From estrogen to androgen receptor: a new pathway for sex hormones in prostate

Abstract

While all three coactivators ARA70, steroid receptor coactivator 1, and RAC3/ACTR can enhance androgen receptor (AR) transcriptional activity at 1 nM dihydrotestosterone, we here demonstrate that only ARA70 can induce AR transcriptional activity >30-fold in the presence of 10 nM 17beta-estradiol (E2), but not diethylstilbestrol. The significance of this newly described E2-induced AR transcriptional activity in DU145 human prostate cancer cells was further strengthened by finding patients with Reifenstein partial-androgen-insensitive syndrome that fail in the E2-AR-ARA70 pathway. Together, our data suggest, for the first time, testosterone/dihydrotestosterone may not be the only ligands for the AR. E2 represents another important natural ligand for AR that may play an essential role for the AR function and the development of the male reproductive system.

Figure 1

The interaction of ARA₇₀ with AR in yeast and mammalian two-hybrid assay and the effect of ARA₇₀ on the E2-mediated AR transcriptional activity. (A) Effects of DHT, Dex, 17β-E2, 17α-E2, and DES on the interaction of ARA₇₀ with wild-type AR by plate nutritional selection in the yeast two-hybrid interaction system. The colonies formed on plates with 1 nM DHT, or 50 nM 17β-E2 but not on 10⁻⁵ M Dex, 17α-E2, or DES. Data were reproducible from two independent transformations. (B) Characterization of E2-mediated interaction between ARA₇₀ and wild-type AR or mutant ARe708k by the mammalian two-hybrid assay. (C) Effects of E2, DES, estrone, estriol, 17α-E2, Tam, ICI, and DHT on the transcriptional activity of AR in the presence or absence of ARA₇₀ in DU145 cells. After transfection, the cells were treated with serial concentrations of E2, DES, estrone, estriol, 17α-E2, Tam, ICI, and DHT (10⁻¹⁰ M: lanes 1, 6, 11, 16, 21, 26, 31, 36; 10⁻⁹ M: lanes 2, 7, 12, 17, 22, 27, 32, 37; 10⁻⁸ M: lanes 3, 8, 13, 18, 23, 28, 33, 38; 10⁻⁷ M: lanes 4, 9, 14, 19, 24, 29, 34, 39; 10⁻⁶ M: lanes 5, 10, 15, 20, 25, 30, 35, 40). The DHT treatments were taken as the positive control. Data represent an average of three independent experiments. The variance is ±15%.

Figure 2

MMTV-CAT reporter was not activated through PR, GR, and ER in the present of ARA₇₀ and 17β-E2. DU145 cells were cotransfected with 3.5 μg MMTV-CAT and 1.5 μg pSG5SR in the presence or absence of ARA₇₀ under the same 10 nM E2 treatment (lanes 6–14). (A) The effect of ARA₇₀ on other steroid receptors with the same reporter gene. The cells were transfected with AR/MMTV-CAT, GR/MMTV-CAT, PR/MMTV-CAT, in the presence of 10 nM DHT, P, or Dex, respectively (lanes 15–20). (B) DES, ICI, and Tam can inhibit the E2-mediated induction of AR transcriptional activity in the presence of ARA₇₀. The induction of AR-ARA₇₀ transcriptional activity by 10 nM E2 was counted as 100% (lane 2). The levels of increased inhibition relative to the concentration of E2 were shown as the following: ICI (lanes 3 and 4: 200-fold and 1,000-fold); Tam (lanes 5 and 6: 200-fold and 1,000-fold); DES (lanes 7 and 8: 200-fold and 1,000-fold).

Figure 3

E2-Ligand binding of AR and ARA₇₀. (A) In vitro synthesized AR, ARA₇₀, and ER were quantitated by [³⁵S]methionine labeling. Equal molar concentrations of ER and AR were used for the [³H]E2 ligand binding assay. Three-fold molar ARA₇₀ was incubated with AR on ice for 1 hr before adding 50 nM [³H]E2. Two hundred-fold unlabeled E2 was used as a competitor to determine the specific binding. (B) E2-specific binding of full-length AR. AR was transcribed and translated in a rabbit reticulocyte lysate system. Aliquots of the lysate were then incubated with 50 nM [³H]E2 (87 Ci/mmol; 1 Ci = 37 GBq) in the presence or absence of 20-fold, 200-fold, and 500-fold unlabeled steroids. The final incubation volume was 100 μl. The values of duplicate assay tubes were within 10% of the average shown in the figure.

Figure 4

The comparison of three coactivators, ARA₇₀, SRC-1, RAC3/ACTR on DHT- and E2- mediated AR transcriptional activity. Fixed amounts of AR and cofactors were used in transfections.

Figure 5

The E2-mediated induction of AR transcriptional activity on mutant ARs. The ARe708k was from the partial-androgen-insensitive syndrome patient. The ARt877a was from LNCaP and prostate cancer patient. Fixed amount of AR and ARA₇₀ were used in transfection. (A) A schematic representation of the helix 3 of AR, PR, and GR showing the location of mutant e708k. (B) The bioprofile of the partial-androgen-insensitive syndrome patient with mutant ARe708 k and the physical defect in the exterior reproductive system of the patient. (C) Effects of DHT and E2 on the transcriptional activity of wild-type AR, and ARe708k in the presence or absence of ARA₇₀ in DU145 cells. (D) Effects of E2 on the transcriptional activity of wild-type AR, mutant ARt877a, and ARe708k in the presence or absence of ARA₇₀.