| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
University Department of Growth and Reproduction (T.H.J., J.M., K.M.M.), Rigshospitalet, DK-2100 Copenhagen, Denmark; Laboratoire de Biologie Endocrinienne et Moléculaire (D.M., Y.M., M.G.F.), Hôpital Debrousse, 69322 Lyon Cedex 05, France; and University Department of Clinical Physiological and Nuclear Medicine (H.D.-P.), Glostrup University Hospital, DK-2600 Glostrup, Denmark
Address all correspondence and requests for reprints to: Trine Holm Johannsen, University Department of Growth and Reproduction, GR-5064, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. E-mail: trinejohannsen{at}rh.dk.
| Abstract |
|---|
|
|
|---|
At referral, increased 17
-hydroxyprogesterone associated with premature pubarche at first suggested a nonclassical 21-hydroxylase deficiency. Sequencing of the CYP21 gene showed both girls only heterozygotes (V281L mutation). This result, combined with SW in infancy, suggested a 3ß-HSD deficiency because of increased dehydroepiandrosterone sulfate levels. Further hormonal studies showed markedly elevated
5-steroids, in particular 17
-hydroxypregnenolone greater than 100 nmol/liter (the clue to the diagnosis) and elevated
5-/
4-steroid ratios. Sequencing of the type II 3ß-HSD gene documented that both girls were compound heterozygotes for T181I and 1105delA mutations. Retrospectively, elevated levels of 17
-hydroxyprogesterone were found on blood spots from Guthries test.
There is no previous report of the combination of SW and premature pubarche due to mutations in the type II 3ß-HSD gene. Because neonatal diagnosis could have prevented life-threatening crises in these girls, this report further supports the benefits for neonatal screening for congenital adrenal hyperplasia whatever the etiology.
| Introduction |
|---|
|
|
|---|
3ß-HSD deficiency is classified into classical and nonclassical forms. Classical 3ß-HSD deficiency is due to mutations in the type II 3ß-HSD (HSD3B2) gene and is characterized by varying degrees of salt wasting (SW). In males it can be associated with varying degrees of pseudohermaphroditism, whereas females exhibit normal sexual differentiation or mild virilization (2, 3). The SW form of 3ß-HSD deficiency is usually diagnosed in the first months of life, in contrast to the non-SW form, which may be diagnosed anytime before puberty (4, 5).
A nonclassical 3ß-HSD deficiency has been described in children with premature pubarche (6) and women with hyperandrogenism (7). Several molecular studies have not been able to demonstrate mutations in either type I or type II 3ß-HSD genes in these phenotypes (8, 9). In hirsute women, nonclassic 3ß-HSD deficiency has been documented by provocative testing (ACTH stimulation test) only (7). Mutations in the HSD3B2 gene were not found in these women with mild to moderate elevations in ACTH-stimulated 17
-hydroxypregnenolone levels (10, 11). However, mutations of the HSD3B1 gene have been found in only a few children with premature pubarche (12). We here report two Danish siblings with considerably delayed diagnosis of SW CAH due to classical 3ß-HSD deficiency.
| Patients and Methods |
|---|
|
|
|---|
Sibling 1 was hospitalized at 20 months of age with a condition of severe sepsis and seizure. Investigations showed plasma sodium 124 mmol/liter (normal range 136146); plasma potassium 5.5 mmol/liter (normal range 3.85.6); plasma C-reactive protein initially 922 mmol/liter, increasing to 3210 mmol/liter (normal range 094); and plasma urea nitrogen 17.5 mmol/liter (normal range 1.85.4). No causes of infection were found, adrenal insufficiency was not suspected, and the sibling was discharged after successful treatment with antibiotics. The following years went uneventful until she was referred to our department at 7 yr 7 months of age because of premature pubarche, slight growth acceleration from a SD score (13) of 0 to +1.6 SD score over 5 yr and advanced bone age (10.7 yr). At examination, she presented with pubic hair Tanner stage 34 but no signs of virilization, clitoromegaly, or labial fusion. Ultrasound showed normal female infantile internal genitalia and eliminated suspicion of abdominal tumor.
Sibling 2, aged 3 yr 8 months, revealed no clinical signs of androgen excess at systematic examination, in particular no genital ambiguity. However, medical history revealed a previous episode with a sepsis-like picture at 19 months of age. Hospital records from this admission documented plasma sodium 124 mmol/liter (normal range 136146); plasma potassium 6.8 mmol/liter (normal range 3.24.7); plasma C-reactive protein initially 1202 mg/liter, increasing to 2448 mg/liter (normal range < 10); plasma urea nitrogen 8.6 mmol/liter (normal range 2.57.5); and basal plasma cortisol (0800 h) of 60 nmol/liter. No causes of infection were found, and after treatment with antibiotics, she was discharged, and the following years were uneventful.
