Role of Androgen Receptor in Progression of LNCaP Prostate Cancer Cells from G1 to S Phase

Background The androgen receptor (AR) plays a critical role in the proliferation of prostate cancer cells. However, its mechanism of action in proliferation remains unknown. An understanding of the mechanism of AR action in proliferation may lead to the development of effective strategies for the treatment of prostate cancer. Methodology/Principal Findings In this study we report that pulse treatment of synchronized LNCaP cells with Casodex, an AR-antagonist, for 4 hours in mid-G1 phase was sufficient to prevent cells from entering S phase. Since the assembly of pre-replication complex (pre-RC) in G1 is required for the progression of cells from G1 to S phase, the effect of Casodex during mid-G1 suggested that the role of AR in proliferation might be to regulate the assembly of pre-RC. To test this possibility, we investigated the interaction between AR and Cdc6, an essential component of pre-RC in LNCaP cells. AR co-localized and co-immunoprecipitated with Cdc6, and Casodex treatment disrupted this interaction. AR-immunoprecipitate (AR-IP) also contained cyclin E and cyclin A, which play a critical role in pre-RC assembly and cell cycle entry into S phase, and DNA polymerase-α, PCNA, and ribonucleotide reductase, which are essential for the initiation of DNA synthesis. In addition, in cells in S phase, AR co-sedimented with components of the DNA replication machinery of cells that entered S phase. Conclusions/Significance Together, these observations suggest a novel role of AR as a component of the pre-RC to exert control over progression of LNCaP cells from G1 to S phase through a mechanism that is independent of its role as a transcription factor.


Introduction
Prostate cancer is the most frequently diagnosed non-skin cancer and second leading cause of cancer deaths in American men [1]. Androgen, by activating androgen receptor (AR), plays an important role in both the development and progression of prostate cancer. Hence, androgen ablation by pharmacological or surgical castration [2] remains frontline therapy for the treatment of locally advanced or disseminated prostate cancer. However, although the disease initially regresses in response to androgen ablation, it eventually relapses to become castration resistant [3,4]. Remarkably however, despite the continued use of this treatment strategy for over 70 years, very little is known about the mechanism by which AR regulates prostate cancer cell proliferation. An understanding of the mechanism of AR action in proliferation may lead to the development of more effective strategies for the treatment of prostate cancer.
Castration-resistant growth of prostate cancer is often associated with increased expression of AR [5,6,7]. AR remains indispensable for the proliferation of prostate cancer cells that have become castrate-resistant; AR-specific shRNA or siRNA blocks prolifera-tion of androgen-sensitive as well as castration-resistant ARpositive prostate cancer cells [8,9,10,11,12]. In addition, microinjection of AR-specific antibody into the nucleus or treatment of cells with AR mRNA hammerhead ribozyme inhibits the proliferation of AR-positive, but not AR-negative, prostate cancer cells [13], indicating a critical role of AR in proliferation of ARexpressing prostate cancer cells. Interestingly, the role of AR in proliferation of prostate cancer cells seems to be independent of its function as a transcription factor, at least in castrate resistant CWR22R3 human prostate cancer cells (derived from the CWR22 xenografts), in which AR-dependent proliferation is ligand independent but AR transcriptional activity remains liganddependent [11]. Collectively, these observations invoke the possibility that, besides its role as a transcription factor, AR may play a direct role in cell cycle regulatory events required for proliferation of prostate cancer cells. A direct role of AR in regulating proliferation contrasts with the notion of an indirect role in which AR transcriptional activity regulates the expression of factors required for cell cycle progression.
Cell cycle progression from G 1 to S phase is fundamental to proliferation. We reported previously that Casodex (bicalutamide), a specific inhibitor of AR, blocks the ability of G 1 phase ARpositive LNCaP prostate cancer cells to enter S phase [14,15], indicating a role of AR in cell cycle progression from G 1 to S phase. Progression of cells from G 1 to S phase requires a cascade of sequential events in G 1 phase that contribute to the assembly of the DNA replication machinery in cells that enter S phase. These events include a) loading of Cdc6, replication licensing factor RLF-B/Cdt1, and mini-chromosome maintenance (Mcm) proteins onto the origin recognition complex (ORC) to form pre-replication complex (pre-RC), and b) unwinding of DNA by helicase (Cdc45) associated with pre-RC, and the assembly of enzymes of DNA synthesis to form mega-complexes required for initiation of DNA synthesis at the origins of DNA replication (see Reddy et al [16] for a review).
