Hydrazinobenzoylcurcumin inhibits androgen receptor activity and growth of castration-resistant prostate cancer in mice

Abstract

There is a critical need for therapeutic agents that can target the amino-terminal domain (NTD) of androgen receptor (AR) for the treatment of castration-resistant prostate cancer (CRPC). Calmodulin (CaM) binds to the AR NTD and regulates AR activity. We discovered that Hydrazinobenzoylcurcumin (HBC), which binds exclusively to CaM, inhibited AR activity. HBC abrogated AR interaction with CaM, suppressed phosphorylation of AR Serine81, and blocked the binding of AR to androgen-response elements. RNA-Seq analysis identified 57 androgen-regulated genes whose expression was significantly (p ≤ 0.002) altered in HBC treated cells as compared to controls. Oncomine analysis revealed that genes repressed by HBC are those that are usually overexpressed in prostate cancer (PCa) and genes stimulated by HBC are those that are often down-regulated in PCa, suggesting a reversing effect of HBC on androgen-regulated gene expression associated with PCa. Ingenuity Pathway Analysis revealed a role of HBC affected genes in cellular functions associated with proliferation and survival. HBC was readily absorbed into the systemic circulation and inhibited the growth of xenografted CRPC tumors in nude mice. These observations demonstrate that HBC inhibits AR activity by targeting the AR NTD and suggest potential usefulness of HBC for effective treatment of CRPC.

Keywords: CTK7A; androgen receptor; calmodulin; castration-resistant prostate cancer; hydrazinobenzoylcurcumin.

Figure 1. Immunohistochemistry of CaM and AR in human prostate tumor and non-tumor tissues

A) Evaluation of CaM expression in non-tumor (benign) and tumor prostate tissues. B) Evaluation of CaM expression in serial sections of prostate tumors with high (AR_3+) and low (AR_1+) AR levels. Immunostaining is as described in Materials and Methods. Images are representative of five each of benign and, AR_3+ and AR_1+ tumor tissue specimens. Table shows percentage of luminal epithelial cells in AR_3+ and AR_1+ tumors with elevated nuclear CaM. Percentage of cells with elevated nuclear CaM was determined by counting more than 300 cells in each tumor tissue.

Figure 2. HBC inhibits proliferation of AR-positive prostate cancer cells

A) Exponentially growing LNCaP and PC-3 cells were treated with increasing concentrations of HBC for 24 hours and the incorporation of pulse labeled ³H-thymidine into DNA was determined as described in Materials and Methods. Each data point is mean and SE obtained from 3 biological replicates, and representative of 2 independent experiments. B) Individual cell lines were plated at a density of 5×10⁴ cells/35 mm² dish, treated with solvent alone (control) or 20 μM HBC starting a day after plating, and the number cells of cells in each dish was determined at day 6 as described in Materials and Methods. Each column is mean and SE of values obtained from triplicate dishes, and the data is representative of 2 independent experiments. C) Exponentially growing LNCaP cells were treated with increasing concentrations of HBC for 24 hours; cell extracts were prepared and Western blot analysis was performed as described in Materials and Methods. Band densities were determined by using the ImageJ program and β-actin-normalized relative densities are presented below each blot.

Figure 3. HBC inhibits AR transcriptional activity

A) Exponentially growing LNCaP cells were treated with increasing concentrations of HBC for 24 hours, and AR, PSA and β-actin levels in cell extracts were determined. B) Exponentially growing LNCaP cells were treated with 40 μM HBC or Casodex for 24 hours and RT-PCR performed to evaluate PSA, NKX3.1 and GAPDH mRNA expression. The relative density of GAPDH normalized PSA and NKX3.1 bands was determined using EagleSight software (version 3.2; Stratagene). C) HBC disrupts AR-CaM interaction. Exponentially growing LNCaP cells were treated with 40 μM HBC or Casodex for 24 hours, AR and CaM immunoprecipitates were prepared and Western blot analysis was performed as described in Materials and Methods. Input represents 10% of the protein used for immunoprecipitation. Observations are representative of 3 independent experiments. D) HBC inhibits AR binding to androgen response elements. EMSA reactions were carried out in the presence of 40 μM HBC or Casodex (CDX), or vehicle (control) and subjected to non-denaturing gel electrophoresis. EMSA was performed as described in Materials and Method. The data is representative of 2 independent experiments. Density of bands in each lane were determined by using the ImageJ program and relative densities were calculated after subtracting the background density of bands in DU145 lane from the density of bands in LNCaP lanes.

