Abstract Despite availability of improved androgen receptor (AR) inhibitors for the treatment of castration-resistant prostate cancer (CRPC), resistance to treatment develops, and has been traced to activation of multiple signaling pathways suppressed by the AR. We previously showed that the AR maintained castration sensitivity in prostate cancer (PCa) by transcriptional regulation of the E3 ubiquitin ligase Nrdp1, which degrades the receptor tyrosine kinase (RTK) ErbB3; whereas AR inhibition suppressed Nrdp1 levels, thereby activating ErbB3. This resulted in CRPC growth and AR stimulation, but in CRPC, the AR was unable to regulate Nrdp1 or suppress ErbB3, causing uncontrolled progression. Here, we investigate the mechanism by which the AR regulates Nrdp1 transcription and why this regulation is lost in CRPC. Immunohistochemical studies in human PCa tissue and in PCa mouse models demonstrated Nrdp1 localization in both nucleus and cytoplasm. In vitro studies determined cytoplasmic Nrdp1 as 36kDa while nuclear Nrdp1 is 28kDa, the 36kDa form, but not the 28kDa form, negatively correlated with ErbB3 levels, but both forms positively correlated with AR. We demonstrate that Nrdp1 is a direct transcriptional target of the AR in androgen-dependent cells expressing a truncated form of the structural protein Filamin A (FlnA) in the nucleus, but in CRPC cells which have lost nuclear FlnA expression, the AR is no longer able to bind to the Nrdp1 promoter. Restoration of nuclear FlnA restored the ability of AR to regulate Nrdp1 transcription. Thus dual targeting of ErbB3 and AR will be effective in patients whose tumors express nuclear FlnA. Citation Format: Rosalinda M. Savoy, Liqun Chen, Salma Siddiqui, Frank U. Melgoza, Blythe Durbin-Johnson, Mohana Roy, Maitreyee K. Jathal, Swagata Bose, Yu Wang, Benjamin Mooso, Leandro D'Abronzo, William H. Fry, Kermit L. Carraway, Paramita M. Ghosh. Androgen receptor transcriptionally targets the ErbB3 regulator Nrdp1 in the presence of nuclear Filamin A. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3578. doi:10.1158/1538-7445.AM2013-3578

Abstract Recurrent prostate cancer (PCa) is treated with androgen deprivation therapy (ADT, chemical castration), but patients on this treatment frequently relapse, indicative of the advent of castration resistant prostate cancer (CRPC). Due to the multifocal nature of the disease, any one treatment may not affect all pathways altered in CRPC; therefore, we set out to delay CRPC by increasing the apoptotic rate during ADT. We previously published that Filamin A (FlnA) is localized to the nucleus in androgen-dependent cells, whereas its expression is lost from the nucleus in CRPC. Restoration of FlnA nuclear expression induced apoptosis in the absence of androgens. We now show that genistein combined polysaccharide (GCP), a combination of isoflavones, can induce FlnA translocation to the nucleus. GCP and its main component genistein previously failed to prevent progression in pre-prostatectomy patients with low-grade PCa. However, our new data, both in vitro and in an animal model, show that GCP impedes relapse following castration, and induces apoptosis in PCa cells undergoing ADT, by promoting FlnA cleavage and nuclear translocation. Investigation of the mechanism of GCP's effects on FlnA showed that ADT phosphorylates FlnA and prevents its cleavage, whereas GCP inhibits FlnA phosphorylation, thereby promoting its cleavage and nuclear translocation. Additional data show that nuclear localization of FlnA induces apoptosis during ADT by preventing androgen receptor (AR) stabilization and by suppression of the ErbB3/PI3K/Akt axis. Taken together, our results identify GCP as a putative therapeutic module to prolong the effectiveness of ADT in patients with metastatic prostate cancer. Citation Format: Benjamin A. Mooso, Maria Mudryj, Ralph W. deVere White, Paramita M. Ghosh, Rosalinda M. Savoy, Jean P. Cheung, Ruth L. Vinall, Clifford G. Tepper, Yu Wang, Salma Siddiqui, Roble G. Bedolla, Margarita Mikhailova. Using GCP to improve initial response to androgen deprivation in prostate cancer [abstract]. In: Proceedings of the AACR Special Conference on Advances in Prostate Cancer Research; 2012 Feb 6-9; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2012;72(4 Suppl):Abstract nr C11.

