The progestational and androgenic properties of medroxyprogesterone acetate: gene regulatory overlap with dihydrotestosterone in breast cancer cells

Abstract

Introduction: Medroxyprogesterone acetate (MPA), the major progestin used for oral contraception and hormone replacement therapy, has been implicated in increased breast cancer risk. Is this risk due to its progestational or androgenic properties? To address this, we assessed the transcriptional effects of MPA as compared with those of progesterone and dihydrotestosterone (DHT) in human breast cancer cells.

Method: A new progesterone receptor-negative, androgen receptor-positive human breast cancer cell line, designated Y-AR, was engineered and characterized. Transcription assays using a synthetic promoter/reporter construct, as well as endogenous gene expression profiling comparing progesterone, MPA and DHT, were performed in cells either lacking or containing progesterone receptor and/or androgen receptor.

Results: In progesterone receptor-positive cells, MPA was found to be an effective progestin through both progesterone receptor isoforms in transient transcription assays. Interestingly, DHT signaled through progesterone receptor type B. Expression profiling of endogenous progesterone receptor-regulated genes comparing progesterone and MPA suggested that although MPA may be a somewhat more potent progestin than progesterone, it is qualitatively similar to progesterone. To address effects of MPA through androgen receptor, expression profiling was performed comparing progesterone, MPA and DHT using Y-AR cells. These studies showed extensive gene regulatory overlap between DHT and MPA through androgen receptor and none with progesterone. Interestingly, there was no difference between pharmacological MPA and physiological MPA, suggesting that high-dose therapeutic MPA may be superfluous.

Conclusion: Our comparison of the gene regulatory profiles of MPA and progesterone suggests that, for physiologic hormone replacement therapy, the actions of MPA do not mimic those of endogenous progesterone alone. Clinically, the complex pharmacology of MPA not only influences its side-effect profile; but it is also possible that the increased breast cancer risk and/or the therapeutic efficacy of MPA in cancer treatment is in part mediated by androgen receptor.

Figure 1

Chemical structures of progesterone, MPA, R5020 and DHT. The pharmaceutical name and manufacturer are shown italic text. DHT, dihydrotestosterone; MPA, medroxyprogesterone acetate.

Figure 2

Transcriptional activities of MPA, R5020 and DHT under the control of PR and PR isoforms. (a) T47Dco breast cancer cells with wild-type equimolar levels of PR-A and PR-B. T47Dco cells were transfected with PRE₂/luciferase and a Renilla luciferase internal control. Cells were treated for 20 hours with MPA, R5020, or DHT, at concentrations ranging from 0.01 to 1000 nmol/l. Luciferase activity was quantitated as a percentage of 1 μmol/l R5020 activity. The study was performed in triplicate and data are reported as mean ± standard error. *P < 0.05, 1 μmol/l dose of hormone versus 1 μmol/l R5020. (b) HeLa cervicocarcinoma cells stably expressing PR-A. (c) PR-negative T47D-Y breast cancer cells transiently transfected with PR-A. (d) HeLa cells stably expressing PR-B. (e) T47D-Y cells stably expressing PR-B. Transient transfection and transcription studies were performed as described above for panel a. DHT, dihydrotestosterone; MPA, medroxyprogesterone acetate; PR, progesterone receptor.

Figure 3

Expression profiles of endogenous PR-regulated genes. Shown are expression profiles of PR-regulated endogenous genes in breast cancer cell lines, with physiological progesterone versus MPA concentrations compared. (a) Venn diagrams showing gene number per condition. PR-positive T47Dco cells (CO) and control PR-negative T47D-Y (Y) cells were treated with 10 nmol/l progesterone or MPA for 6 hours in triplicate, time-separated experiments. Microarray analysis using Affymetrix U-133 2 plus gene chips, displaying about 47,000 genes, was performed. Data were analyzed using Gene Spring software and statistical analysis was conducted using a one-way analysis of variance, with P < 0.05 considered statistically significant for genes regulated at least 2.0-fold. Detailed tables showing complete gene lists are available as Additional files 1 and 2. (b) Dendrograms of PR-regulated genes showing relationships among treatment groups. Gene Spring 6.0 was used to classify all genes 'present' and all MPA- and progesterone-regulated genes. On the left is shown the relatedness among groups for all genes. On the right is a heat map of progestin regulated genes: black = average expression; red = above average; and green = below average. Each column represents a single gene; each row is a cell type and treatment group. DHT, dihydrotestosterone; MPA, medroxyprogesterone acetate; PR, progesterone receptor.

Figure 4

Glucocorticoid and androgenic activity of the synthetic progestins and Dexamethasone. HeLa cells (GR⁺, PR^-, AR^-) were transfected with or without an AR expression vector, PRE₂/luciferase construct and Renilla control, and treated for 20 hours with 10 nmol/l MPA, R5020, DHT, or dexamethasone. Relative luciferase activity was measured. Stippled bars are GR⁺ cells lacking AR; grey bars are GR⁺ and AR⁺ cells. AR, androgen receptor; DEX, dexamethasone; DHT, dihydrotestosterone; GR, glucocorticoid receptor; MPA, medroxyprogesterone acetate; PR, progesterone receptor; RLU, relative luciferase units.

