Expression of two functionally different androgen receptors in a patient with androgen insensitivity

Recently, we demonstrated a previously unknown high rate of de novo mutations of the androgen receptor (AR) gene in androgen insensitivity syndrome (AIS) with some resulting in somatic mosaicism of mutant and wild type AR alleles. However, data on the genotype-phenotype relationship in the latter patients are sparse. We present here a 46,XY newborn with ambiguous genitalia carrying a mosaic of an 866 GTG (Val) --> ATG (Met) mutation with the wild type AR gene. This mutation has usually been associated with complete AIS. Accordingly, we found markedly impaired transactivation due to the mutant Met866 AR. Essential information arose from Scatchard analysis of methyltrienolone binding on cultured genital skin fibroblasts. We demonstrated for the first time the expression of two functionally different ARs (Kd1: 5.58 nM = mutant, Kd2: 0.06 nM = wild type) in one AIS individual. This finding not only represents an important confirmation for the presence of the somatic mosaicism in the patient, it also indicates the most likely molecular mechanism responsible for the unexpectedly strong virilization of the patient: Androgen action through the wild type AR expressed by part of the somatic cells. The present case clearly demonstrates the molecular mechanism by which somatic mosaicism of the androgen receptor gene can modulate in vivo androgen action. It underlines the importance of particular notice on somatic mosaicism in all androgen insensitivity syndrome patients carrying de novo mutations of the androgen receptor gene.

Complete androgen insensitivity caused by a splice donor site mutation in intron 2 of the human androgen receptor gene resulting in an exon 2-lacking transcript with premature stop-codon and reduced expression

Various mutations within the human androgen receptor gene have been documented to cause defective sexual differentiation in karyotypic male individuals. In this study, we report a previously undescribed point mutation at the donor splice-site of the second intron of the androgen receptor gene in a patient with a completely female phenotype. The sequence alteration was detected by single-strand-conformation-analysis-PCR and genomic sequencing. Applying competitive reverse transcribed PCR, cDNA sequencing and Western blotting, we could demonstrate considerable aberrations of structure and concentration of the transcript and its translation product in the patient's fibroblasts from the genital region. (1) In the transcript, exon 1 and 3 are directly linked to each other, the complete second exon is skipped. The mRNA predictively suffers a codon frame-shift in exon 3 associated with a premature termination between codons 598 and 599, leading to a truncated androgen receptor protein lacking any in vivo function. (2) Steady-state concentration levels of transcript and protein are abnormally low. Our observations highlight the influence of exon-flanking intron sequences on proper expression and function of gene products.

Inherited and de novo androgen receptor gene mutations: investigation of single-case families

OBJECTIVE

The objective of this study was to assess somatic and inherited androgen receptor gene mutations in families with only one affected individual.

METHODS

Molecular genetic analysis of the androgen receptor gene in DNA derived from blood leukocytes from 30 families with single-strand conformation analysis, direct sequencing, and restriction fragment analysis was performed.

RESULTS

In 22 families the mothers and all investigated grandmothers were heterozygous carriers. However, within the sisters and aunts, both heterozygous carriers and noncarriers were present. In eight families a de novo mutation was characterized. In three of these patients indication for somatic mosaicism was found.

CONCLUSIONS

De novo mutations occur at a high rate within the androgen receptor gene (8 of 30 = 26.7%); a high proportion (3 of 8) arise after the zygote stage. Thus only direct analysis of the underlying mutation of the androgen receptor gene in the proband and his or her family can provide the basis for genetic counseling.

Physiology and pathophysiology of androgen action

Knowledge of the physiology of male sexual differentiation and the clinical presentation of androgen insensitivity syndromes (AIS) has led to an increasing understanding of the mechanisms of androgen action. Androgens induce their specific response via the androgen receptor (AR), which in turn regulates the transcription of androgen-responsive target genes. The androgen-dependent development of male genital structures and the induction of the normal male phenotype depends on the presence of an intact AR. Structural alterations leading to malfunction of the AR are associated with variable inhibition of virilization despite normal or even supranormal serum levels of androgens. The mapping, cloning and sequencing of the AR gene have facilitated new insights into the study of androgen action. Functional investigation of the normal and the mutant AR in vivo as well as in vitro has led to the characterization of the distinct molecular steps involved in the normal androgen action pathways that are inhibited in the androgen insensitivity syndrome.