Abstract Background: The ErbB receptor tyrosine kinase family regulates proliferation and survival in prostate cancer (PCa). We recently showed that ErbB3 plays a significant role in increasing androgen receptor (AR) transcriptional activity and in causing castration-resistant PCa (CRPC). Reciprocally, AR maintained castration sensitivity by suppressing ErbB3 levels through transcriptional regulation of E3 ubiquitin ligase Nrdp1, while loss of AR regulation of Nrdp1 resulted in an unrestricted surge in ErbB3 levels, and cell growth. Here, we investigate whether Nrdp1 is a transcriptional target of the AR in PCa cells. Results: The promoter region of Nrdp1 contains three androgen response elements (AREs) - one located 215 aa upstream of the transcriptional start site (ARE.03), and two within an internal promoter (ARE.01 and ARE.02). Chromatin immunoprecipitation (ChIP) studies revealed AR binding in PCa cells to ARE.03. AR binding to ARE.03 was found to be androgen regulated in androgen-dependent LNCaP cells, whereas no AR binding to ARE.01 or ARE.02 has been detected in this cell line. Luciferase assay to determine AR transcriptional activity on ARE.03 showed significant response to androgens, whereas full mutation of ARE.03 abolished AR transcriptional activity. However, the AR failed to bind to ARE.03 in C4-2 and LNCaP-AI cell lines, CRPC sublines of LNCaP cells, although no mutations in these regions were identified, while AR continued to bind the AR to PSA ARE in these cells, indicating that the AR was still active. Thus the AR binds to PSA ARE in both androgen-dependent and –independent cells, but to ARE.03 only in androgen-dependent but not –independent cells. Luciferase assay in LNCaP-AI cells showed decreased AR transcriptional activity on ARE.03 when compared to LNCaP, supporting decreased AR binding to this region. Hence we investigated the cause for the differential binding of the AR in these two areas. The structural protein Filamin A (FlnA) was earlier shown to cleave to a 90 kDa fragment (FlnA16-24) whose nuclear localization maintained androgen dependence. We investigated whether this protein played a role in regulation of AR binding to ARE.03 in PCa cells. Transfection of FlnA16-24 in C4-2 cells resulted in localization of FlnA 16-24 to the nucleus, and restored AR binding to ARE.03. FlnA 16-24 significantly increased luciferase activity on ARE.03, while suppressing AR activity on PSA indicating that this AR binding protein is required for AR-mediated transcription of Nrdp1. Conclusions: Our results indicated that Nrdp1 is regulated in androgen sensitive cells, but not in CRPC cells, by AR binding to ARE.03, whereas no such difference in AR binding to PSA ARE exists. We show that AR binding to ARE.03 is abolished in CRPC cells because of a decreased expression of nuclear FlnA, and that restoration of nuclear FlnA will restore androgen-sensitive cell growth. Our data indicate that the AR-coregulator FlnA is required for AR binding to the Nrdp1 promoter and that loss of this protein results in the failure of AR to regulate Nrdp1 transcription, resulting in unrestricted increase in ErbB3. These data identified Nrdp1 as a novel target of AR transcriptional activity in androgen-dependent but not in CRPC cells.

As prostate cancer (CaP) is regulated by androgen receptor (AR) activity, metastatic CaP is treated with androgen deprivation therapy (ADT). Despite initial response, patients on ADT eventually progress to castration-resistant CaP (CRPC), which is currently incurable. We previously showed that cleavage of the 280 kDa structural protein Filamin A (FlnA) to a 90 kDa fragment, and nuclear localization of the cleaved product, sensitized CRPC cells to ADT. Hence, treatment promoting FlnA nuclear localization would enhance androgen responsiveness. Here, we show that FlnA nuclear localization induced apoptosis in CRPC cells during ADT, identifying it as a treatment tool in advanced CaP. Significantly, the natural product genistein combined polysaccharide (GCP) had a similar effect. Investigation of the mechanism of GCP-induced apoptosis showed that GCP induced FlnA cleavage and nuclear localization and that apoptosis resulting from GCP treatment was mediated by FlnA nuclear localization. Two main components of GCP are genistein and daidzein: the ability of GCP to induce G2 arrest was due to genistein whereas sensitivity to ADT stemmed from daidzein; hence, both were needed to mediate GCP's effects. FlnA cleavage is regulated by its phosphorylation; we show that ADT enhanced FlnA phosphorylation, which prevented its cleavage, whereas GCP inhibited FlnA phosphorylation, thereby sensitizing CaP cells to ADT. In a mouse model of CaP recurrence, GCP, but not vehicle, impeded relapse following castration, indicating that GCP, when administered with ADT, interrupted the development of CRPC. These results demonstrate the efficacy of GCP in promoting FlnA nuclear localization and enhancing androgen responsiveness in CaP.

