Available at www.sciencedirect.com journal homepage: www.elsevierhealth.com/journals/clnu ORIGINAL ARTICLE Lycopene differentially induces quiescence and apoptosis in androgen-responsive and -independent prostate cancer cell lines $ Nikita I. Ivanov a,1 , Simon P. Cowell a,1 , Paula Brown b , Paul S. Rennie a , Emma S. Guns a,2 , Michael E. Cox a, ,2 a Prostate Research Centre at Vancouver General Hospital, University of British Columbia, 2660 Oak Street, Vancouver, BC, Canada V6T 1Z4 b Herbal Evaluation and Analysis Laboratory, British Columbia Institute of Technology, Canada Received 25 July 2006; accepted 7 January 2007 KEYWORDS Lycopene; Prostate cancer; Cell-cycle; Insulin-like growth factor; PTEN; Retinoblastoma Summary Background & aims: Lycopene has been credited with a number of health benefits including a decrease in prostate cancer risk. Our study investigates the molecular mechanism underlying anti-cancer activity of lycopene-based products in androgen- responsive (LNCaP) and androgen-independent (PC3) cells. Methods: The effect of lycopene-based agents on prostate cancer growth and survival were examined using proliferation assays, bromodeoxyuridine incorporation and flow cytometric analysis of cellular DNA content. Biochemical effects of lycopene treatment were investigated by immunoblotting for changes in the absolute levels and phosphoryla- tion states of cell cycle regulatory and signalling proteins. Results: LNCaP and PC3 cells treated with the lycopene-based agents undergo mitotic arrest, accumulating in G0/G1 phase. Immunoblot screening indicated that lycopene’s antiproliferative effects are likely achieved through a block in G1/S transition mediated by decreased levels of cyclins D1 and E and cyclin dependent kinase 4 and suppressed ARTICLE IN PRESS 0261-5614/$ - see front matter & 2007 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2007.01.002 Abbreviations: IGF, insulin-like growth factor; IGF-BP IGF, binding protein; IGF-IR IGF, type 1 receptor; CDK, cyclin dependent kinase; HPLC, high performance liquid chromatography; SDS, sodium-dodecyl sulphate; PAGE, polyacrylamide gel electrophoresis; BrdU, bromodeoxyur- idine; BSA, bovine serum albumin; Rb, retinoblastoma. $ This work was supported in part by funding from the Prostate Cancer Research Foundation of Canada, the Canadian Prostate Cancer Bioresearch Network, Canadian Prostate Cancer Research Initiative, The Michael Smith Foundation for Health Research and the National Cancer Institute of Canada. Corresponding author. Tel.: +1 604 875 4818; fax: +1 604 875 5654. E-mail address: [email protected] (M.E. Cox). 1 These authors contributed equally to this work. 2 These authors co-directed this work. Clinical Nutrition (]]]]) ], ]]]–]]] Please cite this article as: Ivanov NI, et al. Lycopene differentially induces quiescence and apoptosis in androgen-responsive and - independent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/j.clnu.2007.01.002
Transcript
ARTICLE IN PRESS
Available at www.sciencedirect.com
Clinical Nutrition (]]]]) ], ]]]–]]]
0261-5614/$ - see frdoi:10.1016/j.clnu.
Abbreviations: IGhigh performance liidine; BSA, bovine s
$This work wasBioresearch NetworCancer Institute of�Corresponding aE-mail address:
Lycopene differentially induces quiescence andapoptosis in androgen-responsive and -independentprostate cancer cell lines$
Nikita I. Ivanova,1, Simon P. Cowella,1, Paula Brownb, Paul S. Renniea,Emma S. Gunsa,2, Michael E. Coxa,�,2
aProstate Research Centre at Vancouver General Hospital, University of British Columbia, 2660 Oak Street,Vancouver, BC, Canada V6T 1Z4bHerbal Evaluation and Analysis Laboratory, British Columbia Institute of Technology, Canada
SummaryBackground & aims: Lycopene has been credited with a number of health benefitsincluding a decrease in prostate cancer risk. Our study investigates the molecularmechanism underlying anti-cancer activity of lycopene-based products in androgen-responsive (LNCaP) and androgen-independent (PC3) cells.Methods: The effect of lycopene-based agents on prostate cancer growth and survivalwere examined using proliferation assays, bromodeoxyuridine incorporation and flowcytometric analysis of cellular DNA content. Biochemical effects of lycopene treatmentwere investigated by immunoblotting for changes in the absolute levels and phosphoryla-tion states of cell cycle regulatory and signalling proteins.Results: LNCaP and PC3 cells treated with the lycopene-based agents undergo mitoticarrest, accumulating in G0/G1 phase. Immunoblot screening indicated that lycopene’santiproliferative effects are likely achieved through a block in G1/S transition mediated bydecreased levels of cyclins D1 and E and cyclin dependent kinase 4 and suppressed
Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
h factor; IGF-BP IGF, binding protein; IGF-IR IGF, type 1 receptor; CDK, cyclin dependent kinase; HPLC,hy; SDS, sodium-dodecyl sulphate; PAGE, polyacrylamide gel electrophoresis; BrdU, bromodeoxyur-retinoblastoma.y funding from the Prostate Cancer Research Foundation of Canada, the Canadian Prostate Cancerte Cancer Research Initiative, The Michael Smith Foundation for Health Research and the National
Please cite this article as: Ivanovindependent prostate cancer cell lin
Retinoblastoma phosphorylation. These responses correlated with decreased insulin-likegrowth factor-I receptor expression and activation, increased insulin-like growth factorbinding protein 2 expression and decreased AKT activation. Exposure to lycopene at dosesas low as 10 nM for 48 h induced a profound apoptotic response in LNCaP cells. In contrastPC3 cells were resistant to apoptosis at doses up to 1 mM.Conclusions: Lycopene exposure can suppress phosphatidylinositol 3-kinase-dependentproliferative and survival signalling in androgen-responsive LNCaP and androgen-independent PC3 cells suggesting that the molecular mechanisms for the cytostatic andcytotoxic actions of lycopene involve induction of G0/G1 cell cycle arrest. This studysupports further examination of lycopene as a potential agent for both the prevention andtreatment of prostate cancer.& 2007 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rightsreserved.