Immunohistochemistry and in situ hybridization of the androgen receptor in the developing human prostate

As it is suggested that the androgen receptor mechanism is required for prostatic development, we attempted to determine the appearance, expression and distribution of the androgen receptor in embryonic, infantile and pubertal human prostate. Using mono- and polyclonal antibodies and a digoxigenin-labeled 713 bp riboprobe, the androgen receptor expression in paraffin sections of fetal, infantile, and pubertal prostates was studied at the protein and RNA level. Under highly standardized conditions, application of the polyclonal antibodies resulted in a weak cytoplasmic and nuclear labeling of the epithelium of fetal glands. No immunoreaction was obtained with monoclonal antibodies. Applying the polyclonal antibody to pubertal and adult specimens, immunoreactivity of the androgen receptor was positive in nuclei of adluminal and basal epithelial cells, in interstitial and vascular smooth muscle cells and vascular endothelium, whereas ganglionic cells and enteroendocrine cells were negative. In situ hybridization with the digoxigenin-labeled riboprobe gave clear positive results already in epithelium of very young fetal specimens. A semiquantitative visual evaluation of in situ hybridizations showed that intermediate intensity of expression was increased in pubertal and adult specimens, whereas strong expression was reduced in prostatic epithelium.

CONCLUSIONS

The essential findings are: (1) an early expression of androgen receptor mRNA in the fetal prostate; (2) no immunoreaction of monoclonal antibodies against the androgen receptor in the same specimens, (3) a decrease of androgen receptor mRNA expression, but increase in immunoreactivity of the androgen receptor protein with the onset of glandular maturation during puberty.

Mosaicism due to a somatic mutation of the androgen receptor gene determines phenotype in androgen insensitivity syndrome

Premature stop codons of the human androgen receptor (AR) gene are usually associated with a complete androgen insensitivity syndrome. We, however, identified an adult patient with a 46,XY karyotype carrying a premature stop codon in exon 1 of the AR gene presenting with signs of partial virilization: pubic hair Tanner stage 4 and clitoral enlargement. No other family members were affected. A point mutation at codon position 172 of the AR gene was detected that replaced the original TTA (Leu) with a premature stop codon TGA (opal). Careful examination of the sequencing gel, however, also identified a wild-type allele, indicating a mosaicism. In addition, elimination of the unique AflII recognition site induced by the mutation was incomplete, thus confirming the coexistence of mutant and wild-type AR alleles in the patient. Normal R1881 binding and a normal 110/112-kDa AR doublet in Western immunoblots consolidated the molecular genetic data by demonstrating the expression of the wild-type AR in the patient's genital skin fibroblasts. Transfection analysis revealed that only relatively high plasmid concentrations carrying the mutated AR complementary DNA lead to expression of a shortened AR due to downstream reinitiation at methionine 189. Thus, reinitiation does not play a role in the presentation of the phenotype; rather, the partial virilization is caused by the expression of the wild-type AR due to a somatic mosaic. We conclude that somatic mosaicism of the AR gene can represent a substantial factor for the individual phenotype by shifting it to a higher degree of virilization than expected from the genotype of the mutant allele alone.