Androgen ablation therapy is effective in treating androgen-dependent prostate tumors; however, tumors that can proliferate in castrate levels of androgen eventually arise. We previously reported that in CWR22Rv1 (Rv1) cells, the protease calpain 2 can cleave the androgen receptor (AR) into a constitutively active ∼80,000 low molecular weight (LMW) form. In this study, we further dissect the mechanisms that produce the AR LMW forms using Rv1 cells and the related CWR22-R1 (R1) cells. The 39-amino acid insertional mutation in the Rv1-AR (E3DM-AR) sensitizes this AR to calpain 2 proteolysis. R1 cells encode the same AR molecule as the parental CWR22 xenograft. Using calpain 2 small interfering RNA and calpeptin, we find that calpain 2 plays a role in the generation of the LMW-AR in R1 cells. Furthermore, LMW-AR expression is regulated by the activation of calpain 2 by ERK 1 and 2. Inhibition of ERK phosphorylation or small interfering RNA-mediated decrease of ERK expression reduces LMW-AR levels in R1 cells. Conversely, activation of the MAPK pathway results in increased ERK phosphorylation and increased levels of LMW-AR. Finally, analyses of human tumor samples found that LMW-AR levels are higher in tumors that have an increased calpain/calpastatin ratio and/or increased levels of phospho-ERK (pERK). This suggests that a higher calpain/calpastatin ratio collaborates with activated ERK to promote the generation of the LMW-AR.

157 Background: Multiple signaling pathways are involved in the development of CRPC. We previously showed that the mTOR pathway is activated in CRPC cell lines while inhibition of this pathway results in upregulation of androgen receptor (AR) signaling (Wang et al, Oncogene. 2008). Simultaneous blockade of the mTOR and AR pathways synergize in inducing PCa cell death and delaying tumor formation in mouse models. We hypothesize that simultaneous blockade of the AR and mTOR pathways in CRPC patients with bicalutamide and everolimus will result in improved efficacy compared to bicalutamide alone. Methods: A phase II clinical trial with a lead-in safety phase was designed to determine the efficacy and tolerability of the bicalutamide and everolimus combination in CRPC patients compared with bicalutamide alone. Patients must have histologically confirmed disease and demonstrated disease progression (either by PSA or radiographically) while on androgen deprivation therapy. At the lead-in phase, all patients receive both agents. At the phase II stage, patients are randomized to bicalutamide +/− everolimus. The primary endpoint is PSA response. The second endpoints include progression-free survival, time-to-treatment failure, overall survival and toxicity. Here, we report the results of the lead-in phase. Results: Eight patients were recruited at the lead-in phase. The bicalutamide/everolimus combination was well tolerated with no unexpected toxicities. Six of 8 patients have had PSA response after at least 8 weeks of therapy and the remaining two patients had stable PSA response. The median time to disease progression was 6.8 months. Nine patients have been recruited at the phase II stage so far. This clinical trial is being subcontracting to the other sites of the California Cancer Consortium. Tumor and blood specimens are being collected for molecular correlative studies of mTOR pathway markers. Conclusions: The rational combination of bicalutamide and everolimus appears to have promising anti-tumor activity and an acceptable toxicity profile. The randomized phase of the clinical trial is currently ongoing and will be reported. Supported by Novartis. [Table: see text]

