Cryptotanshinone suppresses androgen receptor-mediated growth in androgen dependent and castration resistant prostate cancer cells

Abstract

Androgen receptor (AR) is the major therapeutic target for the treatment of prostate cancer (PCa). Anti-androgens to reduce or prevent androgens binding to AR are widely used to suppress AR-mediated PCa growth; however, the androgen depletion therapy is only effective for a short period of time. Here we found a natural product/Chinese herbal medicine cryptotanshinone (CTS), with a structure similar to dihydrotestosterone (DHT), can effectively inhibit the DHT-induced AR transactivation and prostate cancer cell growth. Our results indicated that 0.5 μM CTS effectively suppresses the growth of AR-positive PCa cells, but has little effect on AR negative PC-3 cells and non-malignant prostate epithelial cells. Furthermore, our data indicated that CTS could modulate AR transactivation and suppress the DHT-mediated AR target genes (PSA, TMPRSS2, and TMEPA1) expression in both androgen responsive PCa LNCaP cells and castration resistant CWR22rv1 cells. Importantly, CTS selectively inhibits AR without repressing the activities of other nuclear receptors, including ERα, GR, and PR. The mechanistic studies indicate that CTS functions as an AR inhibitor to suppress androgen/AR-mediated cell growth and PSA expression by blocking AR dimerization and the AR-coregulator complex formation. Furthermore, we showed that CTS effectively inhibits CWR22Rv1 cell growth and expressions of AR target genes in the xenograft animal model. The previously un-described mechanisms of CTS may explain how CTS inhibits the growth of PCa cells and help us to establish new therapeutic concepts for the treatment of PCa.

Fig. 1

CTS selectively inhibits DHT-mediated AR transactivation, but not the ER, PR, and GR activity. A: Structure similarity of DHT and CTS. Left: DHT structure Right: CTS structure, similar to DHT, contains three cyclohexane rings (designated as rings A, B, and C in the left panel) and one furan ring (the D ring) composed of seventeen carbon atoms. B & C: Inhibition of CTS on the androgen-induced AR transcriptional activity in HEK 293 cells. D-F: No effects of CTS on the transcriptional activities of estrogen-induced ERα, Dex-induced GR, and progesterone-induced PR in HEK 293 cells. G: Inhibition of CTS on the DHT-induced AR transcriptional activity of a gain-function mutant AR (T877A) in prostate cancer LNCaP cells. H: Inhibition of CTS on the DHT-induced AR transcriptional activity in CWR22Rv1 cells. MMTV-Luc or ERE-Luc activities were determined. Data represent mean ± SD from three independent experiments. I & J: Inhibition of CTS on the R1881 induced AR transcriptional activity in LNCaP and CWR22Rv1 cells.

Fig. 2

Differential growth inhibition effects of CTS on different prostate cells. Cells were treated with EtOH and CTS (0.5 µM) in the absence or presence of 1 nM DHT. Medium with indicated treatments was refreshed every 2 days (the half-life of CTS is 22.5±1.5 hr, data not shown) for a total of 7 days. A: CTS inhibits the DHT-induced growth of LNCaP cells. B: CTS inhibits the DHT-induced growth of CWR22Rv1 cells. C: No effects of CTS on the growth of AR-negative PC-3 cells. D: No effects of CTS on non-malignant prostate RWPE1 cells. Data represent mean ± SD of three independent experiments with three replicates in each experiment.

Fig. 3

IC₅₀ of CTS in different prostate cancer cells. We determined the cell half-inhibition (IC₅₀) of CTS in LNCaP, CWR22rv1, and PC3. Cells were seeded on 24 well plates in medium with 10%FBS for 24 hr. Medium was then refreshed to medium with 10 %CS-FBS for another 24 hr, and cells were treated with serial concentrations of CTS with or without 1 nM DHT for 2 days. Cells growth and IC₅₀ value were determined by MTT assay. Half-inhibition of CTS was shown in Fig. 3A–3C when cells were treated with 10 nM DHT, and in Fig. 3D–3F when cells were not treated with DHT. Data represent mean ± SD of two independent experiments with three replicates in each experiment.

