Novel 5 alpha-steroid reductase (SRD5A3, type-3) is overexpressed in hormone-refractory prostate cancer

Abstract

Prostate cancer often relapses during androgen-depletion therapy, even under conditions in which a drastic reduction of circulating androgens is observed. There is some evidence that androgens remain present in the tissues of hormone-refractory prostate cancers (HRPC), and enzymes involved in the androgen and steroid metabolic pathway are likely to be active in HRPC cells. We previously carried out a genome-wide gene expression profile analysis of clinical HRPC cells by means of cDNA microarrays in combination with microdissection of cancer cells and found dozens of transactivated genes. Among them, we here report the identification of a novel gene, SRD5A2L, encoding a putative 5 alpha-steroid reductase that produces the most potent androgen, 5 alpha-dihydrotestosterone (DHT), from testosterone. Liquid chromatography-tandem mass spectrometry analysis following an in vitro 5 alpha-steroid reductase reaction validated its ability to produce DHT from testosterone, similar to type 1 5 alpha-steroid reductase. Because two types of 5 alpha-steroid reductase were previously reported, we termed this novel 5 alpha-steroid reductase 'type 3 5 alpha-steroid reductase' (SRD5A3). Reverse transcription-polymerase chain reaction and northern blot analyses confirmed its overexpression in HRPC cells, and indicated no or little expression in normal adult organs. Knockdown of SRD5A3 expression by small interfering RNA in prostate cancer cells resulted in a significant decrease in DHT production and a drastic reduction in cell viability. These findings indicate that a novel type 3 5 alpha-steroid reductase, SRD5A3, is associated with DHT production and maintenance of androgen-androgen receptor-pathway activation in HRPC cells, and that this enzymatic activity should be a promising molecular target for prostate cancer therapy.

Figure 1

(a) Alignment of the amino acid sequence of human SRD5A3 (type 3; GenBank accession no. NP_078868) with those of human SRD5A1 (type 1; GenBank accession no. GenBank accession no. NP_001038), and human SRD5A2 (type 2, NP_000339). The underlined sequence of SRD5A3 is a hydrophobic region that was deleted in SRD5A3‐ΔN (1–58) and the arrow indicates H296, which is one of the most conserved residues in all three 5α‐steroid reductases. (b) Semiquantitative reverse transcription–polymerase chain reaction validated that SRD5A3 expression was upregulated in the microdissected hormone‐refractory prostate cancer (HRPC) cells, compared with hormone‐naïve prostate cancer (HNPC) cells and normal prostatic epithelial (NP) cells, which were also microdissected. SRD5A1 expression was also upregulated in HRPC cells, whereas SRD5A2 expression was suppressed in HNPC and HRPC cells. Expression of ACTB served as the quantitative control. (c) Northern blot analysis showed a high level of SRD5A3 expression in five prostate cell lines (lanes 1–5), whereas its expression was hardly detectable in vital adult organs, including heart, lung, liver, kidney, brain, and prostate (lanes 6–11). (d) Multiple tissue northern blot analysis showed faint expression only in the normal pancreas and no or very low expression of SRD5A3 in adult normal organs.

Figure 2

In vitro 5α‐reductase activity. (a) The enzyme source was obtained from COS7 cells transfected with each of five hemagglutinin (HA)‐tagged expression vectors (SRD5A1, wild‐type SRD5A3, SRD5A3‐ΔN, SRD5A3‐H296A, and Mock). Western blot analysis using HA‐antibody validated their expression. (b) The liquid chromatography‐tandem mass spectrometry chromatogram demonstrated 5α‐dihydrotestosterone (DHT) production in SRD5A3‐transfected cells (panel 3) as well as SRD5A1‐transfected cells (panel 2). Panel 1 shows DHT production in Mock‐transfected COS7 cells as the background. The mutant SRD5A3 (SRD5A3‐ΔN and SRD5A3‐H296A) did not produce DHT (panels 4 and 5).

Figure 3

Effect of SRD5A3 knockdown by small interfering RNA (siRNA) on the growth of prostate caner cells. (a) The siRNA expression vectors specific to the SRD5A3 transcript (si#524 and si#790) or scrambled siRNA (si#SC) and enhanced green fluorescent protein (EGFP) siRNA expression vector (si#EGFP) as negative controls were transfected into 22Rv1 cells. The knockdown effect on the SRD5A3 transcript was validated by reverse transcription–polymerase chain reaction, with ACTB expression as a quantitative control. Transfection with si#524 and si#790 showed a significant knockdown effect, whereas si#SC or si#EGFP showed no effect on the level of SRD5A3 transcript. (b) Transfection with si#524 or si#790 resulted in a drastic reduction in the number of viable cells evaluated by colony‐formation assay, compared with the cells transfected with either of the other siRNA expression vectors that showed no knockdown effect on SRD5A3 expression. (c) Transfection with si#524 or si#790 resulted in a drastic reduction in the number of viable cells measured by MTT assay (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.

Figure 4

Suppression of SRD5A3 expression attenuated 5α‐dihydrotestosterone (DHT) production in C4‐2B cells. (a) Reverse transcription–polymerase chain reaction validated the knockdown effect of SRD5A3 expression and almost unchanged expression of SRD5A1 in C4‐2B cells by SRD5A3 small interfering RNA (siRNA). SRD5A2 was not expressed in C4‐2B cells. ACTB expression served as a quantitative control. (b) The ratio of the total amounts of DHT and testosterone in prostate caner cells and their media. Transfection of the RNA duplex to SRD5A3 (si#790) significantly suppressed the conversion from testosterone to DHT in C4‐2B cells, compared to the control RNA duplex si#EGFP (P < 0.05, Student's t‐test). The y‐axis reveals the ratio of the total DHT to the total testosterone (T).