Abstract Purpose: We previously showed that nuclear localization of the actin-binding protein, filamin A (FlnA), corresponded to hormone-dependence in prostate cancer. Intact FlnA (280 kDa, cytoplasmic) cleaved to a 90 kDa fragment which translocated to the nucleus in hormone-naïve cells, whereas in hormone-refractory cells, FlnA was phosphorylated, preventing its cleavage and nuclear translocation. We have examined whether FlnA localization determines a propensity to metastasis in advanced androgen-independent prostate cancer. Experimental Design: We examined, by immunohistochemistry, FlnA localization in paraffin-embedded human prostate tissue representing different stages of progression. Results were correlated with in vitro studies in a cell model of prostate cancer. Results: Nuclear FlnA was significantly higher in benign prostate (0.6612 ± 0.5888), prostatic intraepithelial neoplasia (PIN; 0.6024 ± 0.4620), and clinically localized cancers (0.69134 ± 0.5686) compared with metastatic prostate cancers (0.3719 ± 0.4992, P = 0.0007). Cytoplasmic FlnA increased from benign prostate (0.0833 ± 0.2677), PIN (0.1409 ± 0.2293), localized cancers (0.3008 ± 0.3762, P = 0.0150), to metastases (0.7632 ± 0.4414, P < 0.00001). Logistic regression of metastatic versus nonmetastatic tissue yielded the area under the receiver operating curve as 0.67 for nuclear-FlnA, 0.79 for cytoplasmic-FlnA, and 0.82 for both, indicating that metastasis correlates with cytoplasmic to nuclear translocation. In vitro studies showed that cytoplasmic localization of FlnA induced cell invasion whereas nuclear translocation of the protein inhibited it. FlnA dephosphorylation with the protein kinase A inhibitor H-89 facilitated FlnA nuclear translocation, resulting in decreased invasiveness and AR transcriptional activity, and induced sensitivity to androgen withdrawal in hormone-refractory cells. Conclusions: The data presented in this study indicate that in prostate cancer, metastasis correlates with cytoplasmic localization of FlnA and may be prevented by cleavage and subsequent nuclear translocation of this protein.

2760 In the initial stages, prostate tumors are dependent on androgens for growth, and hence, recurrent prostate cancer is primarily treated by androgen deprivation therapy. Patients respond initially to such treatment but ultimately relapse indicative of the development of androgen-independent prostate cancer (AIPC). The mTOR inhibitor rapamycin prevented the growth of both androgen-dependent and androgen-independent cell lines in short-term studies, and hence was promising as treatment for prostate cancer. However, long-term treatment with rapamycin induced resistance to this drug. Here we show that mTOR forms two complexes mTORC1 (mTOR/raptor/GβL) and mTORC2 (mTOR/rictor/sin1/GβL), both of which are inhibited by rapamycin. However, over time, rapamycin depleted rictor levels, thereby decreasing mTORC2 and forcing the drug to act through mTORC1 alone. Inhibition of mTORC1 by rapamycin inhibited cell growth but also stimulated Akt phosphorylation and AR transcriptional activity. Stimulation of AR transcriptional activity promoted cell growth and diminished the growth-inhibitory effects of rapamycin. On the other hand, rapamycin inhibition of mTORC2 inhibited cell growth as well as AR transcriptional activity. Therefore, depletion of rictor and rapamycin inhibition of mTORC1 alone caused the delayed resistance to rapamycin. To overcome rapamycin-resistance caused by AR transcriptional activity stimulated by the inhibition of mTORC1, we used the AR antagonist bicalutamide. Simultaneous treatment with bicalutamide inhibited AR transcriptional activity, thereby sensitizing the cells to the growth inhibitory effects of rapamycin. Our data shows that a combination of rapamycin and bicalutamide would be effective therapy for the treatment of prostate cancer.

Recurrent prostate cancer (PC) is usually treated with androgen deprivation therapy, which, despite initial success, eventually fails due to the development of androgen-independent PC. Androgen deprivation stimulates a significant increase in the phosphorylation (activation) of Akt, a serine/threonine kinase, which regulates cell growth and survival. Hence, we asked whether the increase in Akt phosphorylation contributes to the development of androgen independence. Akt regulates transcriptional activity of the androgen receptor (AR), and our data show that Akt-stimulated AR transcriptional activity is dependent on androgen-binding to the AR. PC proliferation has both androgen-sensitive and insensitive components. The androgen sensitive component is Akt-dependent, while the androgen-insensitive is not. However, Akt-induced cell survival is largely AR independent, suggesting that the cell stimulates Akt phosphorylation when subjected to androgen deprivation as an alternate pathway to maintain survival.