Introduction
Therapeutic options for patients with advanced prostatecancer (PCa) are limited, consequently there is a need toexplore new therapies and pursue the biochemical mechan-isms underlying positive pre-clinical and clinical data. Lackof curative therapies and their potential for adverse effectsare factors that have led many men to turn to a potpourri ofherbal preparations and dietary supplements in the hopethat these will help them prevent or overcome PCa.1 Thesepreparations and supplements cover a range of compoundclasses including vitamins, (e.g., vitamin E), minerals, (e.g.,selenium), herbal preparations, (e.g., Saw Palmetto) andphytochemicals, (e.g., lycopene).
Lycopene (c, c-carotene) is an unsaturated carotenoidwith a total of 13 double bonds, 11 of which are conjugated,thus contributing to the compound’s potent antioxidantproperties. Studies in vitro have shown lycopene to be twiceas potent as b-carotene and ten times more potent than a-tocopherol in terms of its singlet oxygen quenching ability.2
Serum lycopene levels were shown to increase significantlyinto the mM range after the consumption of tomato productsand supplements, with a concomitant decrease in thebiomarkers of oxidation including the oxidation of serumlipids, low density lipoprotein cholesterol, serum proteinsand DNA.3
The beneficial properties of tomato products were largelyattributed to lycopene by one of the earliest epidemiologi-cal studies showing an inverse relationship between theconsumption of tomatoes and/or tomato products and PCarisk.4 Over the last decade numerous epidemiological,experimental and tissue culture studies and reviews havebeen published which describe an association betweenlycopene supplementation or a tomato rich diet anddecreased positive markers for PCa.5–7 In addition threephase II clinical studies have been reported recently, twoconducted in India which report favorable outcomesregarding PSA decline in patients with either metastaticdisease (10mg Lycored softules daily for 3 months) or whowere undergoing concurrent orchiectomy (2mg twice dailyfor 3 months); and one dose escalation study (15–120mg/kgdaily for 1 yr) conducted in the US in patients with recurrentdisease which concluded no significant impact of Lyc-O-MatoTM (6% lycopene) on PSA.8–10 Further clinical study isrequired to establish whether the timing of lycopene
NI, et al. Lycopene differentiallyes. Clin Nutr (2007), doi:10.1016
intervention and/or combination with hormone withdrawalare important factors regarding outcome, as well as doseand formulation of lycopene. Much of the literature makesuse of uncharacterized lycopene containing tomato extractsrather than pure lycopene, therefore, perhaps wronglyassuming consistent lycopene stability while also honing inon lycopene as the active ingredient.11 Due to its highlyunsaturated anti-oxidant properties lycopene readily under-goes trans- to cis-isomerization when heated and whenexposed to oxygen, especially if present in solid orpowdered form.12 In order to overcome these stabilityissues, we have used two lycopene-based products, both ofwhich have now been characterized through independentquality control testing and contain 3% and 38% lycopeneby weight.
Beyond its antioxidant properties, the biological actionsof lycopene are not fully delineated, although several cellculture studies suggest effects on cell cycle progression.Lycopene has been indicated to have antiproliferativeeffects on prostate13–16 and breast cancer cell lines.17,18 Inbreast cancer models, reduced expression of cell cycleregulatory proteins, such as cyclins D1 and E and the cyclindependent kinases 2 and 4, as well as suppression of insulin-like growth factor (IGF-I) action have been correlated withlycopene’s effects on proliferation.17–20
Both of the cell lines used in this study lack the tumorsuppressor protein, phosphate and tensin homologue(PTEN). PTEN is among the most commonly mutated genesin epithelial cancers21 and its function is lost in the majorityof lethal PCas.22 PTEN is a phospholipid phosphatase thatdephosphorylates the 30 position of phosphatidylinositol3,4,5-trisphosphate (PIP3), thus diminishing a major sub-strate for plextrin homology domain-containing kinasesincluding AKT/PKB, PDK1 and PDK2/ILK, all of which serveas positive effectors of phosphatidylinositol 3-kinase (PI3 K)signalling.23 Pro-tumorigenic activity results from thedysregulation of these downstream signalling events whichpromote tumor cell growth and elicit protection fromcytotoxic therapies.
