help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Murillo, H.
Right arrow Articles by Tindall, D. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Murillo, H.
Right arrow Articles by Tindall, D. J.
Endocrinology Vol. 142, No. 11 4795-4805
Copyright © 2001 by The Endocrine Society

ARTICLES

Role of PI3K Signaling in Survival and Progression of LNCaP Prostate Cancer Cells to the Androgen Refractory State

Horacio Murillo1, Haojie Huang1, Lucy J. Schmidt, David I. Smith and Donald J. Tindall

Departments of Molecular Pharmacology and Experimental Therapeutics (H.M.), Urology Research (H.M.,, H.H., L.J.S., D.J.T.), Laboratory Medicine and Pathology (D.I.S.), and Biochemistry and Molecular Biology (D.J.T.), Mayo Clinic and Foundation, Rochester, Minnesota 55905

Address all correspondence and requests for reprints to: D. J. Tindall, Ph.D., Department of Urology Research, Guggenheim 1701, Mayo Foundation, 200 First Street SW, Rochester, Minnesota 55905. E-mail: tindall.donald{at}mayo.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The mechanisms by which prostate cancer (PCa) cells progress to a hormone refractory state are poorly understood. The progression process under androgen ablation conditions involves the survival of at least a portion of malignant cells and their eventual proliferation in an androgen-independent manner. The goal of this study was to investigate the role of PI3K signaling in such a progression. Using an in vitro model of androgen ablation, we show that after removal of androgen support, the human PCa cell line LNCaP initially arrested in G1 and trans-differentiated into neuroendocrine-like cells that eventually resumed androgen-independent proliferation. Both acute and chronic androgen ablation resulted in an increase in basal levels of PI3K and Akt activity, which were sustained throughout the progression process. Under these conditions, inhibition of PI3K, pharmacologically or with ectopic expression of PTEN, arrested cell proliferation and blocked progression to the androgen-independent state. In contrast, LNCaP cells in the presence of androgens were marginally sensitive to PI3K inhibition. During the chronic stage of androgen deprivation, androgen-independent proliferation correlated with diminished p27kip1 protein levels, whereas PI3K and Akt activity remained elevated. At this stage, PI3K inhibition rapidly triggered accumulation of p27kip1, cell cycle arrest, and cell death. PI3K modulated p27kip1 levels at least in part by regulating its rate of degradation. Taken together, these data show that androgen ablation alone can increase PI3K-Akt activation, which supports survival after acute androgen ablation and proliferation during chronic androgen deprivation. Successful progression to the androgen-independent state in the LNCaP cell line model requires intact PI3K signaling.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
SURGICAL OR PHARMACOLOGICAL castration, with the intent of decreasing androgen action, remains the mainstay treatment for advanced and aggressive prostate cancer (PCa) (1, 2, 3). This form of endocrine therapy elicits dramatic growth arrest and apoptotic and nonapoptotic cell death in sensitive prostate cells (4, 5, 6). The combined effects lead to prostatic involution and a dramatic response in PCa tumors (4, 5, 6). Although three-fourths of PCa tumors initially respond to this form of treatment, the success of hormonal manipulation is cut short by the emergence of androgen refractory (androgen-independent) disease (5, 6).

Growth factors and their cognate receptors have long been implicated in PCa progression (7, 8). In fact, androgen removal promotes the increased expression of epidermal growth factor receptor family members and ligands, among others (7), thereby contributing to the formation of the abnormal autocrine loops seen in advanced PCa. In addition, loss of androgenic regulation of important metalloproteases such as neutral endopeptidase and maspin promote multiple growth factor and growth factor receptor signaling, which can contribute to progression (9, 10, 11, 12). Thus, androgen removal may increase growth factor signaling, which could contribute to acquisition of the androgen-independent phenotype.

Additional evidence for the importance of growth factor-dependent support for the progression and survival of PCa cells comes from the findings that the phosphatase PTEN, a major regulatory mechanism of PI3K action, is commonly lost through inactivating mutations or loss of expression in PCa cell lines and tumors (13, 14, 15, 16). PTEN loss in malignant cells facilitates dysregulated PI3K activity (15, 16). In addition, it is conceivable that the increased abnormal growth factor loops seen in PCa can enhance growth and mitogenic signaling through PI3K and other effector pathways even in the presence of wild-type PTEN (7, 8). Thus, a reliance on growth factor support in the absence of androgens may be a favored pathway (17) for PCa progression, which may be mediated at least in part through PI3K signaling. However, there are no studies that show direct evidence for such a hypothesis during the process of progression.

In the presence of androgen ablation, successful progression by PCa requires an ability to overcome two major roadblocks. The most important one is the survival of at least a portion of malignant cells during the acute onset of androgen deprivation. The second is the ability to proliferate in an androgen-independent manner, thereby overcoming the growth arrest instituted by the androgen withdrawal upon androgen-responsive PCa cells (18). Given the association of abnormal growth factor signaling and PCa progression, we investigated the role of the PI3K signaling pathway in the human PCa cell line LNCaP (19) during progression to the androgen-independent state. Here we show that removing androgen support from LNCaP cells triggers a series of events, including cell cycle arrest and increased PI3K-Akt activity, culminating in the eventual acquisition of the androgen-independent phenotype. Under such conditions, despite the presence of growth factor support, inhibition of PI3K signaling during the acute phase of androgen deprivation completely blocked progression and hypersensitized relapsed cells to spontaneous apoptosis. In the absence of androgens, LNCaP cells became hyperdependent on PI3K for growth, survival, and progression to and maintenance of the androgen-independent state.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and treatment conditions
The LNCaP cell line was purchased from the American Type Culture Collection (Manassas, VA) and maintained in RPMI 1640 medium (BioWhittaker, Inc., Walkersville, MD) containing 10% FBS (Life Technologies, Inc., Grand Island, NY) with 100 U/ml penicillin and 100 µg/ml streptomycin. LNCaP cells between passages 27 and 65 were used in these studies. The LNCaP C4-2 cell subline was purchased from UroCor, Inc. (Oklahoma City, OK), cultured in T-media (Life Technologies, Inc.), and used between passages 29 and 40. Androgen-deprived medium was routinely prepared by adding 10% charcoal-stripped serum (CSS) instead of untreated FBS [whole FBS-containing medium (WS)] and antibiotics as described above. The androgen deprivation protocol was initiated by plating LNCaP cells at 30% confluence in serum-free medium for 2 d to control for interference from androgens remaining from the prior culture in WS. Next, cells received WS-medium (normal control), CSS plus 1 nM R1881-containing medium (control), WS plus Casodex (Zeneca Pharmaceuticals, Wilmington, DE) (Cx)-containing medium (treated), or CSS-only medium (CSS-medium; treated). The androgen refractory LNCaP subline LNCaP-Rf was established by culturing and passaging LNCaP in CSS-medium for more than 10 wk by the above-described protocol. All cells were maintained as monolayers in a humidified atmosphere containing 5% CO2 at 37 C and passaged at confluence by trypsinization. When drugs were added, cells in serum-free medium were treated with the respective drug concentration for 30 min before changing the medium that contained the drug at the given concentration. Drug-treated cultures were maintained in the dark at all times. Unless specified, all treatment conditions included serum (CSS, CSS plus 1 nM R1881, WS, or WS plus Cx as indicated).

CSS was prepared as follows. A solution containing 5% (wt/vol) activated charcoal (Sigma, St. Louis, MO) and 0.5% (wt/vol) dextran T70 (Pharmacia Biotech, Piscataway, NJ) was prepared in 1 mM HEPES buffer (pH 7.4). The mixture was stirred gently at 4 C for 3 h and centrifuged at 2500 x g for 10 min. At room temperature, 5 g dextran-treated charcoal were added to 500 ml FBS, mixed gently for 1 h, and centrifuged at 2500 x g for 10 min. The collected supernatant (serum) was subjected to another cycle of dextran-treated charcoal treatment, filtered through 0.2-µm porous membranes (Nalgene, Rochester, NY), aliquoted, and maintained at -20 C until used.