In this study, we investigated whether AR interacts with the components of pre-RC and DNA replication machinery. Our studies demonstrate for the first time that AR is required in mid-G 1 phase, at the time when pre-RC assembly is known to begin, in order for LNCaP cells to enter S phase, and that AR is associated with cell cycle regulatory proteins and enzymes of DNA synthesis in a multienzyme complex isolated from S phase cells. These observations suggest AR involvement in proliferation through its interaction with cell cycle regulatory proteins and enzymes of DNA synthesis required for the onset of DNA synthesis and the progression of LNCaP cells from G 1 to S phase.
RNA Isolation, RT-PCR, and Real-time RT-PCR (qRT-PCR) RNA was isolated from control and Casodex treated cells and subjected to RT-PCR to determine PSA and GAPDH expression as described previously [14]. The primer sequences used for PSA are: 59-gcacccggagagctgtgt (forward) and 5-gatcacgcttttgttcctgat (reverse) primers, and for GAPDH are 59-gagatccctccaaaatcaagtg (forward) and 59-ccttccacgataccaaagttgt (reverse). GAPDH was used as an internal control. The densitometry of the RT-PCR bands was performed as described previously [14]. qRT-PCR was performed on an Applied Biosystems 7500 Fast Real-Time PCR System (Applied Biosystems, Foster City, CA) by using Taqman gene expression assays for PSA (Assay Id: Hs02576345_m1) and 18S RNA (Assay Id: Hs03928985_g1) as described previously [17]. The relative level of PSA expression was quantified by using comparative DDC t with 18S RNA as internal control.

Synchronization of LNCaP Cells by Isoleucine Deprivation
Isoleucine deprivation was carried out by our previously published method [15]. Cells were grown to 60% to 70% confluence, and then the medium was replaced with isoleucinefree RPMI 1640 (Life Technologies, Carlsbad, CA) supplemented with 6% dialyzed FCS (Life Technologies, Carlsbad, CA), 2.5 mM glutamine, 100 mg/ml streptomycin, and 100 units/ml penicillin. Cells were maintained in isoleucine-free medium for 36 hours in an incubator as above. Cells were released from isoleucine-block by replacing the medium with the complete medium containing 10% FCS and treated with Casodex (bicalutamide, gift from Astra Zeneca, England, UK) as indicated.
Since entry of cells into S phase is marked by the onset of DNA synthesis, the progression of synchronized cells from G 1 to S phase was monitored by determining the ability of cells to incorporate 3 H-thymidine into DNA as described previously [14,15]. At regular intervals after release from isoleucine-block, cells were pulse-labeled with 2 mCi/ml 3 H-thymidine (ICN Biomedicals, Inc., Costa Mesa, CA) for 30 minutes at 37uC in a humidified incubator. The radioactivity incorporated into acid-precipitable material was then determined as described [15].

Preparation of Cell Extracts and Immunoprecipitation
Exponentially growing LNCaP cells harvested by scraping into phosphate-buffered saline (PBS) were pelleted by centrifugation for 10 minutes at 4uC at 2,000 rpm in a Sorvall RT7 centrifuge and resuspended in immunoprecipitation (IP) buffer (50 mM Tris-HCl, pH 7.4, 0.1% Triton X-100, 5 mM EDTA, 250 mM NaCl, 2 mM CaCl 2 , 50 mM NaF, and 0.1 mM Na 3 VO 4 ) supplemented with protease inhibitor cocktail (P-8340, Sigma Chemical Co., St Louis, MO) at a density of approximately 1610 7 cells/ml. The cell suspension was then subjected twice to 30 pulses of sonication using a Branson Sonifier 250 (Branson Sonic Power Co., Danbury, CT), set at an output control of 2 and a duty cycle of 20, with intermittent cooling on ice. The sonicated cell extract was cleared by centrifugation in an Eppendorf Centrifuge 5415R at 6,000 rpm for 5 minutes at 4uC. Protein concentration in cleared extracts was assessed using BioRad Protein Assay reagent (Bio-Rad Laboratories, Hercules, CA).