The family was Caucasian and the parents not consanguineous. An informed consent was obtained from the parents for the hormonal studies and a written informed consent for the molecular studies in the family.
Hormone analyses
Both serum testosterone (Coat-a-Count, Diagnostic Products Corp., Los Angeles, CA) and serum estradiol (Pantex, Santa Monica, CA) were determined by RIA. Serum LH and serum FSH were measured by time-resolved immunofluorometric assays (Delfia, Wallac, Turku, Finland). Dehydroepiandrosterone sulfate (DHAS) was measured by RIA directly on diluted serum. Other unconjugated steroid hormones were measured by specific RIAs after purification on Celite chromatography (14). Plasma ACTH was measured using a RIA (Nichols Institute Diagnostics, San Clemente, CA) in basal conditions.
ACTH stimulation test.
Plasma cortisol and serum 17
-hydroxyprogesterone (17-OHP) were measured before (time 0) and 30 and 60 min after an iv injection of 0.25 mg tetracosactid (Synacthen, Novartis Healthcare, Copenhagen, Denmark).
Plasma renin concentration was determined by a method using the principle of antibody trapping (15), as modified by Millar et al. (16).
Urine steroid metabolites
All individual urinary steroids were determined by gas chromatography. The identification and values of
5-pregnenetriol, pregnanetriol, and cortisol metabolites were confirmed by mass spectrometry (17).
Molecular studies
Selective amplification of HSD3B2 fragments was performed as described (18), using three different primer pairs for amplification of the coding region and the exon-intron splicing junction boundaries. Direct sequencing of PCR products was done as described (18, 19). PCR products were purified with microspin S400-HR columns (Pharmacia Biotech, Uppsala, Sweden) to remove salt, residual primers, and unincorporated deoxynucleotide triphosphates. Approximately 80100 ng of PCR products were directly sequenced using the AmpliTaq FS dye terminators kit (Applied Biosystems, Foster City, CA). Each exon was sequenced on both strands, using either initial primers or internal primers especially for the exon 4. After 25 cycles in 9600 GeneAmp (Applied Biosystems) (30 sec at 95 C and 4 min 30 sec at 60 C), the reaction products were purified on Sephadex G50 microspin columns, dried under vacuum, and dissolved in 4 µl of a (5:1) formamide/EDTA mix. The electrophoresis was performed with a 7% acrylamide/bis acrylamide 19/1 sequencing gel during 10 h with a 373A model automatic sequencer, and the data were analyzed using Sequed software (Applied Biosystems).
Filter paper blood samples
Retrospective neonatal screening was performed by measuring 17-OHP by Delfia (Wallac) on a spot of capillary blood taken postpartum d 5 for neonatal metabolic and thyroid screenings.
| Results |
|---|
|
|
|---|
Primary hormone results in sibling 1 showed a serum testosterone concentration of 1.62 nmol/liter (normal range 0.111.1),
4-androstenedione (
4-A) 4.81 nmol/liter (normal range 0.774.3), and DHAS 18,000 nmol/liter (normal range 2703,000). Plasma ACTH was 40 pmol/liter (normal range 211). An ACTH test revealed maximal plasma cortisol concentration of 245 nmol/liter (normal response > 500 nmol/liter), whereas serum 17-OHP level was 71 nmol/liter (normal range < 2.8), not increasing during the test. Furthermore, in sibling 1 plasma renin concentration was 271 mIU/liter (normal range 660).
An ACTH test was not performed in sibling 2 before treatment with hydrocortisone. Because of fever she was hospitalized, administered hydrocortisone (Solu-Cortef 14 mg/m2 x 4). After 1-d treatment with hydrocortisone, her androgen hormonal status was investigated, revealing levels of serum testosterone concentration of 1.24 nmol/liter, serum
4-A of 3.22 nmol/liter, and DHAS of 8,600 nmol/liter. Serum 17-OHP was 74 nmol/liter. After treatment with hydrocortisone for 4 months, plasma renin was 342 mIU/liter, and plasma ACTH was 2 pmol/liter.
