| |||||||||||||||||||||||||||||||
The gene responsible for this disorder, DAX1, is on the short arm of the X chromosome11 and encodes a 470-amino-acid member of the nuclear hormone receptor superfamily. The DNA-binding domain consists of amino acid repeats rather than the zinc fingers characteristic of other nuclear hormone receptors.11 It has no known ligand.11,12 DAX1 is expressed in the adrenal cortex, gonads, hypothalamus, and pituitary.13 Little is known about the function of DAX1 or the genes that it regulates. The majority of reported DAX1 mutations are nonsense or frame-shift mutations, suggesting that severe alterations are necessary to cause the clinical symptoms. The few reported missense mutations of the DAX1 gene are in the presumptive ligand-binding domain.1,7,9,10 No relation between the onset of adrenal insufficiency and the site of the mutation has been demonstrated.
We conducted molecular genetic and clinical studies in a kindred with X-linked adrenal hypoplasia congenita and hypogonadotropic hypogonadism. The two affected family members were brothers who were hemizygous for a DAX1 mutation; their unaffected mother was heterozygous for the mutation. The mutation, a nonsense mutation in the DNA-binding domain that should prevent synthesis of a functional protein, was also found in their maternal grandfather, who was unaffected. Their maternal aunt, who had isolated hypogonadotropic hypogonadism, was homozygous for the mutation. This homozygosity apparently resulted from gene conversion, the nonreciprocal transfer of DNA from one parental allele to the other. This family illustrates an extraordinary range of phenotypes associated with the same DAX1 mutation. Moreover, the apparently spontaneous gene conversion in the aunt illustrates a novel molecular cause of isolated hypogonadotropic hypogonadism and suggests a new molecular mechanism by which X-linked recessive disease may occur in a female.
Case Report
The proband (Subject III-3 in Figure 1) was seen at the age of 16 days because of vomiting, lethargy, dehydration, hyponatremia, and hyperkalemia. His genitalia were normal, but the scrotum was hyperpigmented. No 17-ketosteroids or pregnanetriol was detected in a 24-hour urine collection. He was treated with cortisone acetate and desoxycorticosterone acetate. He presented to our clinic at the age of 16 years with delayed puberty. Physical examination revealed small testes (volume, 4 to 6 ml each) and Tanner stage 3 pubic hair. Serum gonadotropin and testosterone concentrations were in the prepubertal range (Table 1). The boy was treated with testosterone.
|
|
The boys' maternal aunt (Subject II-5) was evaluated at the age of 16 years for delayed puberty and primary amenorrhea. She was given a diagnosis of hypogonadotropic hypogonadism and treated with estrogen and progestin. She never had symptoms or signs of adrenal insufficiency. At the age of 50 years, she was evaluated at our clinic. She was tall (174 cm), with eunuchoidal proportions (an arm span of 182 cm and a ratio of the upper to the lower segment of 0.68). There was no hyperpigmentation. Five weeks after the discontinuation of estrogen therapy, serum gonadotropin concentrations were low (Table 1) and did not increase in response to gonadotropin-releasing hormone. Adrenal function was normal both basally (Table 1) and in response to corticotropin (peak serum cortisol concentration, 27 µg per deciliter [745 nmol per liter]). Computed tomography of the abdomen revealed normal adrenal glands, and magnetic resonance imaging of the head revealed a small pituitary gland (4 mm in the largest dimension). Transvaginal ultrasonography revealed a small uterus and small ovaries (right, 1.1 by 1.2 by 1.2 cm; left, 1.0 by 0.9 by 1.0 cm).
Blood was collected from five other family members (the boys' sister, mother, maternal uncle, and maternal grandparents), none of whom had a history of adrenal insufficiency or delayed puberty. Adrenal function was normal in the grandfather (morning serum cortisol concentration, 18 µg per deciliter [497 nmol per liter]).
The protocol was approved by the institutional review board, and informed consent was obtained from each family member studied.
