The Prostate 71:915 ^ 928 (2011)

ElevatedGnRHReceptor Expression PlusGnRHAgonist Treatment Inhibits theGrowthof a Subsetof

Papillomavirus18 -ImmortalizedHumanProstateCells

Kevin Morgan,* Emmanouil Stavrou, Samuel P. Leighton, Nicola Miller,Robin Sellar, and Robert P. Millar

Medical ResearchCouncilHumanReproductive SciencesUnit,TheQueen’sMedical Research Institute,Little FranceCrescent,Edinburgh,UK

BACKGROUND AND AIMS. Human metastatic prostate cancer cell growth can be inhibitedby GnRH analogs but effects on virus-immortalized prostate cells have not been investigated.METHODS. Virus-immortalized prostate cells were stably transfected with rat GnRH receptorcDNA and levels of GnRH binding were correlated with GnRH effects on signaling, cell cycle,growth, exosome production, and apoptosis.RESULTS. High levels of cell surface GnRH receptor occurred in transfected papillomavirus-immortalized WPE-1-NB26 epithelial cells but not in non-tumourigenic RWPE-1, myoepithelialWPMY-1 cells, or SV40-immortalized PNT1A. Endogenous cell surface GnRH receptor wasundetectable in non-transfected cells or cancer cell lines LNCaP, PC3, and DU145. GnRHreceptor levels correlated with induction of inositol phosphates, elevation of intracellular Ca2þ,cytoskeletal actin reorganization, modulation of ERK activation and cell growth-inhibition withGnRH agonists. Hoechst 33342 DNA staining-cell sorting indicated accumulation of cells in G2following agonist treatment. Release of exosomes from transfected WPE-1-NB26 wasunaffected by agonists, unlike induction observed in HEK293[SCL60] cells. Increased PARPcleavage and apoptotic body production were undetectable during growth-inhibition in WPE-1-NB26 cells, contrasting with HEK293[SCL60]. EGF receptor activation inhibited GnRH-inducedERK activation in WPE-1-NB26 but growth-inhibition was not rescued by EGF or PKCinhibitor Ro320432. Growth of cells expressing low levels of GnRH receptor was not affectedby agonists.CONCLUSIONS. Engineered high-level GnRH receptor activation inhibits growth of a subsetof papillomavirus-immortalized prostate cells. Elucidating mechanisms leading to clone-specific differences in cell surface GnRH receptor levels is a valuable next step in developingstrategies to exploit prostate cell anti-proliferation using GnRH agonists. Prostate 71: 915–928,2011. # 2010 Wiley-Liss, Inc.

KEY WORDS: prostate; GnRH receptor; agonist; anti-proliferation; actin cytoskeleton;exosomes; apoptosis; signaling; inositol phosphate; mitogen activatedprotein kinases

The authors have no conflicts of interest to declare.

Author contributions: KM designed and performed experiments,analyzed data and wrote manuscript. ES and SPL designed andperformed experiments. NM and RS performed experiments. RPMcontributed intellectually.

Grant sponsor: Medical Research Council, UK.

*Correspondence to: Kevin Morgan, The Queen’s Medical ResearchInstitute, Little France Crescent, Edinburgh EH164TJ, UK.E-mail: k.morgan@hrsu.mrc.ac.ukReceived 16 August 2010; Accepted 18 October 2010DOI 10.1002/pros.21308Published online 6 December 2010 in Wiley Online Library(wileyonlinelibrary.com).

� 2010 Wiley-Liss, Inc.

INTRODUCTION

Autocrine/paracrine gonadotropin releasing hor-mone (GnRH) and GnRH receptor systems have beenhypothesized to exist in a variety of tissues, although Gprotein coupled GnRH receptor expression is difficultto detect outside the pituitary gland and certain regionsof the brain [1–3] where it mediates a specializedneuroendocrine role in the regulation of the reproduc-tive system. Human prostate cells (including epithelialsubtypes, mesenchyme stromal cells, hyperplastic,dysplastic, or malignant cells) may express endoge-nous GnRH or GnRH receptor [4–8] but the physio-logical relevance of such receptors is not clear.Nevertheless, the GnRH receptor has been consideredas a potential target for the pharmacological treatmentof prostate cancer [4,9–12] and studies have analyzedthe inhibitory effects of GnRH analogs on the prolifer-ation of metastatic human prostate cancer cell linesLNCaP, PC3, and DU145 isolated from advanced casesof disease [9,10,13–18].

Paradoxically, the growth of prostate cancer celllines cultured in vitro or grown as tumor xenografts canbe inhibited either directly by GnRH receptor activa-tion or by GnRH receptor blockade [19]. The intra-cellular effects elicited by GnRH receptor agonists andantagonists involve quite different mechanisms whichcontinue to be dissected at the cellular/molecular leveldownstream from the receptor [20–24]. However, lowlevels of GnRH receptor expression occur in thesetransformed cell lines, influenced by culture conditionsand cell passage number, often making interpretationof data difficult [13,18].

Therefore, more cell models are required to comple-ment these studies in order to elucidate how GnRHreceptor activation affects prostate epithelial cellgrowth and function. Normal prostate epithelial cellsproduce prostate secretions and serve as a protectivelining of the glandular ducts. They commonly becomeinfected by sexually transmitted viruses [25]. Forexample, papillomavirus infection has long beenassociated with incidences of prostate cancer, and itmay contribute to the development of intraepithelialneoplasia [26–31], particularly when viral DNAbecomes integrated into host cell chromosomes. Dur-ing disease progression, some prostate epithelial celltypes may undergo differentiation into either myo-epithelial cells or neuroendocrine-like cells under theinfluence of locally produced cytokines or other factors[32–35].

Exosomes (prostasomes) released from epithelialcells into prostate fluid may be of diagnostic value forstudies of altered epithelial cell function [36]. Physio-logically, up-regulation of exosome release fromstressed tissue may elicit an anti-cancer immune

response via contribution to elevated antigen presen-tation [37–39]. The composition of prostate exosomeshas been characterized by proteomics studies [40]and the data is consistent with their production byexocytosis of multivesicular bodies (MVBs). Exocytosisof MVBs can be increased by elevated intracellularCa2þ, by cellular stress associated with apoptosis or bycell senescence [38,41,42], all driven through effects onthe actin cytoskeleton [43]. Whether GnRH receptoractivation affects prostate exosome release has notbeen investigated, although GnRH receptor activationdoes impact upon levels of intracellular Ca2þ [14],cytoskeletal function [16], and apoptosis [15,22].

Study of cell lines stably expressing different levelsof GnRH receptor is possible for some cell types andsuch studies can be mechanistically informative [20–22]. However, this approach has not been applied toprostate cells to date. Recently, a variety of humanprostate epithelial cell lines have become availableand these provide an opportunity to investigate theeffects of GnRH receptor activation on cell function inmore detail. Human prostate epithelial cell lines weregenerated by immortalization with papillomavirus-18DNA (RWPE-1 cells) [44] or SV40 DNA (PNT1A cells)[45]. These cells are non-tumorigenic in nude mice.RWPE-1 cells gave rise to a number of derivatives byspontaneous differentiation in vitro, including somewith myoepithelial properties (WPMY-1) [46]. Aseries of tumourigenic RWPE-1 derivatives wereisolated following treatment with a DNA alkylatingcarcinogen (methylnitrosourea, MNU), includingWPE-1-NB26 cells which are highly tumorigenic innude mice [47].