988 Recurrent prostate cancer is treated with androgen deprivation, which includes suppression of androgen production alone or in combination with androgen receptor (AR) antagonists (including flutamide or bicalutamide (Casodex)). Patients usually respond initially to such treatment but frequently relapse indicative of the development of hormone refractory prostate cancer. The treatment options for such patients are limited, and while Phase I trials reveal 20-24% increased survival with docetaxel, options to prevent the progression of this disease have not been identified. Prostate cancer is characterized by frequent deletion or loss of heterozygosity in the tumor suppressor PTEN leading to activation of downstream components of the P13K pathway, including the Akt and mTOR kinases. The mTOR inhibitor rapamycin has been reported to inhibit metastatic prostate tumor growth and angiogenesis in in vivo mouse models and is currently in Phase I clinical trials. While rapamycin inhibited the growth of both androgen dependent and independent prostate cancer, it also has been shown to increase AR levels and AR transcriptional activity in vitro, resulting increased expression of prostate specific antigen (PSA), a serum biomarker which is commonly used to track changes in prostate cancer patients. In contrast, the antiandrogen Casodex has been shown to inhibit tumor growth as well as PSA expression in androgen dependent tumors, whereas androgen independent tumors were refractory to Casodex treatment. Since the effects of these two drugs on PSA levels are opposing in vitro, we investigated the effects of using the two in combination. The androgen dependent cell line LNCaP and its androgen independent sublines C4-2 and LNCaP-AI were treated with 100 nM rapamycin, 10 μM Casodex or both. Rapamycin substantially inhibited proliferation in all cells examined, whereas Casodex inhibited the growth of LNCaP cells by 50% but did not affect the growth of the androgen in dependent sublines. However, in combination, the two together had a synergistic effect and inhibited the growth of both androgen dependent and independent lines to a greater extent than either alone. While neither drug caused a substantial increase in apoptosis rates, the combination of rapamycin and Casodex caused substantial amount of apoptosis even in androgen independent cells. In addition, rapamycin increased AR and PSA levels in all 3 cell lines, whereas in combination with Casodex, AR and PSA levels were decreased in all cells examined. These results suggest that Casodex and rapamycin in combination may be of better therapeutic value than each drug individually.

Prostate cancer is initially dependent on androgens for growth; hence, recurrent prostate is treated with androgen ablation which may result in progression to androgen independence characterized by a resistance to such therapy. Androgens bind to and activate the androgen receptor (AR), a member of the nuclear steroid receptor family of transcription factors, which regulates prostate cancer cell proliferation and survival in androgen-independent, as well as -dependent, tumors. Another pathway regulating proliferation and survival is the phosphatidylinositol 3-kinase (PI3K)/Akt pathway. Here we analyze reports in the literature indicating that these two pathways cooperate to regulate prostate tumor development and progression. Studies show that AR transcriptional activity and expression are regulated by Akt. In addition, androgens regulate the Akt pathway by both genomic and non-genomic effects. This explains why prostate tumors subjected to androgen ablation experience an increase in Akt phosphorylation, and suggest that the tumor compensates for the loss of one pathway with another. Different modes of interaction between the two pathways, including direct interaction, or regulation via downstream intermediates, such as the wnt/GSK-3beta/beta-catenin pathway, NF-kappaB, and the FOXO family of transcription factors, will be discussed. In addition, we will discuss the role of Akt in the interaction of the AR with upstream regulators of Akt phosphorylation, such as receptor tyrosine kinases of the EGF and IGF-1 receptor families and the tumor suppressor PTEN.