Chemical reagents
Drug stocks (20–50 mM) of R1881 (DuPont Merck Pharmaceutical Corp., Boston, MA), Cx (Zeneca Pharmaceuticals), LY294002 [2-(4-morpholinyl-4H-1-benzopyran-4-one)], PD98059 (2'-amino-3'-methoxyflavone), H89 [N-(2-[(p-bromocinnamyl)amino]ethyl)-5-isoquinoline-sulfonamide] from Calbiochem (La Jolla, CA), and cycloheximide (Sigma-Aldrich Corp., St. Louis, MO) were dissolved in ethanol. Stocks of wortmannin (1 mM) (Calbiochem) were prepared in dimethylsulfoxide. All drugs were kept in the dark at -20 C, except for wortmannin, which was kept at -80 C.

Northern analysis
Primers for RT-PCR of neuron-specific enolase (NSE; GenBank X51956), prostate-specific antigen (PSA; GenBank M26663), neurotensin (NT; GenBank U91618), and p27kip1 (GenBank AY004255) were as follows: NSE: forward, 5'-GTTCTGAACGTCTGGCTAAATAC-3'; reverse, 5'-CATTGAGTTATGGGGAAATGA-3'; PSA: forward, 5'-AGCCACAGCTTCCCACAC-3'; reverse, 5'-CAGTATTCCCCAGGACACAG-3'; and NT: forward, 5'-CGGACTTGGCTTGTTAGAA-3'; reverse, 5'-TTGTAGAAGAGACAGATAAGTGTGTT-3'. For p27kip1 Northern analysis, an 881-bp insert from the image clone 854668 (GenBank AA630082) in pBlueSK was used after sequence confirmation. Similarly, the RT-PCR products for NSE (662 bp), PSA (657 bp), and NT (651 bp) were cloned into the pCR-Blunt vector (Invitrogen, Carlsbad, CA), and confirmed by sequencing. Total cellular RNA was isolated using the TRIzol reagent (Life Technologies, Inc., Grand Island, NY), and aliquots (10–15 µg) were electrophoresed on 1.2% denaturing formaldehyde-agarose gels and transferred onto positively charged nylon membranes. Filters were hybridized with [32P]deoxy-CTP-labeled probes (NSE, NT, PSA, p27kip1, and GAPDH) for Northern analysis.

Western immunoblotting
Protein samples were prepared by lysing cells over ice in ice-cold RIPA buffer (50 mM Tris-HCl, pH 7.4, 1% NP-40, 0.1% SDS, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EGTA, 1 mM PMSF, 1 µg/ml leupeptin, 1 µg/ml aprotinin, 1 µg/ml pepstatin, 1 mM Na3VO4, and 1 mM NaF). Cell lysates were passed 3-times through a 27-gauge needle and centrifuged at 14,000 x g at 4 C for 15 min. Protein content was determined using the Bio-Rad Laboratories, Inc. DC-protein assay kit (Bio-Rad Laboratories, Inc., Hercules, CA). Equal amounts of protein (15–50 µg) from cell lysates were denatured in sample buffer, subjected to LDS/NuPAGE on 4–20% gels (Novex/Invitrogen, San Diego, CA), and transferred to nitrocellulose membranes. The blots were probed with specific primary antibodies as recommended by the suppliers. Appropriate HRP-conjugated secondary antibodies were used (1:5,000) and visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ). Antibodies were purchased from the following sources: PY20, CDK4, ERK2, and P27Kip1 mouse monoclonal antibodies from Santa Cruz Biotechnology, Inc.; p16INK4a and p21cip1 mouse monoclonal antibodies from Calbiochem (La Jolla CA); anti-whole Akt and Phospho-Akt (Ser473) from New England Biolabs, Inc. (Beverly, MA); anti-androgen receptor (AR) monoclonal antibody was purchased from PharMingen (San Diego, CA); and anti-PI3K p85 subunit was from Upstate Biotechnology, Inc. (Lake Placid, NY).

PI3K kinase assay
The PI3K activity assay was performed as described by Endemann et al. (20) and Huang et al. (21). Briefly, cells were washed twice with ice-cold PBS and lysed in RIPA buffer. The lysates were centrifuged, and the protein content was determined. Five-hundred µg aliquots from each sample were immunoprecipitated with 40 µl aliquots of pre-conjugated, monoclonal anti-phosphotyrosine (PY20) agarose beads (Santa Cruz, CA) by incubating overnight at 4 C in 500 µl of immunoprecipitation buffer (190 mM NaCl, 50 mM Tris-HCl pH 7.4, 6 mM EDTA, 2.5% Triton X-100). All subsequent steps were exactly as described (21). 30 µl aliquots from the kinase reaction assays were separated on thin layer Silica Gel 60 chromatography plates (EM Separations Technology), dried at room temperature, and autoradiographed.

Cell transfection
Wild-type PTEN cDNA (a kind gift from Dr. C. David James, Mayo Clinic and Foundation) was subcloned into the pcDNA3.1(+) vector (Invitrogen, Carlsbad, CA). The PTEN open reading frame was confirmed by sequencing of both strands. Transient transfections of LNCaP cells and LNCaP-Rf cells were performed by electroporation. Briefly, total plasmid DNA was kept constant (20 µg) for all transfections by the addition of empty vector (pcDNA-3.1(+) or pCMV6) to pcDNA-PTEN or pCMV6-CA-Akt. The DNA was ethanol- precipitated and resuspended in 50 µl of serum-free media. Cells (1 x 107 per transfection) in 350 µl of serum-containing media were mixed with the plasmid DNA to a total of 400 µl. The DNA-cell mixture was transferred to a 4-mm cuvette (BTX Inc., San Diego, CA) and electroporated with a single 10-ms, 300-V pulse using a BTX T820 square wave electroporator (BTX Inc., San Diego, CA). Transfection efficiency was assessed by enhanced green fluorescent protein expression (CLONTECH Laboratories, Inc. Palo Alto, CA) cotransfected with each sample.

Cell cycle analysis
Control and treated cells were collected by trypsinization, fixed in freshly prepared ice cold 70% ethanol for 30 min, and placed at -20 C overnight. After washing with 1x PBS, the cells (1 x 106) were stained with 10 µg/ml propidium iodide and 50 µg/ml RNase A, and incubated for 30 min before flow cytometry analysis using a FACSCalibur flow cytometer (Becton Dickinson and Co., San Jose, CA). Data were analyzed using ModFit V.1.2 software.

Morphometric cell death assay
Adherent and floating cells were collected and fixed with a solution of 1.5% formaldehyde, 40% methanol, and 10% acetic acid. Bis-benzimide (Sigma-Aldrich Corp.) was added to a final concentration of 1 µg/ml, and the cells were incubated for 10 min at room temperature. Ten- microliter aliquots were placed on slides and viewed under UV and phase contrast illumination (Carl Zeiss Axiophot). Cells scored as "dead" exhibited signs of hyperfluorescent chromatin (condensation) and/or nuclear fragmentation (apoptotic), or hypofluorescence with cell lysis and/or karyolysis (necrotic). Cells in three to four field quadrants representative of the sample were identified as normal, apoptotic, or necrotic and quantified to obtain the percent of dead vs. normal cells in each sample in three independent experiments.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
In vitro androgen-ablation modeling using LNCaP cells
Our studies were aimed at investigating the role of PI3K signaling in prostate cancer cells given its central role in growth factor-mediated signaling and the extensive implication of growth factors in PCa progression to androgen independence. To this end, we used the LNCaP cell line and developed an in vitro system of androgen ablation that exhibits many of the features of in vivo progression (19). A major concern that we first addressed was the assertion that changes or responses triggered in the cells by our experimental manipulation were specifically due to the ablation of androgen action (Fig. 1Go). This model, like those of others (22, 23, 24), mimics many of the features of PCa progression, and gives rise to androgen-independent cell sublines.