Methods
DNA Extraction, Amplification, and Sequencing of the DAX1 Gene
Genomic DNA was prepared from peripheral-blood leukocytes, skin, and urinary sediment according to standard procedures.14 The DAX1 gene was amplified by the polymerase chain reaction (PCR) with the use of previously described primer pairs.4 The PCR products were sequenced from both strands by the dideoxy method with an automated fluorescence sequencing system (model 373A, Applied Biosystems, Foster City, Calif.).15
Subcloning
Primers 3 and 44 were used to amplify DNA from peripheral-blood leukocytes from the proband's mother and maternal aunt. The PCR products were subcloned into pCR2.1 vector (Invitrogen, San Diego, Calif.). The resulting construct was used to transform Escherichia coli strain JM109 (Promega, Madison, Wis.). Both strands of positive clones were sequenced with the use of primers 3 and 4 according to the dideoxy method.
Sequence-Specific Oligonucleotide Hybridization
Genomic DNA was amplified by PCR with use of primer 4 in conjunction with primer 5'GCAGCATCCTCTACAGCTT3'.9 The concentration of sodium dodecyl sulfate in the hybridization solution was 0.2 percent. Filters were hybridized at 52°C for 20 hours. The wild-type and mutant oligonucleotides were 5'ggcgcgtggtggaccgctcct3' and 5'GGCGCGTGGTGAGACCGCTCCT3', respectively. After hybridization, each filter was washed three times at room temperature for 10 minutes and once at 72°C for 7 minutes in a solution containing 6x saline sodium citrate (1x saline sodium citrate is 0.15 M sodium chloride and 0.015 M sodium citrate) and 0.2 percent sodium dodecyl sulfate.
Hormonal Measurements
Plasma luteinizing hormone, follicle-stimulating hormone, estradiol, testosterone, and dehydroepiandrosterone sulfate were measured by radioimmunoassay by Covance Laboratories (Vienna, Va.). Plasma cortisol was measured by fluorescence polarization immunoassay (Abbott Laboratories, Abbott Park, Ill.). Plasma aldosterone, corticotropin, and renin activity were measured by radioimmunoassay by Mayo Medical Laboratories (Rochester, Minn.).
Results
The proband and his affected brother both had an adenine substituted for guanine in exon 1 of the DAX1 gene (Figure 2). This nonsense mutation introduces a stop codon at position 172. The boys' mother was heterozygous for this nonsense mutation, indicating a carrier state. Thus, as expected for an X-linked disorder, the mother was heterozygous and unaffected, whereas her affected sons were hemizygous for the mutation.
|
We then subcloned the aunt's genomic DNA obtained from lymphocytes. All 15 colonies had the mutation, and 6 had both the mutation and the polymorphism, confirming that two alleles were present in this XX female. One DAX1 allele had both the mutation and polymorphism, whereas the other allele had only the mutation (Figure 1).
Molecular analysis of DAX1 was also performed in the boys' maternal grandparents. Their grandfather carried the mutation but not the polymorphism, whereas their grandmother was heterozygous for the polymorphism and did not have the mutation (Figure 1). To search for evidence of mosaicism, we obtained genomic DNA from other tissues in the aunt, grandmother, and grandfather. In the aunt, analysis of lymphocytes, urinary sediment, and skin all revealed homozygosity for the DAX1 mutation and heterozygosity for the polymorphism. There was no evidence of mosaicism. In the grandmother and grandfather, the results of mutational analysis of DNA from urinary sediment were similar to those in leukocyte DNA. Subcloning analysis of the grandmother's genomic DNA was also performed; eight colonies did not have the nonsense mutation, and two had the polymorphism. In addition, no other base-pair substitutions were detected in the coding region and exon–intron boundaries of DAX1 in any of the family members studied.
To search for further evidence of mosaicism, we used sequence-specific oligonucleotide hybridization to analyze DNA samples from all available family members (including lymphocytes, urinary sediment, and skin from the aunt). In this technique, DNA is bound to nitrocellulose filters and hybridized under stringent conditions to radiolabeled oligonucleotide probes that either match the wild-type sequence or contain a specific single-base-pair substitution. We found no evidence of mosaicism in any of the family members (data not shown).