Functional assessment and clinical classification of androgen sensitivity in patients with mutations of the androgen receptor gene. German Collaborative Intersex Study Group

In the genetic male, mutations of the androgen receptor (AR) gene cause phenotypes ranging from female to subfertile male. Binding assays on genital skin fibroblasts and DNA analysis alone provide incomplete information about receptor function. We used the sex hormone-binding globulin (SHBG) response to stanozolol as a measure of AR function and correlated the results with phenotypes which were classified according to the degree of defective masculinization. Of the 34 patients investigated, 9 had complete, and 14 had partial androgen insensitivity syndrome (AIS) with predominantly female, ambiguous, or predominantly male phenotype. Eleven subjects served as controls. Mutations were characterized using polymerase chain reaction-single strand conformation polymorphism analysis and direct DNA sequencing. DNA analysis revealed two major deletions, two minor defects leading to premature stop codons in exon 1, and 19 point mutations in the DNA- and hormone-binding domains of the AR gene. After stanozolol, SHBG remained unchanged in patients with complete AIS (102.0 +/- 3.8 [SE]%; range 92.4%-129% of the initial value). The SHBG decrease was diminished in partial AIS with predominantly female (83.8% +/- 1.7%; range 81.3%-87.0%), ambiguous (80.4% +/- 4.4%, range 68.4%-89.1%), and predominantly male (mean 65.9% +/- 4.9%, range 48.6%-80.8%) phenotypes, and normal in controls (51.4% +/- 2.1%, range 35.6%-62.1%). Differences between controls and each AIS group were statistically significant (P < 0.05 - < 0.0001). A close correlation was found between the degree of undermasculinization (AIS phenotype) and the SHBG response.

CONCLUSIONS

The SHBG test provides functional information about the severity of the receptor defect in vivo and hence adds to the structural information provided by DNA analysis. It detects receptor defects due to mutations within the entire gene, including the DNA-binding domain, and is a rapid, simple, and cost effective procedure. It may provide useful information for the diagnosis and management of affected children.

The clinical and molecular spectrum of androgen insensitivity syndromes

Androgen insensitivity syndromes (AIS) are due to end-organ resistance to androgenic steroids in males leading to defective virilization of the external genitalia. The phenotype encompasses a wide array of genital ambiguity and may range from completely female to undervirilized but unequivocally male with infertility. This disorder is caused by mutations of the androgen receptor and is an X-linked recessive trait. We have studied 47 patients with AIS and have characterized the underlying molecular abnormality in the androgen receptor gene. Twenty patients had complete AIS and twenty-seven had partial AIS. Of the latter, 11 were of predominantly female phenotypic appearance and gender was assigned accordingly, while 16 were raised as males. Within the group of complete AIS, two patients had gross deletions within the gene, one had a small deletion, and one had an insertion. In the other patients with complete AIS, as well as all individuals with partial AIS, single nucleotide substitutions within the coding region were detected, each leading to an amino acid alteration. Seven codons were involved in more than one mutation in different cases. In addition, in one patient with spinal and bulbar muscular atrophy, an elongation of a glutamine-repeat was characterized. We conclude that mutations in the androgen receptor gene may be present throughout the whole coding region. However, our study provides evidence that several mutational hot spots exist.

Hormonally induced changes in apocrine secretion of transglutaminase in the rat dorsal prostate and coagulating gland

Coagulating gland and dorsal prostate of the rat are peculiar in secreting transglutaminase, a protein-cross linking enzyme that is released in an apocrine fashion. To elucidate whether or not the intracellular pathway and the unusual extrusion mechanism proceed constitutively or were differentially regulated, transglutaminase immunoreactivity was studied both at the light and electron microscopic levels. In addition, ultrastructural morphometry and scanning densitometry were applied to quantitate hormone-dependent distribution of transglutaminase. Coagulating glands and dorsal prostate, respectively, from sexually active rats were compared to those from sexually inactive, castrated, estradiol-treated or testosterone-substituted castrated animals. In intact, sexually active animals, no labeling of the cisternae of rough endoplasmic reticulum was seen, but instead the hyaloplasm was labeled. In the supranuclear portions of the cells an increase in labeling density of the hyaloplasm subjacent to the plasma membrane was found, whereas no labeling of either Golgi stacks or vesicles was observed. Apical blebs projecting into the acinar lumen were densely labeled. In castrated animals, epithelium showed a reduction of rough endoplasmic reticulum, loss of secretory blebs, and a decrease in cell size. Morphometric analysis of immunolabeling of coagulating gland epithelium from experimental animals resulted in a highly significant reduction of labeling of the hyaloplasm and apical blebs which was reversed by testosterone supplementation of castrated animals. After estrogen treatment, the reduction in immunolabeling was less pronounced, but morphology of apical blebs was obviously changed. Results from scanning densitometry of Western blots correlated with quantitative immunoelectron microscopical findings. Northern blot analysis using a secretory transglutaminase cDNA probe showed characteristic changes at the RNA levels. Our results indicate that apocrine secretion of transglutaminase in rat coagulating gland and dorsal prostate is a hormonally controlled process, where androgen deprivation results in impaired biosynthesis and release of transglutaminase, whereas estradiol treatment only partially inhibits secretion, but changes morphological features of the glandular epithelium, especially apocrine bleb formation.