View larger version (116K):
[in this window]
[in a new window]
  		Figure 1. Ablation of androgen action in LNCaP cells triggers morphological and molecular changes that are blocked by androgen addition. A, LNCaP cells growing in WS-medium alone (WS), WS-medium with 5 µM Cx (WS+Cx), CSS-medium alone (CSS), or CSS-medium with 1 nM R1881 (CSS+R1881) were photographed 3 d after treatment initiation under phase contrast illumination at a x300 magnification. B–D, Dose- and time-dependent experiments were used for total RNA isolation and Northern analysis of androgen and neuroendocrine markers. GAPDH mRNA detection was used to illustrate equal loading of samples. Blots were sequentially probed with intermittent stripping for NSE, NT, PSA, and GAPDH. Representative results are shown for samples from dose-dependent, 3-d treatment with Cx (B), time-dependent cultures in CSS-medium (C), and androgen add-back samples from 3 d of treatment (D). Similar results were obtained in two additional independent experiments.

 
LNCaP cells were plated initially in serum-free RPMI media for 48 h to remove long-term androgenic effects, after which time the media was changed to include one of the following: whole serum (WS) which contained low levels of androgens, WS plus the androgen receptor antagonist Cx (WS+Cx), CSS to remove androgens, or CSS plus 1 nM of the synthetic, nonmetabolizable androgen R1881 (CSS+R1881). Treatment of cells with CSS alone induced a time-dependent cell cycle arrest (Table 1Go) and neuroendocrine morphological changes that included soma compaction and the growth of two or more long slender processes connecting with neighboring cells (Fig. 1AGo). The same was seen in cells treated with WS+Cx (Fig. 1AGo). In contrast, cells receiving CSS+R1881 showed none of these changes and behaved like those receiving WS alone (Fig. 1AGo).


View this table:
[in this window]
[in a new window]
  		Table 1. Effects of acute androgen ablation on the cell cycle distribution of LNCaP cells

 
At the molecular level, ablation of androgen action (antagonism with Cx or removal by CSS culture) was confirmed by monitoring the levels of androgen-regulated PSA mRNA (25) (Fig. 1Go, B and C). In addition, androgen removal resulted in the induction of the neuronal marker, NSE mRNA (10), which correlated with the morphological changes seen in response to androgen removal (Fig. 1Go, B and C). Moreover, the induction of NT mRNA was observed with both Cx and CSS treatment (Fig. 1Go, B–D). The latter was in agreement with previous reports of NT protein induction under similar conditions (26). Also, just as the cell cycle arrest and neuroendocrine changes could be blocked by adding R1881, the changes in PSA, NSE, and NT mRNA levels could also be blocked by the addition of R1881 to the medium (Fig. 1DGo). Therefore, in this model the growth, morphological, and molecular changes triggered by the removal or antagonism of androgens were due specifically to the ablation of androgen action.

The dramatic cell cycle arrest and neuroendocrine changes are appreciable within 72 h of androgen deprivation. We call cells at this stage LNCaP-NEL, for LNCaP with neuroendocrine-like features. However, after prolonged culture in the absence of androgens (>10 wk), cells exhibited intermediate neuroendocrine changes and reestablished normal PSA and AR expression (data not shown). In addition, the cell fraction traversing the cell cycle increased (i.e. S-phase >=6–12% vs. <=3% in the LNCaP-NEL stage) in an androgen-independent manner (data not shown). We call cells at this stage LNCaP-Rf for hormone refractory LNCaP. Because the system is amenable to manipulation and exhibits many of the features seen in PCa progression (19), it allows for studies aimed at defining the temporal role of the PI3K signaling pathway in the androgen-independent growth and survival of PCa cells during progression.

PI3K and Akt activity are increased during and after acute and chronic androgen ablation
To determine the role of PI3K during progression to the androgen-independent state, we first asked what effects, if any, acute or chronic androgen deprivation had on the PI3K activity of LNCaP or LNCaP-Rf cells, respectively. As hormone-depleted medium (CSS) can mimic the antiandrogenic effects of Cx, we used CSS-medium to deprive LNCaP cells of androgenic support for the remaining studies. We assessed levels of PI3K activity both directly and indirectly. For direct measurements, we assayed PI3K activity in antiphosphotyrosine (PY20) immunoprecipitates from cells under various treatment conditions. As shown in Fig. 2Go, A and B, comparison of PI3K activity in LNCaP cells in the presence of WS or CSS+R1881 (control) vs. the absence of androgens (CSS) consistently showed a 50% increase within the first 96 h of acute androgen ablation. This activity was highest in LNCaP-Rf cells (~4.3-fold compared with parental; Fig. 2Go, A and B). Importantly, the increased PI3K activity during the acute or chronic culture conditions could be inhibited by addition of R1881. Furthermore, this activity was sensitive to the PI3K inhibitor LY294002 (Fig. 2Go, A and B).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 2. Androgen ablation of LNCaP cell lines increases PI3K and Akt activation. A, LNCaP and LNCaP-Rf cells were cultured under the indicated conditions for 4 d, with the exception of LY294002 (LY), which was added 16 h before harvesting of cells. For PI3K activity assays, a TLC-based assay was employed using phosphatidylinositol as a substrate (20 21 ). PIP, Phosphatidylinositol monophosphate. B, The PIP signals were quantitated by densitometry, and the relative fold changes in PI3K activity are shown. C and D, Similar protein samples as those in A were analyzed by immunoblotting with anti-phospho-Ser473 Akt (p-Akt) and whole Akt protein antibodies, and the relative fold change in phospho-Ser473-Akt signal was quantitated by densitometry. E and F, The time-dependent increase in Akt activation (p-Akt) in a typical LNCaP experiment of acute androgen ablation is shown. The 0-d sample was from cells under the androgen ablation protocol but in the presence of 1 nM R1881 for 3 d and is representative of basal Akt activity in serum-starved or normally growing LNCaP cells. Whole Akt protein immunoreactivity illustrates equal loading. The data shown are representative of three independent experiments with similar findings.

 
For indirect measurements of PI3K activity, we used antiphosphoserine-473-Akt antibodies that can detect active, phosphorylated Akt (27). As shown in Fig. 2Go, C and D, removal of androgens resulted in increased levels of active phosphorylated Akt in both LNCaP and LNCaP-Rf cells. Furthermore, the increase in active Akt under acute androgen deprivation was time dependent and consistent with the 50% increase in PI3K activity seen at 4 d of androgen deprivation (Fig. 2Go, E and F). No changes were seen in Akt protein levels during the acute phase period examined. Taken together, these findings show that acute androgen deprivation of LNCaP cells triggered an increase in PI3K and Akt activity. The heightened PI3K-Akt activation was maintained throughout the progression and during the refractory, androgen-independent state of LNCaP-Rf cells.

PI3K is required for growth and survival of LNCaP cells during acute androgen ablation
Next, we asked whether this increased PI3K-Akt activity is functionally important for the survival of LNCaP cells undergoing acute androgen ablation. To address this question, we subjected LNCaP cells to the androgen ablation protocol and treated them with the PI3K inhibitor LY294002. We assessed cell growth, survival, and the stereotypical progression of LNCaP cells to the NEL stage. Under acute androgen deprivation, PI3K inhibition triggered apoptotic cell death, as determined by morphological changes in chromatin condensation and nuclear fragmentation (Fig. 3AGo, inset). As shown in Fig. 3AGo, LY294002 treatment resulted in a dose-dependent increase in spontaneous apoptosis within 48 h in either the presence or absence of androgens. However, LY294002 had a more potent killing effect on cells subjected to the androgen ablation protocol. These effects were also time dependent. As shown in Fig. 3BGo, most of the cells subjected to the CSS protocol and treated with LY294002 (20 µM or more) succumbed by 96 h (>=98% cell death). Cells cultured in CSS+R1881 plus LY294002 had 20% or less cell death (Fig. 3BGo). Therefore, increased PI3K-Akt activation during the acute phase of androgen ablation is functionally significant for the survival of LNCaP cells. Under these conditions, inhibition of PI3K hypersensitizes LNCaP cells to apoptosis and blocks their transition beyond the NEL stage. This effectively blocked the emergence of androgen-independent cells in these cultures (Fig. 3BGo).



