Dissection of androgen receptor-promoter interactions: steroid receptors partition their interaction energetics in parallel with their phylogenetic divergence

Abstract

Steroid receptors comprise a homologous family of ligand-activated transcription factors. The members include androgen receptor (AR), estrogen receptor (ER), glucocorticoid receptor (GR), mineralocorticoid receptor (MR), and progesterone receptor (PR). Phylogenetic studies demonstrate that AR, GR, MR, and PR are most closely related, falling into subgroup 3C. ER is more distantly related, falling into subgroup 3A. To determine the quantitative basis by which receptors generate their unique transcriptional responses, we are systematically dissecting the promoter-binding energetics of all receptors under a single "standard state" condition. Here, we examine the self-assembly and promoter-binding energetics of full-length AR and a mutant associated with prostate cancer, T877A. We first demonstrate that both proteins exist only as monomers, showing no evidence of dimerization. Although this result contradicts the traditional understanding that steroid receptors dimerize in the absence of DNA, it is fully consistent with our previous work demonstrating that GR and two PR isoforms either do not dimerize or dimerize only weakly. Moreover, both AR proteins exhibit substantial cooperativity between binding sites, again as seen for GR and PR. In sharp contrast, the more distantly related ER-α dimerizes so strongly that energetics can only be measured indirectly, yet cooperativity is negligible. Thus, homologous receptors partition their promoter-binding energetics quite differently. Moreover, since receptors most closely related by phylogeny partition their energetics similarly, such partitioning appears to be evolutionarily conserved. We speculate that such differences in energetics, coupled with different promoter architectures, serve as the basis for generating receptor-specific promoter occupancy and thus function.

Keywords: AR; ER-α; ER-β; GR; HRE; MR; PR-A; PR-B; androgen receptor; estrogen receptor alpha isoform; estrogen receptor beta isoform; glucocorticoid receptor; hormone response element; mineralocorticoid receptor; progesterone receptor A isoform; progesterone receptor B isoform; protein–DNA interactions; quantitative footprinting; steroid receptors; thermodynamics; transcriptional regulation; wild type; wt.

Figure 1. Steroid receptor phylogenetic tree and promoter binding energetics

(A) Phylogenetic tree representing divergence of the steroid receptor family. Filled circle represents the ER-like common ancestor for subfamily 3A (ER-α and ER-β) and subfamily 3C (PR, AR, GR and MR). The two PR isoforms are not shown since they are generated from the same gene via alternative transcriptional or translational start sites. (B) Representative dimer-binding pathway for receptor-HRE₂ assembly. Dimers are formed through microscopic equilibrium constant k_dim and pre-formed dimers bind response elements with intrinsic affinity k_int. Schematic representing the overall reaction for receptor monomer assembly at a palindromic response element regardless of pathway. Total affinity is represented by K_tot. (C) Potential inter-site cooperativity is represented by k_c.

Figure 2. Purification of full-length human AR and characterization using sedimentation velocity and native-PAGE

(A) Baculovirus-expressed wt AR and T877A were purified as described in Materials and Methods. Purified receptors were resolved using 10 % SDS-PAGE and Coomassie-Blue staining. Molecular mass markers are indicated on left. (B) Sedimentation velocity data for 1.4 μM wt AR in 500 mM NaCl. Open circles represent absorbance data collected at 230 nm as a function of time and radial distance. Solid lines represent best fit from c(s) analysis as implemented in the program Sedfit. For clarity, only every seventh data point is displayed. (C) c(s) distributions for three concentrations of wt AR and T877A in 500 mM NaCl. Thick solid line (1.4 μM), dashed line (0.7 μM) and thin solid line (0.14 μM). (D) Immunoblot of wt AR using native-PAGE. Lower molecular mass species is indicated by large arrow; larger mass species indicated by small arrow.

Figure 3. Sedimentation equilibrium analysis of wt AR and T877A in 500 mM NaCl

(A) wt AR sedimentation equilibrium data plotted as absorbance versus r²/2 for three different concentrations of AR. From left to right: 0.6 μM, 0.4 μM, and 0.3 μM. Symbols represent three different rotor speeds: 14,000 rpm (open circles), 17,000 rpm (inverted triangles), and 21,000 rpm (open squares). Solid lines represent best global fit to a two species non-interacting model. Standard deviation of global fit was 0.0039 absorbance units. Residuals are plotted below the data and best-fit lines. For clarity, only every third data point is displayed. (B) T877A sedimentation equilibrium data plotted as absorbance versus r²/2 for three different concentrations. From left to right: 1.3 μM, 0.5 μM, and 0.3 μM. Symbols, corresponding rotor speeds and solid lines are identical to those described above. The standard deviation of global fit was 0.0043 absorbance units. Residuals are plotted below. For clarity, only every third data point is displayed.

Figure 4. Sedimentation velocity analysis of wt AR and T877A in 100 mM NaCl

(A) c(s) distributions for three concentrations of wt AR (top). Thick solid line (1.4 μM), dashed line (0.7 μM) and thin solid line (0.14 μM). (B) c(s) distributions for three concentrations of T877A. Thick solid line (1.1 μM), dashed line (0.7 μM) and thin solid line (0.11 μM).

Figure 5. Quantitative DNase footprint titrations and individual site-binding isotherms of wt AR and T877A assembly at the HRE₂ promoter

(A) Representative autoradiogram of wt AR binding at the HRE₂ promoter in 100 mM NaCl. AR concentration increases from left to right. Positions of site 1 (solid rectangle) and site 2 (open rectangle) are depicted to right. (B) Individual site-binding isotherms constructed from analysis of wt AR and T877A footprint titration images. Filled red squares represent binding to site 1 and open red squares represent binding to site 2 of the HRE₂ promoter. Open blue circles represent binding to site 2 of the HRE_1- promoter. Red and blue lines represent best global fits to all isotherms using Eqs 5 and 6. The fit lines for each HRE₂ site overlay since the sequences of sites 1 and 2 are identical.

Figure 6. Microstate energetics of steroid receptor dimer assembly at the HRE₂ promoter

Circles represent receptor dimerization affinity (k_dim) and triangles represent inter-site cooperativity (k_c). Since dimerization was not observed for wt AR, T877A and GR, downward arrows have been added to indicate that plotted values represent lower limits. Error bars represent 67% confidence intervals.

Figure 7. Probabilities of receptor dimer formation and fully-ligated HRE₂ promoter microspecies under standard state conditions

Probabilities determined from experimentally measured energetics or lower limit estimates. Blue lines represent the probabilities of receptor dimers and red lines represent the probabilities of the fully-ligated HRE₂ promoter. Because dimers have not been observed for wt AR, T877A and GR, blue lines for these proteins represent limits.