It is therefore important to identify agents that canovercome the therapeutic resistant properties of PTENdeficient tumor cells. With this study we investigate themolecular basis for lycopene’s antiproliferative action in twoPTEN-null PCa cell lines through use of two characterizedlycopene-based products. We show that lycopene-induced
induces quiescence and apoptosis in androgen-responsive and -/j.clnu.2007.01.002
Lycopene-induced cell cycle arrest and death of prostate cancer cells 3
G1/S phase mitotic arrest in PCa is associated withdecreased IGF-1R activation and PI3 K signalling and thatthis differentially induces quiescence or apoptosis in thesePTEN-null PCa models.
Materials and methods
Lycopene-based products
Lyophilized, water dispersible, tomato powder containing3% lycopene by weight (LycopenTM) was obtained fromLycoRed Natural Products (Beer-Sheva, Israel). A secondtomato extract was prepared for this study by WatersSolutions (Caldwell, ID) in a proprietary process thatproduces a stable and concentrated lycopene liquid extract,LycoTrueTM. A 90% lycopene reference material was pur-chased from LKT (St. Paul, MN) for the purpose ofconducting high performance liquid chromatography (HPLC)quantitation of the LycopenTM and LycoTrueTM material.
Lycopene preparation
LycopenTM was dissolved in cell culture media containing10% fetal bovine serum prewarmed to 37 1C to yield asolution of 200 mM lycopene (3.579mg/ml). The mixture wasvortexed and sonicated before being filter sterilized. Theconcentration stated on the results represents the theore-tical lycopene concentration. LycoTrueTM was supplied as an8mM solution in tetrahydrofuran (THF) containing 0.05%butylated hydroxytoluene (BHT). This was serially dilutedinto media to achieve the working concentration range.
Product characterization by HPLC
To determine chromatographic purity, LycoTrueTM was diluted1/10 in the THF/BHT solvent and immediately analyzed byHPLC using a C18 column, isocratic conditions and a mobilephase containing 95:5 methanol: methyl tertiary butyl ether(MTBE). Its chromatographic purity relative to other carote-noids at 502nm (16nm bandwidth) was determined to be92.4%. The actual concentration of lycopene in the crudematerial was calculated to be 37.8% w/w once corrected forthe non-ionic low molecular weight surfactant used as astabilizer. A lycopene calibration standard was obtained fromLKT (Catalog] L9609, lot] 23920304).
Cell lines and culture
LNCaP and PC3 cells24 were maintained in RPMI 1640 or DMEMdefined medium (GibcoBRL, Burlington, ON, Canada) supple-mented with 10% fetal bovine serum (FBS) in a 37 1C, 5% CO2
atmosphere, unless otherwise stated. For the various studiescells were plated at 40% confluency, treated at around 50–60%and typically harvested between 70% and 90% confluency.
Crystal violet assay
Cell proliferation was assessed using crystal violet assay.25
Cells were plated on 96-well plates and treated withLycopenTM at lycopene concentrations of 0.1–10 mM for
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/
24 h before fixing, washing and staining with crystal violet.After washing off excess stain, cell associated dye wasreleased by solubilizing the cells and the optical density wasmeasured at 592 nm and viable cell density was expressed asa percentage relative to vehicle treated cells.
MTS assay
The CellTiter 96s AQueous One Solution Cell ProliferationAssay (Promega, Madison WI) was used as per the manu-facturer’s instruction. This MTS {3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tet-razolium}-based assay relies upon the catalytic activity ofviable cells to produce a formazan dye that can bequantified to yield a measure of relative cell number aftera period of treatment. Cells were plated at 4000 per well in96 well plates and incubated for 24 h prior to treatment withLycoTrueTM at lycopene concentrations of 0.01–10 mM, orTHF solvent control for 24, 48 or 72 h. After densitometricanalysis, viable cell density was expressed as a percentageof the respective vehicle treated cells.
LNCaP cells grown on 24 well plates were transfected usinglipofectin (GIBCO BRL) with the androgen responsive fireflyreporter gene construct ARR3-luc and pRLTK (Promega) as atransfection control.26 Transfected cells were treated with aconcentration range of 0–100 mM lycopene in LycopenTM
containing media for 48 h before cells were harvested andlysed with passive lysis buffer (Promega). Lysates wereassayed using dual-luciferase reagents (Promega) in aBerthold plate luminometer. Androgen responsive fireflyluciferase values were normalized using the renilla lucifer-ase response to account for lycopene cytotoxicity andexpressed relative to vehicle treated controls.
Immunoblotting
A protein screen was performed by Kinexus (KinexusBioinformatics Corporation, Vancouver, Canada) using Ki-networkss immunoblot technologies. The relative levels ofphosphorylated forms of 50 different cell signaling andregulatory proteins were determined for LNCaP cells treatedwith 0.1 mM lycopene in LycopenTM or vehicle for 24 h. Theresults of this screen were used as a guide for subsequentvalidation by immunoblot.