We analyzed the capacity of these cell lines to stablyexpress functionally significant levels of GnRH recep-tor and studied the consequences of receptor activationin several transfected RWPE-1, PNT1A, WPMY-1, andWPE-1-NB26 subclones. Of these cell lines, only twoWPE-1-NB26 cell clones stably expressed high levelsof GnRH receptor at the cell surface following trans-fection. They were profoundly sensitive to GnRHagonist treatment. The results indicate that a raresubset of papillomavirus 18-immortalized prostatecells can be targeted by anti-proliferative GnRHagonists following over-expression of the receptor.

MATERIALS ANDMETHODS

Most chemicals were purchased from Sigma–Aldrich, UK, including D-Trp6-GnRH-I (Triptorelin),GnRH receptor antagonist (Antide), sulforhodamine Band Fura-2/AM. Other synthetic peptides were custommade by EZ Biolab, USA. Rabbit polyclonal antibodieswere from New England Biolabs, UK (anti-pERK,

The Prostate

916 Morgan et al.

-cPARP), Sigma, UK (anti-beta actin, alkaline phospha-tase conjugated goat anti-rabbit), Chemicon Inter-national (anti-neutral endopeptidase, NEP) or Abcamplc, UK (anti-annexin-II). EGF receptor inhibitor[4-(4-Benzyloxyanilino)-6,7dimethoxyquinazoline]and protein kinase C (PKC) inhibitor (Ro320432) werepurchased from Calbiochem-Merck, UK.

Cell Culture

Immortalized human prostate cell lines and prostatecancer cell lines were purchased from LGC Promo-chem, UK or the European Collection of Cell CulturesUK (ECACC for PNT1A cells). Cells were culturedaccording to instructions provided by suppliers. RWPEand WPE-1-NB26 cells were grown in keratinocytemedium supplemented with bovine pituitary extract(BPE) and recombinant epidermal growth factor (EGF,Invitrogen, UK). Other cells were cultured in Dulbec-co’s Modified Eagle’s Medium (DMEM, Sigma) with10% fetal calf serum.

Transfections were performed using rat GnRHreceptor cDNA in pcDNA3.1(þ) neo (Invitrogen)mixed with Fugene 6 (Roche Diagnostics, UK) inOptimem 1 (Invitrogen). Cell clones were isolatedfollowing selection with 0.5 mg/ml G418 (PAA, UK),using cloning cylinders from Sigma.

GnRHBindingAssay

Cells were grown to confluence in 12- or 24-wellplates and 125I-GnRH binding analyses were per-formed in using peptide dilutions in HEPES-DMEM-0.1%BSA at 48C applied in triplicate as describedpreviously [20–22]. Data were adjusted for non-specificbinding in the presence of 1 mM unlabeled GnRH-I andaccording to cell number per well (determined usinga hemocytometer). Reproducibility was tested usingcells grown to different degrees of confluence (20–100%), with number of cells/well ranging from0.05� 106 to 1.0� 106.

Inositol PhosphateAssay

Cells were grown to 50–100% confluence in 12- or24-well plates, labeled with 3H-myo-inositol (Perkin–Elmer, UK) in Special-DMEM (GIBCO, Invitrogen)overnight and 3H-inositol phosphate (IP) productiongenerated following GnRH receptor activation wasmeasured in triplicate using Dowex resin ion-exchangechromatography as described previously [22]. Datawere adjusted for cell number and basal 3H-inositolphosphate production.

Cell GrowthAssay

Cells were seeded into 12-well plates and growthwas determined in triplicate at sequential time points

following treatment with vehicle or active agents, usingtrichloro-acetic acid fixation, sulforhodamine B stain-ing and measurement of absorbance at 540 nm asdescribed previously [22].

FluorescenceActivatedCell Sorting

Approximately 0.5� 106 cells were seeded into10 cm diameter dishes and incubated overnight priorto treatment with 100 nM Triptorelin for 96 hr. ThenHoechst 33342 DNA stain was added to 10 mg/mland incubated for another 1 hr at 378C. Cells were thenharvested by trypsinization, centrifugation, and resus-pension in 1 ml cell growth medium. Samples weresubjected to flow cytometry using a BD LSRFortessacell analyzer (BD Biosciences, San Jose, CA).

Intracellular Ca2þMeasurements

Cells were grown in black 96 well view-plates(Perkin–Elmer) until confluent (50,000–150,000 cellsper well). Growth medium was removed and cells wereincubated in growth factor-free/serum-free 10 mMHEPES buffered-DMEM for 30 min at 378C. Mediumwas then replaced with HEPES-DMEM containing10 mM Fura-2/AM and cells were incubated for 1 hr at378C prior to one wash with fresh HEPES-DMEM.Levels of intracellular Ca2þ were monitored for up to40 min using dual wavelength fluorescence excitation(340 and 380 nm) and emission at 510 nm in a Novostarplate reader (BMG Labs, UK) following injection ofGnRH analog or vehicle.

F-ActinAssay

Measurement of intracellular filamentous actin(F-actin) was performed according to a methodpublished elsewhere [48]. Cells were seeded into 96well black view-plates (Perkin–Elmer) and allowed tobecome 90–100% confluent in 100 ml medium prior totreatment with vehicle or Triptorelin (added as a 10 mlspike) for different lengths of time prior to fixation(fresh 2% paraformaldehyde in phosphate bufferedsaline, PBS, 48C for 30 min), permeabilization (0.1%Triton X-100 in PBS for 15 min), two washes with PBS,binding with 500 nM phalloidin-Oregon Green514(Molecular Probes, UK)-diluted from a 6.6 mM stockin methanol (stored at�208C) in PBS, two washes withPBS and fluorescence measurement in a Novostar platereader (BMG Labs) (excitation filter 511–10 nm, emis-sion filter 530–12 nm). Fluorescence due to phalloidin-Oregon green 514 binding was calculated by subtrac-tion of auto-fluorescence from non-stained cells treatedin parallel. The number of cells per well was calculatedusing a hemocytometer following trypsinization ofcells grown in parallel in spare wells. All treatments

The Prostate

GnRHReceptor inHuman Prostate Cells 917

were performed in triplicate and reproducibility ofdata was confirmed on separate occasions.

Apoptotic Body Isolation

For time-course studies associated with analysesof cell growth-inhibition cells were seeded into 10 cmdiameter dishes (matrigel-coated for HEK293 cells)with 20 ml medium per dish (one-dish per time pointper individual experiment). Cell culture supernatantwas harvested and subjected to low-speed centrifuga-tion (18 ml tube, 6,000 rpm 10 min 48C, AH-627 swing-out rotor, Sorvall, USA) to pellet apoptotic bodies. Thepellet was dissolved in 100 ml Laemmli SDS–PAGEsample buffer and stored at �208C prior to analysis byWestern blotting. A second centrifugation step con-firmed that all apoptotic bodies had been removedfrom the culture supernatant (no visible pellet and noWestern blot signal).