View larger version (36K):
[in this window]
[in a new window]
  		Figure 3. Specific hyperdependence for intact PI3K signaling in LNCaP cells for growth and survival under acute androgen ablation conditions. LNCaP cells were subjected to acute androgen ablation in CSS-medium with or without 1 nM R1881 and treated with varying concentrations of LY294002 (added every 24 h), a PI3K inhibitor, the MAPK inhibitor PD98059, or the PKA inhibitor H89. Cell death was scored 48 h later by a morphometric assay of apoptosis (inset; membrane blebbing, chromatin condensation, and nuclear fragmentation) and quantified as a percentage of the total cell counts compared with control cells. The arrow points to the typical cell nuclei scored as apoptotic. Results are representative of two independent experiments performed in triplicate. Each column represents approximately 1800 cells scored; bars represent the SD. B, PI3K inhibition hypersensitizes LNCaP cells to apoptosis and blocks their transition beyond the NEL stage. Cells were subjected to the androgen ablation protocol and were treated without (-LY) or with 20 µM LY294002 (+LY) for 7 d under the indicated culture conditions in six-well plates. Cells were photographed (x300) under phase contrast, fixed, and stained with Coomassie blue. The lower pictures shown below the plate are from the geometrically oriented wells, respectively. Data shown are representative of three independent experiments.

 
As multiple kinase cascades can serve both growth and survival functions (28, 29), we next considered the specificity of this dependence on PI3K. Of note, under certain conditions, nonhydrolyzable cAMP analogs promote neuroendocrine differentiation of LNCaP cells (30, 31), suggesting a role for PKA-dependent pathways in such a process. In addition, neuroendocrine survival has been shown to require PI3K and MAPK activities (32, 33). Given the striking neuroendocrine-like morphology exhibited by LNCaP cells under androgen deprivation, we investigated the functional significance of the MAPK and PKA pathways in our model. As shown in Fig. 3AGo, unlike the effects of PI3K inhibition, pharmacological inhibition of MAPK or PKA with PD98059 and H89, respectively, had minimal effects on the survival of LNCaP cells under acute androgen deprivation. Cell cultures receiving CSS+R1881 (controls) were also minimally affected by MAPK or PKA inhibition. Therefore, these findings suggest that the survival of LNCaP cells upon acute androgen deprivation is predominantly due to the PI3K signaling pathway and not to the MEK-ERK or PKA pathways.

Requirement of PI3K for growth and survival during the chronic, androgen-deprived, refractory phase
We next examined whether the heightened PI3K-Akt activity in the androgen-independent LNCaP-Rf cells was likewise required for their growth and survival in the chronic state of androgen deprivation. We further considered whether inhibition of PI3K by transfection of wild-type PTEN in the LNCaP-Rf cells had the same effects as pharmacological inhibition with LY94002. As shown in Fig. 4AGo, transfection of PTEN into LNCaP and LNCaP-Rf cells cultured in CSS-medium resulted in a dose-dependent increase in apoptosis. However, LNCaP-Rf cells were more sensitive to the killing effects of PI3K inhibition by PTEN. Similar results were observed with the pharmacological inhibition of PI3K with LY294002 (Fig. 4BGo). The slightly lower cell death in the PTEN-transfected cells (Fig. 4AGo) compared with LY-treated cells (Fig. 4BGo) may be due to transfection efficiency (~70%, as determined by cotransfected enhanced green fluorescent protein expression). Taken together, these findings demonstrate that intact PI3K signaling is functionally important for the growth and survival of LNCaP-Rf cells during the chronic stages of androgen deprivation as it is during the acute phase (LNCaP-NEL).



View larger version (32K):
[in this window]
[in a new window]
  		Figure 4. Hyperdependence of LNCaP-Rf cells on intact PI3K activity in the refractory state. A, LNCaP and LNCaP-Rf cells were transfected with increasing amounts of wild-type PTEN expression plasmid while cultured in CSS-medium. Medium was changed 24 h after transfection to remove dead cells and debris from electroporation effects, and cell death was scored 48 h after. The data shown are representative of three independent experiments performed in duplicate; bars represent the SD. B, LNCaP and LNCaP-Rf cells cultured in CSS-medium were treated with vehicle or 20 µM LY294002 (LY) for 72 h. LY294002 was added every 24 h. The data shown are from a representative of three independent experiments performed in duplicate; bars represent the SD. C, Growth curves of LNCaP and LNCaP-Rf cells with or without androgen and with or without LY294002. LNCaP and LNCaP-Rf cells were plated on six-well plates in CSS-medium and allowed to grow to 60% confluence. Cells were then treated with or without 20 µM LY294002 and with or without androgen (1 nM R1881). Cell numbers present in the wells were determined by manual counting using a hemocytometer at the time of treatment initiation (0 h) and at 24-h intervals as indicated. Vehicle (-LY) or LY294002 (+LY) was added every 24 h, and R1881 was added fresh when medium was changed (72 h). The data points shown are the average of three samples from a representative experiment performed in triplicate. D and E, LNCaP C4-2 cell growth under androgen ablation conditions with or without PI3K inhibition. C4-2 cells were subjected to the androgen ablation protocol and growth conditions as indicated. After 7 d of LY294002 (+LY) or vehicle treatment (-LY), cultures were fixed in Coomassie (D), or cell numbers present in the wells were manually counted with a hemocytometer (E). The cell numbers shown are the average from an experiment performed in triplicate. Error bars represent the SD. The dashed line represents the baseline cell numbers present in the wells at the time of treatment (14.5 ± 1 x 104).

 
However, throughout our studies we noticed that the hypersensitization to apoptosis due to PI3K inhibition under androgen-deprived conditions is influenced by the length of androgen deprivation before PI3K inhibition. This appears to be consistent with the long-acting effects of androgens. Nonetheless, as shown in Fig. 4CGo, inhibition of PI3K is growth inhibitory to LNCaP cells regardless of whether androgens are present in the medium. The absence of androgens hypersensitizes cells to spontaneous apoptosis induced by LY294002 treatment. These characteristics are not limited to in vitro-derived LNCaP sublines. As shown in Fig. 4Go, D and E, the in vivo-derived LNCaP-C4-2 subline (24), when cultured under conditions of androgen ablation, exhibits similar dependence on PI3K signaling and susceptibility to apoptosis in the absence of androgens.

Modulation of p27kip1 levels by PI3K during androgen-independent proliferation of LNCaP-Rf cells
A remarkable change in the LNCaP-Rf compared with LNCaP-NEL cells is the ability to proliferate in an androgen-independent manner. Because androgen-independent proliferation would be expected to be a rate-limiting step in PCa progression, we next examined the role of PI3K in the androgen-independent proliferation of LNCaP-Rf cells. Studies have shown that androgen ablation has dramatic growth inhibitory effects in PCa cells and have implicated the cyclin-dependent kinase inhibitor p27kip1 as a major mediator of this growth inhibition (18, 22, 23). During acute androgen deprivation, cell cycle arrest in our model correlated with increased p27kip1 protein levels and PI3K activation in LNCaP-NEL cells (Figs. 2CGo and 5AGo). However, during progression to androgen-independent proliferation (LNCaP-Rf), PI3K activation remained highly elevated, yet p27kip1 protein levels were markedly diminished (Figs. 2CGo and 5AGo). This occurred despite the fact that LNCaP-Rf cells have a relatively slower proliferation rate than LNCaP cells (data not shown). Therefore, we sought to determine the relationship of p27kip1 protein levels and PI3K action in the refractory state. First, we examined the effects of inhibiting PI3K on the cell cycle in LNCaP-Rf cells. As shown in Fig. 5BGo, 20 µM LY294002 induced cell cycle arrest in LNCaP-Rf cells within 24 h of treatment. The cell cycle arrest due to LY294002 treatment is distinctly different from that induced by androgen deprivation (compare Table 1Go and Fig. 5BGo). The LY294002-induced arrest is rapid and characterized by G1- and G2/M-enriched fractions. Although a similar effect is seen in parental LNCaP cells in the presence of androgens, the kinetics are slower (~40 h), and they do not exhibit the hypersensitivity to apoptosis seen in LNCaP-Rf cells ( Figs. 3–5GoGoGo and data not shown). Thus, we next examined the effects of PI3K inhibition on the levels of cyclin-dependent kinase inhibitors in LNCaP-Rf cells. As shown in Fig. 5Go, C and D, PI3K inhibition with LY94002 triggered an increase in p27kip1 levels in a dose- and time-dependent manner in exponentially growing LNCaP-Rf cells. In contrast to changes in p27kip1 levels, p21cip1 and p16INK4a levels decreased with PI3K inhibition. As expected, the blocked phosphoactivation of Akt after LY treatment confirmed the inhibition of PI3K. Taken together, these findings show that changes in the levels of p27kip1 are finely regulated by PI3K activity in LNCaP-Rf cells.