Sibling 1 presented with abnormal urine excretion of steroid metabolites, compared with urine excretion of 32 Danish healthy girls aged 5.811.8 yr with breast development Tanner stage 1 (shown in parentheses): Sibling 1 presented with
5-pregnenetriol being 12.88 mg per 24 h (median 0.00, range 0.000.06) and dehydroepiandrosterone (DHA) metabolites being 11.65 mg per 24 h (median 0.00, range 0.000.04). Metabolites of 17-OHP were 15.27 mg per 24 h (median 0.055, range 0.000.50), corticosterone 10.69 mg per 24 h (median 0.13, range 0.000.28), and cortisol 2.77 mg per 24 h (median 0.54, range 0.121.37). Urine analysis of sibling 2 was not performed.
To establish a firm diagnosis, a new ACTH stimulation test was performed in both siblings after 3 d off hydrocortisone and fludrocortisone (Table 1
). The
5-steroids were highly elevated at baseline, without further significant increase, in particular 17
-hydroxypregnenolone,
5-OHP (275 and 210 nmol/liter, respectively). Also the
4-steroids, i.e. progesterone, 17-OHP,
4-A, and testosterone, were highly increased in basal conditions. None of the
4-steroids or cortisol did increase further during the ACTH test. In sibling 1, cortisol levels were within the normal range for age, whereas the response in sibling 2 was below normal range.
|
First, because premature pubarche was associated with an increased 17-OHP (71 nmol/liter), a nonclassical form of 21-hydroxylase deficiency was suggested, leading to the sequencing of the CYP21 gene. Both girls and the father were heterozygotes for the V281L mutation.
To confirm the diagnosis of 3ß-HSD deficiency, the four exons and the exon/intron boundaries of the HSD3B2 gene were sequenced in both sisters. Two mutations were identified in exon 4 (Fig. 1
). The mutation 1105delA was a deletion of the first nucleotide A of codon 368 located at the end of the coding sequence. This mutation is a frameshift mutation, abolishing the stop codon and should be rendering a protein enlarged by 124 amino acids. The other mutation T181I was a nucleotide substitution from C to A in codon 181, changing a conserved hydrophilic amino acid, threonine (ACT), to a hydrophobic amino acid, isoleucine (AAT). The mother was heterozygote for the mutation T181I and the father for 1105delA. Thus, the two sisters were compound heterozygous for T181I1 and 1105delA.
|
Retrospective analysis of capillary blood collected for neonatal screening of phenylketonuria and hypothyroidism on d 5 after birth showed elevated levels of 17-OHP, e.g. 117 nmol/liter for sibling 1 and 143 nmol/liter for sibling 2 (normal range < 75 nmol/liter).
Evolution
Substitutive replacement therapy was started with hydrocortisone with salt supplementation in the food. In both girls, although blood electrolytes were normal, fludrocortisone treatment was added 5 months later because of the quite high renin levels revealing a compensated salt loss. During follow-up treatment, the
5-steroids, i.e. in particular
5-OHP and DHAS, decreased drastically (although not yet to normal), as illustrated in Table 2
in a recent check-up given as example. Substitutive replacement therapy at this time was hydrocortisone (10 mg in four doses in sibling 1 and 6.25 mg in three doses in sibling 2) and fludrocortisone (0.06 mg in two doses in sibling 1 and 0.05 mg in two doses in sibling 2).
|
| Discussion |
|---|
|
|
|---|
The diagnosis of 3ß-HSD deficiency was made primarily on the basis of the elevation of
5-steroid hormones before and after ACTH stimulation test, subsequently confirmed by urinary analysis and proved by molecular biology.
The especially high levels of
5-OHP, but also of DHA and DHAS, as well as the relatively higher increase in
5-OHP, compared with that of 17-OHP, were of diagnostic value. The best biological criterion of 3ß-HSD deficiency is the finding of
5-OHP levels above 100 nmol/liter whether in basal or ACTH-stimulated condition (20).
The relatively high levels of 17-OHP,
4-A, and testosterone are explained by normal activity of the peripheral isoenzyme HSD3B1, capable of metabolizing elevated levels of
5-steroids into
4-steroids. Levels of
5-steroids did not increase much during the ACTH test because endogenous ACTH already was high. The elevated levels of renin indicated a partial defect in the biosynthesis of aldosterone, explaining the salt loss crises in stress conditions.
The 24-h urine steroid profile showed high concentrations of
5-OHP and DHA metabolites, confirming the diagnosis of 3ß-HSD deficiency. The high levels of 17-OHP and corticosterone metabolites were explained by the normal activity of HSD3B1.