Exosome Isolation

Cells were grown to confluence in 3� 6 well plates,washed twice with PBS and incubated with 1 ml/wellserum-free medium. The cell culture supernatant washarvested, pooled, and subjected to sequential lowspeed and medium speed centrifugation (18 ml tube,three times 6,000 rpm spins for 10 min each, swing-outrotor followed by two times 14,000 rpm spins, 10 mineach) to remove apoptotic cell debris. The supertnatantwas then subjected to high-speed centrifugation(27,000 rpm, 35 min, AH-627 rotor) in order to pelletexosomes. The exosome pellet was dissolved in 60 mlLaemmli sample buffer and 10 ml aliquots wereanalyzed by SDS–PAGE and Western blotting.

Western Blotting

Cells were grown in six-well plates for short-termstudies (up to 6 hr treatment) or in 6 cm diameter dishesfor long term studies (from one to several days).Cell lysates were harvested following treatmentsand aliquots were subjected to Western blotting asdescribed previously. Experiments were performed intriplicate and measurements were standardized tolevels of beta-actin using ImageQuant software (GEHealthcare Life Sciences, UK).

Statistical Analysis

Analyses of quantitative data were performed usingonline tools available at http://easycalculation.com/statistics/standard-deviation.php and http://www.quantitativeskills.com/sisa/ or using Microsoft Excel(t-test). All data were generated in triplicate on at leasttwo separate occasions. Graphs were plotted usingcalculations of mean� standard error of mean (SEM)

and analyzed by non-linear regression modeling usingPrism software (GraphPad, USA). t-Test P values<0.05were considered to be statistically significant.

RESULTS

Virus-immortalized human prostate cell lines weretransfected with a rat GnRH receptor cDNA expressionconstruct and colonies resistant to G418 were isolated.Cell clones expressing rat GnRH receptor at the cellsurface were identified using a 125I-radiolabeled GnRHligand binding assay. Approximately 20–30 clones foreach cell line were assessed for levels of cell surfaceGnRH receptor. Relatively few clones exhibited sig-nificant levels of cell surface receptor-specific binding(1–2 clones for each cell line) and only 2 transfectedWPE-1-NB26 cell clones (WPE-1-NB26-3 and -8)displayed a capacity for maintenance of high levels ofreceptor comparable to the levels (1.6-fold higher)observed in a well-characterized stably transfectedHEK293 cell clone (HEK293[SCL60])(20-22) (Fig. 1A,B).Specific binding of the 125I-labeled ligand was propor-tional to the number of cells per well (Fig. 1C) and non-transfected cells did not possess detectable specificbinding. Endogenous cell surface receptor binding wasnot detectable in metastatic prostate cancer cell linesLNCaP, PC3, and DU145 (Fig. 1D).

The specificity of the binding assay was verifiedusing competitive displacement of the 125I-GnRH tracerwith several synthetic GnRH analogs (Fig. 1E) and itwas possible to generate estimates of the bindingaffinities for these analogs which were consistent withprevious studies (Table I) [21]. His5D-Trp6Trp7Tyr8

GnRH-I, a GnRH analog possessing amino acid residuesubstitutions equivalent to those found in GnRH-II(positions 5, 7, and 8), bound to the rat GnRH receptorwith lower affinity than D-Trp6-GnRH-I (Triptorelin),a GnRH super-agonist.

Presence of functional receptor was verified byassaying for production of 3H-inositol phosphates(IP) following stimulation with GnRH-I, Triptorelin(Fig. 2A) or other GnRH analogs (Fig. 2B). The levels ofIP production correlated with levels of GnRH bindingat the cell surface and binding affinity. Differencesin half maximal excitatory concentration (EC50) forseveral GnRH analogs were determined followinggeneration of dose–response curves (Table I).

Activation of the GnRH receptor caused transient(approximately 2 min duration) and dose-dependentelevations in levels of intracellular Ca2þ according towhich GnRH analog was added to WPE-1-NB26-3 cells(Fig. 2C,D). These transients were similar to resultsobtained with HEK293[SCL60] cells (data not shown).

The growth of cells during stimulation with GnRHagonist (100 nM Triptorelin) was monitored over a

The Prostate

918 Morgan et al.

period of 5 days using sulforhodamine B staining ofcells following in situ fixation. Cells expressing highlevels of GnRH receptor at the cell surface (e.g., clonesWPE-1-NB26-3 and -8, data not shown for clone-8)were growth-inhibited (detectable within 2 days oftreatment with Triptorelin using this assay) compared

to vehicle-treated cells or to cells expressing low levelsof GnRH receptor at the cell surface (clone WPE-1-NB26-14 for example) (Fig. 3A,B). Other GnRH analogagonists were also inhibited cell growth. However,Triptorelin-mediated cell growth-inhibition in WPE-1-NB26-3 cells could not be rescued by PKC inhibitor

The Prostate

Fig. 1. A:Detectionofcell surfaceGnRHreceptorusingbindingdisplacementofradiolabeled125I-His5D-Tyr6 -GnRH-I to stably transfectedcellcloneswithincreasingconcentrationsofunlabeledD-Trp6 -GnRH-I(Triptorelin).HumanprostateepithelialcellclonesWPE-1-NB26-3andWPE-1-NB26-8express similar levelsofreceptor(higher thanstably transfectedHEK293cells).Thescaleusedmeans that someerrorbarsaredifficult tovisualize.B:Relative levels of cell surfaceratGnRHreceptor expressed in stably transfectedhumanprostate cell lines determinedby125I-GnRHbinding assay performed in triplicate.Non-specific bindingwas determined in thepresence of1mMGnRH-I andresidual countssubtractedfromtotalcountsperminute(cpm).WPE-1-NB26-3cells expressedthehighestlevelsofGnRHreceptor.Relativelevelsweredeter-minedbyadjusting fornumberofcellsperwell.C:Specificbindingof125I-GnRHwasproportionaltonumberofcellsperwell.D:Non-transfectedhuman cells, including all prostate cell lines examined in this study, possessed very low level 125I-GnRHbinding that was poorly displaced byunlabeledGnRHanalogpeptides.Representativebindingdataproducedin triplicate forcells at 30 ^90%confluencein24-wellplates.E:BoundHis5D-Tyr6 -125-I-GnRH-I tracercouldbedisplacedfromtransfectedcellsbydifferentsyntheticGnRHanalogs.

GnRHReceptor inHuman Prostate Cells 919

Ro320432 (100 nM, Fig. 3C), unlike the situation instably transfected HEK293[SCL60] cells [22]. A GnRHreceptor antagonist (antide, used at high dose, 1 mM)did not inhibit cell growth (Fig. 3D) but was ableto rescue cells from growth inhibition by agonists(Fig. 3E).

Analysis of cellular DNA content by fluorescenceactivated cell sorting (FACS) after Hoechst 33342staining indicated that cells accumulated in the G2

phase of the cell cycle following treatment withTriptorelin (Fig. 4A). The proportion of WPE-1-NB26-3 cells in G2 phase increased 1.92-fold followingtreatment with 100 nM Triptorelin. This was a smallereffect than seen in HEK293[SCL60] cells, where theproportion of cells in G2 phase increased 2.76-fold.

