Differential gene-expression patterns in genital fibroblasts of normal males and 46,XY females with androgen insensitivity syndrome: evidence for early programming involving the androgen receptor

Background  

Androgen insensitivity syndrome (AIS) comprises a range of phenotypes from male infertility to complete feminization. Most individuals with AIS carry germline mutations of the androgen receptor (AR) that interfere with or ablate its function. As genital fibroblasts retain expression of the AR in vitro, we used genital skin fibroblasts from normal males and 46,XY females with complete AIS due to known AR mutations to gain insights into the role of the AR in human genital differentiation.     

Androgen insensitivity in forty-nine patients: classification based on clinical and androgen receptor phenotypes

Androgen receptor binding was studied in genital skin fibroblasts from 49 patients with androgen insensitivity syndrome (AIS) classified as complete (CAIS) or partial (PAIS) based on the clinical phenotype. The majority (64%) of CAIS and a minority (7%) of PAIS patients were receptor negative. Only 3 receptor-positive AIS cell strains of 30 studied failed to show an increase in specific receptor binding after prolonged androgen exposure in vitro. The gene coding for the androgen receptor in such patients appears intact.     

Replacement of arginine 773 by cysteine or histidine in the human androgen receptor causes complete androgen insensitivity with different receptor phenotypes

We have discovered two different point mutations in a single codon of the X-linked androgen-receptor (AR) gene in two pairs of unrelated families who have complete androgen insensitivity (resistance) associated with different AR phenotypes in their genital skin fibroblasts. One mutation is a C-to-T transition at a CpG sequence near the 5' terminus of exon 6; it changes the sense of codon 773 from arginine to cysteine, ablates specific androgen-binding activity at 37 degrees C, and eliminates a unique KpnI site at the intron-exon boundary. The other mutation is a G-to-A transition that changes amino acid 773 to histidine and eliminates an SphI site. This mutant AR has a normal androgen-binding capacity at 37 degrees C but has a reduced affinity for androgens and is thermolabile in their presence. Transient transfection of COS cells with cDNA expression vectors yielded little androgen-binding activity at 37 degrees C from Arg773Cys and abundant activity with abnormal properties from Arg773His, thereby providing the pathogenicity of both sequence alterations. This conclusion coincides with the following facts about evolutionary preservation of the position homologous to Arg773 in the AR: it is occupied by Arg or lysine in the progesterone, glucocorticoid, and mineralocorticoid receptors, and it is within a 14-amino-acid region of their steroid-binding domains that share approximately 85% amino acid identity.     

Ketoconazole binds to the human androgen receptor.

Ketoconazole, an imidazole anti-fungal agent, has often produced features of androgen deficiency including decreased libido, gynecomastia, impotence, oligospermia, and decreased testosterone levels, in men being treated for chronic mycotic infections. Based on these potent effects on gonadal function in vivo as well as previous work in vitro demonstrating affinity of ketoconazole for receptor proteins for glucocorticoids and 1,25(OH)2 vitamin D3 and for sex steroid binding globulin (SSBG), the binding of ketoconazole to human androgen receptors (AR) in vitro was also examined. Ketoconazole competition with [3H]methyltrienolone (R1881) for androgen binding sites in dispersed, intact cultured human skin fibroblasts was determined at 22 degrees C. Fifty percent displacement of [3H]R1881 binding to AR was achieved by 6.4 +/- 1.8 (SE) x 10(-5) M ketoconazole. Additional binding studies performed with ketoconazole in the presence of increasing amounts of [3H]R1881 showed that the interaction of ketoconazole with AR was competitive when the data were analyzed by the Scatchard method. It should be noted, however, that the dose of ketoconazole required for 50% occupancy of the androgen receptor is not likely to be achieved in vivo, at least in plasma. Finally, androgen binding studies performed with other imidazoles, such as clotrimazole, miconazole, and fluconozole, revealed that in this class of compounds only ketoconazole appears to interact with the androgen receptor. Ketoconazole appears to be the first example of a non-steroidal compound which binds competitively to both SSBG and multiple steroid hormone receptors, suggesting that the ligand binding sites of these proteins share some features in common.     