Discussion
We describe two boys with typical adrenal hypoplasia congenita (neonatal adrenal insufficiency with hypogonadotropic hypogonadism) and a nonsense mutation in the DAX1 gene. This mutation has previously been described in two related males with hypogonadotropic hypogonadism in whom adrenal insufficiency was diagnosed at six weeks and five years of age.1 As expected, our patients' mother was an unaffected carrier.
Two family members had unusual genotypic and phenotypic findings. The boys' unaffected grandfather was hemizygous for the nonsense mutation, indicating a lack of penetrance of the mutation. Their maternal aunt, who had isolated hypogonadotropic hypogonadism, was homozygous for the mutation and heterozygous for a nearby polymorphic marker. The presence of the polymorphism indicates that she inherited two different DAX1 alleles, one from each of her parents, and that the nonsense mutation in her paternal allele was probably introduced into the maternal allele through spontaneous gene conversion early in embryogenesis.
That the same mutation could result in two brothers with the complete syndrome, an unaffected grandfather, and an aunt with only hypogonadotropic hypogonadism is a remarkable discrepancy in genotype–phenotype relations. This discrepancy might be explained by the presence of undetected mosaicism in the grandfather and aunt. We cannot determine whether mosaicism explains the phenotype of the grandfather and aunt because we were not able to obtain adrenal, hypothalamic, or pituitary tissue from them. Alternatively, this variable expressivity might reflect other proteins that serve as DAX1 surrogates during development or epigenetic phenomena that modulate the phenotype of patients with DAX1 mutations.
Although gene conversion during mitosis of somatic cells should result in mosaicism, conversion early in embryogenesis could produce a person without mosaicism. At approximately three to four days of gestation, the majority of cells differentiate into the trophoblasts destined to become extraembryonic tissue, whereas less than 40 percent of cells evolve into the inner cell mass, which is destined to become embryonic tissue.16 Gene conversion before this unequal division of cells could produce a person without mosaicism.
Gene conversion represents one mechanism by which an X-linked recessive disease may occur in a female. Interallelic gene conversion has been described as a source of allelic diversity at the HLA loci, with a frequency of approximately 1 in 10,000 sperm.17 Gene conversion during mitosis involving the transfer of genetic material between maternal and paternal alleles, as may have occurred in our patient, has been described for the autosomal genes causing epidermolysis bullosa18 and Fanconi's anemia.19
Isolated hypogonadotropic hypogonadism may occur in families or sporadically and is five to seven times as common in males as in females.20 Males with this condition may have Kallmann's syndrome owing to a mutation of another X-chromosome gene, KAL1.21 Defects in KAL1 have been found in 52 percent of patients with X-linked cases of isolated hypogonadotropic hypogonadism22 and in 5 percent of patients with sporadic cases.23 Although DAX1 mutations have been thought to cause both adrenal insufficiency and hypogonadotropic hypogonadism, our findings suggest that alterations in the DAX1 gene or functionally related genes may cause isolated hypogonadotropic hypogonadism.
We are indebted to Howard P. Levy, M.D., Ph.D., for insightful discussions.
Source Information
From the Developmental Endocrinology Branch, National Institute of Child Health and Human Development (D.P.M., T.T., J.B., G.B.C.); and the U.S. Public Health Service (D.P.M., J.B.) — both in Bethesda, Md.
Address reprint requests to Dr. Merke at DEB/NICHD/NIH, Bldg. 10, Rm. 10N262, 10 Center Dr., MSC 1862, Bethesda, MD 20892-1862, or at deborah_merke{at}nih.gov.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
A) (solid arrow) in the proband's DNA. His mother was heterozygous for this nonsense mutation, indicating a carrier state. Only the mutation (no wild-type sequence) was found in his aunt, who was also heterozygous for a polymorphism 18 bp upstream from the mutation (open arrow). His grandmother did not have the nonsense mutation and was heterozygous for the polymorphism, whereas his grandfather had the mutation, but not the polymorphism. Automated fluorescence sequencing was performed with the use of PCR products as templates.