Differences between WPE-1-NB26-3 cells andHEK293[SCL60] cells treated with GnRH agonist werealso evident following Western blot analysis. Elevated

The Prostate

TABLE I. Aff|nityofGnRH-AnalogBindingtoRatGnRHReceptorExpressedattheCellSurfaceofHEK293andWPE-1-NB26Cells (CompetitiveDisplacementTitrationMeasurements Encompassing a1,000-FoldUnlabeledAnalogConcentrationRange PerformedinTriplicate onThree SeparateOccasions,�SEM.Doses of Analog Required to Elicit Half-Maximal Inositol Phosphate Production (EC50)WereCalculated forAgonists FollowingDose^ResponseAssays Performed inTriplicate onThree SeparateOccasions)

Cell line GnRH analog Binding IC50 (nM) IP production IC50 (nM)

HEK293[SCL60] D-Trp6-GnRH-I 0.4� 0.2 1.2� 0.6His5D-Trp6Trp7Tyr8GnRH-I 3.4� 0.5 15� 7.0

WPE-1-NB26-3 D-Trp6-GnRH-I 0.7� 0.4 0.9� 0.3His5D-Trp6Trp7Tyr8GnRH-I 2.6� 0.3 12� 2.0

Antide �2.6 Antagonist

Fig. 2. A: Production of 3H-radiolabeled inositolphosphates (IP) in stably transfected cells followingGnRHreceptor activation for 60minwith 1mMGnRH-I. The level of IP production correlates with cell surface GnRH receptor expression levels (Fig. 1A).B: Synthetic GnRH-Ianalogs also elicited IP production, butwith different EC50s.C:Triptorelin induceddose-dependent transient elevations in intracellular Ca2þ

in Fura2-loaded stably transfectedprostate cells.D:Transient elevations in intracellular Ca2þ correlatedwithGnRHbinding affinity (Fig.1D)andinductionof inositolphosphateproduction(B).

920 Morgan et al.

levels of cleaved PARP were difficult to detect in theTriptorelin-treated prostate cell line in contrast toHEK293[SCL60] cells (Fig. 4A). Furthermore, productionof apoptotic bodies, detected using centrifugal isolationand annexin-II immunoblotting, was not elevatedin WPE-1-NB26-3 cell cultures following treatmentwith Triptorelin, in contrast to marked elevation inHEK293[SCL60] cultures (Fig. 4A).

Changes in intracellular levels of filamentous actin(F-actin) following GnRH receptor activation weremeasured by staining fixed and permeabilized cellswith fluorescently tagged phalloidin (Fig. 4B). Triptor-elin induced rapid (within a few minutes) and transientpolymerization of actin into filaments (a time courselasting several hours). Cell type-specific differencesin magnitude and dynamics were observed, withHEK293[SCL60] cells exhibiting a more pronouncedand more prolonged effect than WPE-1-NB26-3 cells(Fig. 4B).

The cell type-specific differences in apoptotic bodygeneration and cytoskeletal actin response to Triptor-elin correlated with differences in liberation of exo-somes. Production of exosomes was determined using

high speed centrifugal isolation of microvesicles fromclarified cell culture supernatants followed by Westernblotting for neutral endopeptidase (NEP) and beta-actin (both constituents of exosomes) [40]. WPE-1NB-26-3 cells constitutively released exosomes regardlessof GnRH receptor activation, whilst HEK293[SCL60] cellswere induced to liberate exosomes following GnRHreceptor activation (Fig. 4B).

Protein kinase-mediated intracellular signalingevents following GnRH receptor activation withTriptorelin were examined using Western blotting.Cell type-specific effects of GnRH receptor activationon levels of phosphorylated ERK1/2 (pERK1/2) werediscovered (Fig. 5A–C). Transient alterations in thelevels of pERK1/2 were detectable in clones WPE-1-NB26-3, RWPE-1-20, WPMY-1-14, and HEK293[SCL60]

but not in PNT1A-1 or WPE-1-NB26-14.In WPE-1NB-26-3 cells, Triptorelin induced a rapid

and transient decrease in the levels of pERK1/2 whencells were grown in the presence of EGF (Fig. 5B,C).The effect of Triptorelin on pERK1/2 levels couldbe reversed (i.e., it elicited transiently elevatedlevels of pERK1/2) when the prostate cells were

The Prostate

Fig. 3. A:Effectsof100nMTriptorelinoninvitrocellgrowth.Growth-inhibitionofWPE-1-NB26-3cells treatedwithTriptorelincomparedtovehicletreatedcells(left)wasdetectablewithin2daysusingsulforhodamineBstaining.B:GrowthofWPE-1-NB26-14cellsexpressinglowlevelsof rat GnRH receptor was unaffected byTriptorelin (right).C: Growth-inhibition of WPE-1-NB26-3 cells treated withTriptorelin couldnot be rescued by treatment with protein kinase C inhibitor Ro320432 (100nM).D: Different GnRH-I agonist analogs inhibited growth ofWPE-1-NB26-3 cells in line with binding affinity (Fig. 1D) but GnRH antagonist, antide, did not inhibit cell growth. E: Growth-inhibitionofWPE-1-NB26-3cells treatedwithTriptorelincouldberesucedby1mMantide.

GnRHReceptor inHuman Prostate Cells 921

The Prostate

Fig. 4.

922 Morgan et al.

pre-equilibrated in medium lacking EGF or in thepresence of EGF receptor (EGFR) tyrosine kinaseinhibitor 4-(4-Benzyloxyanilino)-6,7dimethoxyquina-zoline (2.5 mM). These effects were reproducible andstatistically significant (Fig. 5, Bar graphs). Similarcross-talk involving the EGFR was not apparent inHEK293[SCL60], where Triptorelin induced elevatedlevels of pERK1/2 irrespective of the presence ofserum, EGF, or EGFR inhibitor (Fig. 5).

GnRH receptor function in WPE-1-NB26-3 was lostafter extended passage (�26–27 passages from gen-eration of primary frozen stocks). This was observed asa dramatic decline in the magnitude of agonist-inducedinositol phosphate production whilst high levels ofbinding were maintained. The cells stopped growingand died at passage 27.

DISCUSSION

The ability of prostate epithelial cells to express cell-surface GnRH receptor and the subsequent effects ofGnRH receptor activation on cell growth and functionneed to be determined in order to evaluate the potentialto improve therapeutic applications of GnRH analogsfor the treatment of early-stage cancer. Therefore wesurveyed the ability of stably transfected immortalizedhuman prostate epithelial (RWPE-1, WPE-1NB26,PNT1A) and myoepithelial cell lines (WPMY-1) toexpress the rat GnRH receptor at the cell surface andanalyzed the effects of receptor activation on cellproliferation, intracellular signaling and liberationof exosomes. These studies entailed confirmation ofGnRH receptor expression in terms of ligand bindinglevels and specificity, induction of inositol phosphateproduction and transient elevation of intracellularCa2þ. Further studies determined the effects of GnRHreceptor activation on cell cycle arrest, the induction ofapoptosis, changes in the actin filament component ofthe cytoskeleton and effects on protein kinase signal-ing. Rather than documenting a series of unrelatedobservations, the studies examine inter-related cellularprocesses and establish a benchmark for comparisonwith the effects of GnRH receptor activation in

metastatic prostate cancer cells (see Table II) and othercell types.

It was possible to isolate cell clones stably expressinghigh levels of GnRH receptor with reasonable fre-quency (2 from 25 clones) in one (WPE-1-NB26) out of 4immortalized cell lines tested (RWPE-1, PNT1A, andWPMY-1). Relatively low levels of cell surface GnRHreceptor were detectable in other transfected cellclones. No significant cell surface receptor levels weredetectable in non-transfected cells, including meta-static prostate cancer lines LNCaP, PC3, and DU145derived from advanced disease.