For immunoprecipitation, cell extracts were diluted 5-fold in IP buffer supplemented with protease inhibitor cocktail and incubated at 4uC overnight with 4 mg/ml anti-AR antibodies (AR-N20 or AR-441) or anti-Cdc6 antibodies (H-304) (Santa Cruz Biotech, Santa Cruz, CA). Immune complexes were then adsorbed to Pierce Protein A/G Agarose beads (Thermo Scientific, Philadelphia, PA) for 2 hours at 4uC with gentle agitation. The adsorbed complexes were washed three times with IP buffer by centrifugation in an Eppendorf Centrifuge 5415R at 6,000 rpm for 5 minutes at 4uC, and then eluted with PAGE loading buffer (Bio-Rad, Richmond, CA). Control immunoprecipitates were prepared by using 4 mg/ml purified mouse or rabbit IgG (Antibodies Incorporated, Davis, CA) in place of anti-AR or anti-Cdc6 antibodies.

Isolation of DNA Replication Complex Fraction from Nuclear Lysate
Synchronized LNCaP cells in G 1 phase (1 hour after release from isoleucine-block) or in S phase (24 hours after release from isoleucine-block) were harvested and the nuclear lysate was prepared by a slight modification of the method of Subramanyam et al [18]. Harvested cells were pelleted by centrifugation for 10 minutes at 4uC at 2,000 rpm in a Sorvall RT7 centrifuge and resuspended in buffer A [0.16 M sucrose, 50 mM Tris-HCl (pH 7.6), 25 mM KCl, 10 mM MgCl 2 , 70 mM HEPES (pH 7.6), 0.025 mM CaCl 2 , 1 mM phenylmethylsulfonyl fluoride (PMSF), 2 mM DTT, 1 mM EDTA] at a density of 2610 7 cells/ml. The cell suspension was then homogenized in a motor-driven Wheaton homogenizer (Wheaton Co., Wheaton, IL) until ,90% of the cells were stained with trypan blue dye (usually three strokes at a rotation setting of 3). The cytosolic supernatant was then separated from the nuclear pellet by centrifugation in an Eppendorf Centrifuge 5415R at 6,000 rpm for 5 minutes at 4uC. The nuclear pellet was suspended in a volume of buffer A equal to that used for homogenization and subjected twice to 30 pulses of sonication with a Branson Sonifier 250 (Branson Co., Danbury, CT) set at an output control of 2 and a duty cycle of 20, with intermittent cooling on ice. The sonicated nuclear extract was cleared by centrifugation in an Eppendorf Centrifuge 5415R at 6,000 rpm for 5 minutes at 4uC. The DNA replication complex fraction was then isolated by subjecting the nuclear lysate to sucrose density gradient centrifugation essentially as described by Reddy and Pardee [19]. Nuclear lysate (0.5 ml) was layered over a 4.8 ml linear gradient of 20-40% (wt/vol) sucrose in buffer A, which in turn was layered over a 66% sucrose pad (0.5 ml) and centrifuged in a Beckman SW 50.1 rotor at 4uC for 16 hours at 35,000 rpm. The gradient was then resolved into 0.5 ml fractions and the rapidly sedimenting fraction that contained DNA polymerase activity is referred to as the replitase complex fraction as described by Murthy and Reddy [20]. The complex fraction from G 1 or S phase cells was then subjected to Western blot analysis to assess the presence of enzymes of DNA synthesis and cell cycle regulatory proteins.

Immunofluorescent Staining and Confocal Microscopy
LNCaP cells grown on glass slides were washed once with PBS, followed by fixation in 3.7% (vol/vol) formaldehyde for 20 minutes at 22uC. Cells were permeabilized in 0.5% (vol/vol) Triton X-100 for 15 minutes and blocked for 1 hour in 2% (wt/vol) BSA at 22uC. Slides were then incubated for 1 hour at 22uC with antibodies against AR (AR-N20 or AR-441) and Cdc6 or DNA polymerase-a followed by both goat-anti-mouse-fluorescein isothiocyanate (FITC)-and goat-anti-rabbit-tetramethylrhodamine B isothiocyanate (TRITC)-labeled secondary antibodies (Sigma-Aldrich, St. Louis, MO). After four washes with PBS, slides were mounted with Aqua-Poly/Mount (Polysciences Inc., Warrington, PA). Confocal laser scanning microscopy was performed with a Zeiss LSM 410 upright confocal microscope. Images were processed, and colocalization analysis (which is colored yellow) was performed with a Zeiss LSM 410 Meta Excitation system with collection at 488 and 543 nm. Intensity Correlation Quotient (ICQ) analysis to quantify AR colocalization with Cdc6 or DNA polymerase-a was performed by using ImageJ software (NIH, Bethesda, MD).