Proteins from whole cell lysates of LNCaP and PC3 cellswere separated by SDS–PAGE, transferred to nitrocellulosemembranes, blocked with 5% (w/v) skim milk in 20mM Tris,pH 7.4, 0.05% triton X-100 (TBS-T) for 1 h at roomtemperature and probed by incubation overnight at 4 1Cwith primary antibody in TBS-T with 1% BSA. After washingwith TBS-T membranes were incubated with 1:5000 solutionof appropriate secondary anti-IgG antibody (DAKO, Glostrup,Denmark) for 1 h at room temperature, rewashed in TBS-T,and bands were detected using ECLTM reagent (AmershamPharmacia, Piscataway, NJ) or a LI-COR Odyssey fluores-cence imager.
induces quiescence and apoptosis in androgen-responsive and -j.clnu.2007.01.002
Antibodies used were specific for total IGF-IRb, total Rb,cyclin D1, cyclin E, cdk 2 and cdk 4 from Santa Cruzbiotechnologies (Santa Cruz, CA), cdk 1, cdk 5, MAPK andIGFBP2 from Upstate Biotechnologies (Lake Placid, NY) andsite specific phosphorylated forms of Rb [i.e., at S780 ]9307,S795 ]9301, S807/811 ]9308] total or phosphorylated AKT,and Phospho-IGF-IR (Tyr1131)/ Insulin receptor (Tyr1146)were from Cell Signaling Technologies (Beverly, MA), BrdU(Roche, Nutley, NJ) and vinculin (Sigma, St. Louis, MO).Immunoblots shown are representative of three or morederived from independent experiments. Immunoblot bandintensities on radiographic film were quantified using aMolecular Dynamics Typhoon 9410 scanner and Imagequant5.0 software. Representative immunoblots of experimentsindependently performed at least 3 times are presented.
Bromodeoxyuridine (BrdU) incorporation assay
The mitotic index of LNCaP of PC3 cells grown on glass coverslips to 60% confluence was measured. After treating withLycoTrueTM at the indicated lycopene concentrations for 48 hthe cells were pulsed with 2 ng/ml BrdU for 2.5 h beforeprocessing for immunocytochemical analysis as described.27
Mitotic index was calculated as the % BrdU-positive cellsamongst at least 200 individual cells examined per treat-ment from 6 independent experiments.
Flow cytometry
LNCaP or PC3 cells were plated onto 100mm diameter tissueculture plates in RPMI or DMEM media, respectively, containing10% FBS. The cells were incubated at 37 1C in 5% CO2 for 24h toallow adherence to the plates. The media and non-adherentcells were aspirated from the plates. The cells were treated intriplicate with lycopene or control treatment as indicated. Thecells were incubated for 48h at 37 1C in a 5% CO2 atmospherefollowing which they were harvested by the addition of 5mMEDTA and gently washed off the plate. The cells were pelletedalong with the previously collected media. Cell pellets werefixed with 70% ethanol for 30min on ice and than rinsed threetimes with 1ml of 0.1% glucose in PBS (20mM NaH2PO4,150mM NaCl), repelleted and resuspended in propidiumiodide (PI) staining solution (10ml of 10mg/ml RNase A, 5mlof 10mg/ml PI per 1ml of PBS with 0.1% glucose). After 30minthe cells were analyzed using a Coulter Epics xl flowcytometer.
Statistical analysis
Unless otherwise stated, all values were presented as means7 SEM. Statistical significance was evaluated using theStudent’s t-test for paired comparison; Po0.05 was con-sidered significant.
Results
HPLC analysis of LycopenTM versus LycoTrueTM
quality and purity
Lycopene formulations and deliverable doses have been aconstant point of controversy.11 LycopenTM in media was
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016
assayed at the time of use by 5-point calibration with alycopene calibration standard obtained from LKT that was atleast 90% pure (as verified using HPLC profile). It was notedthat the activity in the LycopenTM material was unstable anddisappeared rapidly from the media. HPLC analysis of theLycopenTM in media over time revealed that the lycopenecontent was decreasing upon storage and that the lycopenecontent of the powdered LycopenTM material was alsodecreasing albeit more gradually than the media prepara-tions. This lycopene decomposition correlated with anobserved decrease in the in vitro efficacy of both storedand freshly prepared LycopenTM in media over time.Consequently, we investigated LycoTrueTM as an alternativesource for lycopene. LycoTrueTM is prepared by WatersSolutions via a proprietary process which allows them toextract high concentrations of lycopene into organic solventor oil and maintain stability for at least 6 months. Ouranalysis indicates that this preparation has improvedstability compared with LycopenTM both as stock solutionin THF/BHT solvent and in media. The correlation betweenloss of lycopene from the LycopenTM preparation and loss ofefficacy in vitro provides further evidence that lycopenerepresents the active ingredient in our assays. The potentantioxidant properties of BHTensure longer shelf life for theLycoTrueTM preparations.
Water dispersible LycopenTM causes growth arrestin prostate cells
The reported ability of lycopene to arrest the growth ofnumerous cell lines including those from breast, endome-trial and PCa18,19,28,29 inspired this investigation into themolecular mechanism of lycopene treatment in PCa cellmodel systems. We initially studied the effects of Lyco-penTM, a commercial dietary supplement containing 3%lycopene (w/w), on the growth of the androgen responsivecell line, LNCaP, using crystal violet staining, which reflectsthe net outcome of treatment on proliferative activityversus cytotoxic effects (Fig. 1(A)). This lycopene containingmixture produced a significant decrease in cell proliferationat a concentration of 1.0 mM lycopene with an observedreduction in cell number reduced by more than 80% at 10 mMafter 24 h.