Prenatal prediction of androgen insensitivity syndrome using exon 1 polymorphism of the androgen receptor gene.

Exon 1 polymorphism of the androgen receptor (AR) gene is characterized by a (CAG)n(CAA) repeat at position 172 following the translation start codon. The aim of this study was to determine whether AR gene exon 1 polymorphism could be used to perform prenatal diagnosis in high risk families with complete or partial androgen insensitivity syndrome. After enzymatic amplification of a 1 kilobase exon 1 fragment, each DNA was simultaneously digested by MspI and PstI restriction enzymes. After electrophoresis on a 15% electrophoresis on a 15% acrylamide gel or a 6% Nusieve gel, we measured the size of the obtained fragments and determined the number of CAG repeats since a 282 basepair fragment corresponds to 21 CAG. We previously showed that the number of CAG repeats within the AR gene exon 1 in 23 families with complete or partial androgen insensitivity syndrome was 19 +/- 4. By this method, we detected heterozygosity in 50% of the mothers. We present here 2 exclusion prenatal diagnoses using exon 1 polymorphism of the AR gene. Family A presented a boy with a severe form of partial androgen insensitivity syndrome. The mother had 2 uncles with ambiguous genitalia. In family B, the affected child had a complete androgen insensitivity syndrome. In both families, analysis of the AR gene exon 1 polymorphism of the trophoblastic DNA showed the presence of the normal maternal X chromosome. The parents decided to carry on the gestation. In family A, the newborn had normal male external genitalia. In family B, sonography confirmed the presence of normal male external genitalia. These data suggest that exon 1 polymorphism of the AR gene could be prenatally used to predict androgen insensitivity syndrome.     

Molecular defects of the androgen receptor

Defects of the androgen receptor cause a wide spectrum of abnormalities of phenotypic male development, ranging from individuals with mild defects of virilization to those with complete female phenotypes. In parallel with this phenotypic spectrum, a large number of different mutations have been identified that alter the synthesis or functional activity of the receptor protein. In many instances, the genetic mutations identified lead to an absence of the intact, full-length receptor protein. Such defects (splicing defects, termination codons, partial or complete gene deletions) invariably result in the phenotype of complete androgen insensitivity (complete testicular feminization). By contrast, single amino acid substitutions in the androgen receptor protein can result in the entire phenotypic spectrum of androgen resistant phenotypes and provide far more information on the functional organization of the receptor protein. Amino acid substitutions in different segments of the AR open-reading frame disturb AR function by distinct mechanisms. Substitutions in the DNA binding domain of the receptor appear to comprise a relatively homogeneous group. These substitutions impair the capacity of the receptor to bind to specific DNA sequence elements and to modulate the function of responsive genes. Amino acid substitutions in the hormone-binding domain of the receptor have a more varied effect on receptor function. In some instances, the resulting defect is obvious and causes an inability of the receptor to bind hormone. In other instances, the effect is subtler, and may result in the production of a receptor protein that displays qualitative abnormalities of hormone binding or from which hormone dissociates more rapidly. Often it is not possible to correlate the type of binding defect with the phenotype that is observed. Instead, it is necessary to measure the capacity of the receptor that is synthesized in functional assays in order to discern any type of correlation with phenotype. Finally, two types of androgen receptor mutation do not fit such a categorization. The first of these—the glutamine repeat expansion that is observed in spinal and bulbar muscular atrophy—leads to a reduction of receptor function that can be measured in heterologous cells or in fibroblasts established from such patients. The expression of ARs containing such expanded repeats in men is associated with a degeneration of motor neurons in the spinal cords of affected patients. Likewise, the alterations of androgen receptor structure that have been detected in advanced forms of prostate cancer also behave as gain-of-function mutations. In this latter type of mutation, the exquisite specificity of the normal androgen receptor is relaxed and the mutant receptors can be activated by a variety of steroidal and non-steroidal ligands.