The ligand binding specificity of the heterologouslyexpressed GnRH receptor was consistent with dataproduced elsewhere [21]. The sensitivity of the bindingassay was sufficient to enable detection of both func-tionally significant and sub-threshold levels of GnRHreceptor. The results of screening a range of subclonessuggest that expression of supra-threshold levels ofGnRH receptor at the cell surface may be quite rare inimmortalized prostate epithelial cells. Our failure todetect significant levels of endogenous GnRH receptorin the cancer cell lines LNCaP, PC-3, and DU145suggests that very particular culture conditions may berequired for these cells to express significant amountsof the receptor, consistent with previous reports [14,18](see Table II).

Expression of high levels of cell surface GnRHreceptor in clone WPE-1-NB26-3 (and in WPE-1-NB26clone-8, not shown in detail) correlated with a largeinduction of inositol phosphate production, transientelevations in intracellular Ca2þ and with the ability ofTriptorelin to inhibit cell proliferation in vitro. WPE-1-NB26-3 cells are immortalized by chromosomal inte-gration of human papillomavirus-18 DNA [44,47],and perhaps this may influence their ability to expressGnRH receptor and some of their responses to GnRHreceptor activation [27].

Low levels of GnRH receptor expression at the cellsurface could not be utilized to inhibit cell growth usingTriptorelin (e.g., in WPE-1-NB26-clone 14). This resultis consistent with previous observations [22]. A GnRHreceptor antagonist, antide, used a high dose (1 mM) did

The Prostate

Fig. 4. A: Fluorescence activated cell sorting ofWPE-1-NB26-3andHEK293[SCL60]cellswithDNAstainedwithHoechst33342indi-catedthatGnRH-inducedcellgrowthinhibitionwasassociatedwithaccumulationofcells in theG2phaseof thecellcycle.B: CleavageofPARPduringcellgrowthinhibitionwasdifficulttodetectbyWesternblotting in WPE-1-NB26-3 cells treated with 100nM Triptorelin(upper panel) in contrast to the situation in transfected HEK293cells. Levels of apoptotic body production detected by centrifugalisolation followedbyWesternblotting for annexin II (lowerpanel)were unaffected by treatment with GnRH agonist (Triptorelin) inWPE-1-NB26-3 cells in contrast to the large increase detectableinHEK293[SCL60] cells.Levels of spontaneous apoptosiswerehigher

in untreated WPE-1-NB26-3 cells compared to untreatedHEK293[SCL60] cells.C: Rapid and transient changes in intracellularlevels of filamentous actinwere inducedby GnRHreceptor activa-tion in stably transfected WPE-1-NB26-3 and HEK293 cells,determined by measuring binding of fluorescently tagged phalloi-din-Oregon Green. Measurements were performed in triplicateandrepeatedonseparate occasions.D:Exosomeproductionmeas-ured using centrifugal isolation followed by Western blotting forneutralendopeptidase(NEP)andbeta-actinappearedtobeconstit-utive in WPE-1-NB26-3 cells and was not affected by treatmentwithGnRH(100nMTriptorelin) incontrast toinductionofexosomeproductionin transfectedHEK293cells.

GnRHReceptor inHuman ProstateCells 923

The Prostate

Fig. 5.

924 Morgan et al.

not affect the growth of the prostate cells or stablytransfected HEK293[SCL60] cells, but was able to rescuecells treated with Triptorelin.

Interestingly, growth-inhibition in WPE-1-NB26-3cells was not rescued by the protein kinase C(PKC) inhibitor Ro320432, unlike the situation inHEK293[SCL60] cells [22] and in contrast to the pro-apopototic effects of PKC activation in the LNCaP cellline [49].

FACS analysis of cellular DNA content indicatedthat WPE-NB26-3 cells treated with Triptorelin becamearrested in the G2 phase of the cell cycle, similarto HEK293[SCL60] cells [20] (Fig. 4). However, usingWestern blotting it was not possible to demonstrateinduction of elevated levels of cleaved PARP (a markerof apoptosis) in WPE-1-NB26-3 cells following treat-ment with Triptorelin, in contrast to HEK293[SCL60] cells(Fig. 4). This was also the case in stably transfected rat

The Prostate

TABLE II. Details of GnRHEffects onHumanProstate Cell Lines

Cell line Refs.

Metastatic cancer cell linesLNCaP Autocrine inhibition of cell growth [9]

100 nM Triptorelin stimulated growth in medium containing bovine serum100 mM Triptorelin inhibited growth in medium with bovine serum or

charcoal-stripped/steroid free bovine serum[13]

GnRH agonist counteracts EGFR [17]DU145 Autocrine inhibition of cell growth [10]

Increased p-JNK, EGFR-independentIncreased p-ERK1/2, EGFR-dependentIncreased apoptosis [15]Increased F-actin [16]GnRH agonist counteracts EGFR [17]

PC3 Increased p-JNK, EGFR-independentIncreased p-ERK1/2, EGFR-dependentIncreased apoptosis [15]Un-responsive to GnRH agonist, high level expression following

transfection followed by activation inhibited growth[13,14]

Immortalized cell linesWPE-1-NB26-3 No detectable endogenous cell surface GnRH receptor, high levels cell

surface GnRH receptor following transfectionThis study

Receptor activation induces high levels inositol phosphate production,and transient elevation in intracellular Ca2þ

Triptorelin transiently increased F-actin, induced G2 arrest but no effecton exosome production and no increase in apoptosis

Antagonist (Antide) did not affect cell growth, but did rescue cells fromagonist-induced growth inhibition

Protein kinase C inhibitor did not prevent growth inhibitionGnRH receptor activation transiently decreased p-ERK1/2 in presence of

EGF-MAPkinase signaling influenced by EGF receptorReceptor expression extinguished after extended passages (�27)

WPE-1-NB26-8 Similar to WPE-1-NB26-3RWPE-1, WPMY-1,

and PNT1ANo detectable endogenous receptor, low levels of receptor after transfection,

variable effects of p-ERK1/2

Fig. 5. A: PhosphorylationofERK1/2 (pERK1/2) followingGnRHreceptor activation in stably transfected cell clones. All cell lineswereculturedin theabsenceof serumorexogenousgrowthfactorsfor16hr prior to stimulation.Highest levels of pERK1/2 occurred inWPE-1-NB26-clone3andtransfectedHEK293cells.B:Phosphory-lation of ERK1/2 in response to GnRH receptor activation wascontext-dependent in WPE-1-NB26-3 cells. Levels of phosphory-lated ERK1/2 decreased transiently in response to GnRHreceptoractivationwhenWPE-1-NB26-3 cellsweregrowingin thepresence

of EGF (upper panel). Levels of phosphorylated ERK1/2 increasedwhenWPE-1-NB26-3 cellswereincubatedin the absence ofexoge-nous growth factor (‘‘starving’’ or in the presence of 2.5mM EGFreceptor inhibitor, lower panel). In contrast, exogenous EGF didnot affect the ERK1/2 response in transfectedHEK293 cells (lowerpanel). C: Levels of phosphorylated ERK1/2 were quantified intriplicate and statistically significant differences are indicated withanasterisk (t-testP< 0.05).