LycoTrueTM mimics the effects of LycopenTM inprostate cells
To test the antiproliferative activity of LycoTrueTM we usedthe MTS quantitative cell staining method. After treatmentfor 48 h or longer a significant decrease in cell number wasobserved in LNCaP cells (Fig. 1(B)). Significant suppressionof cell proliferation was observed at concentrationsX500 nM at time points beyond 48 h, but no effect oftreatment was seen in cells of either line treated withlycopene at concentrations below 250 nM or after only 24 hexposure, as compared with the THF/BHT control treat-ment. We also investigated whether these effects could beduplicated in an androgen independent PCa cell line. Theresponse of PC3 cells to lycopene was similar to that ofLNCaP cells with half maximal growth suppression occurringat 750 nM measured 48 h after treatment and maximal
induces quiescence and apoptosis in androgen-responsive and -/j.clnu.2007.01.002
Figure 1 (A) LycopenTM treatment reduces proliferation ofLNCaP cells. LNCaP cells were treated with LycopenTM atlycopene concentrations corresponding to 0.01–10 mM for 24 hand Crystal Violet stained as described in Materials andMethods. Data is expressed as the change in viability relativeto vehicle treated control cells (% control viability). *Denotes astatistically significant (Po0.05) decrease in cell number aftertreatment. (B) LycoTrueTM treatment reduces cell proliferationin both LNCaP (solid line) and PC3 cells (dashed line). Cells weretreated with LycoTrueTM at lycopene concentrations corre-sponding to 0.01–10 mM for 48 h before being assayed by MTS asdescribed in Materials and Methods. Data is expressed as thechange in viability relative to vehicle treated control cells (%control viability). *Denotes a statistically significant differencebetween treatment and vehicle control (Po0.05).
Lycopene-induced cell cycle arrest and death of prostate cancer cells 5
suppression occurring at 2.5 mM. However, maximal growthsuppression of PC3 cells (25–30% of mock treated PC3 cells)was less that that observed in LNCaP cells (10% of mocktreated LNCaP cells). While the LNCaP cell density 48 h aftertreatment with X2.5mM was less than the seeded celldensity, the final PC3 cell density remained unchanged. Thisobservation suggested that lycopene treatment may becytotoxic to LNCaP cells but only cytostatic to PC3 cells.
Mitotic and apoptotic responses of LNCaP cells tolycopene treatment
The MTS assay indicated that lycopene was a potentinhibitor of proliferation on both PC3 and LNCaP cells, and
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/
also suggested that lycopene was cytotoxic to LNCaP cells whilebeing cytostatic to PC3 cells. In order to examine the directeffect of lycopene on mitotic and apoptotic indexes in thesecell lines, we used PI staining flow cytometry to determine thecellular DNA content and cell cycle stage distribution of the cellpopulations after treatment; in conjunction with BrdU incor-poration into DNA to determine the rate of cell entry intoS-phase. Since the LycoTrueTM offered both a more concen-trated and stable source of lycopene, these experiments wereperformed with this formulation.
By plotting the ratio of the cells in G0+G1 against cells inG2/M+S-phase it is apparent that LNCaP cells undergo G0 orG1-phase cell cycle arrest at all lycopene concentrationstested (Fig. 2(A)). Evidence of G0/G1 arrest was alsoobserved in PC3 cells, but was not significantly differentfrom untreated controls at lycopene doses o400 nM and wasstill less than 2-fold even at 800 nM. However, from theoriginal fluorescence intensity traces (not shown) itappeared that lycopene treatment was dose-dependentlydecreasing the rate cells entered S-phase in both cell lines.We therefore used BrdU incorporation into treated cells todirectly monitor the effect of lycopene treatment on ratesof entry into S-phase, (Fig. 2(B)). In both cell lines, lycopenetreatment significantly lowered de novo DNA synthesis ratesin a dose-dependent fashion. LNCaP cells appeared moresensitive with suppression of mitotic index first evident at80 nM lycopene, significantly different at 4200 nM and amaximal (approximately 10% of control cells) at 400 nM. InPC3 cells, mitotic index suppression was first significantlydifferent from control cells at 200 nM and was maximallysuppressed (to 20% of control cells) at 800 nM. These resultsconfirm that decreased proliferation resulting from lyco-pene treatment is due to suppression of mitotic activity inLNCaP and PC3 proliferation.
To address the differences between the mitotic ratio(G1+G0: S+G2/M) and mitotic index (BrdU) responses ofLNCaP and PC3 cells, the flow data was analyzed todetermine whether cells were becoming senescent orundergoing apoptosis in response to lycopene treatment bycalculating the percentage of the cells that contained lessthan G0/G1 DNA content (Fig. 2(C)). LNCaP cells werereadily induced to undergo apoptosis in response tolycopene. A greater than 2-fold increase in apoptotic indexover controls was observed at 80 nM and saturated at400 nM. This response directly mirrored the ability oflycopene to induce cell cycle arrest in these cells. Weconclude that this combination of reduced mitotic activityand induction of apoptosis accounts for the observedcytotoxic effect of lycopene on LNCaP cell populations,even when cultured under ideal media conditions. Incontrast, no change in basal apoptotic index was observedin PC3 cells at lycopene concentrations of up to 800 nM.These results indicate that the cytostatic effect of lycopeneon PC3 cell populations is due to suppression of mitoticactivity alone.
Figure 2 Lycopene differentially affects mitosis and apoptosis in LNCaP and PC3 cells. Untreated cells or cells treated withLycoTrueTM at lycopene concentrations of 80–800 nM or THF/BHT vehicle in media for 48 h were harvested, stained with PI andanalyzed by flow cytometry. (A) The number of cells in each phase of the profile was determined and ratios of cells in resting phase(G0/G1) versus those undergoing mitosis (S, G2/M) were plotted for LNCaP and PC3 cells. (B) Cells treated as in (A) were pulse-labeled with BrdU, fixed and stained with anti-BrdU antibody for immunofluorescence imaging. The number of BrdU positive cellswere scored manually and plotted as percent of total cells. (C) Apoptotic index was determined by calculating the percentage ofcells exhibitingoG0/G1 DNA content by PI staining flow cytometry. *Denotes a statistically significant difference between treatmentand vehicle control (Po0.05).