Immunofluorescent Detection of Bromodeoxyuridine (BrdU) Incorporated into DNA
LNCaP cells grown on glass slides were labeled with 10 mM BrdU (Sigma-Aldrich, St. Louis, MO) for 30 minutes, washed once with PBS and fixed in 3.7% formaldehyde for 20 min at 22uC. Cells were permeabilized in 0.5% (v/v) Triton X-100 for 15 min and blocked for 1 hour in 2% (w/v) bovine serum albumin at 22uC. Cells were then stained with mouse monoclonal antibodies against BrdU (Sigma-Aldrich, St. Louis, MO), followed by goat-anti-mouse-FITC-labeled secondary antibody. Slides were mounted with Vectashield Mounting Medium containing DAPI (Vector Laboratories, Burlingame, CA) and photomicrography was performed with a Zeiss Axiofot microscope equipped for epifluorescence.

Results
Use of Casodex to Determine the Role of AR in the Cell Cycle of Synchronized LNCaP Cells AR knockdown strategies have been used to demonstrate that AR is required for proliferation of prostate cancer cells [8,9,10,11,12]. However, AR knockdown by siRNA or shRNA requires treatment of prostate cancer cells over a period of several days. Thus, this approach can not be used to identify when in the cell cycle (i.e., G 1 , S, or G 2 /M phases) AR is required for proliferation of LNCaP cells, as each phase of the cell cycle lasts no more than few hours. Therefore, we used the non-steroidal antiandrogen Casodex (bicalutamide) to selectively and acutely inhibit AR in order to determine the role of AR in progression of synchronized LNCaP cells through G 1 and S phases.
Casodex is usually used at 10 to 20 mM to inhibit androgeninduced AR transcriptional activity; but in those experiments, the cells are in charcoal-stripped serum (CSS)-containing medium [21,22]. However, synchronized cells released into CSS-containing medium will not progress through the cell cycle. We compared the effectiveness of Casodex on the expression of prostate specific antigen (PSA), a specific AR-target gene [23], in LNCaP cells in FCS-vs. CSS-medium. Androgen-deprivation of cells in CSSmedium decreased PSA expression to ,50% of that in control cells in FCS-medium and Casodex caused a further decrease (Figs. 1A and 1B). Casodex at 75-100 mM in FCS-medium had the same effect on PSA expression as 25-50 mM Casodex in CSSmedium (Figs. 1A and 1B). Notably, when the dose-dependent response to Casodex is plotted relative to its own control (in FCS vs. CSS), the curves overlap, indicating that the IC 50 to inhibit AR activity is <50 mM Casodex in either FCS or CSS (Fig. 1C). Cells treated with 100 mM Casodex in FCS-medium were morphologically similar to those treated with 20 mM Casodex in CSSmedium, and 100 mM Casodex in FCS-medium caused no noticeable cell death, as determined by the Trypan Blue exclusion method (data not shown). Therefore, we used 100 mM Casodex, a maximally effective concentration to inhibit AR activity, to determine the role of AR in proliferation of LNCaP cells in FCSmedium. This concentration is similar to that used to inhibit PSA expression or growth of LNCaP cells in the presence of FCS [24,25,26].