The findings of elevated 17-OHP levels associated with salt loss were suggestive of 21-hydroxylase deficiency, but sequencing of the CYP21 gene demonstrated only heterozygosity for the Val281Leu mutation in the girls and the father. The Val281Leu mutation is a mutation leaving a residual activity (2050%) and usually associated with a nonclassical form of 21-hydroxylase deficiency (21). In addition, the absence of virilization in combination with a severe salt loss is rare in 21-hydroxylase deficiency, except in patients compound heterozygous for the I172N point mutation in CYP21 and an additional severe mutation, in which salt loss occurs only during stressful events (22, 23). Thus, 21-hydroxylase as the primary etiology of symptoms in the siblings was eliminated.
The correct etiological diagnosis was confirmed by molecular study of the HSD3B2 gene, revealing compound heterozygosity for 1105delA and T181I mutations in the siblings. Clinical and biological data suggested SW. Nevertheless, some residual activity must have been present to explain the unusually late onset of salt crises (20 and 19 months of age), which were precipitated by serious infections.
The frameshift mutation 1105 delA results in a longer protein with 496 amino acids, compared with the wild-type HSD3B2 protein with 372 amino acids. This protein differs from the wild-type enzyme by its last five amino acids and an additional C-terminal extremity of 124 amino acids. This gene product can be expected to have retained lower enzyme activity than the wild type, as reported by Pang et al. (24). In vitro, this reported Stop373C mutation, which results in a protein containing the entire sequence of human HSD3B2 and an additional 95 amino acids long C-terminal, had some residual activity in intact cells but none in homogenates, and it was not detectable on Western blot. Our novel 1105 delA mutant protein is expected to be more unstable due to its longer C-terminal extremity (124 vs. 95 amino acids) and to have a weaker residual activity due to the different nature of the last five amino acids (368372).
The missense T181I mutation should be considered a severe mutation because the phenotype was a classical form of 3ß-HSD deficiency associated with increased renin and clinical salt crisis during stress. Threonine 181 has been conserved in all species (8) and located in a putative substrate-binding domain (25, 26). Two other mutations, P186L and L173R, have previously been described in this region. The P186L mutation was associated with a severe form with salt loss, and the mutant protein has no activity (27). The L173R was associated with male pseudohermaphroditism with normal renin (28) and has in vitro a residual activity (20). These data suggested that threonine 181 plays a role in substrate binding.
It was possible to detect an increased level of 17-OHP in a filter blood sample taken 5 d postpartum. There is one previous report on 3ß-HSD deficiency detected by neonatal screening of 21-hydroxylase deficiency (23), and recently a similar case has been detected (Morel, Y., unpublished data). Neonatal screening for CAH is not a routine in Denmark, but it could have prevented the serious, and potentially life-threatening, SW crises. Thus, our case report supports not only the beneficial effect of neonatal screening for 21-hydroxylase deficiency but also its ability to detect less frequent enzymatic defects in CAH.
The further medical management of these girls follows the principles of hydrocortisone and fludrocortisone replacement as for all other nonhypertensive forms of CAH, except that monitoring of hormonal control requires periodic measurements of ACTH,
5-OHP, DHA, and DHAS (29).
To our knowledge there are only two reports on pubertal development in genetic females with severe 3ß-HSD deficiency. One patient presented with minimal breast development at 14.7 yr of age and required estrogen replacement therapy (30, 31). Without estrogen supplementation menses ceased, and she developed ovarian cysts (32). The other patient underwent spontaneous breast development at the age of 89 yr, followed by menarche and evidence of progesterone secretion (32). The mutations in the HSD3B2 genes of these patients were different from the mutations described in our report. Thus, we cannot predict the possible requirement for sex hormone replacement in our patients. Sibling 1 has started puberty with breast development (Tanner stage 2) at last examination.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online January 25, 2005
1 T.H.J. and D.M. contributed equally to this work and should both be considered first authors. ![]()
Abbreviations:
4-A,
4-Androstenedione; CAH, congenital adrenal hyperplasia; DHA, dehydroepiandrosterone; DHAS, dehydroepiandrosterone sulfate; 3ß-HSD, 3ß-hydroxysteroid dehydrogenase; HSD3B1, 3ß-HSD isoenzymes, type I; HSD3B2, 3ß-HSD isoenzyme type II; 17-OHP, 17
-hydroxyprogesterone; SW, salt wasting.
Received July 14, 2004.
Accepted January 18, 2005.
| References |
|---|
|
|
|---|
5-
4-isomerase in human adrenals and gonads. Mol Endocrinol 5:11471157[Abstract]
5-
4 isomerase deficiency. In: Mason JI, ed. Genetics of steroid biosynthesis and function. Modern Genetics Series. Newark, NJ: Harwood Academic Publishers; 209258
5-steroids. Fertil Steril 74:553557[CrossRef][Medline]
5-steroid dehydrogenase. J Biol Chem 268:1850718512
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |