Oligospermic infertility associated with an androgen receptor mutation that disrupts interdomain and coactivator (TIF2) interactions

Abstract

Structural changes in the androgen receptor (AR) are one of the causes of defective spermatogenesis. We screened the AR gene of 173 infertile men with impaired spermatogenesis and identified 3 of them, unrelated, who each had a single adenine-->guanine transition that changed codon 886 in exon 8 from methionine to valine. This mutation was significantly associated with the severely oligospermic phenotype and was not detected in 400 control AR alleles. Despite the location of this substitution in the ligand-binding domain (LBD) of the AR, neither the genital skin fibroblasts of the subjects nor transfected cell types expressing the mutant receptor had any androgen-binding abnormality. However, the mutant receptor had a consistently (approximately 50%) reduced capacity to transactivate each of 2 different androgen-inducible reporter genes in 3 different cell lines. Deficient transactivation correlated with reduced binding of mutant AR complexes to androgen response elements. Coexpression of AR domain fragments in mammalian and yeast two-hybrid studies suggests that the mutation disrupts interactions of the LBD with another LBD, with the NH2-terminal transactivation domain, and with the transcriptional intermediary factor TIF2. These data suggest that a functional element centered around M886 has a role, not for ligand binding, but for interdomain and coactivator interactions culminating in the formation of a normal transcription complex.

Figure 1

(a) SSCP analyses of family members of 2 probands. Exon 8 fragments were amplified from genomic DNA and electrophoresed on a PAGE gel to display SSCP mobility shifts. The probands, CML (lanes 2 and 6) and KLH (lanes 1 and 5), have a mutant DNA strand (lower arrow) that migrates faster than the WT allele (upper arrow) from the unaffected siblings (sister of CML, lane 4; brother of KLH, lane 8). Mothers of CML (lane 3) and KLH (lane 7) have both mutant and WT alleles, indicating that both women are carriers. (b) Sequencing autoradiogram of a portion of AR exon 8 from patient CML compared with the normal. Three patients had the same A→G substitution. (c) Restriction analyses of family members of 2 patients (CML, KLH). Exon 8 fragments were amplified from genomic DNA and restricted with BbrP1. The M886V mutation creates a new BbrP1 site such that enzymatic digestion results in 2 fragments (B and C) in the probands (lane 3, CML; lane 5, KLH), whereas normal alleles (lane 1, sister of CML; lane 4, normal fertile man; lane 7, brother of KLH) have only 1 fragment (A), measuring 347 bp. Mothers of CML (lane 2) and KLH (lane 6) display all 3 fragments, indicating their heterozygous status. Outer lanes are 123-bp DNA ladders (L).

Figure 2

Dissociation kinetics of ARs in genital skin fibroblasts. Normal (C, open circles and bold lines) and mutant (CML, filled circles and normal lines; KLH, filled squares and thin lines) fibroblast monolayers were exposed to 2 nM [³H]MB (top), 3 nM [³H]DHT (middle), or 3 nM [³H]T (bottom) at 37°C (left) or 42°C (right) for 2 hours. The radiolabeled medium was discarded and replaced with one containing 200-fold excess unlabeled androgen; replicate samples were removed at the indicated times and assayed for [³H]androgen that was still receptor bound. Each data point, the mean of 4 replicates, is expressed as a percentage of maximum binding at time 0. Vertical axes are on the same logarithmic scale, except for bottom right panel.

Figure 3

Transactivation activity of M886V AR with high doses of androgens. WT (open circles and solid lines) or mutant (filled circles and dotted lines) receptors were transiently expressed in COS-7 cells and exposed to increasing doses (nM) of T (a), DHT (b), or MB (c). Transactivation activity was expressed as fold increase in luciferase activity compared with cells not exposed to androgen. β-galactosidase activity and protein content were used to normalize for transfection efficiency and cell numbers, respectively. Each data point represents the mean ± SE of 4 replicates.

Figure 4

(a) Transactivation activity of M886V AR with increasing doses of AR cDNA. CV1 cells were cotransfected with the indicated amounts of WT or M886V AR cDNA and the reporter plasmid pMAM-LUC. Each data point, the mean of triplicates, represents the fold increase in luciferase activity of cells exposed to 30 nM MB compared with those without androgen. Bars are ± SE. (b) Immunoblot of WT (W) or mutant (M) receptors. CV1 cell extracts (10 μg total protein each) depicted in a were electrophoresed on an SDS-PAGE gel, and AR protein identified with a specific antibody (PG-21). Films were overexposed to enhance signal for the low-abundance AR protein.