LNCaP Cell Entry into S Phase Requires Functional AR during Mid-G 1 Phase
We have shown previously that isoleucine deprivation causes LNCaP cells to arrest in G 0 /G 1 phase [15,27]. Upon release from isoleucine-block, i.e., transfer into complete medium containing FCS (FCS-medium), these cells progress through G 1 and enter S phase 12 hours after release, and reach a peak of S phase 10-12 hours thereafter [15]. We have also shown that Casodex treatment for 24 hours starting from the time of release from isoleucine-block abrogates the ability of cells to enter S phase [14,15]. While this suggested a role of AR in the cell cycle, that experimental design cannot distinguish between a role of AR in G 1 vs. S phase. Therefore, in order to distinguish between an inhibitory effect of Casodex in G 1 vs. in S phase, we treated synchronized LNCaP cells with Casodex starting at 0, 4, 8, 12, or 16 hours after release from isoleucine-block, and determined the ability of cells to incorporate 3 H-thymidine ( 3 H-TdR) at 20 hours after release, a time when control cells are in S phase (see schematic of Fig. 2A). As shown in Fig. 2A, Casodex maximally inhibited 3 H-TdR incorporation if added any time during G 1 phase (i.e., 0, 4, or 8 hours after release), indicating a role of AR in G 1 . By contrast, the Casodex effect was blunted if treatment was delayed until 12-16 hours after release from isoleucine-block, a time when the cells have already committed to enter S phase. Casodex treatment of cells that had already entered S phase (16 hours after release from isoleucine-block) had very little effect on 3 H-TdR incorporation ( Fig. 2A). Thus, the inhibitory effect of Casodex occurs in G 1 phase, when cells are preparing to enter S phase, and not in S phase. Therefore, AR is required in G 1 for cells to enter S phase.
We then determined when in G 1 phase Casodex is most effective in suppressing cell entry into S phase. We pulse treated synchronized LNCaP cells with Casodex during the first, second, or third 4-hour interval following release from isoleucine-block, corresponding to early-G 1 , mid-G 1 , or late-G 1 , respectively, and determined the rate of 3 H-TdR incorporation into DNA at 4-hour intervals for 24 hours (see schematic of Fig. 2B). We have shown previously that 3 H-TdR incorporation offers a reliable and reproducible method to monitor the progression of synchronized cells from G 1 to S phase and is corroborated by flow cytometry analysis as well as the expression of S phase-specific cyclins and the activation of cyclin-dependent kinases [14,15]. As shown in Fig. 2B, treatment with Casodex during mid-G 1 (4-8 hours after release from isoleucine-block) was more effective at blocking entry into S phase than was treatment during early-G 1 (0-4 hours after release from isoleucine-block) or late-G 1 phase (8-12 hours after release from isoleucine-block). This suggests AR involvement in cell cycle regulatory events in mid-G 1 phase that are critical for LNCaP cells to progress from G 1 to S.

AR Interacts with Cdc6, a Critical Component of Pre-RC Required for the Assembly of Replication Machinery in S Phase LNCaP Cells
Since Casodex treatment in mid-G 1 blocked the ability of synchronized LNCaP cells to enter S phase (Fig. 2), we tested whether AR interacts with cell cycle regulatory proteins involved in the assembly of pre-RC, as assembly of pre-RC in mid-to late-G1 is required for cells to enter S phase and initiate DNA synthesis. Cdc6 binding to the origin recognition complex (ORC) is the first step in the process of pre-RC assembly [16]. We observed that an immunoprecipitate prepared using Cdc6-specific antibodies (Cdc6-IP) contained AR, suggesting AR interaction with Cdc6 (Fig. 3A). This interaction is specific since one of the most abundantly expressed proteins, prostate-specific antigen (PSA), in LNCaP cells was not associated with Cdc6-IP (Fig. 3A). We have shown previously that the interaction between cell cycle regulatory proteins and enzymes of DNA synthesis in a variety of cells leads to the assembly of a multienzyme complex called replitase or DNA replication machinery [28,29,30]. These complexes are readily detectable in cells that are in S phase, but not any time during G1 phase [29,30]. Therefore, we tested whether AR-Cdc6 interaction is a cell cycle-dependent phenomenon. As shown in Fig. 3B, although Cdc6 was present at a low level in G 1 phase cells, its association with AR-IP was seen in synchronized LNCaP cells that were in S phase but not in those that were in G 1 phase. By comparison, lamin B, a nuclear protein, showed very little difference in its association with AR-IP from G 1 vs. S phase cells (Fig. 3B). Thus, AR exhibits a cell cycle-dependent interaction with Cdc6.

