Unexpected hormonal activity of a catechol equine estrogen metabolite reveals reversible glutathione conjugation

Abstract

4-Hydroxyequilenin (4-OHEN) is a major phase I metabolite of the equine estrogens present in widely prescribed hormone replacement formulations. 4-OHEN is autoxidized to an electrophilic o-quinone that has been shown to redox cycle, generating ROS, and to covalently modify proteins and DNA and thus potentially to act as a chemical carcinogen. To establish the ability of 4-OHEN to act as a hormonal carcinogen at the estrogen receptor (ER), estrogen responsive gene expression and proliferation were studied in ER(+) breast cancer cells. Recruitment by 4-OHEN of ER to estrogen responsive elements (ERE) of DNA in MCF-7 cells was also studied and observed. 4-OHEN was a potent estrogen, with additional weak activity associated with binding to the arylhydrocarbon receptor (AhR). The potency of 4-OHEN toward classical ERalpha mediated activity was unexpected given the reported rapid autoxidation and trapping of the resultant quinone by GSH. Addition of thiols to cell cultures did not attenuate the estrogenic activity of 4-OHEN, and preformed thiol conjugates added to cell incubations only marginally reduced ERE-luciferase induction. On reaction of the 4OHEN-GSH conjugate with NADPH, 4-OHEN was observed to be regenerated at a rate dependent upon NADPH concentration, indicating that intracellular nonenzymatic and enzymatic regeneration of 4-OHEN accounts for the observed estrogenic activity of 4-OHEN. 4-OHEN is therefore capable of inducing chemical and hormonal pathways that may contribute to estrogen-dependent carcinogenesis, and trapping by cellular thiols does not provide a mechanism of termination of these pathways.

Figure 1

Activation of ERE-luciferase reporter by E₂, 4-OHEN, and 4-OHE₂ in MCF-7 cells. (A) Cells were transfected with ERE-luciferase and pre-treated with or without 10 μM Ro41-0960 (COMT inhibitor) for 24 h, and then treated with different concentrations of E₂ (green circles), 4-OHEN (red open squares), 4-OHE₂ (burgundy triangles), E₂ + COMT inhibitor (orange open circles), 4-OHEN + COMT inhibitor (blue squares) for 18 h. (B) Antiestrogen, ICI 182780 (1 μM), abolished the activity of E₂ (1 nM green), 4-OHEN (100 nM red), and 4-OHE₂ (10 nM burgundy triangle). All data represent relative induction of ERE-luciferase activity compared to 1 nM E₂ set at 100%. Data showed mean and s.d. of three independent experiments analyzed by ANOVA with Newman-Keuls post test (** p<0.001 vs. DMSO control).

Figure 2

Analysis of estrogen-responsive gene expression induced by 4-OHEN in breast cancer cells. Induction of mRNA for: A) trefoil factor 1 (pS2); and, B) progesterone receptor (PR) in MCF7 cells; in addition to C) TGF-α in S30 cells. Measurements from cells treated with E₂ (10 nM), ICI (1 μM), 4-OHEN (1 nM, 10 nM, 100 nM), or combinations as indicated, for 24 h. mRNA levels were quantified using real-time PCR. Assays were performed in three individual experiments and data represent the mean ± s.d; mRNA levels were significantly different (p < 0.05) compared to vehicle control.

Figure 3

Cell proliferation in MCF-7 breast cancer cells. Treatment for 6 days with vehicle (DMSO/ethanol, diamond), E₂ (green circle), 4-OHEN (red square), 4-OHE₂ (burgundy triangle), E₂ + 1 μM ICI (orange circle), 4-OHEN + 1 μM ICI (blue square), 4-OHE₂ + 1 μM ICI (lavender triangle). DNA content was quantified as a measure of cell growth. Assays were performed in three individual experiments and data represent the mean ± s.d.

Figure 4

Chromatin immunoprecipitation (ChIP) assay for liganded ER binding to ERE. PCR products for pS2 ERE and a pS2 upstream negative control from MCF-7 cells cultured in estrogen-free media for 4 days before treatment. Cells were treated with DMSO, 10 nM E₂, or different concentration of 4-OHE₂ or 4-OHEN for 45 min, and crosslinked by 1.5% formaldehyde for 15 min. All samples were pulled down by anti-ERα, IgG, and then incubated with a protein A/G agarose slurry. (1:Input, positive control, which was all DNA fragments without any antibody treatment; 2: rAb ERα, ERα Antibody; 3: rIgG, negative control (as a control for the specificity of antiERα). Images shown are representative of three individual experiments.

Figure 5

Effect of AhR knockdown on XRE- and ERE-luciferase activities in MCF-7 cells. (A) Induction of XRE-luciferase activity by 4-OHEN compared to AhR agonist 3-MC. Cells were transiently transfected with 4 μg of an XRE-luciferase reporter plasmid. After transfection, cells were treated with DMSO, 3-MC (10 nM), 4-OHEN (10 nM, 100 nM, 1 μM) for 24 h. (B) siRNA for AhR does not affect ERE-luciferase activity induced by 4-OHEN. Cells were transfected with ERE-luciferase plasmid as well as with siSC (E. gracilis, as a control for RNA interference performance), siAhR, or siLuc (firefly luciferase, as a cross-check for luciferase induction). Cells were treated with 10 nM E₂ or 1 μM 4-OHEN for 24 h. Assays were performed in three individual experiments and data represent the mean ± s.d. (* p<0.05 vs. DMSO control).

Figure 6

Effect of added thiols on 4-OHEN mediated ERE-luciferase activity in MCF-7 cells. (A) Cells were pre-treated with NAC (solid lines and symbols) or NAP (open symbols and dash lines) for 30 min, followed by addition of 4-OHEN (10 nM, red) or E₂ (1 nM, green) for 18 h. (B and C) 4-OHEN (10 nM) was reacted with thiols at different concentrations for 30 min prior to addition to cells; cells were further incubated for 18 h. E₂ (1 nM) response is shown for comparison. Data show mean and s.d.

Figure 7

(A & B) Non-enzymatic reduction of mono-GSH conjugate. (A) Chromatogram for conjugate (3) prior to addition of NADPH (5 mM). (B) Chromatogram after reaction with NADPH for 5 h at 28-29 °C showing formation of 4-OHEN (1). UV detection was at 250 nm. (C-F) Normalized HPLC peak area vs. time data for the formation of 4-OHEN 1 (squares) and the disappearance of 4-OHEN mono-GSH conjugate 3 (circles) at 250 nm: C, in the presence of 0.5 mM NADPH; D, in the presence of 1 mM NADPH; E, in the presence of 5 mM NADPH. Data were fitted to pseudo-first order rate equations to obtain the rate constant for the formation of 4-OHEN as, respectively, 0.72 × 10^-2 min^-1, 1.04 × 10^-2 min^-1, and 1.72 × 10^-2 min^-1, compiled in panel F.

Scheme 1