5alphaDH-DOC (5alpha-dihydro-deoxycorticosterone) activates androgen receptor in castration-resistant prostate cancer

Abstract

Prostate cancer often relapses during androgen-depletion therapy, even under the castration condition in which circulating androgens are drastically reduced. High expressions of androgen receptor (AR) and genes involved in androgen metabolism indicate a continued role for AR in castration-resistant prostate cancers (CRPCs). There is increasing evidence that some amounts of 5alpha-dihydrotestosterone (DHT) and other androgens are present sufficiently to activate AR within CRPC tissues, and enzymes involved in the androgen and steroid metabolism, such as 5alpha-steroid reductases, are activated in CRPCs. In this report, we screened eight natural 5alphaDH-steroids to search for novel products of 5alpha-steroid reductases, and identified 11-deoxycorticosterone (DOC) as a novel substrate for 5alpha-steroid reductases in CRPCs. 11-Deoxycorticosterone (DOC) and 5alpha-dihydro-deoxycorticosterone (5alphaDH-DOC) could promote prostate cancer cell proliferation through AR activation, and type 1 5alpha-steroid reductase (SRD5A1) could convert from DOC to 5alphaDH-DOC. Sensitive liquid chromatography-tandem mass spectrometric analysis detected 5alphaDH-DOC in some clinical CRPC tissues. These findings implicated that under an extremely low level of DHT, 5alphaDH-DOC and other products of 5alpha-steroid reductases within CRPC tissues might activate the AR pathway for prostate cancer cell proliferation and survival under castration.

Figure 1

Cell growth assays of 5αDH steroids and their precursors. 5α‐Dihydro‐deoxycorticosterone (5αDH‐DOC) (a) and 5αDH‐progesterone (c) at 10⁻⁷ M or lower concentration showed growth‐promoting effect on both androgen receptor (AR)‐positive prostate cancer (PC) cell lines 22Rv1 and LNCaP, but did not on AR‐negative PC cells DU‐145 and PC‐3. (b,d) Their 4‐ene‐3‐oxosteroids (DOC and progesterone) had also some ability to stimulate cell proliferation. Prostate cancer (PC) cells were treated with the indicated concentration of each steroid (x‐axis, 10⁻¹⁰–10⁻⁶ M) and 72 h later, MTT (3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromide) assay was performed. Y‐axis: absorbance (ABS) at 490 nm (MTT assay), and at 630 nm as a reference, measured with a microplate reader. Each assay was tested six times and the means ± SD were plotted.

Figure 2

Luciferase assays of 5αDH steroids and their precursors for androgen receptor (AR) transactivation activity. (a) LNCaP cells and 22Rv1 cells were transfected with pGL3‐PSA and the luciferase activity in the presence of indicated concentration (X‐axis) of 5α‐dihydro‐deoxycorti‐costerone (5αDH‐DOC) and 11‐deoxycorticosterone (DOC) was compared. LNCaP cells and 22Rv1 cells expressed mutant AR, T877A, and H874Y, respectively. (b) Androgen receptor (AR)‐null PC‐3 cells were co‐transfected with wild‐type AR (left), mutant AR‐T877A (right), or mock (lower) and pGL3‐PSA and the luciferase activity in the presence of indicated concentration (x‐axis) of 5αDH‐DOC and DOC was compared. These luciferase activities were blocked by antiandrogen bicalutamide (BCL). Each assay was tested six times and the means ± SD were plotted. *P < 0.05, **P < 0.01 by Student’s t‐test. (c) 22Rv1 cells were transfected with pGL3‐PSA and the luciferase activity in the presence of indicated concentration (x‐axis) of 5αDH‐progesterone and progesterone was compared. (d) Androgen receptor (AR)‐null PC‐3 cells were co‐transfected with wild‐type AR and pGL3‐PSA and the luciferase activity in the presence of indicated concentration (x‐axis) of 5αDH‐progesterone and progesterone was compared. Each assay was tested six times and the means ± SD were plotted. *P < 0.05, **P < 0.01 by Student’s t‐test.

Figure 3

Type 1 5α‐steroid reductase (SRD5A1) was responsible for the production of 5α‐dihydro‐deoxycorticosterone (5αDH‐DOC). (a) COS7 cells were transfected with SRD5A1, SRD5A3, or mock vector and the expression of exogenous SRD5A1 and SRD5A3 were evaluated by western blot analysis using anti‐HA tag antibody. (b) The transfected cells were treated with 10⁻⁶ M 11‐deoxycorticosterone (DOC), and after 1 h of incubation, the conditioned media were harvested. Liquid chromatography‐tandem mass spectrometry (LC‐MS/MS) analysis of the media specifically detected the production of 5αDH‐DOC which was converted from DOC in COS7 cells overexpressing SRD5A1 (middle panel), but not in COS7 cells overexpressing SRD5A3 (lower panel) or mock cells (upper panel). These experiments were performed in duplicate (right and left panels). (c) 1‐μM Dutasteride treatment inhibited 5αDH‐DOC production in COS7 cells transfected with SRD5A1 vector. 5α‐Dihydro‐deoxycorticosterone (5αDH‐DOC)was detected by LC‐MS/MS and these experiments were performed in duplicate (right and left panels).

Figure 4

(a) Reverse transcription (RT)‐PCR confirmed knockdown effect on type 1 5α‐steroid reductase (SRD5A1) expression by siSRD5A1 in 22Rv1 cells. β‐Actin (ACTB) was used to quantify the input RNAs. (b) Knockdown of SRD5A1 expression by siSRD5A1 suppressed the proliferation of 22Rv1 cells, compared with the control RNA duplex siEGFP (P < 0.01, Student’s t‐test). Y‐axis, absorbance (ABS) at 490 nm (MTT assay), and at 630 nm as a reference, measured with a microplate reader. (c) Suppression of SRD5A1 expression by siSRD5A1 reduced 5α‐dihydro‐deoxycorticosterone (5αDH‐DOC) production in 22Rv1 cells (P < 0.01, Student’s t‐test). 5α‐Dihydro‐deoxycorticosterone (5αDH‐DOC) in the media was measured by liquid chromatography‐tandem mass spectrometry (LC‐MS/MS) analysis. (d) Detection of 5αDH‐DOC and DHT in 13 clinical CRPC tissues. 5α‐Dihydro‐deoxycorticosterone (5αDH‐DOC) was detected by sensitive LC‐MS/MS in clinical CRPC tissues. 5α‐Dihydrotestosterone (DHT) was also measured by sensitive LC‐MS/MS in the same tissues, and the level of 5αDH‐DOC was inversely correlated with the low level of DHT concentration (Pearson r = −0.5727, P < 0.05).