An androgen receptor mutation in the MDA-MB-453 cell line model of molecular apocrine breast cancer compromises receptor activity

1. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 1

   Sequencing and functional analysis of the AR in MDA-MB-453 cells. (A) Partial nucleotide sequence of AR exon 7 cDNA in the MDA-MB-453 (left panel) and T-47D (right panel) breast cancer cell lines. The arrowhead indicates nucleotide position 3719, where the G-T nucleotide substitution is located in MDA-MB-453 cells, resulting in the substitution of a glutamine residue for histidine at amino acid 865. (B) Partial nucleotide sequence of AR exon 7 genomic DNA in MDA-MB-453 cells. (C and D) MDA-MB-453 cells were transiently transfected with wtAR (blue curves) or AR-Q865H (red curves) expression constructs and the ARR3-tk-luc reporter construct followed by incubation with increasing concentrations (0.01–100 nM) of (C) DHT or (D) MPA or vehicle (0.1% ethanol, control) as indicated. Cell lysates were assayed for luciferase reporter gene activity. (E and F) PC-3 cells were transiently transfected and assayed for luciferase reporter gene activity in response to (E) DHT and (F) MPA as described for panels C and D. Points represent the average ±s.e.m. of quadruplicate wells. *P<0.05, two-way ANOVA, AR-Q865H vs wtAR. (G and H) AR protein levels in transiently transfected PC-3 cell lysates from the experiments shown in panels E and F were detected by western blot analysis. GAPDH was used as a loading control.
2. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 2

   Effect of the Q865H substitution on ligand and intramolecular interactions and AR stability. (A) Stability of ligand–AR complexes. COS-1 cell lysates transiently transfected with wild-type AR (blue curves) or the AR-Q865H variant (red curves) were incubated with DHT (left panel) or MPA (right panel) at 37 °C for the indicated times. Points represent the average ±s.e.m. of at least three independent experiments. (B) Homology modeling of the holo wtAR (blue) and AR-Q865H (yellow) AF2 region. Diagrams showing the α-helical and β-sheet backbone of the AR-LBD were generated using the MOLSCRIPT/Raster3D program. Important amino acid side chains are depicted in stick form and predicted hydrogen bonds are shown as green dotted lines. Wild-type (Q) and variant (H) residues at position 865 are indicated. (C) Induction of the AR N/C interaction by DHT and MPA was measured using a mammalian two-hybrid assay. COS-1 cells were transiently transfected with the pVP16-AR-NTD(wt) and the pM-AR-LBD(wt) or pM-AR-LBD(Q865H) expression constructs and the pGK1 reporter. Cells were treated with increasing concentrations of DHT (0.1–100 nM, black bars), MPA (1–1000 nM, gray bars), or vehicle (0.1% ethanol, control) as indicated for 48 h and cell lysates were assayed for luciferase reporter gene activity. Bars represent the average ±s.e.m. of quadruplicate wells. *P<0.05, two-way ANOVA, DHT vs MPA, ^#P<0.05 AR-Q865H vs wtAR. (D) Ligand-induced stabilization of the endogenous AR-Q865H variant in MDA-MB-453 cells. Cells were treated with vehicle control (veh, 0.1% ethanol), DHT (1 nM, 10 nM), or MPA (1 nM, 10 nM) for 24 h and then AR and GAPDH protein levels were measured by western blotting.
3. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 3

   Response of endogenous AR in MDA-MB-453 breast cancer cells to androgenic and non-androgenic ligands and antagonists. (A) Partial sequence alignment of steroid receptor family members. The gray shading surrounds residues homologous to AR amino acid 865. (B) Cells were transiently transfected with the MMTV-luciferase reporter plasmid followed by incubation with vehicle (veh, 0.1% ethanol, control) or 10 nM of DHT, MPA, progesterone (Prog), dexamethasone (Dex), hydroxyflutamide (OHF), bicalutamide (Bic), or 17β-estradiol (E₂). Cell lysates were then assayed for luciferase reporter gene activity. (C, D and E) Cells were treated with the above ligands and then the mRNA levels for (C) FKBP5, (D) UGT2B28, and (E) C1ORF116 were measured by quantitative RT-PCR. Bars represent the average ±s.e.m. of triplicate samples. *P<0.05, one-way ANOVA, hormone vs vehicle control.
4. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 4

   Transactivation activity of the endogenous AR in MDA-MB-453 breast cancer cells in response to DHT and MPA. Cells were transiently transfected with the (A) MMTV-luciferase, (B) PSA-luciferase, or (C) ARR3-tk-luciferase reporter plasmids followed by incubation with vehicle (0.1% ethanol, control), DHT (0.01–100 nM, black curves), or MPA (0.01–100 nM, gray curves). Cell lysates were then assayed for luciferase reporter gene activity. Points represent the average ±s.e.m. of quadruplicate wells. *P<0.05, two-way ANOVA, DHT vs MPA.
5. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 5

   Microarray analysis of DHT- and MPA-regulated genes in MDA-MB-453 cells. (A) Volcano plots showing distribution of gene expression for DHT (left panel) or MPA (right panel) compared to vehicle control. Dashed lines represent the P<0.05 statistical cutoff, dotted lines represent the 1.2-fold change (FC, inner lines), and 1.5 FC (outer lines) cutoffs. Black points represent genes with FC>1.5 (P<0.05), dark gray points represent genes with FC=1.2–1.5 (P<0.05), and light gray points represent genes with FC<1.2 (P<0.05) and/or P≥0.05. (B) Euler diagrams showing the number of genes uniquely or commonly upregulated or downregulated >1.2-fold by DHT (1 nM) and MPA (100 nM) compared with vehicle (P<0.05).
6. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 6

   Comparison of DHT- and MPA-regulated genes in MDA-MB-453 (black bars) and MFM-223 (gray bars) cells. Cells were treated with vehicle control (veh, 0.1% ethanol), DHT (1 nM), or MPA (100 nM) for 6 h. mRNA levels for target genes were measured by quantitative RT-PCR. (A) C1ORF116, (B) SLC15A2, (C) FKBP5, (D) RANBP3L, (E) SEC14L2, (F) DUSP10, (G) PCDH20, and (H) TMPRSS2. Points represent the average ±s.e.m. of triplicate samples. *P<0.05, one-way ANOVA, hormone vs vehicle control.
7. View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   Figure 7

   Association of DHT- and MPA-regulated genes in MDA-MB-453 with E₂-regulated genes in MCF-7 breast cancer cells and genes in the Wnt signaling pathway as assessed by GSEA. The change in expression of each gene in MDA-MB-453 in response to DHT treatment is shown as a heatmap (top panel). Probe sets in the datasets were collapsed to gene level, assigned a score based on a signal-to-noise ratio algorithm, and rank-ordered by this score. E₂ upregulated genes from (A) the Cicatiello dataset or (B) the GEMS dataset, in the ordered MDA-MB-453 dataset are shown as black lines (middle panel), and the running enrichment score (ES) is plotted (bottom panel). The maximum ES and P value are indicated. (C) GSEA results comparing DHT-regulated genes in MDA-MB-453 with genes in the KEGG Wnt signaling pathway.

• © 2012 Society for Endocrinology