N.I. Ivanov et al.6
Please cite this article as: Ivanov NI, et al. Lycopene differentially induces quiescence and apoptosis in androgen-responsive and -independent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/j.clnu.2007.01.002
Lycopene-induced cell cycle arrest and death of prostate cancer cells 7
exhibit clear evidence of mitotic arrest upon treatment.Inhibition of prostate cell proliferation by lycopene has beenreported by other investigators, in both benign30 andmetastatic19 cancer cells, however, the underlying mechan-ism for this effect has yet to be elucidated. To investigatewhether lycopene treatment interfered with proliferativesignaling, LNCaP cells treated with 100 nM lycopene fromLycopenTM in media containing 10% FBS for 48 h werescreened for the relative levels (50 proteins) and thephosphorylation state (28 protein phospho sites) of cellsignaling and regulatory proteins using Kinexus Kinetworkss
assays. These screens indicated that the LycopenTM treat-ment was affecting a number of mitogenic signaling path-ways. Most notable was a 5- to 10-fold decrease inexpression of the cyclin-dependent kinases (cdks) 1, 4, 5and 7 and a �5-fold suppression in Rb phosphorylation atS807/S81 and �2-fold suppression at S780 (Tables 1 and 2).These results suggested that lycopene treatment wasdirectly interfering with growth signaling events that controlG1/S phase cell cycle transition.
Effects of lycopene treatment on insulin-likegrowth factor axis components expression
IGFs, their type I receptor tyrosine kinase receptor (IGF-1R)and their binding proteins (IGFBPs) are potent promoters ofcell proliferation and have been correlated with PCa risk anddisease progression.31–38 Primary down-stream signalling
Table 1 Blot intenstities for Kinexus Kinetworkss assays of LNmedia containing 10% FBS for 48 h screened for the relative leve
Table 2 Blot intensities for Kinexus Kinetworkss assays of LNmedia containing 10% FBS for 48 h screened for the phosphorylat
Full name of protein Site of phosphorylation
Cyclin-dependent kinase 1 (cdc2) Y15Protein kinase B alpha (Akt1) T308Protein kinase B alpha (Akt1) S473Retinoblastoma 1 S807/S811Retinoblastoma 1 S780
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/
from IGF-1R includes activation of the Ras/MAPK andphosphatidylinositol 3-kinase (PI3 K)/AKT growth and survi-val pathways. In the Kinexus screens, we saw evidence thatlycopene treatment resulted in decreased expression ofERK1 MAPK and its upstream MAPK kinase kinase, BRaf1. Wealso detected evidence of suppression of the parallel MAPKpathways, p38Hog1 and JNK and their MAPK kinases MEK2,MKK4 and 6. We also say evidence of partial suppression ofPI3K signalling as measured by decreased levels of activationsite phosphorylations on AKT and decreased expression of thedown-stream ribosomal S6 kinase effectors, RSK1 and 2.
Since we have previously characterized IGF axis compo-nent expression and activation as important contributors toLNCaP growth and survival signalling33,39, and our initialstudies above indicated that LNCaP cells underwent cellcycle arrest and induction of apoptosis when treated withlycopene, we investigated whether lycopene-inducedchanges in cell cycle regulation implicated by the Kinexusscreen were associated with alterations in this growth factoraxis (Fig. 3). Treatment with LycopenTM at doses equivalentto 500 nM lycopene for 48 h showed suppressed IGF-1Rexpression and corresponding IGF-I-mediated activation inLNCaP cells cultured in the presence or absence ofandrogens. Previous studies have indicated that in LNCaPcells, IGFBP2 is the predominant expressed binding proteinfor IGF-I.33 Exposure of LNCaP cells to LycopenTM asdescribed above was found to increase IGFBP2 expressionapproximately 10-fold in both the presence and absence ofandrogen.
CaP cells treated with 100 nM lycopene from LycopenTM inls of cell signalling and regulatory proteins.
Figure 3 LycopenTM treatment alters IGF axis component expression and activation in LNCaP cells. The effects on the expressionand activation of IGF-IR and expression of IGFBP2 in LNCaP cells was assessed in cells treated with LycopenTM, androgen, and/or IGF-I. Cells were treated for 48 h with a 500 nM lycopene equivalent71 nM synthetic androgen, R1881. IGF-I (100 mg/ml) was added to theindicated samples 10min prior to harvest. For the upper 2 panels IGF-1R was immunoprecipitate from treated cell lysates was probedwith polyclonal antibody against total IGF-IRb-subunit (top panel) or against IGF-IRb phosphorylated at tyrosine 1131 (second panel),respectively. For the lower 2 panels, immunoblots of cell lysates were probed with polyclonal antibody against IGFBP2 (third panel)or MAPK (fourth panel; used as a loading control). The numbers indicate the relative densities of the IGFBP2 bands compared theMAPK control as calculated by densitometric scanning.
Exposure duration 1 Hour 4 Hours
Lycopene, µµM 0 00.01 0.10 1 0.01 0.10 1
Total Rb
PS795 Rb
PS780 Rb
Figure 4 LycoTrueTM treatment alters Rb phosphorylation inLNCaP cells. The effect of lycopene treatment on thephosphorylation of Rb in LNCaP cells was performed on cellstreated for 1 or 4 h with LycoTrueTM at lycopene concentrationsof 0.01–1 mM before lysis and immunoblot analysis. Total celllysates were probed with antibodies generated against total Rb(lower panel) or one of two different serine phosphorylation
N.I. Ivanov et al.8
Effects of lycopene treatment on androgen signaling
These changes in IGF-1R and IGFBP-2 expression areconsistent with our previous observation of adaptiveresponses of LNCaP cells to androgen withdrawal leadingus to postulate that the effects of lycopene on LNCaP cellsmight be mediated through disruption of androgen receptor(AR) activation. Using LNCaP cells transiently transfectedwith a firefly luciferase reporter gene under the control ofan AR responsive element and treated with varyingconcentrations of lycopene and the synthetic androgen,R1881, for 16–24 h (before overt lycopene cytotoxicity wasobserved) we observed no effect of lycopene treatment onandrogen-stimulated expression of firefly luciferase (datanot shown). We conclude that the cytotoxic effect oflycopene in LNCaP cells is not a consequence of interferencewith AR activity.
sites (S780, upper panel; S795, middle panel) of Rb.
Effects of lycopene on G1/S cell cycle progression
We therefore turned to characterization of the effect oflycopene treatment on direct regulators of cell cycleprogression indicated by the Kinexus screen. Rb contains16 consensus sites for phosphorylation40 and we examinedthe modification of two key serine residues known to bealtered during cell cycle progression in more detail (Fig. 4).LNCaP cells treated for 1 and 4 h with LycoTrueTM at finallycopene concentrations of between 10 nM and 1mM andexamined for changes in Rb phosphorylation status relativeto total cellular level. While no change in relative Rbphosphorylation was observed after 1 h, Rb Serine 780phosphorylation was suppressed in cells treated with 0.1and 1 mM lycopene after 4 h. Rb Serine 795 phosphorylation
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016
was also decreased at the highest dose (1 mM) 4 h afterlycopene treatment. Total cellular Rb was decreasedapproximately 2-fold at 1 h after 10 mM lycopene treatmentand at 4 h after 0.1 and 1 mM lycopene consistent withapoptotic induction.41 This observation indicates that inLNCaP cells induction of G1/S-phase cell cycle arrest ismediated through decreased Rb phosphorylation in responseto LycoTrueTM exposure. Together with the previous resultsdemonstrating cytostatic effect of LycoTrueTM on PC3 cells,these observations suggest that lycopene may directlysuppress growth factor mediated signaling independent ofandrogen sensitivity of prostate cancer cells.
Rb is in turn phosphorylated by cdks 2, 4 and 6 in complexwith their conjugate activating cyclins.42 Among other sites
induces quiescence and apoptosis in androgen-responsive and -/j.clnu.2007.01.002
Lycopene-induced cell cycle arrest and death of prostate cancer cells 9
Rb Serine 780 has been identified as being preferentiallyphosphorylated by cdk4 in complex with its associatedcyclins D1, D2 and D343,44 and activation of cdk 4 in complexwith cyclin D1 is associated with initiation of S-phaseprogression.45 Since growth of both LNCaP and PC3 cells issuppressed by lycopene treatment, we investigated whetherlycopene treatment affected expression levels of thesemitogenic regulators similarly in these two cell lines(Fig. 5(A)). Normalized to expression of MAPK, immunoblotanalysis indicated that cyclin D1 expression was similarlyreduced in LNCaP and PC3 cells treated for 48 h cells withlycopene at 200 nM and was essentially undetectable atdoses X400 nM. The cdk 2-cyclin E complex has also beenshown to be able to phosphorylate Rb and drive cell cycleprogression from G1 into S phase46,47 and expression levelsof cyclin E and cdk 2 were also substantially decreased incells treated with 4200 nM lycopene for 48 h. Expression ofcyclin D3, cdk 1, cdk 4 and cdk 5 was found to be unaffectedby lycopene treatment at concentrations of up to 1 mM inboth cell lines (data not shown). These observations indicatethat lycopene induces cell cycle arrest by altering expres-sion and activation of the same key G1/S phase cell cycleregulatory factors in both androgen-responsive and -inde-pendent PCa cells.
Effects of lycopene treatment on IGF-1R and AKT
Since the effect of LycopenTM treatment on IGF-1R andIGFBP2 expression in LNCaP cells did not correlate with achange in AR activation, yet both cells exhibited similargrowth arrest and decreased cyclin/cdk expression profiles,we next asked whether lycopene similarly altered IGF-1Rexpression and activity in PC3 cells (Fig. 5(B)). LNCaP andPC3 cells were treated with LycoTrueTM to achieve lycopeneconcentrations of 80–800 nM for 48 h. As shown above usinglycopene derived from LycopenTM, LycoTrueTM sourcedlycopene exposure caused a dose dependent decrease inIGF-IR protein level in both LNCaP and PC3 cells at lycopeneconcentrations 4200 nM. Further investigation revealedthat at these concentrations, lycopene also reduced thesteady-state phosphorylation status of AKT, a prominentdownstream target of the IGF pathway in both prostate celllines that is basally hyperphosphorylated due to the lack ofPTEN expression in these cell lines.23 These observationsindicate that lycopene treatment decreases IGF-1R expres-sion in both androgen-responsive and -independent PCa cellsand suggests that disruption of this crucial growth factorpathway is important in mediating lycopene-induced mitoticarrest.