GnRHReceptor inHuman Prostate Cells 925

B35 neuroblastoma cells [22]. It is possible that WPE-1-NB26-3 cells may undergo necrosis (organelle swelling)or premature senescence rather than apoptosis (organ-elle compaction) following treatment with Triptorelin.Interestingly, the basal level of apoptosis in WPE-1-NB26-3 cell cultures was higher relative toHEK293[SCL60] cultures (annexin-II blotting in Fig. 4)but apoptotic body production was not altered inthe prostate cell cultures following treatment withTriptorelin.

Changes in the actin cytoskeleton elicited by GnRHreceptor activation (detected by phalloidin-OregonGreen 514 binding to filamentous actin) wereless marked in WPE-1-NB26-3 cells compared toHEK293[SCL60] cells. Production of filamentous actinat the plasma membrane can be associated withchanges in cell shape, movement, exocytosis, celldivision, and apoptosis [50–53].

Understanding how changes in actin assembly occurin response to GnRH receptor signaling will becomeclearer as more of the molecular interactions betweenreceptor-activated G proteins and protein scaffoldsare elucidated. Regional alterations in levels ofphosphatidylinositol 4,5-bisphosphate (PIP2) in sub-compartments of the plasma membrane influenced byactivation of heterotrimeric Gq/11-phospholipase C[54–57] may influence juxta-membrane cytoskeletalresponses to GnRH receptor activation. Elevation ofintracellular Ca2þ by inositol trisphosphate action alsoinfluences cytoskeletal function downstream fromreceptor activation, in addition to the likely activationof Rho GTPases caused by liberation of activated Gprotein beta-gamma subunits.

Interestingly, there was no effect of GnRH receptoractivation on exosome release from WPE-1-NB26-3cells, in contrast to HEK293[SCL60] cells (Fig. 4B). Thiscell-specific difference may be due to alternativeregulatory mechanisms affecting exocytosis of multi-vesicular bodies: constitutive, Ca2þ-regulated, andapoptosis-induced [36,41,42]. Further studies arerequired to understand how exosome release isregulated in WPE-1-NB26 cells, but we note thatexosome production correlated with the degree ofapoptosis occurring in the cell cultures (compareannexin-II blots with exosome production in Fig. 4).

Building models to explain how GnRH receptoractivation leads to differential effects on mitogen-activated protein kinase (MAPkinase) activation is achallenging puzzle in view of the intricate effects ofMAPkinase activities on cell proliferation [58–60].Interestingly, the MAPkinase responses elicited byGnRH receptor activation differed in WPE-1-NB26-3cells and HEK293[SCL60] cells (Fig. 5). EGF and EGFreceptor inhibitor were able to modulate the effects ofGnRH on elevation of pERK1/2 levels in the prostate

cells but not in HEK293[SCL60] cells. Competitionbetween EGF receptor and GnRH receptor signalingin WPE-1-NB26 cells is consistent with other descrip-tions of crosstalk to the MAPkinase cascade, includingeffects in LNCaP and DU145 cells [61].

The mechanism leading to transient ERK1/2dephosphorylation in WPE-1-NB26-3 cells is notunderstood. It may involve GnRH receptor mediatedactivation of a specific ERK phosphatase or it mayinvolve induced dissociation of EGF receptor fromsignaling scaffold molecules such as Grb2 and Sos1[62–64], perhaps via a pathway involving PKC-p90RSK-Sos [65]. A more detailed analysis of thefunctional status of MAP kinase signaling scaffoldproteins in these cells will be required to answer thisproblem.

In summary, our results suggest that a small pro-portion of transfected papillomavirus-immortalizedprostate cells can efficiently maintain high levels ofGnRH receptor at the cell surface. More studies arerequired to determine whether other prostate celltypes can express functionally significant levels ofGnRH receptor (neuroendocrine-like cells would be anobvious candidate) and whether endogenous GnRHreceptor levels are elevated at particular stages ofprostate disease.

How GnRH receptor activation leads to cell cycleG2-arrest and how the transition from G2 arrest toapoptosis is averted in the face of robust GnRH receptoractivation in virally immortalized cells [66–68] are twoareas of investigation which can be addressed using thenew prostate cell models described here.

In the future, development of techniques to specif-ically elevate levels of GnRH receptor in disease cellsmay potentially be utilized to more effectively inhibittheir proliferation or re-growth using GnRH agonisttherapy.

ACKNOWLEDGMENTS

Thanks to Ronnie Grant and Ted Pinner for graphics.

REFERENCES

1. Wolfe AM, Wray S, Westphal H, Radovick S. Cell-specificexpression of the human gonadotropin-releasing hormone genein transgenic animals. J Biol Chem 1996;271:20018–20023.

2. Jennes L, Eyigor O, Janovick JA, Conn PM. Brain gonadotropinreleasing hormone receptors: Localization and regulation.Recent Prog Horm Res 1997;52:475–490.

3. Granger A, Ngo-Muller V, Bleux C, Guigon C, Pincas H, Magre S,Daegelen D, Tixier-Vidal A, Counis R, Laverriere JN. Thepromoter of the rat gonadotropin-releasing hormone receptorgene directs the expression of the human placental alkalinephosphatase reporter gene in gonadotrope cells in the anteriorpituitary gland as well as in multiple extrapituitary tissues.Endocrinology 2004;145:983–993.

The Prostate

926 Morgan et al.

4. Qayum A, Gullick W, Clayton RC, Sikora K, Waxman J. Theeffects of gonadotrophin releasing hormone analogues inprostate cancer are mediated through specific tumour receptors.Br J Cancer 1990;62:96–99.

5. Qayum A, Gullick WJ, Mellon K, Krausz T, Neal D, Sikora K,Waxman J. The partial purification and characterization ofGnRH-like activity from prostatic biopsy specimens andprostatic cancer cell lines. J Steroid Biochem Mol Biol 1990;37:899–902.

6. Halmos G, Arencibia JM, Schally AV, Davis R, Bostwick DG.High incidence of receptors for luteinizing hormone-releasinghormone (LHRH) and LHRH receptor gene expression inhuman prostate cancers. J Urol 2001;163:623–629.

7. Tieva A, Stattin P, Wikstrom P, Bergh A, Damber JE. Gonado-tropin-releasing hormone receptor expression in the humanprostate. Prostate 2001;47:276–284.

8. Straub B, Muller M, Krause H, Schrader M, Miller K. Real-timequantitative reverse transcriptase-polymerase chain reaction forluteinizing hormone-releasing hormone receptor gene mRNAexpression in human prostate cancer. Urology 2003;62:172–176.

9. Limonta P, Dondi D, Moretti RM, Maggi R, Motta M.Antiproliferative effects of luteinizing hormone-releasing hor-mone agonists on the human prostatic cancer cell line LNCaP.J Clin Endocrinol Metab 1992;75:207–212.

10. Dondi D, Limonta P, Moretti RM, Marelli MM, Garattini E, MottaM. Antiproliferative effects of luteinizing hormone-releasinghormone (LHRH) agonists on human androgen-independentprostate cancer cell line DU145: Evidence for an autocrine-inhibitory LHRH loop. Cancer Res 1994;54:4091–4095.

