Testosterone metabolites inhibit proliferation of castration- and therapy-resistant prostate cancer

Abstract

Novel treatments for castration-resistant prostate cancer (CRPC) such as abiraterone acetate (AA) or enzalutamide effectively target the androgen pathway to arrest aberrant signalling and cell proliferation. Testosterone is able to inhibit tumour cell growth in CRPC. Estrogen receptor-beta (ERβ) binds the testosterone-metabolites 3β-androstanediol and 3α-androstanediol in parallel to the canonical estradiol. In the prostate it is widely accepted that ERβ regulates estrogen signalling, mediating anti-proliferative effects. We used the prostate cancer cell lines LNCaP, PC-3, VCaP, and the non-neoplastic BPH-1. VCaP cells were treated with 1 nmol/L testosterone over 20 passages, yielding the cell line VCaPrev, sensitive to hormone therapies. In contrast, LNCaP cells were grown for more than 100 passages yielding a high passage therapy resistant cell line (_hiPLNCaP). VCaP and _hiPLNCaP cell lines were treated with 5 μmol/L AA for more than 20 passages, respectively, generating the AA-tolerant-subtypes VCaP^AA and hiPLNCaP^AA. Cell lines were treated with testosterone, dihydrotestosterone (DHT), R1881, and the androgen-metabolites 3β-androstanediol and 3α-androstanediol. 3β-androstanediol or 3α-androstanediol significantly reduced proliferation in all cell lines except the BPH-1 and androgen receptor-negative PC-3 and markedly downregulated AR and estrogen receptor alpha (ERα). Whereas ERβ expression was increased in all cell lines except BPH-1 or PC-3. In summary, 3β-adiol or 3α-adiol, as well as DHT and R1881, significantly reduced tumour cell growth in CRPC cells. Thus, these compounds represent novel potential therapeutic approaches to overcome drug-resistance in CRPC, especially with regard to AR-V7 function in therapy resistance. Furthermore, these data confirm the tumour suppressor properties of ERβ in CRPC.

Keywords: 3a-androstendiol; 3b-androstendiol; AKR1C1; AKR1C2 and AKR1C3; castration resistant prostate cancer.

Figure 1. Proliferation after treatment with 3α-adiol

Several prostate cancer cell lines were treated with 0, 0.01, 0.1, 1, 5, 10, 100, 250, or 1000 nmol/L 3α-adiol. Proliferation was assessed by performing BrdU-ELISA. 3α-adiol led to a significant reduction in proliferation in various LNCaP- and VCaP-derived cell lines, whereas androgen receptor (AR)-negative PC-3 and non-neoplastic BPH-1 cells showed no changes in proliferation (n.s. = not significant, ^*P < 0.05, ^**P < 0.005, ^***P < 0.0005).

Figure 2. Proliferation after application of 3β-adiol

Several prostate cancer cell lines were treated with 0, 0.01, 0.1, 1, 5, 10, 100, 250, or 1 nmol/L 3α-adiol. Proliferation was assessed by performing BrdU-ELISA. 3β-adiol led to a significant reduction in proliferation in LNCaP and VCaP derivatives, whereas PC-3 and BPH-1 showed no changes in proliferation (n.s. = not significant, ^*P < 0.05, ^**P < 0.005, ^***P < 0.0005).

Figure 3. Cell cycle changes in various CRPC-cells after the application of 3α-adiol, 3β-adiol, and R1881

Prostate cancer cell lines were treated with 1 μmol/L 3α-adiol, 1 μmol/L 3β-adiol or 10 nmol/L R1881. Cell cycle was assessed using the commercial Muse™ Cell Cycle assay kit. In VCaP and VCaPAA cells, 3a-, 3β-adiol significantly increased cell numbers in the G0/G1 phases, whereas cells in the S or G2/M phases were significantly reduced. R1881 showed an opposite effect with a reduced number of cells in the G0/G1 phases and increased number of cells in the S and G2/M phases. In VCapAA cells, no significant changes were detected (A–F). In LNCAP 3a- and 3β-adiol application led to increases in cells in the G0/G1 phases, whereas the number of cells in S and G2/M phases was significantly reduced. R1881 resulted in inconsistent changes (G–I) (n.s. = not significant, ^*P < 0.05, ^**P < 0.005, ^***P < 0.0005).

Figure 4. Androgen receptor (AR), Oestrogen receptor (ER)a, and ERβ expression after treatment with 3α-adiol, 3β-adiol, and R1881

Western blotting was performed to determine protein expression. 3α-adiol (1 μmol/L) or 3β-adiol (1 μmol) considerably reduced AR expression in VCaPrev cells, whereas 10 nmol/L R1881 resulted in a slight induction of AR expression. In addition, no significant changes in ERa or ERβ expression were detectable (A). In VCaP cells, 1 μmol/L 3α-adiol or 1 μmol/L 3β-adiol reduced AR expression, whereas 10 nmol/L R1881 resulted in elevated AR expression. R1881, but not 3α-adiol and 3β-adiol, stimulated the expression of ERa or ERβ (B). 3α-adiol (1 μmol/L) or 3β-adiol (1 μmol/L) markedly reduced AR expression in VCaPAA cells, whereas 10 nmol/L R1881 resulted in a slight increase in AR expression. 3α-adiol (1 μmol/L) or 3β-adiol (1 μmol/L), as well as androgen stimulation with R1881, reduced ERa expression and increased ERβ expression (C). The opposite regulation of protein expression (downregulation for ERa and upregulation for ERβ) by 1 μmol/L 3α-adiol or 1 μmol/L 3β-adiol was even more pronounced for LNCaP derivatives (D–F).

Figure 5. ARV7 expression in VCaP under increasing androgen deprivation

With increasing androgen deprivation therapy (ADT), the expression of full-length androgen receptor (flAR) continuously increased (as shown in Figure 4A–4C), which was accompanied by a stepwise increase in specific antibody staining for the AR splice variant 7 (ARV7), but only with maximum ADT (A). The strong ARV7 signal in the presence of abiraterone acetate was reversible upon additional treatments of VCaP AA cells with 1 μmol/L 3α-adiol or 3β-adiol (B).