Figure 5

(a) DNA mobility gel shift assay. WT (W) or mutant (M) receptors were expressed in COS-7 cells and exposed to 10 nM (lanes 1 and 2) or 100 nM (lanes 3–12) DHT. Equivalent quantities of immunoreactive AR from the cell extracts were added to binding reactions containing ³²P-labeled synthetic ARE (lanes 1–10) or ERE (lanes 11 and 12) oligonucleotide sequences. Excess unlabeled ARE (lanes 5 and 6), ERE (lanes 7 and 8), or Oct (lanes 9 and 10) oligonucleotides were added as competitor DNA to demonstrate the specificity of the binding reaction. The dark band at the bottom represents unbound ³²P-labeled DNA. (b) Immunoblot of WT or mutant receptors used in gel shift assay. Five microliters of representative cell extract (used in the gel shift assay depicted in Figure 5a) was exposed to either 10 or 100 nM of DHT and was separated on an SDS-PAGE gel. AR protein was identified with a specific antibody (PG-21). (c) Quantification of binding to AREs. Receptors were expressed in COS cells, exposed to [³H]MB and equivalent quantities (50,000 dpm) of [³H]MB-AR complexes incubated with 150 pmol of either biotin-labeled natural ARE (an ARE from MMTV-LTR) or synthetic ARE in 2 independent series of experiments. Streptavidin-biotin–bound AREs were collected by centrifugation, and [³H]MB-labeled receptor bound to the AREs was quantified by scintillation counting. Known DNA binding–domain mutants (ΔF582, ΔR615) with severe impairment of DNA binding were used for comparison. In the right panel, background counts were lowered by treating the lysate with dextran-coated charcoal and by centrifugation at 100,000 g for 1 hour. Assays using mock-transfected cells showed minimal background activity. Each data point was the mean of 2 experiments, and bars indicate their range.

Figure 6

(a) Transcription activity of AR fragments. Deletion constructs encoding the ARTAD (amino acids 1–504) or the ARLBD (amino acids 507–919) were transiently transfected into CV-1 cells. Transcriptional activity (measured in RLU) of the WT or M886V LBD fragments, alone or coexpressed with TAD (1 μg), was measured with a pMAM-LUC in the presence and absence of indicated amounts DHT (nM) (left). The experiments were repeated on another 2 occasions using independent plasmid preparations, and the results were expressed as fold increase in luciferase activity in the presence and absence of 3 nM DHT (right). (b) Interactions of LBD and TAD fusion proteins in the mammalian two-hybrid assay. The fusion proteins VP16AD-ARTAD and GAL4DBD-ARLBD were coexpressed in HeLa cells, and receptor TAD-LBD interactions were measured with (17m)₅-E1bTATA-Luc reporter plasmid. Cells were exposed to increasing doses of pVP16AD-ARTAD or MB as indicated. Data points represent mean ± SE of at least 3 replicates and reflect fold increase in luciferase activity of cells exposed to MB over those not exposed to the androgen.

Figure 7

(a) Effect of TIF2 on AR activity in HeLa cells. Mutant or WT AR plasmids were cotransfected with the indicated amounts of cDNA encoding full-length TIF2, with (+) or without (–) 0.01 nM MB (left). Cells were transfected with 50 ng TIF2 cDNA and exposed to increasing doses of MB (right). AR activity was measured with a multimeric AR reporter gene (ARE-TATA-Luc). (b) Interactions of TIF2 and ARLBD fragments in the mammalian two-hybrid assay. The fusion proteins VP16AD-TIF2 and GAL4DBD-ARLBD were coexpressed in HeLa cells, and protein-protein interactions were measured with (17m)₅-E1bTATA-Luc reporter plasmid. Amino acid positions of TIF2 and ARLBD fragments are numbered. Vertical bars within the TIF2 fragment indicate nuclear receptor–interacting box motifs (LXXLL), and numbers indicate the first L of each consensus motif. In the left panel, cells were exposed to increasing doses of MB; in the right panel, cells were exposed to increasing doses of the VP16AD-TIF2 expression vector. Data are expressed as fold increase in luciferase activity with or without 1 nM MB. Bottom panels show an immunoblot of WT (lanes 1–3) and mutant (lanes 4–6) GAL4DBB-ARLBD fusion proteins (∼51 kDa) from representative cell lysates (10 μg protein per lane) and specific [³H]MB-binding activity (fmol/mg protein) of cells transfected with WT or mutant GAL4DBB-ARLBD vector. (c) Effect of TIF2 on TAD-LBD interactions in the mammalian two-hybrid assay. WT or mutant GAL4DBD-ARLBD fusion protein was coexpressed with VP16AD-ARTAD in HeLa cells, and TAD-LBD interactions were measured with a (17m)₅-E1bTATA-Luc reporter plasmid (as in b). The effect of cotransfecting increasing doses of a fourth vector, pSG5-TIF2, encoding the full-length TIF2 protein, was measured as fold increase in luciferase activity with and without 0.1 nM MB. Data were the mean ± SE of at least 3 replicates.