The androgen receptor depends on ligand-binding domain dimerization for transcriptional activation

Abstract

Whereas dimerization of the DNA-binding domain of the androgen receptor (AR) plays an evident role in recognizing bipartite response elements, the contribution of the dimerization of the ligand-binding domain (LBD) to the correct functioning of the AR remains unclear. Here, we describe a mouse model with disrupted dimerization of the AR LBD (AR^Lmon/Y ). The disruptive effect of the mutation is demonstrated by the feminized phenotype, absence of male accessory sex glands, and strongly affected spermatogenesis, despite high circulating levels of testosterone. Testosterone replacement studies in orchidectomized mice demonstrate that androgen-regulated transcriptomes in AR^Lmon/Y mice are completely lost. The mutated AR still translocates to the nucleus and binds chromatin, but does not bind to specific AR binding sites. In vitro studies reveal that the mutation in the LBD dimer interface also affects other AR functions such as DNA binding, ligand binding, and co-regulator binding. In conclusion, LBD dimerization is crucial for the development of AR-dependent tissues through its role in transcriptional regulation in vivo. Our findings identify AR LBD dimerization as a possible target for AR inhibition.

Keywords: androgen receptor; chromatin binding; dimerization; ligand-binding domain; transcriptional activation.

Figure 1. Disrupting AR LBD dimerization

1. A

   Crystal structure of the human AR LBD core dimer based on PDB 5JJM (1 monomer in gray and 1 monomer in pink). W752, located at the interface of the AR LBD dimer, is involved in hydrogen bond formation with T756 of the neighboring LBD and thereby stabilizes the interface. R752 disrupt this stabilization by steric hindrance and charge repulsion. The side chains of both W and R are shown at position 752. A close‐up of the LBD‐LBD interface is given. The ligand (DHT) is depicted as spheres.

2. B

   Acceptor photobleaching FRET after transfection of Hep3B cells with labeled WT LBD or labeled W752R‐mutated LBD. Representative confocal images of Hep3B cells transiently expressing WT or W752R AR in the presence of 10 nM DHT are shown below the bars. Scale bar = 10 µm. The bar graphs show means ± SEM (biological replicates, n = 54 (WT LBD) and n = 65 (W752R LBD), unpaired two‐tailed Student’s t‐test, ***P < 0.001).

3. C

   Representative pictures of the AGD of 13‐week‐old WT male (upper left), WT female (lower left), AR^Lmon/Y (upper right), and AR^−/Y (lower right) mice.

4. D

   Upper panel: a representative picture of the urogenital tract of a WT male and an AR^Lmon/Y mouse. Lower panel: a representative picture of the testis of a WT male and an AR^Lmon/Y mouse. Scale bar = 1 cm.

5. E, F

   Serum levels of T (E) and LH (F) in WT males, and AR^Lmon/Y and AR^−/Y mice at the age of 13 weeks. The bar graphs show means ± SEM (biological replicates, n = 8, one‐way ANOVA with Tukey’s multiple comparisons test, **P < 0.01, ***P < 0.001, ns = not significant).

Figure EV1. Molecular dynamics of the AR LBD dimerization interface and yeast two‐hybrid assay

1. A

   Molecular dynamics simulations based on the WT AR LBD dimer crystal structure.

2. B

   Molecular dynamics simulations based on the W752R AR LBD dimer model.

3. C

   Superposed model of WT LBD dimer (pink) and W752R LBD dimer (white).

4. D

   Molecular dynamics simulations based on the W752A AR LBD dimer model.

5. E

   Superposed model of WT LBD dimer (pink) and W752A LBD dimer (purple).

6. F

   Yeast two‐hybrid assay on human WT LBD and human W752R LBD. RFP signal normalized to 0 nM DHT is shown. The bar graphs show means ± SEM (biological replicates, n = 3, unpaired two‐tailed Student’s t‐test, ***P < 0.001, ns = not significant).

Figure EV2. Detailed evaluation of the AR^Lmon/Y phenotype

1. A

   Evolution of the anogenital distance (AGD) over time. Average is shown, and shaded areas represent SEM (biological replicates, n ≥ 10).

2. B

   Body weight followed over time. Average is shown, and shaded areas represent SEM (biological replicates, n ≥ 10).

3. C

   Total amount of fat at 12 weeks of age determined by EchoMRI normalized to body weight. The bar graphs show means ± SEM (biological replicates, n ≥ 10, one‐way ANOVA with Tukey’s multiple comparisons test, *P < 0.05, ns = not significant).

4. D

   Testes weight normalized to body weight of 13‐week‐old WT males, and AR^Lmon/Y and AR^−/Y mice. The bar graphs show means ± SEM (biological replicates, n = 8, one‐way ANOVA with Tukey’s multiple comparisons test, **P < 0.01, ***P < 0.001).