Figure 4. Effect of HBC on androgen-stimulated AR

A and B) HBC inhibits androgen-stimulated AR activity. Exponentially growing LNCaP cells (FCS) were hormone-deprived (CSS) and stimulated with 2 nM R1881 in the absence or presence of HBC for 24 hours. AR, PSA and β-actin protein (A) and PSA, NKX3.1, and GAPDH mRNA levels (B) were determined by Western blot and RT-PCR, respectively. Cells treated with 40 μM HBC were used for RT-PCR analysis. Band densities were determined by using the ImageJ program and β-actin- (A) and GAPDH- (B) normalized relative band densities are presented. C) HBC inhibits AR association with PSA- AREs. Hormone-deprived LNCaP cells were treated with 2 nM R1881 in the presence of 40 μM HBC, chromatin-immunoprecipitates were prepared using ant-AR-N20 antibodies (AR-ChIP), and DNA was purified from AR-ChIP and subjected to PCR analysis. ChIP prepared using purified IgG (IgG-ChIP) served as a negative control. Input represents 10% of chromatin used for Immunoprecipitation. D) HBC inhibits phosphorylation of AR Serine 81 residue. Cell extracts prepared from hormone-deprived cells stimulated with R1881 (2 nM) in the presence of HBC (40 μM) for 2, 4, or 6 hours were subjected to Western blot analysis using antibodies against AR-pSer81, AR-N20 and β-actin. All experimental procedures are as described in Materials and Methods. Data in each panel is representative of 2 or more independent experiments.

Figure 5. Analysis and validation of HBC affected androgen-regulated gene expression

Hormone-deprived LNCaP cells were treated with 2 nM R1881 in the absence or presence of 40 μM HBC for 24 hours and androgen-regulated genes whose expression affected by HBC were identified through RNA-Seq as described in Materials and Methods. A and B) Androgen-regulated genes whose expression was significantly (p<0.002) repressed (A) or stimulated (B) by HBC were analyzed for differential expression in cancer vs. normal prostate tissues by using nine reference data sets (Supplementary Table S2) in Oncomine. The heat maps contain individual studies. The heat map intensity corresponds to percentile overexpression (red) or repression (blue) in prostate carcinoma as compared to normal prostate. C) RT-qPCR validation of representative androgen-regulated genes that were repressed (a and b) or stimulated (c) in HBC treated cells. D) HBC suppresses the expression of androgen/AR-regulated genes in 22Rv1 cells as well as in LNCaP cells. Exponentially growing LNCaP and 22Rv1 cells were treated with HBC (40 μM) for 24 hours, and RT-PCR of selected genes in total RNA was performed as described in Materials and Methods. E) Ingenuity Pathway Analysis (IPA) was used to identify AR- and/or CaM-regulated genes among the genes that were affected significantly (p<0.05) by HBC. F) IPA based identification of network interactions between the eight HBC repressed genes that are regulated both by AR and CaM.

Figure 6. Effect of HBC on tumors derived from castration-resistant C4-2B cells in nude mice

A) Concentration-time profiles of HBC in plasma (μg/ml) and in tissues (μg/gm) following intra-peritoneal injection of 100 mg/kg in mice. Each point represents the mean concentration from two mice. B) HBC suppresses the growth of tumors derived from C4-2B cells in nude mice. Nude mice bearing C4-2B tumors were given twice daily (5 days/week) i.p. injection of 100 mg/kg HBC or vehicle and tumor size was measured twice a week as described in Materials and Methods. Data points, mean tumor size (n = 5); bars, SE; *, p<0.05, and **, p<0.01. C) HBC decreases AR and cyclin A levels and increases p27 levels in tumor tissues. Western blot analysis of AR, cyclin A, and β-actin in tissue extracts of tumors from vehicle and HBC treated mice is as described in Materials and Methods. D) HBC suppresses the expression of Ki67 and inhibits histone H3K9 acetylation. Immunostaining of Ki67 and histone H3 acetyl K9 in tumor tissues is as described in Materials and Methods.