Fig. 4

CTS Inhibits the AR target gene expression in LNCaP and CWR22Rv1 cells. Cells were treated with ethanol or CTS (0.5 µM) in the absence or presence of 1 nM DHT for 2 days. We used real-time RT-PCR to analyze the mRNA expressions of AR target genes, PSA, TMPRSS2, and TMEPA1, in LNCaP and CWR22Rv1 cells. The respective mRNA levels of these genes in each treatment group were displayed as fold changes compared to the untreated group. Data are shown as the mean ± SD of three independent experiments with three replicates in each experiment.

Fig. 5

CTS inhibited AR transactivation is not via changing AR protein expression or stability. A and B: Western blot analyses of PSA and AR levels in control and CTS treated LNCaP or CWR22Rv1 cells in the absence or presence of DHT. 50 µg of total protein from cells was applied onto a 10% sodium dodecylsulfate-polyacrylamide gel and subjected to electrophoresis followed by Western blot using anti-AR or anti-PSA antibodies. The values shown in Supplementary Fig. 1 represent changes in density of the bands normalized to α-tubulin using the Image Lab statistics software (the representative graph quantitation are shown). All of the data were validated by three independent experiments. C. CTS inhibits the transactivation of the full-length AR (flAR), but not the constitutive activated N-terminal AR. pSG5-ARN-DBD and pSG5-flAR were transfected into HEK 293 cells. After 24 hrs transfection, cells were treated with or without 10 nM DHT and 2.5 µM CTS. D. CTS cannot effectively bind to AR. We used competitive ligand binding assay to determine whether CTS can specifically bind to the AR. LNCaP cell medium was changed to RPMI with10% CS-FBS for 24 hr and 1 nM ³H-R1881 was then added into culture medium with or without DHT or CTS for 1 hr. Unlabeled DHT with the concentrations of 0.1, 1, 10, 100, 1000 nM were used to compete for the ³H-R1881 binding as positive control (left). CTS at the concentration ranges of 0.1, 0.25, 1, 2.5, 10, 25 µM were used to determine the potential AR antagonist effects (right).

Fig. 6

A. CTS inhibits the interaction of AR N-terminus and C-terminus using mammalian 2-hybrid interaction assay. B. CTS inhibits the interaction of AR and AR coregulator using mammalian 2-hybrid assay. C. CTS inhibits the E2 and Adiol-induced full-length (flAR) and transactivation. MMTV-Luc reporter was activated through AR in the presence of 10 nM DHT, E2 or Adiol (lanes 2–4). 2.5 µM CTS could effectively inhibit the DHT, E2, and Adiol stimulated AR activity (lanes 6–8). The solvent (EtOH) treated AR-baseline transcriptional activity was counted as 1 fold (lane 1). Data were averaged from three independent experiments.

Fig. 7. CTS treatments inhibit cancer growth using the in vivo CWR22Rv1 xenograft PCa model

To test the therapeutic effect of CTS, 1x10⁶ CWR22Rv1 cells per site were injected subcutaneously into nude mice at 7 weeks (W) of age. One week after implantation, when tumors established to the size >50 mm³, mice were i.p. injected with vehicle (DMSO), low dose of CTS (5 mg/kg), or high dose of CTS (25 mg/mice), every two days for 4 weeks (from 8 to 12 weeks old). The mice were then sacrificed and the tumors were collected for data analysis. A. CTS treatments inhibit CWR22Rv1 tumor size in a dose dependent manner. B. CTS treatments inhibit CWR22Rv1 tumor weights (N=6 for each group, P<0.05). C. The mouse body weights were not affected by mock or CTS treatments. The right panel showed an enlarged measurement scale of body weights. As the mock control mice carried a bigger tumor, thus the BW is around 1.2 g slightly heavier than the CTS treated group at the end of experiment.