5. E

   Kidney weight normalized to body weight of 13‐week‐old WT males, and AR^Lmon/Y and AR^−/Y mice. The bar graphs show means ± SEM (biological replicates, n = 8, one‐way ANOVA with Tukey’s multiple comparisons test, *P < 0.05, ns = not significant).

6. F

   Serum levels of FSH in WT males, and AR^Lmon/Y and AR^−/Y mice at the age of 13 weeks. The bar graphs show means ± SEM (biological replicates, n = 8, one‐way ANOVA with Tukey’s multiple comparisons test, **P < 0.01, ns = not significant).

7. G

   Upper panel: H&E staining on testis of a WT male, and AR^Lmon/Y or AR^−/Y mouse. Lower panel: immunofluorescence staining of the AR (green). Nuclei are shown in blue. Orange, blue, and white arrows indicate LC, SC, and peritubular myoid cells, respectively. Scale bar = 50 µm.

8. H–I

   Relative contribution of seminiferous epithelium (H) and interstitium (I) in testis from 13‐week‐old mice. Proportions are expressed relative to total testis volume. The bar graphs show means ± SEM (biological replicates, n = 5, one‐way ANOVA with Tukey’s multiple comparisons test, ***P < 0.001, ns = not significant).

Figure 2. Transcriptome analysis of testes

1. A, B

   Volcano plots visualizing the differentially expressed genes in AR^Lmon/Y (A) and AR^−/Y (B) testes compared with testes from WT males. The dotted horizontal lines represent a q‐value of 0.05, while the dotted vertical lines represent a FC of 1.5. Genes that are downregulated compared with WT are shown using blue dots, while red dots represent the upregulated genes. Hsd17b3 and Star are indicated as yellow and green stars, respectively.

2. C

   Box plots showing testicular expression from the bulk RNA‐seq data (biological replicates, n = 5) of SC‐ and LC‐specific genes extracted from the single‐cell RNA‐seq data described by Green et al (2018). Box plots are composed of a box from the 25th to 75th percentile with the median as a line and whiskers calculated via the Tukey method. Expression levels are normalized to average expression in WT for each gene. Two‐way ANOVA with Tukey’s multiple comparisons test, ***P < 0.001.

3. D

   RT–qPCR analyses of Rhox5 and Insl3 of testes from 13‐week‐old WT males, and AR^Lmon/Y and AR^−/Y mice. Expression levels are normalized to WT. The bar graphs show means ± SEM (biological replicates, n = 6, one‐way ANOVA with Tukey’s multiple comparisons test, ***P < 0.001, ns = not significant).

Figure 3. Steroidogenesis in AR^Lmon/Y and AR^−/Y mice

1. A

   Expression levels of genes involved in steroidogenesis normalized to WT. The bar graphs show means ± SEM (data derived from RNA‐seq analysis of testes, biological replicates, n = 5, one‐way ANOVA with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant).

2. B

   Serum levels of A‐dione in WT males, and AR^Lmon/Y and AR^−/Y mice at the age of 13 weeks. The bar graphs show means ± SEM (biological replicates, n = 8, one‐way ANOVA with Tukey’s multiple comparisons test, ***P < 0.001, ns = not significant, n.d. = not detected; below detection limit of 0.175 nmol/L).

3. C

   Expression levels of Hsd17b3 normalized to WT. The bar graphs show means ± SEM (data derived from RNA‐seq analysis of testes, biological replicates, n = 5, one‐way ANOVA with Tukey’s multiple comparisons test, **P < 0.01, ***P < 0.001).

4. D

   Reporter gene assays in HeLa cells co‐transfected with the mouse Insl3 promoter together with the WT or mutant mouse AR. The bar graphs show means ± SEM (biological replicates, n = 4, unpaired two‐tailed Student’s t‐test, **P < 0.01, ns = not significant).

5. E

   Schematic overview of the feedback mechanism between testes and pituitary in the transcriptional regulation of steroidogenic genes.

Figure 4. Evaluation of the renal androgen response after disrupting AR LBD dimerization

1. A

   The experimental set‐up to study the androgen response in kidney.

2. B

   PCA on the renal transcriptomes of WT ORX, WT ORX + T, AR^Lmon/Y ORX and AR^Lmon/Y ORX + T.

3. C

   Heatmap for the genes that are differentially expressed in kidney between WT ORX and WT ORX + T (q < 0.05; FC > 1.5) and corresponding levels in AR^Lmon/Y ORX and AR^Lmon/Y ORX + T.

