations in SF-1 are sufficient to cause severe phenotypes. The combination of haploinsufficiency of functional domains of variable importance (9, 11, 13, 14, 15) with a possible dominant-negative effect of the mutant protein (12) might contribute to this variability. In addition, our observation showed that the same SF-1 mutation caused different alterations in Leydig and Sertoli cell functions in two subjects, suggesting that modulating factors may influence SF-1 protein activity. SF-1 is known to interact with the promoter region of several steroidogenic enzyme genes, together with numerous coactivators, and with the promoter region of the AMH gene, together with SOX9, WT1, DAX-1, and GATA4 and other coactivators (23); it is tempting to speculate that these proteins may partly compensate for the effect of the loss of functional domains of SF-1. In addition, bioactive sphingolipids and phospholipids have recently been shown to be endogenous ligands for SF-1. Whereas sphingosin has been found to antagonize SF-1 activity, it is likely that multiple bioactive lipids are ligands for SF-1 and that these lipids act differentially to control SF-1 activity in a context-dependent manner (30). Additional functional studies will be needed to understand these points.
In conclusion, although SF-1 gene loss-of-function mutations are a known cause of dysgenetic testis, our observation showed that affected 46,XY subjects may have apparent normal testosterone and AMH production in the neonatal period, which may mistakenly lead to a presumptive diagnosis of PAIS. Notably, in a cohort of 173 individuals diagnosed with partial insensitivity syndrome, mutations in the AR gene were found in only 28% of the cases (31). We therefore suggest that screening for SF-1 mutation should be performed in subjects with apparent PAIS and no mutation in the AR gene.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online May 8, 2007
1 R.C. and D.M. contributed equally to this study. ![]()
Abbreviations: AF-2, Activation function 2; AMH, anti-Mullerian hormone; AR, androgen receptor; DBD, DNA-binding domain; LBD, ligand-binding domain; SF-1, steroidogenic factor-1.
Received January 5, 2007.
Accepted May 2, 2007.
| References |
|---|
|
|
|---|
gene mutation: an unusual expression of McCune-Albright syndrome in a prepubertal boy. J Clin Endocrinol Metab 86:1778–1781This article has been cited by other articles:
![]() |
B. Kohler, L. Lin, I. Mazen, C. Cetindag, H. Biebermann, I. Akkurt, R. Rossi, O. Hiort, A. Gruters, and J. C Achermann The spectrum of phenotypes associated with mutations in steroidogenic factor 1 (SF-1, NR5A1, Ad4BP) includes severe penoscrotal hypospadias in 46,XY males without adrenal insufficiency Eur. J. Endocrinol., August 1, 2009; 161(2): 237 - 242. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Lourenco, R. Brauner, L. Lin, A. De Perdigo, G. Weryha, M. Muresan, R. Boudjenah, G. Guerra-Junior, A. T. Maciel-Guerra, J. C. Achermann, et al. Mutations in NR5A1 Associated with Ovarian Insufficiency N. Engl. J. Med., March 19, 2009; 360(12): 1200 - 1210. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Pelusi, Y. Ikeda, M. Zubair, and K. L. Parker Impaired Follicle Development and Infertility in Female Mice Lacking Steroidogenic Factor 1 in Ovarian Granulosa Cells Biol Reprod, December 1, 2008; 79(6): 1074 - 1083. [Abstract] [Full Text] [PDF] |
||||
Read all eLetters
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |