Constitutive activation of PI3K-Akt and NF-κB during prostate cancer progression in autochthonous transgenic mouse model

The Prostate 64:224 ^239 (2005)

ConstitutiveActivationof PI3K-Akt andNF-kBDuringProstateCancerProgressioninAutochthonous

TransgenicMouseModel

Sanjeev Shukla,1 Gregory T. MacLennan,2,3 Susan R. Marengo,1

Martin I. Resnick,1,3 and Sanjay Gupta1,3*1DepartmentofUrology,CaseWestern ReserveUniversityandUniversityHospitalsof Cleveland,Cleveland,Ohio

2Departmentof Pathology,CaseWestern ReserveUniversityandUniversityHospitalsof Cleveland,Cleveland,Ohio3IrelandCancerCenter,Cleveland,Ohio

BACKGROUND. Cancer progression is usually facilitated by independent growth signalsthat may lead to increased cell survival and evasion of apoptosis. Phosphatidylinositol 30-OHkinase (PI3K)-Akt and transcription factor NF-kB are important signaling molecules and keysurvival factors involved in the control of cell proliferation, apoptosis, and oncogenesis.Although PI3K-Akt and NF-kB have been implicated in the development and progression ofprostate cancer, expression of thesemolecules duringprogression of autochthonousdisease hasnot been elucidated.METHODS. Prostate cancer growth and progression in autochthonous transgenic adenocar-cinoma of the mouse prostate (TRAMP) mice and male non-transgenic littermates wereobserved by magnetic resonance imaging (MRI). Expression patterns of PI3K-Akt, NF-kB, IkB,and associated signaling molecules during different stages of cancer progression in these micewere examined byWestern blot analysis, electrophoretic mobility shift assay (EMSA), enzyme-linked immunoabsorbent assay (ELISA), kinase assay, and immunohistochemistry.RESULTS. Sequential MRI and gross analysis of prostate gland exhibited increasing prostatevolume associated with the development and progression of prostatic adenocarcinoma inTRAMPmice, compared to male non-transgenic littermates. Differential protein expression ofPI3K, phosphorylated-Akt (Ser473), IkBa and its phosphorylation, IKK kinase activity, NF-kB/p65, p50, DNA binding, and transcriptional-regulated genes, viz., Bcl2, cyclin D1, MMP-9, andVEGF were observed during prostate cancer progression in TRAMP mice, compared to malenon-transgenic littermates. Expressions of these molecules were significantly increased duringcancer progression observed at 24 and 32 weeks of age.CONCLUSIONS. Differential expression pattern of PI3K-Akt, NF-kB and IkB during prostatecancer progression in TRAMPmice suggest that these molecules represent potential moleculartargets for prevention and/or therapeutic intervention. Prostate 64: 224–239, 2005.# 2005 Wiley-Liss, Inc.

KEY WORDS: IkB; PTEN; prostate cancer; TRAMP

Abbreviations: PI3K, phosphatidylinositol 30 kinase; NF-kB, nuclearfactor-kappaB; HGPIN, high-grade prostatic intraepithelial neopla-sia; GEM, genetically engineered mouse; NIK, NF-kB-inducingkinase; IKK, IkB kinase; MRI, magnetic resonance imaging; EMSA,electrophoretic mobility shift assay; TRAMP, transgenic adenocarci-noma of the mouse prostate; MMP, matrix metalloprotease; VEGF,vascular endothelial growth factor; PTEN, phosphatase and tensinhomologue deleted on chromosome-10.

Grant sponsor: United States Public Health Services; Grant numbers:RO1 CA 108512, RO3 CA 099049, RO3 CA 107806, R21 CA 109424;Grant sponsor: Cancer Research and Prevention Foundation.

*Correspondence to: Sanjay Gupta, PhD, Department of Urology,The James and Eilleen Dicke Research Laboratory, Case WesternReserve University and University Hospitals of Cleveland, 10900Euclid Avenue, Cleveland, Ohio 44106.E-mail: [email protected] 12 August 2004; Accepted 3 November 2004DOI 10.1002/pros.20217Published online 14 February 2005 in Wiley InterScience(www.interscience.wiley.com).

� 2005 Wiley-Liss, Inc.

INTRODUCTION

Prostate cancer in humans is believed to developfrom a precursor lesion, high-grade prostatic intrae-pithelial neoplasia (HGPIN), usually in the peripheralzone [1,2]. This invasive carcinoma has a variablepropensity to progress locally or to metastasize [1,2].The development of metastatic capability is oftenaccompanied by a shift from androgen dependence toandrogen independence (or insensitivity) resulting inlimited treatment options for advanced stage cancers[3,4]. Elucidation of molecular pathways that lead toprostate cancer development and progression is diffi-cult in human patients, as these studies may requiredecades of observation. Such studies are also hamperedby the need for obtaining repeat biopsies from thepatients under observation. Consequently, it is advan-tageous to evaluate molecular pathways of prostatecancer development and progression in animal modelsthat simulate human disease, with the aim of develop-ing novel preventive and therapeutic strategies formanagement of this very prevalent and devastatinginfirmity.

Development of genetically engineered mouse(GEM) models over the last decade has revolutionizedthe investigation of mechanism(s) of carcinogenesis[5,6]. These models are developed by genetically mani-pulating oncogenes or tumor suppressor genes impli-cated in human cancers. Such models are powerfultools in cancer research, facilitating identification ofpotential molecular target(s) involved in cancer devel-opment and progression [5–7]. A major strength ofthese models is that cancer arises from normal cells intheir natural tissue microenvironment and progressesthrough multiple stages, as does human cancer [6].These animal models need to be evaluated for mole-cular alterations during cancer progression, and thesechanges must be carefully correlated with possiblesimilar alterations in human prostate cancer.

Considerable evidence supports the notion that thephosphatidylinositol 30-OH kinase (PI3K)-Akt survivalsignaling pathway plays a critical role in humancancers [8]. Activation of PI3K leads to an increase inits lipid products, phosphatidylinositol 3,4,5-tripho-sphate and phosphatidylinositol 3,4-biphosphate, inplasmamembranes [9]. These lipid secondmessengersfacilitate the membrane co-localization of Akt withphosphoinositide-dependent kinases, leading to Aktphosphorylation and its subsequent activation. Activa-tion of Akt subsequently results in inhibition ofproapoptotic signals from proteins such as BAD,caspase 9, members of forkhead family; and promotionof cell survival signals by activating Rel/NF-kB trans-cription factors [8–10]. Previous studies have demon-strated increase of NF-kB levels and/or increased

transcriptional activity in transformed cells and inmany solid cancer cell lines [11]. Aberrant NF-kB acti-vation has been observed in several human malignan-cies including Hodgkin’s lymphoma [12], squamouscell carcinoma of head and neck [13], non-small celllung cancer [14], carcinoma of uterine cervix [15],colorectal cancer [16], pancreatic cancer [17], thyroidcancer [18], breast cancer [19], leukemia [20], multiplemyeloma [21], melanoma [22], and esophageal cancer[23]. Studies from our laboratory and elsewhere haveshown constitutive NF-kB/p65 activation in humanprostatic adenocarcinoma, correlating with increasinggrade of cancer [24,25]. In most non-transformed cells,NF-kB complexes as a heterotrimer essentially com-posed of p50 and RelA/p65 subunits normally seques-tered in the cytoplasm bound to an inhibitory protein,IkB [26–28].With the onset of NF-kB activation, the IkBproteins become phosphorylated, ubiquitinated, andeventually degraded. This liberatesNF-kB, allowing itstranslocation to the nucleus, where it enhances thetranscription of specific genes [26–28]. NF-kB isregulated through signaling mechanisms that controltranscriptional regulation of IkB family members con-verging on a high-molecular weight oligomeric pro-tein, the serine-specific IkB kinase, IKK [26]. Previousstudies have shown the involvement of NF-kB-indu-cing kinase (NIK) and IKK as a mechanism of consti-tutive NF-kB activation in human prostate carcinomacells [29,30]. Based on its role as a key regulator of cellsurvival, PI3K-Akt and NF-kB have emerged asimportant factors in tumorigenesis.

In recent years, transgenic adenocarcinoma of themouse prostate (TRAMP) has emerged as an excellentmouse model of prostate cancer [31]. TRAMP malesspontaneously develop prostatic adenocarcinoma thatprogresses through multiple stages and that exhibitsboth histological and molecular features similar to thatof human prostate cancer. Male TRAMP mice expressa PB-Tag transgene consisting of the minimal �426/þ28 bp regulatory element of the rat probasin promoterdirecting prostate-specific epithelial expression of thesimian virus 40 early genes to abrogate the activity ofp53 and Rb tumor suppressor genes [31]. The loss offunctional p53 and Rb predisposes epithelial cells toenhanced survival signals and allows investigation ofmolecular alterations. Prostate cancer progresses in thismodel in an androgen-dependent fashion and is highlyreproducible. By approximately 6weeks, TRAMPmiceexhibit low-grade PIN, which progresses to HGPIN by12 weeks. Focal adenocarcinoma develops between 12and 18 weeks, and progresses to poorly differentiatedcarcinomawithin 24weeks. By 28weeks of age, 100%ofthese transgenic mice, without any treatment, harbormetastatic prostate cancer in liver, lymph nodes, lungs,and occasionally in bone [31,32]. In this study, we

PI3K/Akt andNF-kBExpression in Prostate Cancer Progression 225

evaluated TRAMP mice of various ages to determinewhether activation of PI3K-Akt kinase, NF-kB and IkBare significantly associated with prostate cancer pro-gression.

MATERIALSANDMETHODS

Animals

Male and female heterozygous C57BL/TGNTRAMP mice, Line PB Tag 8247NG were purchasedas breeding pairs from The Jackson Laboratory (AnnArbor, MI). The animals were bred and maintained atthe AAALAC-accredited Animal Resource Facility ofCaseWestern Reserve University. Housing and care ofthe animals was in accordance with the guidelinesestablished by the University’s Animal ResearchCommittee and with the NIH Guidelines for the Careand Use of Laboratory Animals. Transgenic males forthese studies were routinely obtained as [TRAMP XC57BL/6]F1 or as [TRAMP X C57BL/6]F2 offspring.Identity of transgenic mice was established by PCR-based DNA-screening as previously described [33].Tumor progression was followed by palpation andmagnetic resonance imaging (MRI) assessment. Thirty-two male TRAMP mice and 24 male non-transgeniclittermates were sacrificed at approximately 8, 16, 24,and 32 weeks, dorso-lateral and ventral lobes of theprostate were dissected and fixed for immunohisto-chemistry and/or snap frozen for themolecular studiesdescribed below.

Magnetic Resonance Imaging (MRI)

Monitoring of prostate tumor growth and volumewas conductedby randomselectionof sixTRAMPmiceand male non-transgenic littermates each by perform-ing MRI at 16, 20, 24, 28, and 32 weeks of age as pre-viously described [33]. Imaging in these animals wasperformed by using a whole body 1.5 T imager with25 mT/m gradient strength, 150-msec rise time, and acustom-built 1-cm small animal receiver coil. T1-weighted (TR/TE¼ 400 msec/14 msec), double-echoT2-weighted (TR/TE¼ 1900 msec/20, 84 msec), andCISS T2-weighted (TR/TE/Flip angle¼ 12.3 msec/5.9 msec per 708) gradient echo volumetric scans witha field of view between 2 and 5 cm and in plane resolu-tion of 78–200 mmwere obtained with a slice thicknessof 500–2,000 mm. Images were filmed for subjectiveanalysis and/or transferred to a free-standing imagingworkstation for volumetric analysis of prostate tumor.

ElectrophoreticMobility Shift Assay (EMSA)

EMSA for NF-kB; p65 and p50 subunits wereperformed in the nuclear fraction of prostate tissueobtained from TRAMP mice and male non-transgenic

littermates of different age using LightshiftTM Che-miluminiscent EMSA kit (Pierce Biotechnology,Rockford, IL) as previously described [25]. Briefly,DNAwas biotin labeled using the Biotin 3’ end labelingkit (Pierce Biotechnology) in a 50-ml reaction buffer,5 pmol of double stranded NF-kB oligonucleotide(50-AGTTGAGGGGACTTTCCCAGGC-30 and 30-TC-AACTCCCCTGAAAGGGTCCG-50) incubated in amicrofuge tube with 10 ml of 5� TdT (terminal deoxy-nucleotidyl transferase) buffer, 5 ml of 5 mM biotin-N4-CTP, 10 U of diluted TdT, 25 ml of ultra pure water at378C for 30min. The reactionwas stoppedwith 2.5 ml of0.2 M EDTA. To extract labeled DNA, 50 ml ofchloroform:isoamyl alcohol (24:1) was added to eachtube and centrifuged at 13,000g. The top aqueous phasecontaining the labeled DNA was further used forbinding reactions. Each binding reaction contained 1�-binding buffer (100 mM Tris, 500 mM KCl, 10 mMdithiothretol, pH 7.5), 2.5% glycerol, 5 mM MgCl2,50 ng/ml poly (dI-dC), 0.05% NP-40, 2.5 mg of nuclearextract, and 20–50 fm of biotin-end-labeled targetDNA. The contents were incubated at room tempera-ture for 20 min. To this reaction mixture, 5 ml of 5�loading bufferwas added, subjected to gel electrophor-esis on a native polyacrylamide gel, and transferred to anylon membrane. After transfer was completed, DNAwas cross-linked to themembrane at 120mJ/cm2 usinga UV cross-linker equipped with 254-nm bulb. Thebiotin end-labeled DNA was detected using strepta-vidin-horseradish peroxidase conjugate and a chemi-luminiscent substrate. The membrane was exposed toX-ray film (XAR-5 Amersham Life Science, Inc.,Arlington Height, IL) and developed using a Kodakfilmprocessor. For super-shift assay, antibodies againstNF-kB/p65, NF-kB/p50 was added 30min after begin-ning of the reaction, and incubation was continued foran additional 30–45min.DNAprotein complexeswereanalyzed as described above.

Western BlotAnalysis

Prostate tissues from TRAMP mice and male non-transgenic littermates of different age were processedfor total, cytosolic, and nuclear lysates. Briefly, totaltissue lysates were prepared in cold RIPA buffer[50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% TritonX-100, 1% sodium deoxycholate]. For cytosolic andnuclear lysate, tissuewasminced and suspended in ice-cold hypotonic buffer [25 mM Tris-HCl (pH 7.4), 1 mMMgCl2, 5 mM KCl and protease inhibitor mixture(Roche Molecular Biochemicals)] incubated on ice for30 min; mixed with 0.5% NP-40 containing 100 mMNa3VO4 and 1mMPMSF. The lysate was centrifuged at500g for 10 min; the supernatant constitutes the cyto-solic fraction. The pelleted nuclei were extracted with

226 Shukla et al.

30–50 ml of solution containing 10 mM Tris-HCl(pH7.0), 250mMNaCl, 30mMsodiumpyrophosphate,50 mM NaF, 0.05% NP-40, and inhibitors at 48C for20 min with agitation. The extract was centrifuged at12,000g for 20 min at 48C; the supernatant constitutesthe nuclear extract. The protein content was deter-mined using the DC Bio-Rad protein assay kit aspreviously described [33]. For Western blot analysis,25-mg protein was resolved using 4–20% polyacryla-mide gels (Novex, Carlsbad, CA) and transferred to anitrocellulose membrane. The blot was blocked inblocking buffer (5% non-fat dry milk/1% Tween 20; in20 mM TBS, pH 7.6) for 2 hr at room temperature,incubated with mouse monoclonal and polyclonalantibodies of Akt1/2, NF-kB/p65, NF-kB/p50, IKKa,IkBa, Bcl2 (Santa Cruz Biotechnology, Santa Cruz, CA),PI3K, PTEN, p-IkBa, p-Akt (Ser473) (Cell SignalingTechnology, Beverly, MA), cyclin D1, MMP-9, andVEGF (Lab Vision Corp., Fremont, CA) in blockingbuffer for 2 hr at room temperature or overnight at 48Cfollowed by incubationwith anti-mouse IgG secondaryantibody conjugated with horseradish peroxidase(Amersham-Pharmacia, Piscataway, NJ) and detectedby ECL-chemiluminescence and autoradiographyusing XAR-5 film (Eastman Kodak, Rochester, NY). Itis important to emphasize here that for detection ofNF-kB/Rel members, we utilized monoclonal antibodiesthat could detect the activated form of NF-kB/Relproteins by recognizing an epitope overlapping thenuclear localization signal and IkBa binding site of NF-kB/p65 andNF-kB/p50 and are therefore selective andspecific for the activated forms.

IKKaKinaseAssay

For this assay, 200mgof cytoplasmic proteinwaspre-cleared with A-G beads and immunoprecipitated withagarose-conjugated IKKantibody at 48Covernight. Theimmuno-complex thus obtainedwaswashed 2–3 timeseach with buffer, and then incubated with GST-IkBsubstrate with limited antibody and times as pre-viously described [34].

Enzyme-Linked ImmunosorbentAssay (ELISA)Assay forNF-kBActivity

ELISA was performed for NF-kB: p65 and p50activity. The commercially available Trans-AMTM NF-kB assay kit was obtained from Active Motif NorthAmerica (Carlsbad, CA) for assay of NF-kB: p65 andp50 activity according to the vendor’s protocol. Briefly,the assay uses an oligonucleotide containing NF-kBconsensus site (50-GGGACTTTCC-30) that binds to thecell extract and can detect NF-kB, which can recognizean epitope on p65 or p50 activated and bound to itstarget DNA. This assay is specific for NF-kB activationand is highly sensitive.

Immunohistochemistry

Immuno-histochemical staining for NF-kB/p65 andIkBa was performed using Dako EnVision System(Dako Cytomation, Carpentaria, CA) according tomanufacturer’s protocol as previously described [25].Briefly, 4-mm thick paraffin-embedded sections ofdorso-lateral prostate from 24- and 32-week ageTRAMP mice and male non-transgenic littermateswere deparaffinized, rehydrated, immersed in targetretrieval solution, and blocked for endogenous perox-idase activity. The sectionswerepermeabilized inTNB-BB (100 mM Tris, pH 7.5, 150 mMNaCl, 0.5% blockingagent, 0.3%Triton-X, and 0.2% saponin) and incubatedin primarymonoclonal antibodies of NF-kB/p65, IkBa,and p-Akt (Ser473) (1:200 dilution), overnight at 48C.Control sections were incubated with antisera in thepresence of tenfold excess of these antibodies or withisotype-matched IgG normal goat serum. After wash-ing three times in TBS, sections were incubated for 2 hrat room temperature with biotinylated secondaryantibody. Immunoreactive complexes were detectedusing diaminobenzidine substrate-chromagen. Slideswere then counterstained in Mayer’s hematoxylin,mounted in crystal mount media, and dried overnighton a level surface.

Pathological Examination

The H&E and immuno-stained slides were reviewedindependently by three of the authors (S.S., S.G., andG.T.M.) using light microscopy. Prostate tissues wereidentified and graded as normal, PIN, well-differen-tiated adenocarcinoma (WD),moderately-differentiatedadenocarcinoma (MD), and poorly differentiated ade-nocarcinoma (PD) according to the grading systempreviously described for TRAMPmice [32].

Statistical Analysis

Densitometric measurements of the bands in Wes-tern blot analysis were performed using digitalizedscientific software program UN-SCAN-IT purchasedfrom Silk Scientific Corporation (Orem, Utah). Thesignificance between the TRAMP mice and male non-transgenic littermates of various age groups wasperformed by the Student’s t-test and P-values lessthan 0.05 were taken as significant in the experiments.

RESULTS

To monitor prostate tumor growth and progressionin TRAMP mice, we used a recently perfectedtechnique of MRI. As shown in Figure 1A, prostatevolume was measured by MRI at 16, 20, 24, 28, and32 weeks of age in TRAMP mice and male non-transgenic littermates. Prostate volumes of TRAMP


micewere significantly higher thanprostate volumes ofmale non-transgenic littermates (data not shown) at allage groups, consistent with development and progres-sion of prostate tumor. A progressive 2.2-fold increasein prostate volume was observed in 16-week TRAMPmice compared to male non-transgenic littermatewhich further increased to 2.5-fold at 32 weeks of age(Fig. 1B). The prostate tumor growth in TRAMP micewas further evident by abdominal pelvic palpation,and from assessment of weights of the GU-apparatusand prostates of these groups ofmice (data not shown).Previous studies have shown that by 8 weeks of age,80% TRAMP mice developed prostate cancer asobserved by prominent lesions of invasive cancer inthe dorso-lateral prostate compared to ventral prostatelobes [32]. Therefore, we have confined our studies tothe dorso-lateral lobes of prostate from TRAMP miceand male non-transgenic littermates.

Histopathological examination of the dorso-lateralprostates of TRAMP mice at 8 weeks of age exhibitedPIN lesions in the majority (70%) with a low incidenceof cancer (<5%) within the lobe. In 16-week TRAMPmice, cancer was predominantly well differentiated(66%) with fewer PIN lesions (<12%) and normalprostate (<5%). TRAMP mice at 24 weeks of agepredominantly exhibited well-differentiated cancer(70%) but with a full spectrum of pathology (normal<2%; PIN<5%;moderately differentiated cancer<5%;and poorly differentiated cancer 20%); whereas at 32weeks, poorly differentiated cancer was more promi-nent (68%) alongwithwell-differentiated cancer (22%),and moderately-differentiated cancer (<8%), respec-tively, in marked contrast to the benign findings

observed in non-transgenic littermates (data notshown).

To examine the effect of prostate cancer progressionon NF-kB constitutive activation, we performed elec-trophoretic-mobility shift assay (EMSA), in the nuclearfractions obtained from dorso-lateral prostates ofTRAMP mice and age-matched male non-transgeniclittermates. EMSA is a technique used to confirm theincreased nuclear translocation and DNA bindingactivity of NF-kB [25]. As shown in Figure 2, increasedDNA binding activity for NF-kB/p65 and p50 wasobserved in the dorso-lateral prostate of male non-transgenic littermate with increasing age indicatingnormal age-related prostate growth. The increase inDNA binding activity was more significant duringprostate cancer progression in TRAMP mice at 24 and32 weeks compared to age-matched male non-trans-genic. To further confirm the specificity of NF-kBDNAbinding, we performed super-shift assay with anti-bodies specific for NF-kB/p65 and NF-kB/p50. Asshown in Figure 2, a strong super-shift in case of anti-p65, but a weak shift for anti-p50 was observed in bothTRAMP mice and age-matched male non-transgenicsuggesting that the observed NF-kB band consisted ofthese two subunits. Unlabeled oligonucleotide dimin-ished the intensity of NF-kB, indicating that the shiftedcomplex was a NF-kB specific band (data not shown).

We next examined the protein expression of NF-kB/p65, NF-kB/p50 in the nuclear and cytosolic fractionof dorso-lateral prostate in TRAMP mice and age-matched non-transgenic littermates. As shown inFigure 3, immunoblot analysis for NF-kB/p65 andNF-kB/p50 in the nuclear fraction of male non-

Fig. 1. A:Prostate tumorgrowthinTRAMPmicebetween16and32weeksasassessedbymagneticresonanceimaging.Representativeimageof individual TRAMPmice from each age group are shown here.B:Volumetric analysis of prostate inTRAMPmice andmale non-transgeniclittermates.AmarkedincreaseinprostatevolumewasobservedinTRAMPmicecomparedtomalenon-transgeniclittermates.Arrowsindicateprostate gland.The details are described in Materials and Methods.Values representmean of six animals� SE, **P< 0.001, Student’s t-test,TRAMPversusnon-transgenic. [Color figurecanbeviewedin theonlineissue,whichis availableatwww.interscience.wiley.com.]

228 Shukla et al.

Fig. 2. NF-kBDNAbinding activity in the dorso-lateralprostate ofTRAMPmice andmalenon-transgenic littermatesbyEMSA.EMSAwasperformedto identifynuclear translocationofNF-kBdimers and theirbinding toDNA.Anincreasein thenuclear translocationofNF-kB/p65was observed in prostates of TRAMPmice compared to male non-transgenic littermates.The super-shift assay was performed in 32-weekTRAMPmice andmalenon-transgenic littermateswith antibodies specific forNF-kB/p65 andNF-kB/p50. A shift inNF-kB/p65was observedin32-weekTRAMPmice.Thedetails aredescribedinMaterials andMethods.Controls:#1Biotin-EBNAcontrolDNA,#2Biotin-EBNAcon-trolDNAþEBNAextract,#3Biotin-EBNAcontrolDNAþEBNAextractþ 20-foldmolarexcessofunlabeledEBNADNA.Quantitationofbandswasdonebydensitometricanalysis,andisshownasfoldchangeascomparedto8-week-oldTRAMPandnon-transgeniclittermateintheircorrespondinggroupat thebottomof thebands.

Fig. 3. Protein expressionofNF-kB/p65 andNF-kB/p50 in the dorso-lateralprostate ofTRAMPmice andmalenon-transgenic littermates.Cytosolic andnuclear extracts fromprostateswerepreparedandelectrophorsedby SDS^PAGE followedbyWesternblottingwith anti-p65andanti-p50 antibodies in thenuclear andcytosolic fractions.Aprogressiveincreasein theproteinexpressionofNF-kB/p65andNF-kB/p50 inthenuclear andcytosolic fractionswasobservedin theprostatesofTRAMPmicecomparedtomalenon-transgenic littermates.Thedetails aredescribedinMaterials andMethods.Quantitationofbandswasdonebydensitometric analysis, andis shownas foldchangenormalized to actin(cytosolic fraction)andOct-1 (nuclear fraction), comparedto8-week-oldTRAMPandnon-transgenic littermatein thecorresponding group.


transgenic exhibited an increase in protein expressionat 16, 24, and 32 weeks age compared to 8-week-oldmice. In the cytosolic fraction, however, a progressiveincrease in the protein expression of NF-kB/p50 wasobserved. No significant change in protein expressionof NF-kB/p65 was observed in the cytosolic fraction ofthese mice. In contrast, TRAMP mice exhibited aprogressive increase in protein expression of NF-kB/p65 in the nuclear and cytosolic fractions, whichcorrelated with prostate cancer progression withincreasing age in thesemice. Similar increase in proteinexpression of NF-kB/p50 was observed in TRAMPmice at 24 and 32 weeks both in nuclear and cytosolicfractions (Fig. 3). These results were further confirmedby a more sensitive method that can detect NF-kB andrecognize an epitope on p65 and p50 by ELISA. Asshown in Figure 4A, NF-kB/p65 was significantlyactivated in 24- and 32-weekTRAMPmice compared toage-matched non-transgenic littermates. Similarly,NF-kB/p50 was also activated in TRAMP mice comparedto age-matched non-transgenic at 24 and 32 weeks butthe magnitude of activation was less than compared top65 (Fig. 4B).

Next we analyzed the expression of NF-kB inhibi-tory protein, IkBa, and the upstream kinase IKKa in thecytosolic fraction of TRAMP mice and age-matchedmale non-transgenic littermates. As shown in Figure 5,immunoblot analysis for IkBa exhibited a similarpattern of protein expression as previously observedforNF-kB/p65with significantly high levels inTRAMPmice at 24 and 32weeks comparedwith non-transgeniclittermates. Since the rapid turnover of IkBa requires

phosphorylation at its N-terminal serine residues as asignal for the ubiquitination that eventually leads to itsdegradation, we determined the phosphorylation ofIkBa in the dorso-lateral prostate of TRAMP mice andage-matched non-transgenic littermates. For this weused an antibody specific to phosphorylation sites attheN-terminal Ser 32/36 of IkBa. Compared to prostatetissue of 8 and 16 weeks, a progressive increase in p-IkBawith tumor progression was observed in TRAMPmice similar to the previously noted increase in IkBa.No significant change in the protein expression of IKKawas observed in the prostates of TRAMP mice ofvarious age groups (Fig. 5). In contrast, no significantchange in the protein expression of IKKa, IkBa, and itsphosphorylation was observed in the dorso-lateralprostates of age-matched non-transgenic littermates(Fig. 5).

Because phosphorylation of IkBa is mediated byIKKs, we next determined whether IkBa phosphoryla-tion is mediated via IKKa kinase activity in TRAMPmice during prostate cancer progression. We per-formed an in vitro IKKa kinase activity assay in thecytosolic fraction of dorso-lateral prostate of age-matched TRAMP mice and male non-transgenic. Forthis assay, equal amount of cytosolic protein from 24-week-old TRAMP mice and male non-transgeniclittermate was immunoprecipitated with agarose-con-jugated IKKa antibody and the immuno-complex wasthen incubated with IkBa-GST antibody for varioustime intervals. As shown in Figure 6A, a significantincrease in the levels of IKKa kinase activity wasobserved in TRAMP mice in time-dependent fashion.

Fig. 4. ActivationofNF-kB/p65 (A), andNF-kB/p50 (B) in thedorso-lateralprostate ofTRAMPmiceandmalenon-transgenic littermatesinnuclear fraction as observedbyELISA.NF-kB activationwasdeterminedbyELISATrans-AMkit fromActiveMotifNorthAmerica (Carlsbad,CA) for assay of NF-kB/p65 and NF-kB/p50 activity.The details are described under Materials and Methods.Values represent mean of sixanimals� SE, conductedinduplicate. *P< 0.05, **P< 0.001,Student’st-test,TRAMPversusnon-transgenic.

230 Shukla et al.

No significant changes in the level of IKKa kinaseactivitywere observed after incubation for various timeintervals in the prostate of non-transgenic littermate(Fig. 6A). To further confirm that increased IKKakinaseactivity was associated with prostate cancer progres-sion in TRAMP mice, equal amounts of cytosolicprotein from prostates of 24- and 32-week TRAMPmice was immunoprecipitated with agarose-conju-gated IKKa antibody and the immuno-complex was

then incubated with 0.2, 0.4, and 0.8 mg of IkBa-GSTsubstrate for 30 min, kinase assay was then performed.As shown in Figure 6B, addition of increasingconcentration of substrate to assay incubations resultedin significant increase of IKKa kinase activity in dorso-lateral prostate of TRAMP mice. The magnitude ofincrease of IKKa kinase activity was similar in both 24-and 32-week-old TRAMP mice (Fig. 6B). These resultssuggest that activation of NF-kB in TRAMPmice could

Fig. 5. Proteinexpressionof IKKa and IkBa anditsphosphorylation (Ser32/36) in the dorso-lateralprostate ofTRAMPmice andmalenon-transgeniclittermates.Cytosolic fractions fromprostateswerepreparedandelectrophorsedbySDS^PAGEfollowedbyWesternblottingwithanti-IKKa,anti-IkBa,andanti-p-IkBaantibodiesinthecytosolic fractions.AmarkedincreaseintheproteinexpressionofIkBaanditsphosphor-ylationin the cytosolic fractionswereobservedinprostatesofTRAMPmiceof24and32weekscompared tomalenon-transgenic littermates.The details are described in Materials and Methods.Quantitation of bands was done by densitometric analysis, and is shown as fold changenormalizedto actin, comparedto8-week-oldTRAMPandnon-transgenic littermatein thecorrespondinggroup.

Fig. 6. A: IKKa kinaseactivityin thedorso-lateralprostate ofTRAMPmiceandmalenon-transgenic littermate.Cytosolicprotein (200mg)fromprostateof24-week-oldTRAMPmiceandmalenon-transgeniclittermatewereimmunoprecipitatedwithagarose-conjugatedIKKaanti-bodyovernight,andkinaseassayandphosphorylatedexpressionofIkBawereperformedatvarious timeintervalsusingIkBa-GSTas substrate.AhighkinaseactivitywasobservedinTRAMPmicewherethesubstratewasrapidly finishedwithin5minofincubationcomparedtonon-trans-geniclittermate.B:IKKakinaseactivityinthedorso-lateralprostateofTRAMPmice.Cytosolicprotein(200mg)fromtheprostateof24-and32-weekTRAMPmicewasimmunoprecipitatedwithagarose-conjugated IKKaantibodyovernightandkinaseassaywasperformedusingvariousconcentration of IkBa-GSTas substrate. A progressive increase in kinase activity was observedwith increasing concentrations of IkBa-GSTsubstrate.The details are described inMaterials andMethods.Quantitation of bandswas donebydensitometric analysis, and is shown as foldchangenormalizedtoactincomparedto thecorrespondingcontrol.


be due to phosphorylation and faster turnover of IkBain the cytosolic fraction, which correlates with prostatecancer progression.

Since prostate cancer progression requires increasedcell survival signals which could be mediated by PI3K-Akt signaling pathway [8–10], we next analyzed theprotein expression of PI3K, native/activated Aktlevels, and evaluated the significance of these proteinsduring prostate cancer progression in TRAMPmice. Asshown in Figure 7, immunoblot analysis for regulatorysubunit of PI3K (p85) exhibited significant increase inprotein expression in the dorso-lateral prostates ofTRAMPmice observed at 16, 24, and 32weeks of age. Asimilar pattern of protein expression for activated Akt(p-Akt; Ser473) was observed as previously shown forPI3K with significantly high levels in TRAMP mice insimilar age group. However, no significant variation inthe protein expression of native Akt (Akt1/2) wasobserved in these tissues. In contrast, no significantalteration in the protein expression of PI3K, Akt1/2,and p-Akt (Ser473) was observed in male non-transgenic littermates in various age groups (Fig. 7).

We next examined the protein expression of NF-kB/p65, IkBa, andp-Akt (Ser473) by immunohistochemicalanalysis in the dorso-lateral prostates of TRAMP miceand male non-transgenic littermates. Staining for NF-kB/p65, IkBa, and p-Akt are very weak and predomi-nantly distributed in the stroma of the prostate of 32week male non-transgenic (Fig. 8A,E,I). Similarly,negative to very weak staining patterns for NF-kB/p65, IkBa, andp-Aktwere observed inprostates ofmale

non-transgenic littermates at 8-, 16-, and 24-week age(data not shown). In contrast, weak focal positiveimmunostaining forNF-kB/p65, IkBa, andp-Akt in thePIN lesions was noted in approximately 37% ofTRAMP mice at 8 weeks of age and in 62% in 16-week-old TRAMP mice, whereas no significant immu-nostaining for these proteinswas observed in rest of theanimals (data not shown). Focal NF-kB/p65, IkBa, andp-Akt immunopositivity were evident in approxi-mately 88% of prostates of TRAMP mice at 24 and32 weeks of age. These mice had multiple foci ofadenocarcinoma, and demonstrated prominent NF-kB/p65 staining in 10–15% of the lesions, which arehighly proliferative (Fig. 7B,C). Similar observationswere noted for IkBa (12–15%) (Fig. 8F,G); and p-Akt(8–10%) (Fig. 8J,K) inTRAMPmice of 24 and 32week ofage as were observed for NF-kB/p65.

Since constitutive PI3K/Akt activation may be dueto loss of function and/or stability of PTEN, whichfunctions as a negative regulator of PI3K [9], we furtherdetermined the endogenous expression of PTEN andits phosphorylation in prostates of these mice. PTEN, adual-specificity phosphatase, has been identified as atumor suppressor capable of dephosphorylating bothtyrosine phosphate and serine/threonine phosphateresidue on protein [9]. Recent studies have shown thatnon-catalytic regulatory carboxy terminus of PTENcontains multiple phosphorylation sites includingSer380, and Th382 and Th385, which play critical rolesin regulating PTEN stability [35,36]. As shown inFigure 9A, a progressive increase in endogenous PTEN

Fig. 7. ProteinexpressionofPI3K,Akt1/2,anditsphosphorylation(Ser473)inthedorso-lateralprostateofTRAMPmiceandmalenon-trans-geniclittermates.Tissuelysates fromprostateswerepreparedandelectrophorsedbySDS^PAGEfollowedbyWesternblottingwithanti-PI3K,anti-AKT1/2, andanti-p-Akt (Ser473) antibodies.Aprogressiveincreasein theproteinexpressionofPI3K,PTEN, andp-Aktwere observedinprostates of TRAMPmice compared tomale non-transgenic littermates.The details are described inMaterials andMethods.Quantitation ofbandswasdonebydensitometricanalysis,andis shownasfoldchangenormalizedtoactin,comparedto8-week-oldTRAMPandnon-transgeniclittermatein thecorrespondinggroup.

232 Shukla et al.

protein level was observed in TRAMP mice, whichcorrelated with progressive loss of differentiation andcancer progression,whereas no significant alteration inthe level of PTEN protein was observed in male non-transgenic littermates. To further evaluate the stabi-lity/function of PTEN, we used a phospho-specificantibody; Anti-PTEN [pSpTpS380/382/385] (BioSourceInternational, Camarillo, CA), which has the ability todetect phosphorylation at Ser380, Ser385, and Th382sites of human PTEN. This is also a highly conservedsequence inmice.Using a set of positive control (3T3-L1adipocytes), either stimulated (þLIF) or untreated(�LIF) with leukemia inhibitory factor and throughpeptide competition assay using non-phospho andphospho-PTEN [pSpTpS380/382/385] sample pair, wedetermined the stability/function of PTEN in thesemice. As shown in Figure 9B, compared to 8-, 16-, 24-

week TRAMP mice; PTEN phosphorylation wasobserved in 32-week TRAMP mice, whereas no bandswere observed in non-transgenic littermates. Competi-tion with non-phospho-peptide exhibited proteinbands in both TRAMP mice and non-transgeniclittermates; with higher band intensity observed at 24-and32-weekTRAMPmice. Similarly, theprotein bandswere observed in both TRAMP and non-transgeniclittermates after competition with phospho-peptideindicating the presence of non-phosphorylated PTEN(Fig. 9B).

Since PI3K/Akt and NF-kB have been shown to beinvolved in transcriptional regulation of a number ofgenes whose products are implicated in carcinogenesis[28], we determined the protein expression of Bcl2,cyclin D1, MMP-9, and VEGF, all of which areassociated with progression of prostate cancer. As

Fig. 8. Immunostaining forNF-kB/p65 IkBa andp-Akt (Ser473) inrepresentative samples of the dorso-lateralprostates fromTRAMPmiceandmalenon-transgenic littermates.Aweak stainingpattern forNF-kB/p65(arrowhead),p-Akt, anddiffuse IkBa stainingwasobservedin thedorso-lateralprostate ofnon-transgenic littermates (A,E,I).FocalNF-kB/p65 stainedlesionswere observedin the dorso-lateralprostate of24- and 32-weekTRAMPmice (arrowhead) (B,C).Diffuse occasional staining for IkBawas observed inmalignant epithelial cells of 24-weekTRAMPmice (F);whereasgranularcytoplasmic stainingof IkBawas observedin 32-weekTRAMPmice (arrowhead) (G). Similar focalp-Aktlesions were observed in the dorso-lateral prostate of 24- and 32-weekTRAMPmice as noted for NF-kB/p65 (arrowhead) (J,K).Negativecontrol(D,H,L);magnification200�. [Color figurecanbeviewedin theonlineissue,whichis availableatwww.interscience.wiley.com.]


shown in Figure 10, significant increases in the proteinexpression of Bcl2, MMP-9, and VEGF were observedin the dorso-lateral prostates of male non-transgeniclittermates at 16, 24, and 32 weeks of age. A decrease inthe Bcl-2 protein expression at 24 and 32 weeks of agewas observed, which correlated with normal age-relat-edprostate growth in thesemice. In contrast, significantincreases in the protein expression of Bcl2, cyclin D1,MMP-9, and VEGF were observed in prostates ofTRAMP mice at 24 and 32 weeks of age, whichcorrelated with prostate cancer progression (Fig. 10).

DISCUSSION

The pathologic features and biologic behavior of theTRAMP mouse prostatic neoplasm recapitulate manyof the features of prostate cancer in humans [31,32].Prostatic epithelium in TRAMP mice progresses fromintraepithelial neoplasia to adenocarcinoma that even-tually metastasizes, most often to lymph nodes, lungs,and occasionally to bone, kidney, and adrenal glands[31]. The development of a neuroendocrine phenotype

in advanced prostate cancer has also been observed inthese mice [32]. Because of the similarities betweenhuman prostate cancer and TRAMP mouse prostatecancer, the TRAMP mouse model has been used for awide range of studies including analysis of growthfactors [37,38], assessment of intermediate and end-point biomarkers [39–42], andmarkers of angiogenesis[43]. Such studies aregreatly facilitated by the relativelyshort time course of prostate cancer progression inthese mice. In the present study, we have shown thatPI3K-Akt, NF-kB, and IkB are constitutively activatedduring prostate cancer progression in TRAMP mice.

Prostate cancer progression may be achieved bypromotion of cell survival and/or inhibition of proa-poptotic signal through PI3K-Akt signaling pathway[8–10]. PI3K is a heterodimeric protein composed of acatalytic subunit (p110a) and a regulatory subunit (p85)that participate inmultiple cellular processes includingcell growth, transformation, anddifferentiation [8]. ThePI3K pathway is an essential survival pathway shownto be upregulated by several different mechanisms in anumber of cell types and some forms of human cancers

Fig. 9. ProteinexpressionofPTEN(A)anditsphosphorylation(pSpTpS380/382/385)(B) inthedorso-lateralprostateofTRAMPmiceandmalenon-transgeniclittermates.Tissuelysates fromprostateswerepreparedandelectrophorsedbySDS^PAGEfollowedbyWesternblottingwithanti-PTEN,anti-p-PTEN[pSpTpS380/382/385]antibodies.AprogressiveincreaseintheendogenousproteinexpressionofPTENanditsphosphor-ylationwas observed in prostates of TRAMPmice compared tomale non-transgenic littermates.The details are described in Materials andMethods.Quantitation ofbandswas donebydensitometric analysis, and is shown as fold changenormalized to actin, compared to 8weekoldTRAMPandnon-transgenic littermatein thecorrespondinggroup.