View larger version (52K):
[in this window]
[in a new window]
  		Figure 5. p27kip1, p21cip1, and p16INK4a changes during progression of LNCaP cell lines and inhibition of PI3K in LNCaP-Rf cells. A, Protein samples were prepared for immunoblot analysis of cell cycle regulators from LNCaP-NEL cells at 3 d of acute androgen ablation treatment and from exponentially growing LNCaP and LNCaP-Rf cells in CSS+R1881 and CSS, respectively. Western blots were probed with antibodies to p27kip1, p21cip1, p16INK4a, and CDK4 proteins. Unchanged CDK4 levels illustrate equal loading of lanes. B, Exponentially growing LNCaP-Rf cell monolayers were treated for 24 h with vehicle or 20 µM LY294002 (LY) as indicated. Cells were analyzed by flow cytometry. Results shown are representative of three independent experiments. C and D, Exponentially growing LNCaP-Rf cell monolayers were treated with 20 µM LY294002 (LY) for the indicated times (C) or for 48 h with the indicated concentrations of LY (D). Western blots were probed with specific antibodies to p27kip1, p21cip1, and p16INKa; phospho-Ser473-Akt (p-Akt); and whole Akt protein. Immunoreactivity to whole Akt protein illustrates equivalent loading of samples. The data shown are representative of three independent experiments.

 
PI3K inhibition affects the rate of degradation of p27kip1 in LNCaP-Rf cells
We next examined the mechanism by which p27kip1 is regulated by PI3K in LNCaP-Rf cells. The levels of p27kip1 can be regulated at both transcriptional and posttranscriptional levels (34). Akt, a major PI3K effector, has been shown to phosphorylate and thereby inactivate forkhead-type transcription factors, which regulate p27kip1 at the transcriptional level (35, 36). Therefore, it was plausible that the effects of PI3K could be mediated by Akt via forkhead transcription factors. We treated LNCaP-Rf cells with LY94002 under various conditions or transfected them with increasing amounts of myristolated, constitutively active Akt (37) and examined p27kip1 mRNA levels by Northern analysis. No changes in p27kip1 message levels were seen under such conditions (Fig. 6Go, A and B), whereas parallel samples used for immunoblot analysis consistently showed increased p27kip1 protein levels with PI3K inhibition (Fig. 6Go, A and B). Likewise, increasing active Akt had no effect on p27kip1 message levels.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 6. A PI3K-sensitive step in p27kip1 protein turnover regulates its accumulation in LNCaP-Rf cells. A, Exponentially growing LNCaP-Rf cells were subjected to a dose-dependent treatment with LY294002 (LY) for 48 h or were transiently transfected with empty vector (0 µg) or increasing amounts of constitutively active Akt expression vector (CA-Akt), and total RNA was isolated and subjected to Northern analysis with a specific p27kip1 probe. The blots were reprobed with GAPDH as a control. Parallel cell samples were used for immunoblotting for p27kip1 protein, phospho-Ser473-Akt, whole Akt, and ERK2 as a loading control. B and C, Exponentially growing LNCaP-Rf cells were treated with 30 µM LY294002 (LY) or ethanol (EtOH) for 24 h, followed by 20 µg/ml cycloheximide (time zero) and collected at the indicated times (0–8 h). Samples were subjected to Western immunoblotting for p27kip1, p21cip1, and ERK2 proteins to determine their half-lives under the treatment conditions (B). Multiple chemiluminescence exposures were obtained, and the signal from nonsaturating exposures was used for the quantification of signal intensities. The p27kip1 signal was normalized to the signal obtained at time zero, and the percentage of signal remaining at each time was plotted on a log-linear plot and fitted to a linear curve (C).

 
As p27kip1 mRNA levels were not affected, we next considered posttranscriptional regulation. We compared the half-life of p27kip1 protein in the presence or absence of PI3K inhibition. Exponentially growing LNCaP-Rf cells were first treated with LY94002 for 24 h, followed by cycloheximide treatment (time zero). Cells were then harvested for immunoblotting of p27kip1 and p21cip1 at the indicated time intervals. As shown in Fig. 6Go, C and D, PI3K inhibition increased the half-life of p27kip1 protein from 4 h to greater than 8 h, whereas the half-life of p21cip1 was not affected by the treatment. Taken together, these findings suggest that p27kip1 protein levels are regulated posttranscriptionally by PI3K at a step involving its rate of degradation.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study we investigated the role of the PI3K signaling pathway during in vitro progression of LNCaP cells to the androgen-independent state. Our findings show that LNCaP cells used PI3K signaling to successfully overcome two critical limiting steps for progression to androgen independence, namely survival and androgen-independent proliferation. Paradoxically, androgen ablation alone increased PI3K-Akt activation, which facilitated progression. The increased PI3K signaling was necessary for surviving the acute onset and chronic stages of androgen deprivation. During the acute phase, PI3K inhibition blocked survival and therefore effectively blocked the emergence of androgen-independent cells in the population. In addition, the androgen-independent proliferation of LNCaP-Rf cells became exquisitely sensitive to PI3K inhibition. This correlated with p27kip1 protein levels, which are regulated by PI3K activity posttranscriptionally independent from Akt. Thus, PI3K signaling serves two major functions during the acute and chronic stages of androgen deprivation in LNCaP cells: in the acute phase, its effects on growth and survival are predominant, whereas in the refractory state, it also becomes important for fine modulation of p27kip1 levels, which allows proliferation in an androgen-independent manner.

Our in vitro androgen ablation modeling triggers changes that are specifically due to the ablation of androgen action. Two critical components of this study system are the androgen and serum deprivation for 48 h, followed by the addition of hormone-depleted medium (or WS and AR antagonism) that provides growth signals. However, because PCa cells require androgens to effectively overcome the G1 cell cycle restriction (18), cells eventually accumulate at the G1 phase of the cell cycle. This androgen deprivation induced cell cycle arrest differs from that induced in these cells by PI3K inhibition. The latter is characterized by enriched G1 and G2/M fractions from which cells will undergo apoptosis if the PI3K inhibition is not relieved.

In our model, as in those used by others (18, 22, 23), the CDI p27kip1 seems to be the major mediator of the G1 arrest, because its changes among the fluctuating cyclin-dependent kinase inhibitors correlated with the arrest or resumption of cell cycling throughout the progression process. However, it is possible that the arrest due to acute androgen ablation is different from that due to PI3K inhibition. The acute androgen ablation is mainly a G0/G1 arrest, whereas the PI3K inhibition-mediated arrest is a mix of late G1 and G2/M arrest. Consistent with such a possibility is the finding that p16INK4a, which acts in early G1 (38), does not decrease during the acute androgen ablation G1 arrest, but is decreased beyond its already low levels by the PI3K inhibition-mediated arrest in the chronic and refractory state. In contrast, the cell cycle arrest due to PI3K inhibition appears to be a late G1 event, because p27kip1, which exhibits broader inhibitory roles (39), accumulates, whereas p16INK4a and p21cip1 levels are further diminished. In this context, the decrease in p16INK4a (and possibly p21cip1) allow cells beyond the early G1 phase at the time of PI3K inhibition initiation, to continue cell cycle progression. However, even those cells escaping p16INK4a restrains are affected by the increasing p27kip1 levels and are eventually arrested (i.e. G2/M). For example, the G2/M and S-phase fractions due to acute androgen ablation are equally diminished (<4%). In contrast, the PI3K inhibition-mediated growth arrest in LNCaP-Rf cells exhibited a large G2/M fraction (>16%) and negligible S-phase (<=1%). In either scenario, the resulting cell cycle arrest is distinctively different and consistent with the overall fluctuation in cyclin-dependent kinase inhibitors.

