Steroid receptor coactivator-3 glutamine repeat polymorphism and the androgen insensitivity syndrome.

Defects in androgen signaling can result in complete and partial male undermasculinization (1). The androgen receptor (AR) mediates the actions of the androgens testosterone and 5a-dihydrotestosterone in male embryonic and pubertal development. The AR is a member of the nuclear receptor superfamily of liganddependent transcription factors (2). The AR includes a polymorphic region of 11-31 glutamine repeats (AR[Gln]n). Longer AR[Gln]n tracts are associated with both partial androgen insensitivity (PAIS) in XY males (3) and the rare X-linked spinal and bulbar muscular atropy (SBMA) (4) which is often associated with infertility, testicular atropy and gynecomastia. The recruitment of multiple coactivator proteins is required for optimal AR function (5). It was demonstrated recently that the AR[Gln]n expansion found in SBMA diminishes interactions of the ARA-24 coactivator with the AR, resulting in decreased androgendependent transactivation (6). Similarly, it would be expected that coactivator mutations or polymorphisms which interfere with receptor –coactivator interactions would likely result in decreases in androgen-dependent transactivation. Although some cases of complete and partial androgen insensitivity are caused by mutations within the AR or defects in androgen synthesis, for most cases there is no identifiable defect in androgen production and no AR mutation (7). In these cases it is necessary to investigate other coactivator proteins that interact with the AR and are required for optimal AR function. There is compelling genetic and clinical evidence suggesting that impaired coactivator function may result in partial hormone resistance syndromes. Mouse genetic models reveal that disruption of the steroid receptor coactivator (SRC)-1 (8, 9) and SRC-3 (p/CIP/RAC3/AIB1/ACTR/TRAM-1) (10) can result in multiple nuclear hormone resistance syndromes. Furthermore, recent clinical reports also suggest a possible coactivator basis for hormone resistance syndromes (11, 12). SRC-3 is an AR coactivator (13) which contains a polymorphic glutamine repeat, SRC-3[Gln]n, encoded by a variable track of CAG/CAA codons in SRC-3 (14). Although the functional consequences of differences in SRC-3[Gln]n lengths have yet to be studied, the relationships between this polymorphism and breast (15, 16) and prostate (17, 18) cancer risks have been investigated. In this study we investigated whether variation in the SRC-3[Gln]n may contribute to the etiology of male undermasculinization. A heterogeneous group of undermasculinized patients ðn 1⁄4 65Þ with normal AR binding attributes, wild type AR sequence determined by single strand conformational polymorphism (SSCP) and/or sequence screening (7), and with AR[Gln]n within the normal range were selected from the Cambridge Intersex Database (1). Local ethics committee approval was obtained for the use of patient samples as part of a sexual development disorders research program. To avoid selection bias for ethnicity or geography, control subjects ðn 1⁄4 57Þ were selected from the same database. The control subjects harbored either an AR mutation or a diagnosis of Denys–Drash syndrome. These subjects have been shown to have the same AR[Gln]n profile as found in the normal population (3). Failure to detect a difference between allele distributions in the control and test cohorts would favor the null hypothesis of no contribution between SRC-3[Gln]n variants and undermasculinization. The unpaired t-test was used to compare mean SRC3[Gln]n length between groups. Statistical significance was regarded as P # 0:05: Two rounds of polymerase chain reaction (PCR) were used to amplify the polyglutamine encoding region of the SRC-3 gene. The first round used primer sequences, 50cag agc cga cag gca ctt gaa ttg-30 (forward) and 50gac tga tag atg gat gca gcc tgc g-30 (reverse). The second, nested, round of PCR used primer sequences, 50-tcc gac aac aga ggg tgg cta tg-30 (forward), and 50-tta gga ggt ggg ctg aag gcc tg-30 (reverse), using protocols described previously (14). The reverse primer was labeled with FAM for patients and HEX for control DNA amplifications. Thermal cycling was performed as follows: 95 8C for 5 min, followed by 35 cycles of 94 8C for 30 s, 56 8C for 30 s (oligonucleotide annealing), 72 8C for 90 s (DNA synthesis). The PCR products were mixed with GS-350-TAMRA size standard markers (Applied BioSystems, Warrington, UK) and loading buffer, and were heat denatured and electrophoresed on an ABI 377 sequencer. Several samples were electrophoresed on multiple gels to control for discrepancies European Journal of Endocrinology (2003) 148 277–279 ISSN 0804-4643