934 Prostate cancer is initially dependent on androgens for growth and thus responsive to androgen ablation. However, many patients treated with hormonal therapy ultimately progress to an androgen independent (AI) state. The LNCaP cell line has been widely used as a model for androgen-dependent (AD) prostate cancer. This cell line can be induced to progress to androgen independence by prolonged androgen deprvation and two AI prostate tumor cell lines cloned from LNCaP cells, LNCaP-AI (developed in vitro) and C4-2 (developed in vivo), are often used as models to study androgen independence. Both these LNCaP clones express the androgen receptor (AR), and hence, in this study, we evaluate the role of the AR in their proliferation. Proliferation was determined by MTT assay and by flow cytometry in propidium iodide stained ethanol fixed cells. Protein expression was determined by immunoblotting while AR transcriptional activity was determined by luciferase assay in cells transfected with a human prostate specific antigen (PSA) promoter tagged to a luciferase construct (hPSA-luc). We used siRNA technology and a pharmacological inhibitor to downregulate the expression and activation of the AR, respectively. Both LNCaP-AI and C4-2 cells showed increased proliferation rates compared to LNCaP cells. LNCaP cells were growth arrested by 10 μM Casodex, a competitive inhibitor of androgen binding to the AR, whereas LNCaP-AI and C4-2 cells were not, reflecting their AI nature. C4-2 cells exhibited lower AR expression and higher endogenous express of the AR target gene PSA compared to LNCaP. In contrast, LNCaP-AI cells showed higher AR expression compared to LNCaP but did not express endogenous PSA. Despite this, in LNCaP-AI cells, AR transcriptional activity on hPSA-luc was increased compared to LNCaP and C4-2 cells and was androgen-sensitive, showing that the AR in these cells remained active. Next, we determined whether AR expression mediated AI cell proliferation. Transfection with AR-specific siRNA duplexes downregulated the AR and inhibited proliferation in LNCaP and LNCaP-AI, but not in C4-2 cells. These results indicated that the AR mediated the growth of LNCaP-AI but not C4-2 cells. Despite AR-independent proliferation in C4-2 cells, the AR in this cell line was capable of regulating cell cycle proteins. However, we provide evidence that proliferation in C4-2 cells exhibited increased dependence on an alternate signaling pathway involving Akt, a serine/threonine kinase downstream of phosphatidylinositol 3-kinase. Our data suggest that LNCaP-AI and C4-2 cells represent two different forms of AI prostate cancer - while the LNCaP-AI cells are AR-dependent, the C4-2 cells are AR-independent.

935 Prostate cancer is almost universally detected by increased levels of prostate specific antigen (PSA) in the serum and expression of this serum marker is also used to determine tumor recurrence. However, over the last few years, clinicians have remarked about the lack of correlation between serum PSA levels and tumor volume or progression. The aim of these studies was to compare the pathways leading to PSA expression and proliferation in androgen-dependent LNCaP prostate cancer cells and its androgen-independent clone C4-2, which exhibit increased PSA expression. In androgen dependent cells, both PSA expression and growth is regulated by the activation of the androgen receptor (AR), a ligand activated transcription factor. Casodex, a competitive inhibitor of androgen binding to the AR, inhibited the growth of LNCaP but not C4-2 cells, but decreased PSA levels in both thus suggesting a dissociation between PSA expression and growth in C4-2 cells. Based on these results, we investigated whether signal transduction pathways known to regulate cell proliferation also affected PSA expression. The two best-known pathways regulating growth are those downstream of Ras and phosphatidylinositol 3-kinase (PI3K). Ras activates the mitogen activated protein kinases (MAPK) while the downstream effectors of PI3K include Akt and mTOR. In turn mTOR regulates the activation of p70S6 kinase, 4E-BP1 and e1F4G. The effect of these kinases on PSA expression and proliferation were evaluated by inhibition of the individual components with pharmacological inhibitors and small interfering RNA (siRNA). Proliferation rates were determined by MTT assay and by flow cytometry while protein expression was determined by immunoblotting. Treatment of LNCaP and C4-2 cells with PD98059, which inhibits p42/44MAPK activation, significantly decreased PSA expression but not proliferation. In contrast, the PI3K inhibitor LY294002 and the mTOR inhibitor rapamycin, inhibited cell growth but caused an increase in PSA expression, indicating that activation of PI3K and mTOR mediates proliferation but inhibit PSA expression. Downregulation of Akt by an Akt1-specific siRNA, decreased both growth and PSA expression. The downstream effectors of mTOR were also downregulated by specific siRNA. Inhibition of 4E-BP1, but not p70S6 kinase or e1F4G significantly decreased proliferation rates whereas e1F4G inhibition, but not that of p70S6 kinase or 4E-BP1, caused a significant upregulation of PSA expression, thereby indicating that mTOR9s effects on proliferation are mediated by 4E-BP1 while its effect on PSA expression are mediated by e1F4G. These results demonstrate for the first time that PSA expression and cell proliferation are differentially regulated downstream of AR activation - whereas PSA expression is mediated by Akt and p42/44MAPK activation and is inhibited by e1F4G, proliferation in prostate cancer cells is mediated by Akt and 4E-BP1.

