003 1 -399819413602-0227a03.0010 PEDIATRIC RESEARCH Vol. 36, No. 2, 1994 Copyright O 1994 International Pediatric Research Foundation, Inc. Prirrred it1 U.S. A. A Practical Approach to the Detection of Androgen Receptor Gene Mutations and Pedigree Analysis in Families with X-Linked Androgen Insensitivity C. RIS-STALPERS, T. HOOGENBOEZEM, H. F. B. M. SLEDDENS, M. C. T. VERLEUN-MOOIJMAN, H. J. DEGENHART, S. L. S. DROP, D. J. J. HALLEY, J. C. OOSTERWIJK, M. B. HODGINS, J. TRAPMAN, AND A. 0. BRINKMANN Departments of Endocrinology and Reproduction (C.R. -S., M. C. T. I/. -M.,A. 0.B.], Pediatrics (T.H., H.J. D., S.L. S. D. 1, Pathology [H.F. B. M.S., J. T.I, and Clinical Genetics (D.J. J. H.1, Erasmus University Rotterdam, Rotterdam, The Netherlands; Clinical Genetics Center Leiden, Leiden, The Netherlands [J.C. 0.1; and Department of Dermatology, Glasgow University, Glasgow, United Kingdom [M.B. H.I Androgen insensitivity syndrome (AIS) is an X-linked identify possible carriers for the syndrome in families of disorder in which defects in the androgen receptor gene AIS patients using single-strand conformation polymor- have prevented the normal development of both internal phism and restriction site analysis of PCR products. In one and external male structures in 46,XY individuals. This case, the polymorphic (CAG),(CAA) repeat in exon 1 en- survey reports the analysis of 11 AIS subjects. The andro- coding a polyglutamine stretch was used to identify the gen receptor gene of these subjects was analyzed using mutant allele in a family with X-linked partial androgen polymerase chain reaction (PCR)-single-strand conforma- insensitivity before the identification of the actual genomic tion polymorphism analysis and sequencing or sequencing mutation. PCR-single-strand conformation polymorphism of PCR-amplified androgen receptor gene fragments alone. analysis proved to be a fast and reliable technique to screen In total, 10 single base changes and one partial gene dele- for androgen receptor gene mutations and to study the tion were detected. Seven single base changes resulted in androgen receptor gene of family members of AIS-affected an amino acid change, one resulted in the introduction of a individuals. (Pediatr Res 36: 227-234, 1994) premature stop codon, one event represented a single base insertion resulting in a frame-shift, and one single base Abbreviations change affected a donor splice site. The androgen receptor AR, androgen receptor protein in genital skin fibroblasts from several patients was hAR, human androgen receptor studied with respect to molecular mass after immunopre- AIS, androgen insensitivity syndrome cipitation and SDS-PAGE. Two patients expressed a trun- PCR, polymerase chain reaction cated receptor protein in agreement with the established SSCP, single-strand conformation polymorphism genomic mutation. Pedigree analysis was performed to DHT, Sa-dihydrotestosterone The most common cause of male pseudohermaphrodit- differentiation. In the most severe cases, a 46,XY indi- ism is AIS (1). Androgen insensitivity is a heterogenous vidual is presented with a female phenotype (complete syndrome in which the masculine development of both AIS). In other cases, it may concern a 46,XY phenotypic internal and external structures of a 46,XY individual female with clitoromegaly or ambiguous genitalia (partial may be affected. The clinical phenotype of affected indi- AIS) or a phenotypic male with severe hypospadias (Re- viduals consists of a spectrum of defects in male sexual ifenstein syndrome) or unexplained infertility (2). In all these cases. the clinical svndrome results from dimin- ished or absent androgen action due to a (partly) non- Received July 22, 1993; accepted March 22, 1994. Correspondence and reprint requests: Dr. A. 0. Brinkmann. Department of functioningAR, although testosterone 'ynthesis is unim- Endocrinolow and Reproduction, Erasmus University Rotterdam, P.0. Box paired (1).~. Defining- all the possible mutations in the AR in 1738, 3000 I% otterd dam, The Netherlands. patients with partial androgen insensitivity may eventu- Supported by the Netherlands Organization for Scientific Research (Gebied Medische Wetenschappen), the Dutch Cancer Society. the Sophia Foundation for lead to a better understanding and even a Medical Research. and the Scottish Home and Health Department. better Way of predicting the further development of the 228 RIS-STALPERS ET AL. secondary sex characteristics and fertility of affected medium supplemented with 10% FCS, nonessential children. amino acids, and antibiotics. The AR is a member of the steroidtthyroid hormone1 Specijc androgen binding and Western blot analysis. retinoic acid receptor zinc finger family (3), consists of Scatchard plot analysis was performed in the laboratory 910 amino acid residues (4), and has a molecular structure of origin; relevant references are included in Table 1. homologous with that of the other family members: an Genital skin fibroblasts were incubated with %-labeled N-terminal transcription-regulating domain, a DNA- DHT (New England Nuclear, Boston, MA) in a range of binding domain, and a C-terminal ligand-binding domain 10 concentrations (0-6 nM) for 45 min at 37°C. Nonspe- (5). The number of amino acid residues can vary between cific DHT binding was determined using labeled DHT in individuals because of two polymorphic amino acid the presence of 100 nM unlabeled DHT. After incubation, stretches in the N-terminal domain: a polymorphic GGN cells were harvested and the cell pellet resulting from a repeat encoding a glycine stretch and a polymorphic 500 x g centrifugation step was washed with a Tris-HC1/ (CAG),(CAA) repeat encoding a polyglutamine stretch. EDTA buffer (0.9% NaCl, 10 mM Tris, 10 mM Na2Mo0,, The polymorphic (CAG),(CAA) repeat has a frequency 1 mM EDTA, 0.5 mM Dm, 0.002% NaN,, pH 7.4). The of heterozygosity in females of 0.89 and is extremely supernatant was removed and the cell pellet was stored useful for carrier diagnosis in families with AIS (6). overnight at -70°C. After thawing, 40 mL of 0.6 M KC1 Although most members of the nuclear receptor family were added and the cells were left at 4°C for 1 h. A 0.25% have a corresponding resistance syndrome, a mutant AR charcoal suspension (150 p,L) containing 0.025% dextran gene can cause complete androgen resistance in 46,XY was added, and, after 5 min of shaking, cell debris and individuals because of its location on the X chromosome, charcoal were removed by centrifugation at 1000 x g for whereas for the other members of the nuclear receptor 10 min. One hundred mL of supernatant were used for 'H family the consequences of one mutant receptor allele counting. The protein content of the samples was mea- can be compensated by the other autosomal wild-type sured using the BioRad protein assay kit (BioRad Labo- allele. ratories, Richmond, CA). A Scatchard plot was con- In this study, the localization and characterization of structed using standard statistical routines. Each assay mutations in the hAR gene from 11 unrelated subjects was done in duplicate with four samples for each DHT suffering from different degrees of androgen insensitivity concentration. The two maximum binding capacity and are reported. Several mutations have been identified by k, values were combined, using the squared standard SSCP analysis (7); other mutations could be confirmed by errors as weighing factors. PCR-SSCP analysis. The SSCP analysis technique, in Immunoprecipitation and Western blot analysis of the most cases combined with restriction enzyme analysis of AR protein obtained from approximately 5 x 106 genital PCR products, was also successfully used to screen fam- skin fibroblasts were performed as described previously ily members of AIS patients for heterozygosity of the (8). The AR was immunoprecipitated from whole cell ly- syndrome. The polymorphic (CAG),(CAA) repeat in sates of genital skin fibroblasts with the AR-specific MAb exon 1 of the hAR gene was used to study family mem- F39.4.1. followed by SDS-PAGE and immunoblotting us- bers of an AIS patient who at a later date was diagnosed ing the polyclonal antibody Sp061. The AR protein on as having a genomic deletion of more than 6 kb from immunoblot was visualized by chemiluminescence (12). intron 2 of the hAR gene. DNA isolation and analysis. Genomic DNA was isolated Immunoprecipitation of the AR from genital skin fibro- from genital skin fibroblasts or from blood cells using blast lysates, obtained from either receptor-binding- standard methods (13). PCR reactions were done in a negative or receptor-binding-positive AIS patients, was 100-mL reaction volume as described before (8) using the performed to investigate whether a receptor protein was Perkin-Elmer Thermo Cycler, 2.5 U Taq polymerase present. (ArnpliTaq), and the appropriate reaction buffer and con- ditions as described by the supplier (Cetus, Norwalk, CT). For PCR reactions covering the GGN repeat in exon METHODS 1, deaza deoxyguanosine triphosphate was used instead of deoxyguanosine triphosphate. Oligonucleotides used Clinical subjects. Subjects were initially diagnosed in for PCR amplification of the AR gene and for direct different clinics in the Netherlands, Germany, Canada, sequencing are indicated in Table 2. and the United Kingdom. The diagnosis of partial or Radioactive PCR to determine the length of the complete AIS was made, taking into account the karyo- (CAG),(CAA) repeat in exon 1 or as a basis for SSCP type, the clinical phenotype, laboratory data including were prepared using 22 nM "~a-deoxy~~~in a 15-pL the male hormone levels in plasma when available, and standard PCR reaction. To determine the length of the relevant family history (summarized in Table 1). In- repeat encoding the polyglutamine stretch, PCR products formed consent was obtained from all subjects or their were size fractionated on a 6% polyacrylamide denatur- parents. ing sequence gel accompanied by a standard sequence Cell culture.