Effects of androgen deprivation and estrogen treatment on the structure and protein expression of the rat coagulating gland

The effects of androgen deprivation and estrogen stimulation on rat coagulating gland were determined by immunohistochemistry and morphometric quantification of different tissue compartments. In castrated or estrogen-treated or estrogen-treated castrated animals, the reduction of the glandular lumen is the most obvious morphological alteration, which is accompanied by an increase in stromal tissue, especially within the lamina propria. Regressive changes occur most rapidly in castrated animals (already by the end of the first week), slower in estrogen-treated castrated animals, and still slower in estrogen-treated normal animals. In castrated animals, epithelium shows a reduction of rough endoplasmic reticulum, loss of secretory blebs, and a decrease of cell size and immunoreactivity for secretory transglutaminase. The reduction of glandular lumen results from an impressive increase in connective tissue of the lamina propria. Smooth muscle cells become atrophic in castrated animals, less so in estrogen-treated animals and in castrated estrogen-treated animals. A relative increase in thickness of the smooth muscle cell layer occurs in all experimental groups and is most obvious in estrogen-treated normal animals. The proportion of myofilament and intermediate filament proteins (smooth muscle-specific actin and desmin immunoreactivities) remains nearly unaltered in these cells after hormonal challenge. A redistribution of intermediate filaments occurs forming thicker bundles within the cells. No indication for increased mitotic activity of estrogenized smooth muscle cells has been found. After castration, and after estrogen treatment, the fibroblasts and the smooth muscle cells, respectively, appear responsible for the architectural changes within the coagulating gland. Reactions of the stroma are differentially regulated after estrogen treatment and androgen deprivation. No indication for increased biosynthetic activities of smooth muscle cells has been observed in any of the experimental conditions.

Estrogen-induced morphological and immunohistochemical changes in stroma and epithelium of rat ventral prostate

Prostatic smooth muscle cells have been regarded to play a major pathogenetic role during the development of benign prostatic hyperplasia (BPH) in elderly men. Altered hormonal signals (increased estrogen) have been made responsible for the "metabolic" transformation of prostatic smooth muscle cells, which were thought to produce increased amounts of connective tissue fibers observed in BPH. In order to find out the role of metabolically "activated" smooth muscle cells, hormone stimulation experiments were performed in male rats. The effects of androgen deprivation and estrogen stimulation were recorded by semiquantitative analysis of intermediate and myofilament proteins in stromal smooth muscle cells. In castrated or estrogen-treated or estrogen-treated and castrated animals, the reduction of the glandular lumen is the most obvious morphological alteration, accompanied by an increase in connective tissue. Regressive changes occurred most rapidly in castrated animals (already within the first week), slower in castrated estrogen-treated animals and still slower in normal estrogen-treated animals. Regression of the epithelium was accompanied by a marked decrease in immunoreactivity for prostatic binding protein (PBP) in castrated animals, while PBP immunoreactivity in estrogenized animals was retained for up to 6 weeks. Smooth muscle cells became atrophic in castrated animals. This effect was attenuated in estrogen-treated animals. There was no indication for enhanced collagen synthesis by smooth muscle cells. Actin and desmin-immunoreactivity were only slightly altered in experimental animals and showed a changed distribution pattern. Prostatic smooth muscle cells respond less markedly to hormonal alterations than do the fibroblasts.