Casodex Disrupts AR-Cdc6 Interaction
Since AR is implicated to play a role in regulation of Cdc6 expression [14,31], we tested whether Casodex-induced blockade of the entry of LNCaP cells into S phase is due to a decrease in Cdc6 protein levels. In fact, under the experimental conditions employed in the present study, Casodex had no noticeable effect on either Cdc6 or AR protein levels in exponentially growing LNCaP cells (Fig. 4A, Input). Nonetheless, Casodex disrupted AR interaction with Cdc6 as indicated by the presence of Cdc6 in AR-IP of control cells but not in that of Casodex treated cells (Fig. 4A, AR-IP). This Casodex-induced disruption of AR association with Cdc6 was further supported by confocal microscopy (Fig. 4B); we observed a noticeable decrease in the colocalization of AR with

AR Interacts with S Phase Cyclins
Besides Cdc6, some of the cell cycle regulatory proteins, such as cyclins E and A, also play an important role in the assembly of pre-RC in G 1 [16]. Therefore, we tested whether AR interacts with any of the cyclins that are known to be involved in the progression of cells from G 1 to S phase (viz., cyclins E and A) and G 2 /M (viz., cyclin B) phases in LNCaP cells. We observed that cyclin E and cyclin A, but not a mitotic cyclin, cyclin B, were associated with AR-IP prepared from exponentially growing LNCaP cells (Fig. 5). By comparison, GAPDH, a cytosolic marker, was not associated with AR-IP. Thus, AR exhibits a specific interaction with cell cycle regulatory proteins that are involved in regulation of cell entry into S phase.

AR Interacts with Enzymes of DNA Synthesis
Since Cdc-6-dependent assembly of pre-RC leads to the recruitment of enzymes of DNA synthesis to the sites of DNA replication [16], and since AR interacts with Cdc6 (Fig. 3), we tested whether AR-IP also contains enzymes of DNA synthesis. As shown in Fig. 6, we observed that AR-IP prepared using two different AR antibodies (AR-N20 and AR-441) contained DNA polymerase-a, a key enzyme involved in initiation of DNA synthesis (Fig. 6A). AR interaction with DNA polymerase-a was further corroborated by immunoflourescent confocal microscopy (Fig. 6B); we observed a strong colocalization of AR with DNA polymerase-a (ICQ = 0.4560.016 SD) in the nuclei of LNCaP cells. In addition to DNA polymerase-a, AR-IP prepared using AR-441 and AR-N20 antibodies contained proliferating cell nuclear antigen (PCNA), an auxiliary protein of DNA polymerase, and the catalytic subunit of ribonucleotide reductase (RNR2), which converts ribonucleotides to deoxyribonucleotides required for DNA synthesis (Fig. 6C). In addition to its role in proliferation [8,9,10,11,12], AR is also reported to play a critical role in survival of LNCaP cells, as AR knockdown leads to apoptotic cell death . Casodex inhibits AR-Cdc6 interaction in LNCaP cells: A) Exponentially growing LNCaP cells were treated with Casodex or vehicle (control) for 24 hours and AR-IP prepared from treated and control cells was subjected to Western blot analysis to detect AR and Cdc6. B) LNCaP cells grown on slides were synchronized by isoleucine-deprivation and released into complete medium in the presence of vehicle (control) or Casodex. At 24 hours after release from isoleucine-block (when control cells are expected to enter S phase), cells were fixed and processed for immunoflourescent staining using anti-AR (AR-441) mouse monoclonal and anti-Cdc6 (H-304) rabbit polyclonal antibodies. doi:10.1371/journal.pone.0056692.g004 [32]. Therefore we tested whether the role of AR in regulation of apoptosis also involves its interaction with apoptotic proteins. As shown in Fig. 6A, caspase-3, an apoptotic enzyme, was not detected in AR-IPs. Thus, AR exhibits a selective interaction with the enzymes of DNA synthesis required for the proliferation of LNCaP cells.