The morphological and molecular neuroendocrine changes observed under our experimental manipulations are not solely dependent on intact PI3K signaling (this report and our unpublished data). However, neuroendocrine changes in LNCaP cells can be triggered by a myriad of stimuli (30, 31, 40) and may be a generalized stress response by these cells. These changes are not limited to in vitro-derived, androgen-independent LNCaP sublines because the C4- 2 subline (24) can also be induced to an NEL stage by the in vitro androgen ablation protocol (our unpublished data). Nevertheless, the contribution of neuroendocrine changes to progression remain to be clarified, as PCa tumors exhibit similar changes when androgen deprived (40).

Our findings show that PI3K is required for the growth of LNCaP cells regardless of whether androgens are present. However, the absence of androgens hypersensitizes cells to PI3K inhibition-dependent apoptosis. This latter effect was dependent upon the length of androgen deprivation before PI3K inhibition, as noted by others (41). These effects may be due to the long-acting effects of androgens. The finding that pharmacological or molecular inhibition of PI3K results in apoptotic hypersensitivity under androgen-deprived conditions strongly suggests that increased PI3K activation is functionally significant for the ability of these PCa cells to survive and thus progress to the androgen-independent state.

Addition of androgens within a discrete window of time affords cells some protection from PI3K inhibition-dependent apoptosis. This could also be appreciated in LNCaP-Rf cells despite their growth inhibitory sensitivity to 1 nM R1881, as has been shown in similar LNCaP sublines (22, 23). In this regard, the LNCaP-Rf cells, like similarly derived in vitro cell sublines (22, 23), differ from in vivo-derived LNCaP sublines such as C4-2 (24) in that they are growth inhibited by nanomolar concentrations of androgens. The in vivo-derived, androgen-independent PCa cell sublines and xenografts are apparently not subsequently growth inhibited by androgens (42). However, when C4-2 cells are subjected to androgen ablation conditions as in our studies, they do exhibit some growth inhibition after the addition of androgen, as well as a dependence and sensitivity to PI3K signaling inhibition. Furthermore, prolonged culturing of LNCaP-Rf cells in CSS-medium eventually (~1 yr) gives rise to androgen-independent cells that are not growth inhibited by the addition of 1 nM R1881 (our unpublished data), in agreement with findings in similarly evolved sublines (22). This paradoxical result may in part be related to changes in hormone sensitivity and the bimodal effect on cell proliferation in response to androgens as well as AR expression and stability changes (42).

The PI3K-Akt signaling axis has been described recently as a dominant growth factor survival pathway in PCa cells, which at best can only be partially compensated by a MAPK-sensitive step (43). A similar possibility was suggested by others in LNCaP cells (41). Our studies show that PI3K signaling plays multiple roles throughout the progression and maintenance of the androgen-independent state. Both acute and chronic androgen deprivation triggered an increase in PI3K and Akt activation, which was sustained throughout the progression process. Although it is unclear how the increased PI3K activation ensues in our model, removal of androgen support from PCa cells has been shown to promote the abnormal establishment of autocrine growth factor loops and enhance the engagement of already present growth factor signaling pathways (7, 10). Consistent with this possibility, we saw the induction of neurotensin after androgen ablation (10, 26). It is conceivable that in PCa cells under such conditions, neurotensin could synergize with other growth factors, such as epidermal growth factor and IL-6 (44), thereby heightening PI3K activation. Alternatively, the loss of PTEN, which often occurs in PCa, may also potentiate PI3K activation (16).

Recent studies have shown antagonism between PTEN and the AR’s transcriptional activity as well as AR-dependent growth and apoptosis in PCa cells (45). Our findings are in agreement with such a possibility. Together, these studies suggest that PI3K activity contributes to the AR transcriptional activity and androgen-dependent growth, possibly by increasing overall gene transcription and expression (46). Inhibition of PI3K activity in PTEN-null PCa cells can antagonize androgen action (45). Indeed, an apparently antagonistic relationship between PTEN and androgen action is consistent with our findings of increased PI3K activity upon androgen ablation. Likewise, reintroduction of androgens in the context of androgen ablation decreases PI3K activity, yet does not cause cell death as does direct PI3K inhibition by PTEN or LY294002. The latter is in agreement with the idea that PI3K is required for LNCaP cell growth regardless of androgen status. Although no effects of androgen addition on phospho-Ser473-Akt levels were found by Li et al., (45), this apparent difference from our findings may be due to the differences in experimental conditions used, which include cell culture and the time frames examined. For example, under our experimental conditions, it is the adaptive response of the cells to the androgen ablation stress that increases PI3K-Akt activity above basal levels. It is this relative increase in PI3K-Akt that is decreased or blocked from taking place by the reintroduction of androgen for 96 h. In addition, our findings suggest that in the context of androgen ablation stress, the relationship between the PI3K pathway and androgens is likely to be indirect, through enhanced or antagonized autocrine growth factor loops (7).

Whereas PI3K action is necessary for PCa cell progression as seen in this study model, it remains unclear which PI3K effector(s), in the context of a PCa cell, mediates the critical actions needed for progression. Major individual effectors, such as Akt alone, may not suffice. For example, during the acute ablation of androgen action in this model, PI3K and Akt activities are increased. However, during this very critical period, cell cycle progression is arrested rather than driven forward. This cell cycle quiescence is maintained for months despite the heightened PI3K and Akt activity levels. In addition, the expression of a constitutively active Akt protein does not promote cell cycle progression in the absence of androgens (our unpublished data). In agreement with such findings, Graff et al. (47) reported that LNCaP cells stably expressing a constitutively active form of Akt cannot form tumors in castrated mice. These findings suggest that other major PI3K effectors, the combination of Akt and other PI3K-dependent actions, or a complex interaction of PI3K action and that of other, as yet unknown, signaling mechanisms cooperate in providing the necessary growth and survival support for PCa cells to proliferate in an androgen-independent manner.

The androgen-independent proliferation in LNCaP-Rf cells was accompanied by a profound decrease in p27kip1 protein levels. Surprisingly, although PI3K-Akt inhibition would be expected to influence p27kip1 mRNA levels, possibly via forkhead transcription factors (35, 36), no induction of mRNA levels were observed. Also, increasing active Akt by exogenous expression of constitutively active Akt (37) had no effect on p27kip1 mRNA levels. However, it has recently been shown that endogenous forkhead transcription factors in PTEN-null cells may be deregulated in their localization, which may hinder their transcription capacity (48). In our studies the critical regulation of diminished p27kip1 protein levels in LNCaP-Rf cells, which presumably allows androgen-independent G1-S transition, seems to be dependent primarily on the heightened PI3K activity. This appears to be at the level of protein turnover, because the p27kip1 protein half-life rapidly increased upon PI3K inhibition. Furthermore, the effects of PI3K inhibition were specific and selective for p27kip1, as shown by the accumulation of p27kip1, but not p21cip1. Consistent with these findings, two recent reports have shown that p27kip1 levels can be regulated by PTEN through the ubiquitin E3 ligase SCFSKP2 and its lipid phosphatase, respectively (49, 50).

Our findings suggest one mechanistic explanation for the changes in p27kip1 protein levels seen in PCa tumors. The initial cell cycle inhibitory effects of acute androgen ablation correlate with increased p27kip1 levels, much as is seen in vivo (51, 52, 53, 54). After progression to the androgen-independent state, the heightened PI3K activity in LNCaP-Rf cells appears to be responsible for the steady state profound decrease in p27kip1 protein. In vivo, PCa tumors also exhibit a decrease in p27kip1 with progression (51). It is tempting to speculate that this process would be enhanced during the progressing of PCa to a hormone refractory state due to increased growth factor action, possibly via cyclin E-CDK2 activation and changes in cyclin-p27kip1 complexes (55, 56). Naturally, loss of PTEN would be expected to contribute to this effect, as has been seen in PTEN-null glioblastoma cells (57). The androgen deprivation model used in our studies exhibits striking similarities to PCa in vivo, where p27kip1 is generally decreased and is emerging as a promising prognostic marker (51, 58, 59, 60).

