Role of androgens and the androgen receptor in epithelial-mesenchymal transition and invasion of prostate cancer cells

Abstract

Androgens are functionally required for the normal growth of the prostate gland and in prostate tumor development and progression. Epithelial-mesenchymal-transition (EMT) is an important process during normal development and in cancer cell metastasis induced by factors within the microenvironment, such as transforming growth factor-beta (TGF-beta). This study examined the ability of androgens to influence EMT of prostate cancer epithelial cells. The EMT pattern was evaluated on the basis of expression of the epithelial markers E-cadherin/beta-catenin, and the mesenchymal markers N-cadherin, as well as cytoskeleton reorganization in response to 5alpha-dihydrotestosterone (DHT; 1 nM) and/or TGF-beta (5 ng/ml). Overexpressing and silencing approaches to regulate androgen receptor (AR) expression were conducted to determine the involvement of AR in EMT in the presence or absence of an AR antagonist. Our results demonstrate that androgens induce the EMT pattern in prostate tumor epithelial cell with Snail activation and lead to significant changes in prostate cancer cell migration and invasion potential. Expression levels of AR inversely correlated with androgen-mediated EMT in prostate tumor epithelial cells, pointing to a low AR content required for the EMT phenotype. These findings indicate the ability of androgens to induce EMT by potentially bypassing the functional involvement of TGF-beta, thus contributing to metastatic behavior of prostate cancer cells.-Zhum, M.-L., Kyprianou, N. Role of androgens and the androgen receptor in epithelial-mesenchymal transition and invasion of prostate cancer cells.

Figure 1.

Effect of androgens on EMT of prostate cancer cells. A) Prostate cancer cells (PC-3, LNCaP, and LNCaP TβRII) were treated with DHT (0.1–10 nM) as shown for 72 h. Total cell lysates were analyzed by Western blotting to determine the expression of E-cadherin, β-catenin, and N-cadherin. B) LNCaP, LNCaP TβRII, and C4–2B cells were treated with DHT for 72 h and subjected to subcellular fractionation as described in Materials and Methods. Western blot analysis was performed in cytosolic and nuclear fractions to determine β-catenin levels. GAPDH and PARP served as an internal control for cytosolic and nuclear fractions, respectively. C, D) Combined effect of androgens and TGF-β on EMT markers E-cadherin and β-catenin (C) and Snail protein expression (D). E, F) Results from the real-time PCR analysis of E-cadherin and Snail mRNA levels. LNCaP TβRII and PC-3 cells were treated with DHT (0.1–10 nM) for 24 h, and relative mRNA expression levels of E-cadherin (E) and Snail (F) were evaluated as described in Materials and Methods.

Figure 2.

Androgens regulate cytoskeleton reorganization of prostate cancer cells. A) LNCaP TβRII cells were exposed to DHT (1 nM) and/or TGF-β (5 ng/ml). Expression of actin, talin, cofilin, and tubulin was determined by Western blot analysis. GAPDH served as internal loading control. B) LNCaP TβRII cells were exposed to DHT and subjected to immunofluorescence for actin, talin, and cofilin detection. Level of cytoskeleton proteins was assessed by FACS. C) LNCaP TβRII cells were treated with DHT and/or TGF-β, and F-actin was detected using FITC-phalloidin under fluorescent microscopy. Red arrowheads indicate microvilli formation. D) Actin colocalization of talin and cofilin in response to androgens (yellow arrowheads); after treatment with DHT, LNCaP TβRII cells were subjected to immunofluorescence: red indicates cofilin (top panels) and talin (bottom panels), respectively; green indicates actin; blue indicates nuclear staining; and yellow indicates colocalization.

Figure 3.

Effect of androgens and TGF-β on prostate cancer cell invasive behavior. A) LNCaP TβRII cells were treated with DHT (1 nM) and/or TGF-β (5 ng/ml) for 24, 48, and 72 h, and cell migration was determined. B) Effect of androgens on prostate cancer cell invasion. LNCaP and LNCaP TβRII cells were exposed to DHT for 48 h, and cell invasion was assessed as described in Materials and Methods.

Figure 4.

Detection of AR in PC-3 Cells. A) PC-3 cells were treated with increasing doses (0.1, 1, and 10 nM) of BSA-testosterone. Expression patterns of β-catenin, E-cadherin, and N-cadherin were evaluated by Western blotting. GAPDH was used as a loading control. B) AR expression in PC-3 cells after DHT treatment was detected by Western blot analysis. C) PC-3 cells were treated with DHT (0.1–10 nM) and Casodex (10 μM). Immunoblotting was used to assess expression of E-cadherin, β-catenin, N-cadherin, and vimentin. D) AR expression in PC-3 cells after DHT treatment, as determined by immunofluorescence followed by FACS analysis.

Figure 5.

Relationship between AR status and EMT. A) Results of AR overexpression in prostate cancer cells. Expression vectors encoding the wild-type AR and mutant AR (harboring the LNCaP AR mutation) were transfected in PC-3 cells. Expression of AR was detected in stable clones by Western blot analysis. B) ShAR RNA was transfected into LNCaP cells and CW22 cells, and stable transfectants were generated. Reduction/loss of AR protein was examined by Western blotting. C) PC-3 Zeo, PC-3-hAR, and PC-3-LAR cells were treated with DHT (1 nM) and/or TGF-β (5 ng/ml). Expression of E-cadherin, β-catenin, and N-cadherin were determined by Western blot analysis. GAPDH served as internal control. D) Expression profile of E-cadherin and β-catenin in LNCaP-null vector control and LNCaP ShAR cells; CW22-null vector control and CW22 ShAR cells after treatment with DHT and TGF-β.

Figure 6.

AR involvement in EMT-related cytoskeleton reorganization and cell invasion. A) PC-3 Zeo, PC-3-hAR, and PC-3-LAR cells were treated with DHT, and their invasion ability was assessed. B) Effect of AR loss on the invasion ability of prostate cancer cells. LNCaP-null vector control cells and LNCaP ARSh cells; CW22-null vector control cells and CW22 ShAR-silenced cells were exposed to androgens, and their invasion potential was determined. C) PC-3 Zeo, PC-3-hAR, and PC-3-LAR cells were treated with DHT and subjected to immunofluorescence analysis as described in Materials and Methods. Red indicates cofilin; green indicates actin microfilaments; blue indicates nuclei. D) LNCaP-null vector control cells and LNCaP AR sh cells; CW22-null vector control cells and CW22 ShAR cells were treated with DHT, and immunofluorescence analysis for actin (green), cofilin (red, top panels), and talin (red, bottom panels) was conducted.