Enzymes of DNA Synthesis Co-sediment with AR in Nuclear Lysate of S, but not G 1 , Phase LNCaP Cells Progression of cells from G 1 to S phase is associated with the assembly of enzymes of DNA synthesis into replitase complexes that can be isolated by biochemical fractionation of nuclear lysate [29,30]. We tested whether the entry of LNCaP cells into S phase involves AR interaction with the enzymes of DNA synthesis in replitase complexes. We have shown previously that AR cosediments with DNA polymerase activity in the replitase fraction isolated from exponentially growing LNCaP cells [20]. We now report that other enzymes of DNA synthesis, viz., PCNA and ribonucleotide reductase, and cell cycle regulatory proteins, viz., cyclin A, cyclin E and p27 Kip-1 , also co-sediment with AR (Fig. 7). Importantly, this co-sedimentation was seen in nuclear lysate prepared from cells that were in S phase (20 hours after release from isoleucine-block), but not G 1 phase (1 hour after release from isoleucine block) (Fig. 7). Thus, there is an S phase-specific association of AR with enzymes of DNA synthesis in a replitase complex fraction.

Discussion
The ability of cells to enter S phase is determined by events in G 1 phase. While a role of AR in the control of proliferation of prostate cancer cells is evident from a number of studies that knocked down AR in androgen-sensitive and castration-resistant prostate cancer cells [8,9,10,11,12], the mechanism by which this occurs remains elusive. Our studies demonstrate for the first time that AR is required in mid-G 1 phase for synchronized LNCaP cells to progress from G 1 to S phase and that AR interacts with cell cycle regulatory proteins and enzymes required for initiation of DNA synthesis. While these observations remain to be validated in other AR-positive prostate cancer cell lines, the data presented here suggest AR involvement in proliferation of LNCaP cells through its interaction with components of the pre-RC necessary for the initiation of DNA synthesis and, therefore, for the progression of cells from G 1 to S phase.
We observed that the anti-proliferative effect of Casodex is quite evident in LNCaP cells that were treated with Casodex specifically during mid-G 1 phase but not in those that had already entered S phase (Fig. 2). Interestingly, this is consistent with an earlier report in which cyproterone acetate, another AR antagonist, blocks testosterone-stimulated DNA synthesis in the prostate gland of castrated rat when administered with testosterone but not if cyproterone acetate is delayed until DNA synthesis has already started [33].
Our observations suggest a role of AR in events leading up to the transition of cells from G 1 to S phase. Pre-RC assembly is one such event that occurs during that period and is necessary for the initiation of DNA synthesis and, therefore, for the progression of cells from G 1 to S phase [16]. We observed that AR coimmunoprecipitates with Cdc6, a critical component of pre-RC (Fig. 3). There are other pre-RC components, such as Orc2, Cdt1, and Mcm7, that are also reported to be associated with AR in prostate cancer cells [34,35]. Interestingly, Cdc6 association with AR-IP was abrogated in cells treated with Casodex (Fig. 4), indicating a critical role of AR-Cdc6 interaction in progression of LNCaP cells from G 1 to S phase. Progression of cells from G 1 to S phase requires not only pre-RC assembly but also the recruitment of enzymes of DNA synthesis to pre-RC. DNA synthesis requires the concerted action of a number of enzymes and proteins in megacomplexes, which are referred to as DNA replication machinery, DNA synthesome or replitase [36,37,38]. Pre-RC serves as a ''launching pad'' for the assembly of DNA replication machinery at sites of DNA replication [16]. Our studies revealed AR association with DNA replication machinery specifically in cells that entered S phase, but not in those that were in G 1 phase (Fig. 7). AR integration into DNA replication machinery is also supported by the observation that AR is localized at sites of DNA replication [39]. Thus, AR integration into replication machinery through its interaction with the components of pre-RC as depicted in Fig. 8, may allow AR to exert control over proliferation of prostate cancer cells.
Based on the critical role of AR interaction with Cdc6 in cell cycle progression, which is indicated from the observation that Casodex disrupts AR interaction with Cdc6 (Fig. 4) and inhibits the progression of cells from G 1 to S phase (Fig. 2), we propose that AR plays a direct role in proliferation, independent of its role in regulating the expression of genes required for proliferation. In addition, whereas a 4-hour Casodex treatment was sufficient to block G 1 phase cells from entering S phase (Fig. 2), a similar 4hour treatment with Casodex had no noticeable effect on AR transcriptional activity, as determined by PSA mRNA expression, in LNCaP cells (data not shown). Thus, the Casodex-sensitive role of AR in progression of cells from G 1 to S phase, i.e., proliferation, seems to be independent of its role as a transcription factor. The AR indeed has been shown to exhibit a critical function in proliferation of prostate cancer cells that is distinct from its function as a transcription factor; in CWR22R3 cells, the AR role in proliferation, but not its role as a transcription factor, is androgen-independent [11]. Similarly, Sathya et al [40] showed that the regulation of AR transcriptional activity and the role of AR in proliferation are mechanistically distinct; they developed tissue-specific AR modulators (SARMs) that are very weak as activators of AR transcriptional activity but are as effective as dihydrotestosterone (DHT) in stimulating the proliferation of prostate cancer cells.