In conclusion, our findings are consistent with a major steady shift in growth support from androgens to growth factors during progression of LNCaP cells to the androgen-independent state, much as is seen during PCa progression (7, 8, 10). The PI3K signaling axis appears to be a central mediator of this growth factor action, especially under the stress of androgen ablation. The fact that LNCaP cells are PTEN-null may have allowed this event to be more easily revealed in our studies. These findings suggest that prostate cancer cell growth is controlled by a balanced interrelationship of multiple signaling pathways. Interference of androgen-regulated growth and survival pathways by androgen ablation in LNCaP cells is mainly compensated by PI3K signaling, allowing cells to survive androgen deprivation and the eventual emergence of androgen-independent growth. As PI3K signaling alone does not suffice to sustain androgen-independent proliferation, it remains to be determined what additional changes in PCa cells occur during the transitory phase following acute androgen ablation that eventually make androgen-independent proliferation possible. If indeed the PI3K pathway is a favored pathway for PCa progression (17), targeting the PI3K signaling axis alone or in combination with androgen ablation may prove therapeutically useful.


   Acknowledgments
 
We thank Drs. Larry M. Karnitz, Ralf Janknecht, and Michael P. Fautsch for suggestions and critical reading of this manuscript.


   Footnotes
 
This work was supported in part by a predoctoral MARC fellowship (to H.M.) from NIGMS (GM-18397), the T. J. Martell Foundation, and the NCI, NIH (CA15083 and CA91956).

1 H.M. and H.H. contributed equally. Back

Abbreviations: A-I, Androgen-independent; CDI, cyclin-dependent kinase inhibitor; CSS-medium, charcoal-stripped serum medium; Cx, Casodex; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GF, growth factor; LN-NEL, LNCaP with neuroendocrine-like features; LN-Rf, androgen refractory LNCaP; NSE; neuron-specific enolase; NT, neurotensin; PCa, prostate cancer; PSA, prostate-specific antigen; WS-medium, whole FBS-containing medium.

Received April 9, 2001.