11. Stangelberger A, Schally AV, Nagy A, Szepeshazi K, KanashiroCA, Halmos G. Inhibition of human experimental prostatecancers by a targeted cytotoxic luteinizing hormone-releasinghormone analog AN-207. Prostate 2006;66:200–210.

12. White CD, Stewart AJ, Lu ZL, Millar RP, Morgan K. Anti-proliferative effects of GnRH agonists: Prospects and problemsfor cancer therapy. Neuroendocrinology 2008;88:67–79.

13. Ravenna L, Salvatori L, Morrone S, Lubrano C, Cardillo MR,Sciarra F, Frati L, Di Silverio F, Petrangeli E. Effects of triptorelin,a gonadotropin-releasing hormone agonist, on the humanprostatic cell lines PC3 and LNCaP. J Androl 2000;21:549–557.

14. Franklin J, Hislop J, Flynn A, McArdle CA. Signalling and anti-proliferative effects mediated by gonadotrophin-releasing hor-mone receptors after expression in prostate cancer cells usingrecombinant adenovirus. J Endocrinol 2003;176:275–284.

15. Kraus S, Levy G, Hanoch T, Naor Z, Seger R. Gonadotropin-releasing hormone induces apoptosis of prostate cancer cells:Role of c-Jun NH2-terminal kinase, protein kinase B, andextracellular signal-regulated kinase pathways. Cancer Res2004;64:5736–5744.

16. Enomoto M, Utsumi M, Park MK. Gonadotropin-releasinghormone induces actin cytoskeleton remodeling and affects cellmigration in a cell-type-specific manner in TSU-Pr1 and DU145cells. Endocrinology 2006;147:530–542.

17. Montagnani Marelli M, Moretti RM, Mai S, Januszkiewicz-Caulier J, Motta M, Limonta P. Type I gonadotropin-releasinghormone receptor mediates the antiproliferative effects ofGnRH-II on prostate cancer cells. J Clin Endocrinol Metab2009;94:1761–1767.

18. Angelucci C, Lama G, Iacopino F, Ferracuti S, Bono AV, MillarRP, Sica G. GnRH receptor expression in human prostate cancercells is affected by hormones and growth factors. Endocrine2009;36:87–97.

19. Kim DK, Yang JS, Maiti K, Hwang JI, Kim K, Seen D, Ahn Y, LeeC, Kang BC, Kwon HB, Cheon J, Seong JY. A gonadotropin-releasing hormone-II antagonist induces autophagy of prostatecancer cells. Cancer Res 2009;69:923–931.

20. Miles LE, Hanyaloglu AC, Dromey JR, Pfleger KD, Eidne KA.Gonadotropin-releasing hormone receptor-mediated growthsuppression of immortalized LbetaT2 gonadotrope and stableHEK293 cell lines. Endocrinology 2004;145:194–204.

21. Lopez de Maturana R, Pawson AJ, Lu ZL, Davidson L, MaudsleyS, Morgan K, Langdon SP, Millar RP. Gonadotropin-releasinghormone analog structural determinants of selectivity forinhibition of cell growth: Support for the concept of ligand-induced selective signaling. Mol Endocrinol 2008;22:1711–1722.

22. Morgan K, Stewart AJ, Miller N, Mullen P, Muir M, Dodds M,Medda F, Harrison D, Langdon S, Millar RP. Gonadotropin-releasing hormone receptor levels and cell context affect tumorcell responses to agonist in vitro and in vivo. Cancer Res 2008;68:6331–6340.

23. White CD, Coetsee M, Morgan K, Flanagan CA, Millar RP, LuZL. A crucial role for Galphaq/11, but not Galphai/o or Galphas,in gonadotropin-releasing hormone receptor-mediated cellgrowth inhibition. Mol Endocrinol 2008;22:2520–2530.

24. Festuccia C, Dondi D, Piccolella M, Locatelli A, Gravina GL,Tombolini V, Motta M. Ozarelix, a fourth generation GnRHantagonist, induces apoptosis in hormone refractory androgenreceptor negative prostate cancer cells modulating expressionand activity of death receptors. Prostate 2010;70:1340–1349.

25. Zambrano A, Kalantari M, Simoneau A, Jensen JL, Villarreal LP.Detection of human polyomaviruses and papillomaviruses inprostatic tissue reveals the prostate as a habitat for multiple viralinfections. Prostate 2002;53:263–276.

26. DiPaolo JA, Popescu NC, Alvarez L, Woodworth CD. Cellularand molecular alterations in human epithelial cells transformedby recombinant human papillomavirus DNA. Crit Rev Oncog1993;4:337–360.

27. Garner-Hamrick PA, Fostel JM, Chien WM, Banerjee NS, ChowLT, Broker TR, Fisher C. Global effects of human papillomavirustype 18 E6/E7 in an organotypic keratinocyte culture system.J Virol 2004;78:9041–9050.

28. Leiros GJ, Galliano SR, Sember ME, Kahn T, Schwarz E, EiguchiK. Detection of human papillomavirus DNA and p53 codon 72polymorphism in prostate carcinomas of patients from Argen-tina. BMC Urol 2005;24:5–15.

29. Korodi Z, Dillner J, Jellum E, Lumme S, Hallmans G, Thoresen S,Hakulinen T, Stattin P, Luostarinen T, Lehtinen M, et al. Humanpapillomavirus 16, 18 and 33 infections and risk of prostatecancer: A Nordic nested case–control study. Cancer EpidemiolBiomarkers Prev 2005;14:2952–2955.

30. Pretet JL, Charlot JF, Mougin C. Virological and carcinogenicaspects of HPV. Bull Acad Natl Med 2007;191:611–623.

31. Martinez-Fierro ML, Leach RJ, Gomez-Guerra L, Garza-Gua-jardo R, Pais-Johnson T, Beuten J, Morales-Rodriguez IB,Hernandez-Ordonez MA, Calderon-Cardenas G, Ortiz-LopezR, et al. Identification of viral infections in the prostate andevaluation of their association with cancer. BMC Cancer2010;10:326.

32. Smith PC, Hobisch A, Lin DL, Culig Z, Keller ET. Interleukin-6and prostate cancer progression. Cytokine Growth Factor Rev2001;12:33–340.

33. Guarino M, Rubino B, Ballabio G. The role of epithelial-mesenchymal transition in cancer pathology. Pathology 2007;39:305–318.

The Prostate

GnRHReceptor inHuman Prostate Cells 927

34. Yuan TC, Veeramani S, Lin MF. Neuroendocrine-like prostatecancer cells: Neuroendocrine transdifferentiation of prostateadenocarcinoma cells. Endocr Relat Cancer 2007;14:531–547.

35. Vindrieux D, Escobar P, Lazennec G. Emerging roles of chemo-kines in prostate cancer. Endocr Relat Cancer 2009;16:663–673.

36. Lehmann BD, Paine MS, Brooks AM, McCubrey JA, Renegar RH,Wang R, Terrian DM. Senescence-associated exosome releasefrom human prostate cancer cells. Cancer Res 2008;68:7864–7871.

37. Denzer K, Kleijmeer MJ, Heijnen HF, Stoorvogel W, Geuze HJ.Exosome: From internal vesicle of the multivesicular body tointercellular signaling device. J Cell Sci 2000;113:3365–3374.

38. Nilsson BO, Carlsson L, Larsson A, Ronquist G. Autoantibodiesto prostasomes as new markers for prostate cancer. Ups J Med Sci2001;106:43–49.