Since cells that lack AR can progress flawlessly through the entire cell cycle, it seems remarkable that AR would integrate into the replication machinery to regulate proliferation of AR-positive prostate cancer cells. Interestingly, AR is not the only transcription factor that exerts control over proliferation through integration into the replication machinery. E2F-1 and Rb protein, both of which regulate the expression of genes necessary for cell cycle progression and DNA replication [41,42], also function at replication origins to limit DNA replication by interacting with proteins of the ORC, not by acting as transcription factors [43,44]. Conversely, proteins involved in the assembly of pre-RC, viz., Mcm proteins 2-7, colocalize with RNA polymerase II on actively transcribing genes and are required for transcription elongation [45]. These interactions between transcription factors and proteins involved in the assembly of pre-RC may facilitate the orderly replication of transcriptionally active regions of genomic DNA in S Figure 7. Differential association of AR with cell cycle regulatory proteins and enzymes of DNA synthesis in replitase complex from synchronized LNCaP cells in G 1 vs. S phase: LNCaP cells were synchronized by isoleucine-deprivation and replitase complex (Cpx) fraction was prepared from the nuclear lysate of cells in G 1 phase (1 hour after release from isoleucine-block) or S phase (20 hours after release from isoleucine-block). Equal amounts of protein were subjected to Western blot analysis. RNR2, ribonucleotide reductase (RNR2). doi:10.1371/journal.pone.0056692.g007 phase. Interestingly, the timing of initiation of replication in a genomic region is dictated by the transcriptional activity in that region [46,47]; transcriptionally active regions are known to replicate early in S phase, whereas transcriptionally inactive regions replicate late in S phase [48,49]. Furthermore, origins of DNA replication are enriched around transcription start sites and transcription initiation is reported to influence the initiation of DNA replication [50,51]. Thus, there is a cross-talk between transcription and replication, which may be facilitated by the interactions between proteins involved in DNA replication and transcription. Therefore, we propose that AR associated with proteins and enzymes of pre-RC and replication machinery may play a role in coordinating the replication of transcriptionally active AR-target genes during S phase.
In summary, AR may exert control over proliferation of prostate cancer cells through its interaction with pre-RC and DNA replication machinery. Future studies focusing on identification of protein(s) that interface with AR in pre-RC may lead to the development of new and more effective strategies to suppress AR-dependent growth of prostate cancer. In addition, altered interaction of AR with pre-RC and replication machinery could circumvent the requirement for androgen and account for the continued critical role of AR in proliferation of castration-resistant prostate cancer cells. Therefore, analysis of pre-RC and replication machinery constituents in androgen-sensitive vs. castrationresistant prostate cancer cells may yield valuable information about the mechanism of AR action in prostate cancer cell proliferation and may lead to the identification of potentially more effective therapeutic targets for prostate cancer treatment. Figure 8. Model of AR interaction with pre-RC and replication machinery as cells progress from G 1 to S phase. Pre-RC assembly involves sequential recruitment of Cdc6, Cdt1, and mini-chromosome maintenance (Mcm) proteins to the origin recognition complex (ORC), which serves as a ''launching pad'' for the assembly of DNA replication complexes/machinery at the origin of DNA replication. Cyclin-dependent kinases, such as Dbf4-Cdc7 kinase (not shown), cyclin E-Cdk-2 and cyclin A-Cdk-2, play a role in the recruitment of Mcm proteins and enzymes of DNA synthesis to form DNA replication machinery as cells progress from G 1 to S phase. AR integrates into DNA replication machinery through its interaction with pre-RC components. Events occurring at one of the many origins of DNA replication that exist in a proliferating cell are depicted. doi:10.1371/journal.pone.0056692.g008