Accepted for publication July 12, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Huggins C, Hodges CV 1972 Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. CA Cancer J Clin 22:232–240[Abstract/Free Full Text]
  2. Huggins C 1965 Two principles in endocrine therapy of cancers: hormone deprival and hormone interference. Cancer Res 25:1163–1167
  3. Catalona WJ 1994 Management of cancer of the prostate. N Engl J Med 331:996–1004[Free Full Text]
  4. Westin P, Bergh A 1998 Apoptosis and other mechanisms in androgen ablation treatment and androgen independent progression of prostate cancer: a review. Cancer Detect Prev 22:476–484[Medline]
  5. Fournier G 1996 Treatment of hormone-refractory prostate carcinoma. Eur Urol 30(Suppl 1):32–37
  6. Kyprianou N, Bruckheimer EM, Guo Y 2000 Cell proliferation and apoptosis in prostate cancer: significance in disease progression and therapy. Histol Histopathol 15:1211–1223[Medline]
  7. Russell PF, Bennett S, Stricker P 1998 Growth factor involvement in progression of prostate cancer. Clin Chem 44:705–723[Abstract/Free Full Text]
  8. Wong YC, Wang YZ 2000 Growth factors and epithelial-stromal interactions in prostate cancer development. Int Rev Cytol 199:65–116[CrossRef][Medline]
  9. Papandreou CN, Usmani B, Geng Y, Bogenrieder T, Freeman R, Wilk S, Finstad CL, Reuter VE, Powell CT, Scheinberg D, Magill C, Scher HI, Albino AP, Nanus DM 1998 Neutral endopeptidase 24.11 loss in metastatic human prostate cancer contributes to androgen-independent progression. Nat Med 4:50–57[CrossRef][Medline]
  10. Nelson JB, Carducci MA 2000 Small bioactive peptides and cell surface peptidases in androgen-independent prostate cancer. Cancer Invest 18:87–96[Medline]
  11. Schneikert J, Peterziel H, Defossez P-A, Klocker H, de Launoit Y, Cato ACB 1996 Androgen receptor-Ets protein interaction is a novel mechanism for steroid hormone-mediated down-modulation of matrix metalloproteinase expression. J Biol Chem 271:23907–2391[Abstract/Free Full Text]
  12. Zhang M, Magit D, Sager R 1997 Expression of maspin in prostate cells is regulated by a positive Ets element and a negative hormonal responsive element site recognized by androgen receptor. Proc Natl Acad Sci USA 94:5673–5678[Abstract/Free Full Text]
  13. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J, Miliaresis C, Rodgers L, McCombie R, Bigner SH, Giovanella BC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R 1997 PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 275:1943–1947[Abstract/Free Full Text]
  14. Steck PA, Pershouse MA, Jasser SA, Yung WK, Lin H, Ligon AH, Langford LA, Baumgard ML, Hattier T, Davis T, Frye C, Hu R, Swedlund B, Teng DH, Tavtigian SV 1997 Identification of a candidate tumor suppressor gene, MMAC1, at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet 15:356–362[CrossRef][Medline]
  15. Wu X, Senechal K, Neshat MS, Whang YE, Sawyers CL 1998 The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci USA 95:15587–15591[Abstract/Free Full Text]
  16. Whang YE, Wu X, Suzuki H, Reiter RE, Tran C, Vessella RL, Said JW, Isaacs WB, Sawyers CL 1998 Inactivation of the tumor suppressor PTEN/MMAC1 in advanced human prostate cancer through loss of expression. Proc Natl Acad Sci USA 95:5246–5250[Abstract/Free Full Text]
  17. Hanahan D, Weinberg RA 2000 The hallmarks of cancer. Cell 100:57–70[CrossRef][Medline]
  18. Knudsen KE, Arden KC, Cavenee WK 1998 Multiple G1 regulatory elements control the androgen-dependent proliferation of prostatic carcinoma cells. J Biol Chem 273:20213–20222[Abstract/Free Full Text]
  19. Horoszewicz JS, Leong SS, Chu TM, Wajsman ZL, Friedman M, Papsidero L, Kim U, Chai LS, Kakati S, Arya SK, Sandberg AA 1980 The LNCaP cell line: a new model for studies on human prostatic carcinoma. Prog Clin Biol Res 37:115–132[Medline]
  20. Endemann G, Yonezawa K, Roth RA 1990 Phosphatidylinositol kinase or an associated protein is a substrate for the insulin receptor tyrosine kinase. J Biol Chem 265:396–400[Abstract/Free Full Text]
  21. Huang C, Schmidt PC, Ma W-Y, Schmidt HHO, Dong Z 1997 Phosphatidylinositol-3 kinase is necessary for 12-O-tetradecanoylphorbol-13-acetate-induced cell transformation and activated protein 1 activation. J Biol Chem 272:4187–4194[Abstract/Free Full Text]
  22. Kokontis JM, Hay N, Liao S 1998 Progression of LNCaP prostate tumor cells during androgen deprivation: hormone-independent growth, repression of proliferation by androgen, and role for p27Kip1 in androgen-induced cell cycle arrest. Mol Endocrinol 12:941–953[Abstract/Free Full Text]
  23. Lu S, Tsai SY, Tsai MJ 1999 Molecular mechanisms of androgen-independent growth of human prostate cancer LNCaP-AI cells. Endocrinology 140:5054–5059[Abstract/Free Full Text]
  24. Thalmann GN, Sikes RA, Wu TT, Degeorges A, Chang SM, Ozen M, Pathak S, Chung LW 2000 LNCaP progression model of human prostate cancer: androgen-independence and osseous metastasis. Prostate 44:91–103[CrossRef][Medline]
  25. Montgomery BT, Young CY, Bilhartz DL, Andrews PE, Prescott JL, Thompson NF, Tindall DJ 1992 Hormonal regulation of prostate-specific antigen (PSA) glycoprotein in the human prostatic adenocarcinoma cell line, LNCaP. Prostate 21:63–73[Medline]
  26. Sehgal I, Powers S, Huntley B, Powis G, Pittelkow M, Maihle NJ 1994 Neurotensin is an autocrine trophic factor stimulated by androgen withdrawal in human prostate cancer. Proc Natl Acad Sci USA 91:4673–4677[Abstract/Free Full Text]
  27. Alessi DR, Cohen P 1998 Mechanism of activation and function of protein kinase B. Curr Opin Genet Dev 8:55–62[CrossRef][Medline]
  28. Schlessinger, J 2000 Cell signaling by receptor tyrosine kinases. Cell 103:211–225[CrossRef][Medline]
  29. Lewis T S, Shapiro P, Ahn N G 1998 Signal transduction through MAP kinase cascades. Adv Cancer Res 74:49–139[Medline]
  30. Cox ME, Deeble PD, Lakhani S, Parsons SJ 1999 Acquisition of neuroendocrine characteristics by prostate tumor cells is reversible: implications for prostate cancer progression. Cancer Res 59:3821–3830[Abstract/Free Full Text]
  31. Bang YJ, Pirnia F, Fang WG, Kang WK, Sartor O, Whitesell L, Ha MJ, Tsokos M, Sheahan MD, Nguyen P, Niklinski WT, Myers CE, Trepel JB 1994 Terminal neuroendocrine differentiation of human prostate carcinoma cells in response to increased intracellular cyclic AMP. Proc Natl Acad Sci USA 91:5330–5334[Abstract/Free Full Text]
  32. Yao R, Cooper GM 1995 Requirement for phosphatidylinositol-3 kinase in the prevention of apoptosis by nerve growth factor. Science 267:2003–2006[Abstract/Free Full Text]
  33. Yao H, York RD, Misra-Press A, Carr DW, Stork PJ 1998 The cyclic adenosine monophosphate-dependent protein kinase (PKA) is required for the sustained activation of mitogen-activated kinases and gene expression by nerve growth factor. J Biol Chem 273:8240–8247[Abstract/Free Full Text]
  34. Slingerland J, Pagano M 2000 Regulation of the cdk inhibitor p27 and its deregulation in cancer. J Cell Physiol 183:10–17[CrossRef][Medline]
  35. Medema RH, Kops GJ, Bos JL, Burgering BM 2000 AFX-like Forkhead transcription factors mediate cell-cycle regulation by Ras and PKB through p27kip1. Nature 404:782–787[CrossRef][Medline]
  36. Tang ED, Nunez G, Barr FG, Guan KL 1999 Negative regulation of the forkhead transcription factor FKHR by Akt. J Biol Chem 274:16741–16746[Abstract/Free Full Text]
  37. Ahmed NN, Grimes HL, Bellacosa A, Chan TO, Tsichlis PN 1997 Transduction of interleukin-2 antiapoptotic and proliferative signals via Akt protein kinase. Proc Natl Acad Sci USA 94:3627–3632[Abstract/Free Full Text]
  38. Sherr CJ 1996 Cancer cell cycles. Science 274:1672–1677[Abstract/Free Full Text]
  39. Sherr CJ, Roberts JM 1995 Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 15:1149–1163
  40. Abrahamsson PA 1999 Neuroendocrine differentiation in prostatic carcinoma. Prostate 39:135–148[CrossRef][Medline]
  41. Carson JP, Kulik G, Weber MJ 1999 Antiapoptotic signaling in LNCaP prostate cancer cells: a survival signaling pathway independent of phosphatidylinositol 3'-kinase and Akt/protein kinase B. Cancer Res 59:1449–1453[Abstract/Free Full Text]
  42. Gregory CW, Jhonson Jr RT, Mohler JL, French FS, Wilson EM 2001 Androgen receptor stabilization in recurrent prostate cancer is associated with hypersensitivity to low androgen. Cancer Res 61:2892–2898[Abstract/Free Full Text]
  43. Lin J, Adam RM, Santiestevan E, Freeman MR 1999 The phosphatidylinositol 3'-kinase pathway is a dominant growth factor-activated cell survival pathway in LNCaP human prostate carcinoma cells. Cancer Res 59:2891–2897[Abstract/Free Full Text]
  44. Mitra SP, Carraway RE 1999 Synergistic effects of neurotensin and ß-adrenergic agonist on 3,5-cyclic adenosine monophosphate accumulation and DNA synthesis in prostate cancer PC3 cells. Biochem Pharmacol 57:1391–1397[CrossRef][Medline]
  45. Li P, Nicosia SV, Bai W 2001 Antagonism between PTEN/MMAC1/TEP-1 and androgen receptor in growth and apoptosis of prostatic cancer cells. J Biol Chem 276:20444–20450[Abstract/Free Full Text]
  46. Davies MA, Koul D, Dhesi H, Berman R, McDonnell TJ, McConkey D, Yung WKA, Steck PA 1999 Regulation of Akt/PKB activity, cellular growth, and apoptosis in prostate carcinoma cells by MMAC/PTEN. Cancer Res 59:2551–2556[Abstract/Free Full Text]
  47. Graff JR, Konicek BW, McNulty AM, Wang Z, Houck K, Allen S, Paul JD, Hbaiu A, Goode RG, Sandusky GE, Vessella RL, Neubauer BL 2000 Increased AKT activity contributes to prostate cancer progression by dramatically accelerating prostate tumor growth and diminishing p27Kip1 expression. J Biol Chem 275:24500–24505[Abstract/Free Full Text]
  48. Nakamura N, Ramaswamy S, Vazquez F, Signoretti S, Loda M, Sellers WR 2000 Forkhead transcription factors are critical effectors of cell death and cell cycle arrest downstream of PTEN. Mol Cell Biol 20:8969–8982[Abstract/Free Full Text]
  49. Mamillapalli R, Gavrilova N, Mihaylova VT, Tsvetkov LM, Wu H, Zhang H, Sun H 2001 PTEN regulates the ubiquitin-dependent degradation of the CDK inhibitor p27(KIP1) through the ubiquitin E3 ligase SCF(SKP2). Curr Biol 11:263–267[CrossRef][Medline]
  50. Weng LP, Brown JL, Eng C 2001 PTEN coordinates G(1) arrest by down-regulating cyclin D1 via its protein phosphatase activity and up-regulating p27 via its lipid phosphatase activity in a breast cancer model. Hum Mol Genet 10:599–604[Abstract/Free Full Text]
  51. Macri E, Loda M 1999 Role of p27 in prostate carcinogenesis. Cancer Metastasis Rev 17:337–344
  52. Chen Y, Robles AI, Martinez LA, Liu F, Gimenez-Conti IB, Conti CJ 1996 Expression of G1 cyclins, cyclin-dependent kinases, and cyclin-dependent kinase inhibitors in androgen-induced prostate proliferation in castrated rats. Cell Growth Differ 7:1571–1578[Abstract]
  53. Agus DB, Cordon-Cardo C, Fox W, Drobnjak M, Koff A, Golde DW, Scher HI 1999 Prostate cancer cell cycle regulators: response to androgen withdrawal and development of androgen independence. J Natl Cancer Inst 91:1869–1876[Abstract/Free Full Text]
  54. Myers RB, Oelschlager DK, Coan PN, Frost AR, Weiss HL, Manne U, Pretlow TG, Grizzle WE 1999 Changes in cyclin dependent kinase inhibitors p21 and p27 during the castration induced regression of the CWR22 model of prostatic adenocarcinoma. J Urol 161:945–949[CrossRef][Medline]
  55. Sheaff RJ, Groudine M, Gordon M, Roberts JM, Clurman BE 1997 Cyclin E-CDK2 is a regulator of p27Kip1. Genes Dev 11:1464–1478[Abstract/Free Full Text]
  56. Lane HA, Beuvink I, Motoyama AB, Daly JM, Neve RM, Hynes NE 2000 ErbB2 potentiates breast tumor proliferation through modulation of p27kip1-Cdk2 complex formation: receptor overexpression does not determine growth dependency. Mol Cell Biol 20:3210–3223[Abstract/Free Full Text]
  57. Cheney IW, Neuteboom ST, Vaillancourt MT, Ramachandra M, Bookstein R 1999 Adenovirus-mediated gene transfer of MMAC1/PTEN to glioblastoma cells inhibits S phase entry by the recruitment of p27Kip1 into cyclin E/CDK2 complexes. Cancer Res 59:2318–2323