39. Mignot G, Roux S, Thery C, Segura E, Zitvogel L. Prospects forexosomes in immunotherapy of cancer. J Cell Mol Med 2006;10:376–388.

40. Utleg AG, Yi EC, Xie T, Shannon P, White JT, Goodlett DR, HoodL, Lin B. Proteomic analysis of human prostasomes. Prostate2003;56:150–161.

41. Savina A, Fader CM, Damiani MT, Colombo MI. Rab11 promotesdocking and fusion of multivesicular bodies in a calcium-dependent manner. Traffic 2005;6:131–143.

42. Yu X, Harris SL, Levine AJ. The regulation of exosome secretion:A novel function of the p53 protein. Cancer Res 2006;66:4795–4801.

43. Lanzetti L. Actin in membrane trafficking. Curr Opin Cell Biol2007;19:453–458.

44. Bello D, Webber MM, Kleinman HK, Wartinger DD, Rhim JS.Androgen responsive adult human prostatic epithelial cell linesimmortalized by human papillomavirus 18. Carcinogenesis1997;18:1215–1223.

45. Degeorges A, Hoffschir F, Cussenot O, Gauville C, Le Duc A,Dutrillaux B, Calvo F. Recurrent cytogenetic alterations ofprostate carcinoma and amplification of c-myc or epidermalgrowth factor receptor in subclones of immortalized PNT1human prostate epithelial cell line. Int J Cancer 1995;62:724–731.

46. Webber MM, Trakul N, Thraves PS, Bello-DeOcampo D, ChuWW, Storto PD, Huard TK, Rhim JS, Williams DE. A humanprostatic stromal myofibroblast cell line WPMY-1: A model forstromal–epithelial interactions in prostatic neoplasia. Carcino-genesis 1999;20:1185–1192.

47. Webber MM, Quader ST, Kleinman HK, Bello-DeOcampo D,Storto PD, Bice G, DeMendonca-Calaca W, Williams DE. Humancell lines as an in vitro/in vivo model for prostate carcinogenesisand progression. Prostate 2001;47:1–13.

48. Leira F, Alvarez C, Cabado AG, Vieites JM, Vieytes MR, BotanaLM. Development of a F actin-based live-cell fluorimetricmicroplate assay for diarrhetic shellfish toxins. Anal Biochem2003;317:129–135.

49. Tanaka Y, Gavrielides MV, Mitsuuchi Y, Fujii T, Kazanietz MG.Protein kinase C promotes apoptosis in LNCaP prostate cancercells through activation of p38 MAPK and inhibition of the Aktsurvival pathway. J Biol Chem 2003;278:33753–33762.

50. Hall A. Rho GTPases and the actin cytoskeleton. Science 1998;279:509–514.

51. Malacombe M, Ceridono M, Calco V, Chasserot-Golaz S,McPherson PS, Bader MF, Gasman S. Intersectin-1L nucleotideexchange factor regulates secretory granule exocytosis byactivating Cdc42. EMBO J 2006;25:3494–3503.

52. Yasuda S, Taniguchi H, Oceguera-Yanez F, Ando Y, Watanabe S,Monypenny J, Narumiya S. An essential role of Cdc42-likeGTPases in mitosis of HeLa cells. FEBS Lett 2006;580:3375–3380.

53. Martin-Belmonte F, Gassama A, Datta A, Yu W, Rescher U,Gerke V, Mostov K. PTEN-mediated apical segregation ofphosphoinositides controls epithelial morphogenesis throughCdc42. Cell 2007;128:383–397.

54. Dowal L, Provitera P, Scarlata S. Stable association between Galpha(q) and phospholipase C beta 1 in living cells. J Biol Chem2006;281:23999–24014.

55. Jezyk MR, Snyder JT, Gershberg S, Worthylake DK, Harden TK,Sondek J. Crystal structure of Rac1 bound to its effectorphospholipase C-beta2. Nat Struct Mol Biol 2006;13:1135–1140.

56. Golebiewska U, Scarlata S. Galphaq binds two effectorsseparately in cells: Evidence for predetermined signaling path-ways. Biophys J 2008;95:575–582.

57. Ueda H, Nagae R, Kozawa M, Morishita R, Kimura S, Nagase T,Ohara O, Yoshida S, Asano T. Heterotrimeric G proteinbetagamma subunits stimulate FLJ00018, a guanine nucleotideexchange factor for Rac1 and Cdc42. J Biol Chem 2008;283:1946–1953.

58. Lopez-Ilasaca M. Signaling from G-protein-coupled receptorsto mitogen-activated protein (MAP)-kinase cascades. BiochemPharmacol 1998;56:269–277.

59. Villanueva J, Yung Y, Walker JL, Assoian RK. ERK activity andG1 phase progression: Identifying dispensable versus essentialactivities and primary versus secondary targets. Mol Biol Cell2007;18:1457–1463.

60. Klein EA, Campbell LE, Kothapalli D, Fournier AK, Assoian RK.Joint requirement for Rac and ERK activities underlies the mid-G1 phase induction of cyclin D1 and S phase entry in bothepithelial and mesenchymal cells. J Biol Chem 2008;283:30911–30918.

61. Moretti RM, Marelli MM, Dondi D, Poletti A, Martini L, Motta M,Limonta P. Luteinizing hormone-releasing hormone agonistsinterfere with the stimulatory actions of epidermal growth factorin human prostatic cancer cell lines, LNCaP and DU 145. J ClinEndocrinol Metab 1996;81:3930–3937.

62. Hu Y, Bowtell DD. Sos1 rapidly associates with Grb2 and ishypophosphorylated when complexed with the EGF receptorafter EGF stimulation. Oncogene 1996;12:1865–1872.

63. Sakaguchi K, Okabayashi Y, Kido Y, Kimura S, Matsumura Y,Inushima K, Kasuga M. Shc phosphotyrosine-binding domaindominantly interacts with epidermal growth factor receptorsand mediates Ras activation in intact cells. Mol Endocrinol1998;12:536–543.

64. Shah BH, Farshori MP, Jambusaria A, Catt KJ. Roles of Src andepidermal growth factor receptor transactivation in transientand sustained ERK1/2 responses to gonadotropin-releasinghormone receptor activation. J Biol Chem 2003;278:19118–19126.

65. Frodin M, Gammeltoft S. Role and regulation of 90 kDaribosomal S6 kinase (RSK) in signal transduction. Mol CellEndocrinol 1999;151:65–77.

66. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, HowleyPM. The E6 oncoprotein encoded by human papillomavirusestypes 16 and 18 promotes the degradation of p53. Cell 1990;63:1129–1136.

67. Passer BJ, Nancy-Portebois V, Amzallag N, Prieur S, Cans C,Roborel de Climens A, Fiucci G, Bouvard V, Tuynder M,Susini L, et al. The p53-inducible TSAP6 gene product regulatesapoptosis and the cell cycle and interacts with Nix and the Myt1kinase. Proc Natl Acad Sci USA 2003;100:2284–2289.

68. Terrano DT, Upreti M, Chambers TC. Cyclin-dependent kinase1-mediated Bcl-xL/Bcl-2 phosphorylation acts as a functionallink coupling mitotic arrest and apoptosis. Mol Cell Biol2010;30:640–656.

The Prostate

928 Morgan et al.