234 Shukla et al.

[10]. Amplification of the gene coding for the catalyticsubunit of PI3K has been observed in cervical andovarian cancers [44,45]. Upon activation of cells withgrowth factors or cytokines, PI3K is recruited to theplasmamembrane,where it catalyzes the conversion ofmembrane phosphoinositide 4,5-biphosphate (PIP2) inthe D3 position to generate phosphoinositide 3,4,5-triphosphate (PIP3). The accumulation of PIP3 creates adocking site for Akt at the plasma membrane, whichbinds to PIP3 via the pleckstrin homology domain aswell as its upstream kinases, PDK-1 and the proposedPDK2 (10). These PDKs phosphorylate Akt at Thr308/309 andSer473/474 residues, respectively, causing full-length Akt activation [8–10]. Activation of Akt (knownas protein kinase B) has also been implicated in thegenesis or progression of various human malignancies[8,46]. PI3K targets, AKT1 andAKT2 are amplified andoverexpressed in breast, gastric, and ovarian cancers[47,48]. Akt3 activity is often increased in prostate andbreast cancer [49]. In experimental systems, constitu-tively active PI3K or Akt is oncogenic in cell culturesystems and animal tumor models [50,51]. Severalstudies have shown that Akt/PKB activates thetranscription of a wide variety of genes, especiallythose involved in immune activation, cell proliferation,apoptosis, and cell survival [8–10,50,51]. ActivatedAktprotects cells from apoptotic death by phosphorylating

substrates such as BAD, pro-caspase-9, NF-kB, andfork-head transcription family members [10,46]. In ourstudies, we observed constitutive activation of PI3Kand Akt (Ser473) in TRAMP mice between 18 and 32weeks of age, a time period during which these micedeveloped invasive prostate cancer with metastases.These results suggest that PI3K and Akt may beimportant molecular markers of prostate cancer pro-gression.

Constitutive PI3K/Akt activation in high-gradeprostate adenocarcinoma of TRAMP mice may be dueto autocrine cytokine stimulation, loss of function and/or inactivation of PTEN, or gain of function in Akt andPI3K [8–10]. In previous studies, loss of function and/or mutation of PTEN have been observed in advancedstage human prostate cancer [52,53] androgen-insensi-tive prostate cancer cells, and in xenograft models [54].Loss of PTEN function through PTEN mutations inmurine models has been shown to be associated withneoplasia in multiple tissue types, including endome-trium, liver, gastrointestinal epithelium, thyroid, thy-mus, and prostate [55]. Loss of PTEN function inprostate cancer cells has been shown to be associatedwith increased proliferation, angiogenesis, and tumor-igenesis [56,57]. Introduction of the PTEN gene incancer cell lines has been shown to inhibit cellproliferation and tumor growth, establishing PTEN as

Fig. 10. ProteinexpressionofBcl2, cyclinD1,MMP-9, andVEGF in thedorso-lateralprostateofTRAMPmiceandmalenon-transgenic litter-mates.Tissuelysate fromprostateswerepreparedandelectrophorsedbySDS^PAGE followedbyWesternblottingwithanti-Bcl2, anti-cyclinD1, and anti-MMP-9 and anti-VEGF antibodies. Amarked increase in the protein expression of these proteins was observed in prostates ofTRAMPmiceof24and32weekscomparedtomalenon-transgeniclittermates.ThedetailsaredescribedinMaterialsandMethods.Quantitationofbandswasdonebydensitometric analysis, andis shownas foldchangenormalizedto actin, compared to8-week-oldTRAMPandnon-trans-genic littermatein thecorrespondinggroup.


tumor suppressor gene [58]. The tumor suppressorfunction of PTEN has been associatedwith its ability todephosphorylate phosphoinositides at positions D3 ofthe inositol ring and to antagonize the PI3K-Akt anti-apoptotic pathway [9]. Studies have shown that thecarboxy terminal region is essential for regulatingPTEN stability and enzymatic activity, and unlikemany signaling proteins, which require phosphoryla-tion for activation, phosphorylation of PTEN at itscarboxy terminal inhibits its activity [35,36]. Phos-phorylated PTEN exists in a monomeric closedconformation thereby reducing its ability to bind toother proteins [59]. Unphosphorylated PTEN exists inan open configuration andparticipates in the formationof high molecular weight complexes. Unphosphory-lated PTEN is less stable butmore active and the loss ofprotein stability is offset by a gain of PTEN activity [59].In our studies, we observed an increase in endogenousPTEN protein expression and its phosphorylation inTRAMP mice during prostate cancer progressionsuggesting that stabilization of the corresponding pro-tein occurs followingprotein inactivation. These resultssuggest that mitogenic signals for invasive prostategrowth in TRAMP mice may be due to effectors otherthan PTEN. However, detail studies are required tosubstantiate these findings.

Members of Rel/NF-kB family are important signal-ingmolecules with a variety of targets that play pivotalroles in intracellular signal transduction pathwaysinvolved in cell growth, cellular transformation, andtumorigenesis [26–28]. Constitutive NF-kB activationhas been detected in several types of carcinoma;mutations and rearrangements of NF-kB/IkB familymembers have been observed in some hematologicalcancers [12–23]. Previously, we have shown that NF-kB/p65 is constitutively activated in human prostateadenocarcinoma and that the level of activationincreases in parallel with increasing cancer grade [25].Similar histological findings were observed in TRAMPmice of various ages previously, with the demonstra-tion that TRAMP mice between 20–32 weeks age aretypically characterized by moderately or poorly differ-entiatedprostate adenocarcinoma [31,32].Weobservedincreased DNA binding activity of NF-kB/p65 inTRAMP mice at 24 and 32 weeks of age, comparedwith male non-transgenic littermates. Studies haveshown that Rel A/p65 exhibited strong transactivationpotential as observed by its constitutive activation insome forms of human cancers [60]. Nuclear transloca-tion of Rel A and NF-kB-DNA binding activity arehigher in cancers of the liver, pancreas, stomach, colon,and uterine cervix in humans, compared to normaltissue from the same sites [14–17]. Likewise, increasedNF-kB/p50 expression was observed in both nuclearand cytosolic fractions of TRAMP mice and in male

non-transgenic littermates. Previous studies havedemonstrated that NF-kB/p50 has low transactivationactivity, usually related to normal growth and mayhave a limited role in carcinogenesis [61]. These resultssuggest NF-kB/p65 (Rel A) is an important molecularmarker for prostate cancer progression in TRAMPmice.

Altered expression of IkBa in malignant tissues hasbeen linked to constitutive NF-kB activation viaphosphorylation of IkBa at Ser32/36, resulting in therelease and nuclear translocation of active NF-kB[11,12]. We observed an increase in the proteinexpression of IkBa and its phosphorylation in thedorso-lateral prostate of TRAMP mice, compared tomale non-transgenic littermates, and the level ofprotein expression of IkBa increased in parallel withprostate cancer progression. This marked increase inIkBa protein expression and its phosphorylation inprostates of TRAMP mice may be the consequence offunctional activation of NF-kB in prostate cancer,which is known to cause strong transcriptional upre-gulation of IkBa as a feedback mechanism possiblyoperative in other cancers as well. Our results arein agreement with previous observations that IkBaprotein levels were increased in advanced stagecarcinoma cells derived from androgen-insensitivehuman prostate cancer, which exhibit high cons-titutive NF-kB expression. Previous studies haveshown that the upstream events associated with theconstitutive activation of NF-kB in prostate carcinomacells involve activation of tyrosine kinases; NIK andIKK [29,30]. We observed that IKK kinase activity wasupregulated in TRAMP mice. This may result inphosphorylation and faster turnover of IkBa in thecytosolic fraction,which correlateswithprostate cancerprogression.

Activation of the PI3K-Akt pathway plays animportant role in the activation of NF-kB, whichmodulates transcriptional activity of a number of genesthat contribute to tumorigenesis [10,28]. PI3K/Akt andNF-kB have been shown to be activated by cytokines,epidermal growth factor, receptor expression, Rasactivation, or through oxidative stress and cell damage[28]. Importantly, chronic inflammation and release ofpro-inflammatory cytokines, known to induce NF-kB,may provide selective growth advantage of neoplasticcells [62]. Studies on prostate carcinoma cells haveshown that PI3K/Akt andNF-kB signaling controls thelevels and expression of many genes whose productsregulate proliferation, cell motility, metastasis, andangiogenesis [10,28]. In our studies, we have shownsignificant increase in the protein expression of Bcl2,cyclin D1, MMP-9, and VEGF in TRAMP mice, all ofwhich are associated with prostate cancer progression.These results suggest that these genes and their

236 Shukla et al.

productsmay serve as additionalmolecularmarkers ofcancer progression in TRAMP mice.