In a previous report, we showed that increased activation of Akt, a downstream effector of phosphoinositide 3-kinase (PI3K) together with decreased activation of extracellular-signal-regulated kinase (ERK), a member of the mitogen-activated protein kinase (MAPK) family, predicted poor clinical outcome in prostate cancer (Kreisberg et al. 2004 Cancer Research 64 5232-5236). We now show that Akt activation, but not ERK activation, is correlated with proliferation in human prostate tumors as estimated by the expression of the cell proliferation antigen Ki67. We verified these results in vitro, using the androgen-dependent prostate cancer cell line LNCaP and its androgen-independent clone C4-2 as models of prostate cancer of good and poor clinical outcome, respectively. C4-2 cells expressed higher Akt activation, lower ERK activation and increased proliferation compared with LNCaP cells, similar to cases of poor clinical outcome. The PI3K inhibitor LY294002, but not the MAPK/ERK kinase inhibitor PD98059, induced growth arrest in both cell lines. Transient transfection with constitutively active Akt increased proliferation while dominant negative Akt decreased it, thus showing that Akt plays an important role in prostate cancer proliferation. Akt regulates the expression and activation of the androgen receptor. Androgen receptor inhibition with Casodex induced growth arrest in LNCaP cells, but not in C4-2 cells. Another PI3K downstream effector, p70 S6 kinase, requires prior phosphorylation by mammalian target of rapamycin (mTOR) for complete activation. Activation of p70 S6 kinase was higher in C4-2 compared with LNCaP cells. Rapamycin, an mTOR inhibitor, had a growth-inhibitory effect in C4-2 cells, but not in LNCaP cells. Our data suggest a shift from a Casodex-sensitive proliferation pathway in LNCaP cells to a rapamycin-sensitive pathway in C4-2 cells.

969 Prostate cancer is the second leading cause of cancer-related deaths among men in the United States. Initially, prostate tumors are dependent on androgens for growth, and respond to androgen-deprivation therapy; however, the majority of prostate cancers treated with androgen ablation progress to an androgen-independent phenotype for which there is no established therapy. Filamin A (FlnA) is a 280kDa membrane-associated actin-binding protein consisting of an actin-binding domain followed by 24 100-residue repeats. FlnA is cleaved intracellularly to a 90kDa fragment consisting of repeats 16-24 (FlnA16-24). FlnA16-24 translocates to the nucleus and represses androgen receptor (AR) transcriptional activity. Here, we show that in the androgen-dependent human prostate cancer cell line LNCaP, the 90 kDa fragment translocates to the nucleus, whereas in an androgen-independent clone of LNCaP cells, C4-2, FlnA remains largely in the cytoplasm. Western blotting with an antibody to the C-terminal domain of FlnA which recognizes the 90 kDa fragment shows that C4-2 cells express decreased levels of FlnA16-24 compared to LNCaP. We hypothesize that failure of FlnA cleavage to FlnA16-24 results in the androgen independence of C4-2 cells. Unlike LNCaP cells, C4-2 cells are not growth arrested by the antiandrogen Casodex, reflecting their androgen-independent nature. Transfection of FlnA16-24, but not FlnA1-15, results in Casodex-induced growth arrest in C4-2 cells. The AR is a transcription factor that binds to androgen responsive elements (ARE) in the promoter region of target genes. P21Cip1/Waf1, an inhibitor of cyclin dependent kinases and cell cycle progression, contains an ARE in its promoter region. Casodex fails to induce p21Cip1/Waf1 expression in C4-2 cells. However, transfection with FlnA(16-24), but not FlnA(1-15) stimulated Casodex-induced p21 Cip1/Waf1 expression. These results suggest that FlnA16-24 reverses C4-2 cells to an androgen-dependent phenotype and that FlnA may be a suitable target for therapy in prostate cancer progression.