4. D–F

   mRNA expression levels extracted from RNA‐seq data for Fkbp5 (D), Odc1 (E), and Kap (F) in kidneys used for RNA‐seq. Expression levels are normalized to WT ORX. The bar graphs show means ± SEM (biological replicates, n = 5, one‐way ANOVA with Tukey’s multiple comparisons test, **P < 0.01, ***P < 0.001, ns = not significant).

Figure 5. In vitro analysis of the W752R mutation

1. A

   Transactivation assay in HEK293 cells with an integrated 4xSLP‐HRE2 E1B 12 TATA Luc reporter and transiently transfected with human WT or W752R AR followed by stimulation with increasing DHT concentrations. The bar graphs show means ± SEM (biological replicates, n = 4, two‐way ANOVA with Sidak’s multiple comparisons test, ***P < 0.001, ns = not significant).

2. B

   Specific ligand‐binding assay in COS cells with overload of cold ligand and increasing concentrations of radioactive labeled DHT. Values are normalized to protein expression. Individual data points are shown with nonlinear regression curve fit (n = 2).

3. C

   Western blot on cytoplasmic, nuclear, and whole‐cell extracts derived from HeLa cells transfected with human WT or W752R AR followed by stimulation with increasing concentrations of DHT. Both panels represent the same blot. The blot was cut into two parts, and antibodies against HSP90 and Lamin A/C were used to confirm cellular fractionation of cytoplasmic and nuclear proteins. For AR visualization (upper panel), longer exposure time was used. # = residual AR expression (lower panel).

4. D

   EMSA using nuclear extracts of COS‐7 cells transfected with AR WT or AR W752R and incubated with radiolabeled ARE, more specifically TAT‐GRE (Denayer et al, 2010). Shift (S) occurred through binding of the AR dimer on the ARE. Supershift (SS) occurred after addition of an AR antibody. S = shift and U = unbound ARE. Western blot on AR, depicted on the right, shows similar expression levels for WT and mutated receptors.

Source data are available online for this figure.

Figure 6. Chromatin binding of the AR^Lmon

1. A

   Western blot on whole‐cell extracts and chromatin fractions extracted from kidneys of WT males and AR^Lmon/Y mice. Antibodies against GAPDH and histone 3 were used to confirm cellular fractionation. Relative intensities are indicated, whereby the upper bands (AR) are normalized to the lower bands (GAPDH or histone 3). Both panels represent the same blot. The blot was cut into two parts, and a longer exposure time was used for AR visualization (upper panel).

2. B

   AR occupancy (blue) and levels of the histone marks H3K27ac (green) and H3K27me3 (red) at ARBS within the regulatory regions of Fkbp5, Kap (both androgen‐regulated in kidney), and Tox3 (androgen‐regulated in prostate but not in kidney and hence used as a negative control) were assessed by ChIP‐qPCR in kidneys from 13‐week‐old AR^Lmon/Y mice and WT males. The bar graphs show means ± SEM (biological replicates, n = 6; one‐sample t‐test with the Benjamini–Hochberg correction for multiple testing was used to determine whether a mean Log₂ FC was statistically different from 0, *P < 0.05, **P < 0.01, ***P < 0.001). Fold enrichments are shown in Appendix Fig S6C–E.

3. C

   Representative pictures of immunofluorescence AR staining (green) in kidneys of castrated WT males and AR^Lmon/Y mice supplemented with vehicle or T. Nuclei are shown in blue. Scale bar = 20 µm.

Source data are available online for this figure.

Figure 7. Comparison of WT AR and AR W752R interactomes

1. A

   Double‐hybrid assay in COS cells transfected with WT LBD, W752R LBD, or empty vector (negative control) in the presence of the LxxLL fragment of SRC1. Average is shown, and shaded areas represent SEM (biological replicates, n = 4, Tukey’s multiple comparisons test, **P < 0.01, ***P < 0.001).

2. B

   Confocal fluorescence microscopy images of AR‐WT‐BirA* and AR W752R‐BirA*expressing HEK293 cells treated in the presence of 50 µM biotin with 100 nM DHT or vehicle (ethanol) and with or without 0.03 µg/ml TET as indicated. AR‐BirA*s were detected with anti‐AR (red) and biotinylated proteins with fluorescently labeled streptavidin (green). Nuclei were visualized using DAPI. Scale bar = 20 µm.

3. C

   Heatmap showing the MS spectral counts of high‐confidence interactors (FDR < 0.05 after SAINT analysis) identified with WT AR‐BirA* and AR W752R‐BirA* upon DHT or vehicle exposure. Values for three biological replicates from DHT‐ and vehicle‐exposed samples are shown. GRPEL1 and DBT were the only high‐confidence interactors of AR W752R‐BirA*. On the right, FDRs and spectral count averages are shown for the DHT‐treated samples. Spectral counts have been normalized to those of AR in each sample.