Studies have shown that NF-kB and PI3K/Akt playprominent roles in promotion of cell survival andtumorigenesis [8–11]. Although initially believed tooperate as components of distinct signaling pathways,recent studies have demonstrated that the NF-kB andPI3K/Akt signaling pathways may converge [63–66].This convergence has been demonstrated in somehighly malignant forms of human cancer such as acutemyeloid leukemia, melanoma, squamous cell carci-noma of oral cavity, and pancreatic adenocarcinoma[63–66]. We have observed that PI3K/Akt and NF-kBactivation occur during prostate cancer progression inTRAMP mice, suggesting that these signaling path-ways function through constitutive upregulation toinduce aggressive prostate tumor growth in thismousemodel. The schematic representation of this signalingpathway is depicted in Figure 11. Since constitutiveactivation of PI3K/Akt andNF-kB pathways are consi-dered an important mechanism for prostate cancer

progression in TRAMP mice, further target-basedmolecular approaches and sustained inhibition of thesepathways through prevention and/or therapeuticintervention could be conducted in this mouse modelof prostate carcinogenesis.

ACKNOWLEDGMENTS

This work was supported by grants from UnitedStates Public Health Services RO1 CA 108512, RO3 CA099049, RO3CA107806, R21CA109424 and funds fromCancer Research and Prevention Foundation to S.G.

REFERENCES

1. Abate-Shen C, Shen MM. Molecular genetics of prostate cancer.Genes Dev 2000;14:2410–2434.

2. DeMarzo AM, Nelson WG, Isaacs WB, Epstein JI. Pathologicaland molecular aspects of prostate cancer. Lancet 2003;361:955–964.

3. Feldman BJ, Feldman D. The development of androgen-independent prostate cancer. Nat Rev Cancer 2001;1:34–45.

4. Grossmann ME, Huang H, Tindall DJ. Androgen receptorsignaling in androgen-refractory prostate cancer. J Natl CancerInst 2001;93:1687–1697.

5. Balmain A. Cancer as a complex genetic trait: Tumor suscept-ibility in humans and mouse models. Cell 2002;108:145–152.

6. Hansen K, Khanna C. Spontaneous and genetically engineeredanimalmodels; use inpreclinical cancerdrugdevelopment. Eur JCancer 2004;40:858–880.

7. Weiss B, Shannon K. Mouse cancer models as a platform forperforming preclinical therapeutic trials. Curr Opin Genet Dev2003;13:84–89.

8. Fresno Vara JA, Casado E, de Castro J, Cejas P, Belda-Iniesta C,Gonzalez-Baron M. PI3K/Akt signalling pathway and cancer.Cancer Treat Rev 2004;30:193–204.

9. Vazquez F, SellersWR. The PTEN tumor suppressor protein: Anantagonist of phosphoinositide 3-kinase signaling. BiochimBiophys Acta 2000;1470:21–35.

10. Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKTpathway in human cancer. Nat Rev Cancer 2002;2:489–501.

11. Lin A, Karin M. NF-kappaB in cancer: A marked target. SeminCancer Biol 2003;13:107–114.

12. Younes A, Garg A, Aggarwal BB. Nuclear transcriptionfactor-kappaB in Hodgkin’s disease. Leuk Lymphoma 2003;44:929–935.

13. Ondrey FG, DongG, Sunwoo J, Chen Z,Wolf JS, Crowl-BancroftCV, Mukaida N, Van Waes C. Constitutive activation oftranscription factors NF-(kappa)B, AP-1, and NF-IL6 in humanhead and neck squamous cell carcinoma cell lines that expresspro-inflammatory and pro-angiogenic cytokines. Mol Carcinog1999;26:119–129.

14. Mukhopadhyay T, Roth JA, Maxwell SA. Altered expression ofthe p50 subunit of the NF-kappa B transcription factor complexin non-small cell lung carcinoma. Oncogene 1995;11:999–1003.

15. Nair A, Venkatraman M, Maliekal TT, Nair B, Karunagaran D.NF-kappaB is constitutively activated in high-grade squamousintraepithelial lesions and squamous cell carcinomas of thehuman uterine cervix. Oncogene 2003;22:50–58.

16. Yu HG, Yu LL, Yang Y, Luo HS, Yu JP, Meier JJ, Schrader H,Bastian A, Schmidt WE, Schmitz F. Increased expression of

Fig. 11. Schematic representation for constitutive PI3K/Akt andNF-kB activation in the prostate of TRAMP mice. representsactivation. [Color figure can beviewed in the online issue, which isavailableatwww.interscience.wiley.com.]


RelA/nuclear factor-kappa B protein correlates with colorectaltumorigenesis. Oncology 2003;65:37–45.

17. Wang W, Abbruzzese JL, Evans DB, Larry L, Cleary KR, ChiaoPJ. The nuclear factor-kappa B RelA transcription factor isconstitutively activated in human pancreatic adenocarcinomacells. Clin Cancer Res 1999;5:119–127.

18. Visconti R, Cerutti J, Battista S, Fedele M, Trapasso F, Zeki K,MianoMP,deNigris F,CasalinoL,Curcio F, SantoroM,FuscoA.Expression of the neoplastic phenotype by human thyroidcarcinoma cell lines requires NFkappaB p65 protein expression.Oncogene 1997;15:1987–1994.

19. Biswas DK, Shi Q, Baily S, Strickland I, Ghosh S, Pardee AB,Iglehart JD. NF-kappa B activation in human breast cancerspecimens and its role in cell proliferation and apoptosis. ProcNatl Acad Sci USA 2004;101:10137–10142.

20. Costello R, Lipcey C. Involvement of the different members ofthe NF-kappaB transcription factor family in B-chronic lympho-cytic leukemia. Leukemia 1993;7:764–765.

21. Chauhan D, Uchiyama H, Akbarali Y, Urashima M, YamamotoK, Libermann TA, Anderson KC. Multiple myeloma celladhesion-induced interleukin-6 expression in bone marrowstromal cells involves activation of NF-kappa B. Blood 1996;87:1104–1112.

22. Yang J, Richmond A. Constitutive IkappaB kinase activitycorrelates with nuclear factor-kappaB activation in humanmelanoma cells. Cancer Res 2001;61:4901–4909.

23. Abdel-Latif MM, O’Riordan J, Windle HJ, Carton E, Ravi N,Kelleher D, Reynolds JV. NF-kappaB activation in esophagealadenocarcinoma: Relationship to Barrett’s metaplasia, survival,and response to neoadjuvant chemoradiotherapy. Ann Surg2004;239:491–500.

24. Lessard L,Mes-MassonAM, Lamarre L,Wall L, Lattouf JB, SaadF. NF-kappa B nuclear localization and its prognostic signifi-cance in prostate cancer. BJU Int 2003;91:417–420.

25. Shukla S,MacLennanGT, Fu P, Patel J,Marengo SR, ResnickMI,Gupta S. Nuclear factor-kappaB/p65 (Rel A) is constitutivelyactivated in human prostate adenocarcinoma and correlateswith disease progression. Neoplasia 2004;6:390–400.

26. Gilmore TD. The Re1/NF-kappa B/I kappa B signal transduc-tion pathway and cancer. Cancer Treat Res 2003;115:241–265.

27. Vermeulen L, De Wilde G, Notebaert S, Vanden Berghe W,Haegeman G. Regulation of the transcriptional activity of thenuclear factor-kappaB p65 subunit. BiochemPharmacol 2002;64:963–970.

28. Pahl HL. Activators and target genes of Rel/NF-kappaBtranscription factors. Oncogene 1999;18:6853–6866.

29. Suh J, Payvandi F, Edelstein LC, Amenta PS, Zong WX,Gelinas C, Rabson AB. Mechanisms of constitutive NF-kappaBactivation in human prostate cancer cells. Prostate 2002;52:183–200.

30. Gasparian AV, Yao YJ, Kowalczyk D, Lyakh LA, Karseladze A,Slaga TJ, Budunova IV. The role of IKK in constitutive activationof NF-kappaB transcription factor in prostate carcinoma cells.J Cell Sci 2002;115:141–151.

31. Gingrich JR, Barrios RJ, Morton RA, Boyce BF, DeMayo FJ,Finegold MJ, Angelopoulou R, Rosen JM, Greenberg NM.Metastatic prostate cancer in a transgenic mouse. Cancer Res1996;56:4096–4102.

32. Kaplan-Lefko PJ, Chen TM, Ittmann MM, Barrios RJ, Ayala GE,Huss WJ, Maddison LA, Foster BA, Greenberg NM. Pathobiol-ogy of autochthonous prostate cancer in a pre-clinical transgenicmouse model. Prostate 2003;55:219–237.

33. Gupta S, AdhamiVM, SubbarayanM,MacLennanGT, Lewin JS,Hafeli UO, Fu P, Mukhtar H. Suppression of prostate carcino-genesis by dietary supplementation of celecoxib in transgenicadenocarcinoma of the mouse prostate model. Cancer Res 2004;64:3334–3343.

34. Shukla S, Gupta S. Suppression of constitutive and tumornecrosis factor alpha-induced nuclear factor (NF)-kappaBactivation and induction of apoptosis by apigenin in humanprostate carcinoma PC-3 cells: Correlation with down-regula-tion of NF-kappaB-responsive genes. Clin Cancer Res 2004;10:3169–3178.

