Correlation of clinical, endocrine and molecular abnormalities with in vivo responses to high‐dose testosterone in patients with partial androgen insensitivity syndrome

OBJECTIVE To investigate the responses of two patients previously diagnosed as Reifenstein's syndrome to graded high‐dose testosterone in terms of hormone levels, nitrogen balance and sebum secretion and to attempt to correlate these parameters with the properties of their androgen receptors and mutations in the androgen receptor gene. DESIGN Nitrogen balance was determined by comparing controlled nitrogen intake to the amount excreted. Sebum excretion was measured on the forehead. Patients were studied during control periods (no treatment) and during administration of testosterone propionate. Blood samples were used as a source of genomic DNA and to measure peripheral hormone levels; androgen receptor binding was determined using genital skin fibroblasts. PATIENTS Two patients of XY karyotype, with ambiguous external genitalia and problems of testicular descent who had required mastectomy as teenagers. Normal male controls of proven fertility. MEASUREMENTS Nitrogen balance, sebum excretion rate and peripheral hormone levels (testosterone, dihydrotestosterone, LH and FSH) were studied before and after testosterone therapy (1 or 5 mg/kg/day). Genomic DNA was extracted from peripheral blood leucocytes and regions of the androgen receptor gene amplified by polymerase chain reaction using pairs of specific primers. Mobility of amplified DNA from patients was analysed on denaturing gradient acrylamide gels and fragments differing in mobility from those of normal controls were sequenced. Fibroblasts were cultured from scrotal skin biopsies and androgen receptor binding parameters, subcellular localization and up‐regulation were determined. RESULTS Testosterone therapy resulted in raised plasma testosterone, dihydrotestosterone and oestradiol in both patients. In patient 1 (lesser genital abnormality), LH was suppressed by 5 mg/kg/day testosterone to the upper limit of the normal range but FSH remained low normal. Both LH and FSH were suppressed by testosterone treatment in patient 2 (greater genital abnormality). Nitrogen retention was increased in both patients (4.2 and 3.0 g/24 h respectively); sebum excretion rate increased to normal in patient 1 but showed no change in patient 2. Mutations in the androgen receptor gene were identified in both patients. In patient 1 a single nucleotide change from adenosine to guanosine resulted in the substitution of glycine for glutamic acid at position 772 within the hormone binding domain of the receptor. In patient 2 a single nucleotide mutation from guanosine to adenosine resulted in the substitution of lysine for arginine at position 608 (exon 3) situated in the second zinc finger of the DNA binding domain. Both patients had a normal number of androgen binding sites in genital skin fibroblasts but those in patient 1 showed reduced binding affinity and rapid dissociation of receptor/ligand complexes while those in patient 2 showed defective nuclear localization. CONCLUSION In patients with partial androgen insensitivity syndrome the type of androgen receptor mutation and responses to short‐term androgen treatment can be correlated with the individual’s potential to virilize. If there is a mutation in the androgen receptor DNA binding domain the patient may show little ability to virilize either spontaneously at puberty or after androgen treatment. Sebum excretion appears to be more discriminating than nitrogen balance or gonadotrophin supression as an index of tissue response to androgens.