Figure 5

Transcriptional activities of MPA, R5020, and DHT under AR control. (a) Cells transiently expressing AR and PRE₂/luciferase. AR-negative HeLa and T47D-Y cells were transiently transfected with an AR expression vector. Cells were cotransfected with PRE₂/luciferase construct and Renilla control, treated with varying concentrations of MPA, R5020, or DHT for 20 hours, and luciferase activity was quantified. Data are reported as a percentage of 1 μmol/l DHT (which was set at 100%). *P < 0.05, 1 μmol/l dose of hormone versus 1 μmol/l DHT. (b) Androgenic activity on two different promoter constructs. HeLa cells were transiently cotransfected with and AR expression vector and either PRE₂/luciferase or MMTV/luciferase. Cells were treated with ethanol, or 10 nmol/l and 1 μmol/l R5020, MPA, or DHT for 20 hours, and relative luciferase activity was measured. AR, androgen receptor; DHT, dihydrotestosterone; MPA, medroxyprogesterone acetate; PR, progesterone receptor; RLU, relative luciferase units.

Figure 6

Characterization Y-AR: a new breast cancer cell line stably expressing AR. (a) AR levels and function of Y-AR. Inset: whole cell extracts of Y-AR and MCF-7 cells were subjected to SDS-PAGE, and proteins were transferred to nitrocellulose and probed with the anti-AR antibody PG-21 (Upstate Biotechnology). The 110 kDa AR band is shown. Main figure: Y-AR and control T47D-Y cells were transiently transfected PRE₂/luciferase and treated with DHT, MPA, and R5020 at concentrations from 0.01 to 1000 nmol/l for 20 hours. Luciferase activity is shown as a percentage of 1 mmol/l DHT (which was set at 100%). Data shown are the mean of triplicate determinations ± standard error. (b) AR immunocytochemistry in Y-AR cells. Y-AR breast cancer cells were treated with ethanol or 1 μmol/l DHT for 30 min, or 2 and 20 hours. ARs were immunologically visualized using green fluorescent FITC (row A), and the cell nucleus was labeled with blue DAPI (row B). Cells were visualized using confocal microscopy and photographed at 100×, as described in the Materials and method section. (c) PCR of the PSA transcript. RNA was harvested from Y-AR cells treated 6 and 12 hrs with ethanol, or 1 μmol/l R5020, MPA, or DHT. RT-PCR was performed using PSA-specific primers. GAPDH is shown as a control. Representative results are shown. AR, androgen receptor; DHT, dihydrotestosterone; EtOH, ethanol; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MPA, medroxyprogesterone acetate; PR, progesterone receptor; PSA, prostate-specific antigen.

Figure 7

Expression profile of AR regulated genes. Shown are expression profiles of AR-regulated genes, with physiologic DHT versus physiologic and pharmacologic MPA concentrations in Y-AR cells compared. (a) Venn diagram comparing the number of genes regulated at least 2.0-fold by low-dose DHT versus low and high dose MPA. Y-AR cells were treated with ethanol, or 10 nmol/l progesterone, MPA, or DHT, or 1 μmol/l MPA for 6 hours in triplicate, time separated experiments. RNA was extracted, derivatized, and used to probe Affymetrix U-133 2 plus gene chips interrogating about 47,000 genes. Data were analyzed using Gene Spring software, and statistical analysis was performed using a one-way analysis of variance, with P < 0.05 considered statistically significant for genes regulated at least 2.0-fold. (b) Confirmation of the hormonal regulation of four genes identified in panel a. Four genes shown to be hormone regulated in Y-AR cells (i.e. F3, Maf-B, Krt4 and p57) were chosen for further analysis. Bar graphs: the hormonal regulation of the four selected transcripts in Y-AR cells, as assessed by microarray profiling. RT-PCR: Y-AR cells were treated for 6 hours with ethanol, 10 nmol/l progesterone, MPA, or DHT, or 1 μmol/l MPA and RNA was isolated. Primers directed against the transcripts of the four selected genes were used in RT-PCR reactions, and GAPDH was run as an internal control. (c) Venn diagrams showing number and overlap of genes regulated by 10 nmol/l MPA in AR⁺ Y-AR cells versus PR⁺ T47D cells. Microarray and data analysis were performed as described for Fig. 3a (PR⁺, AR^- cells) and panel a above (AR⁺, PR^- cells), and data for low-dose MPA in the two cell lines were compared. The 88 genes regulated by MPA through either PR or AR were tabulated (Additional files 1 and 2). The top ranking genes are shown in Table 1. AR, androgen receptor; DHT, dihydrotestosterone; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MPA, medroxyprogesterone acetate; PR, progesterone receptor.