35. Miller SJ, Lou DY, Seldin DC, Lane WS, Neel BG. Directidentification of PTEN phosphorylation sites. FEBS Lett 2002;528:145–153.

36. Georgescu MM, Kirsch KH, Akagi T, Shishido T, Hanafusa H.The tumor-suppressor activity of PTEN is regulated by itscarboxyl-terminal region. Proc Natl Acad Sci USA 1999;96:10182–10187.

37. Kaplan PJ, Mohan S, Cohen P, Foster BA, Greenberg NM. Theinsulin-like growth factor axis andprostate cancer: Lessons fromthe transgenic adenocarcinoma of mouse prostate (TRAMP)model. Cancer Res 1999;59:2203–2209.

38. Polnaszek N, Kwabi-Addo B, Peterson LE, Ozen M, GreenbergNM, Ortega S, Basilico C, IttmannM. Fibroblast growth factor 2promotes tumor progression in an autochthonous mousemodelof prostate cancer. Cancer Res 2003;63:5754–5760.

39. Schatten H, Wiedemeier AM, Taylor M, Lubahn DB, GreenbergNM, Besch-Williford C, Rosenfeld CS, Day JK, Ripple M.Centrosome-centriole abnormalities are markers for abnormalcell divisions and cancer in the transgenic adenocarcinomamouse prostate (TRAMP) model. Biol Cell 2000;92:331–340.

40. Bruckheimer EM, Brisbay S, JohnsonDJ, Gingrich JR, GreenbergN, McDonnell TJ. Bcl-2 accelerates multistep prostate carcino-genesis in vivo. Oncogene. 2000;19:5251–5258.

41. Yang D, Holt GE, Velders MP, Kwon ED, Kast WM. Murinesix-transmembrane epithelial antigen of the prostate, prostatestem cell antigen, and prostate-specific membrane antigen:Prostate-specific cell-surface antigens highly expressed in pros-tate cancer of transgenic adenocarcinoma mouse prostate mice.Cancer Res 2001;61:5857–5860.

42. Maddison LA, Huss WJ, Barrios RM, Greenberg NM. Differ-ential expression of cell cycle regulatorymolecules and evidencefor a ‘‘cyclin switch’’ during progression of prostate cancer.Prostate. 2004;58:335–344.

43. Huss WJ, Hanrahan CF, Barrios RJ, Simons JW, Greenberg NM.Angiogenesis and prostate cancer: Identification of a molecularprogression switch. Cancer Res 2001;61:2736–2743.

44. Ebert AD, Wechselberger C, Frank S, Wallace-Jones B, Seno M,Martinez-Lacaci I, Bianco C, De Santis M, Weitzel HK, SalomonDS. Cripto-1 induces phosphatidylinositol 30-kinase-dependentphosphorylation of AKT and glycogen synthase kinase 3beta inhuman cervical carcinoma cells. Cancer Res 1999;59:4502–4505.

45. Philp AJ, Campbell IG, Leet C, Vincan E, Rockman SP, White-head RH, Thomas RJ, Phillips WA. The phosphatidylinositol 30-kinase p85alpha gene is an oncogene in human ovarian andcolon tumors. Cancer Res 2001;61:7426–7429.

46. Blume-Jensen P, Hunter T. Oncogenic kinase signalling. Nature2001;411:355–365.

47. Bellacosa A, de Feo D, Godwin AK, Bell DW, Cheng JQ,Altomare DA, Wan M, Dubeau L, Scambia G, Masciullo V.Molecular alterations of the AKT2 oncogene in ovarian andbreast carcinomas. Int J Cancer 1995;64:280–285.

238 Shukla et al.

48. Staal SP. Molecular cloning of the akt oncogene and its humanhomologues AKT1 and AKT2: Amplification of AKT1 in aprimary human gastric adenocarcinoma. Proc Natl Acad SciUSA 1987;84:5034–5037.

49. Nakatani K, ThompsonDA, Barthel A, SakaueH, LiuW,WeigelRJ, Roth RA. Up-regulation of Akt3 in estrogen receptor-deficient breast cancers and androgen-independent prostatecancer lines. J Biol Chem 1999;274:21528–21532.

50. Chang F, Lee JT,Navolanic PM, SteelmanLS, Shelton JG, BlalockWL, Franklin RA, McCubrey JA. Involvement of PI3K/Aktpathway in cell cycle progression, apoptosis, and neoplastictransformation: A target for cancer chemotherapy. Leukemia2003;17:590–603.

51. Shioi T,McMullen JR,KangPM,DouglasPS,ObataT, FrankeTF,Cantley LC, Izumo S. Akt/protein kinase B promotes organgrowth in transgenic mice. Mol Cell Biol 2002;22:2799–2809.

52. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J,MiliaresisC,RodgersL,McCombieR, Bigner SH,GiovanellaBC,Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R.PTEN, a putative protein tyrosine phosphatase gene mutated inhuman brain, breast, and prostate cancer. Science 1997;275:1943–1947.

53. WhangYE,WuX, SuzukiH, Reiter RE, TranC, Vessella RL, SaidJW, IsaacsWB, SawyersCL. Inactivationof the tumor suppressorPTEN/MMAC1 in advanced human prostate cancer throughloss of expression. Proc Natl Acad Sci USA 1998;95:5246–5250.

54. Vlietstra RJ, vanAlewijk DC,Hermans KG, van Steenbrugge GJ,Trapman J. Frequent inactivation of PTEN in prostate cancer celllines and xenografts. Cancer Res 1998;58:2720–2723.

55. Podsypanina K, Ellenson LH, Nemes A, Gu J, Tamura M,YamadaKM,Cordon-CardoC,CatorettiG, FisherPE, ParsonsR.Mutation of Pten/Mmac1 in mice causes neoplasia in multipleorgan systems. Proc Natl Acad Sci USA 1999;96:1563–1568.

56. Wang S, Gao J, Lei Q, Rozengurt N, Pritchard C, Jiao J, ThomasGV, Li G, Roy-Burman P, Nelson PS, Liu X, Wu H. Prostate-

specificdeletion of themurinePten tumor suppressor gene leadsto metastatic prostate cancer. Cancer Cell 2003;4:209–221.

57. Koul D, Shen R, Garyali A, Ke LD, Liu TJ, Yung WK. MMAC/PTEN tumor suppressor gene regulates vascular endothelialgrowth factor-mediated angiogenesis in prostate cancer. Int JOncol 2002;21:469–475.

58. Ali IU, Schriml LM, Dean M. Mutational spectra of PTEN/MMAC1 gene: A tumor suppressor with lipid phosphataseactivity. J Natl Cancer Inst 1999;91:1922–1932.

59. VazquezF,GrossmanSR,TakahashiY,RokasMV,NakamuraN,Sellers WR. Phosphorylation of the PTEN tail acts as aninhibitory switch by preventing its recruitment into a proteincomplex. J Biol Chem 2001;276:48627–48630.

60. Rayet B, Gelinas C. Aberrant rel/nfkb genes and activity inhuman cancer. Oncogene 1999;18:6938–6947.

61. Phi van L. Transcriptional activation of the chicken lysozymegene by NF-kappa Bp65 (RelA) and c-Rel, but not by NF-kappaBp50. Biochem J 1996;313:39–44.

62. Li Q, Verma IM. NF-kappaB regulation in the immune system.Nat Rev Immunol 2002;2:725–734.

63. Dhawan P, Singh AB, Ellis DL, Richmond A. Constitutiveactivation of Akt/protein kinase B in melanoma leads to up-regulation of nuclear factor-kappaB and tumor progression.Cancer Res 2002;62:7335–7342.

64. Liptay S,WeberCK, LudwigL,WagnerM,AdlerG, SchmidRM.Mitogenic and antiapoptotic role of constitutiveNF-kappaB/Relactivity in pancreatic cancer. Int J Cancer 2003;105:735–746.

65. Nakayama H, Ikebe T, Beppu M, Shirasuna K. High expressionlevels of nuclear factor kappaB, IkappaB kinase alpha and Aktkinase in squamous cell carcinoma of the oral cavity. Cancer2001;92:3037–3044.

66. Birkenkamp KU, Geugien M, Schepers H, Westra J, LemminkHH, Vellenga E. Constitutive NF-kappaB DNA-binding activityinAML is frequentlymediated by a Ras/PI3-K/PKB-dependentpathway. Leukemia 2004;18:103–112.


Documents

Constitutive activation of PI3K-Akt and NF-κB during prostate cancer progression in autochthonous transgenic mouse model