Discussion
The mechanism delineated in this study suggests a realtherapeutic potential of tomato products for PCa patients.Both the 3% tomato preparation and the 38% lycopenepurified from tomato extract possessed antiproliferativeactivity in both androgen-responsive and -independent PCacells via down-regulation of IGF-1R expression and signaltransduction resulting in decreased expression of cyclin/cdkcomplexes critical for phosphorylation of Rb. The resultwould then be stabilization of Rb in complex with E2F
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/
impeding S-phase cell cycle progression. Furthermore incertain cellular backgrounds, exemplified here by theandrogen-responsive LNCaP cell, lycopene induces apoptosisin addition to its anti-mitotic effect, while other cell lines,exemplified by the androgen-independent PC3 cell, exhibitresistance to apoptosis.
We have previously demonstrated that re-expression ofPTEN in LNCaP and PC3 cells promotes mitotic arrest andinduction of apoptosis through an autocrine Fas/FasL-mediated cell death pathway.48 However, those experimentswere performed under serum-free conditions. Here wedemonstrate that lycopene treatment mimics the effect ofPTEN re-expression by suppressing activation of AKT andexpression of down-stream mitogenic targets such as cyclinD1 in the presence of FBS-containing media. The differentialapoptotic response of LNCaP and PC3 cells suggests thatblocking this pathway is sufficient to allow autocrineapoptotic induction in LNCaP cells but that other pro-survival signalling events stimulated by serum factor arecapable of protecting PC3 cells under these cultureconditions.
We have observed an apparent interaction with compo-nents of the IGF signalling pathway, including dephosphor-ylation of AKT which is constitutively active in PTEN nullprostate cells, suggesting a potential therapeutic mechan-ism for this compound.49 This correlates with down regula-tion of total IGF-IR levels, decreases receptorphosphorylation and up regulation of IGFBP2 production inlycopene treated LNCaP cells. These effects are consistentwith the response of LNCaP cells to androgen withdra-wal.33,39 Furthermore, in androgen sensitive LNCaP cells lossof androgen receptor signalling can stop cell division andinitiate apoptosis.39,50 Since PC3 cells lack AR we hadpostulated that the differential apoptotic response seen inLNCaP and absent in PC3 cells might be a product oflycopene interfering with androgen receptor signalling.Androgen responsive luciferase reporter assays howeverdemonstrated that androgen receptor signalling is unaf-fected by lycopene treatment indicating that the differ-ential apoptotic response is not a direct product ofinterference by lycopene with androgen signalling.
It has been reported that IGF-1R, cdk 2 and cdk 4 proteinlevels declined in response to lycopene treatment in breastand endometrial cancer cells and that this may explain theanticancer properties of tomatoes.28 Decreased in Rbphosphorylation, expression of cyclin D1 and activity ofcdk 2 were demonstrated in MCF-7 breast cancer cellstreated with 2 mM lycopene.28 Similarly in normal prostatecells treated with 1mM or more of lycopene, cyclin D1expression was impaired and cell proliferation was slowed ascells accumulated in G1.30 However, in these studieslycopene effects were reported at doses 450-fold higherthan that described here for LycoTrueTM. Although previousreports have indicated that IGFBP3 levels in other cell linescan be modulated by lycopene treatment there was nodetectable IGFBP3 in our PCa cell line samples either beforeor after treatment.
In comparison, regulation of Rb phosphorylation and theassociated kinases has been reported in LNCaP cells treatedwith the Milk Thistle derived polyphenol Silibinin.51 In otherstudies this same naturally derived flavonoid was shown toproduce G1 cell cycle arrest in breast52 and prostate53
induces quiescence and apoptosis in androgen-responsive and -j.clnu.2007.01.002
Figure 5 LycoTrueTM treatment suppresses expression of cyclins D1 and E and cdk2, IGF-1R and activation of AKT in LNCaP and PC3cells. Cells were untreated (U), control treated with THF/BHT (T) or treated with LycoTrueTM at lycopene concentrations of 800, 400,200 or 80 nM for 48 h before immunoblot analysis. (A) Total cell lysates were probed with polyclonal anti-cyclin D1 (upper panel),anti-cyclin E (second panel), anti-cdk2 (third panel) antibodies or monoclonal anti-MAPK (lower panel) as a loading control. (B)Immunoblot analysis of expression of IGF-IR and expression and activation state of its downstream target AKT was performed onLNCaP and PC3 cell lines treated as in (A). Lysates were probed with anti-IGF-1R (upper panel), anti-phospho serine 473 (pAKT,second panel), anti-total AKT (third panel) antibodies or monoclonal anti-vinculin (lower panel) as a loading control.
N.I. Ivanov et al.10
cancer cells via induction of cdk inhibitors and decreasedactivity or expression of cdks. We have shown that lycopenealso affects the levels and activation states of several ofthese proteins and report a decrease in the level of cyclinD1, cyclin E and cdk 2. Although the level of Rb remainsconstant, phosphorylation, particularly at Serine 780, wasalso greatly decreased after lycopene treatment. Theimportance of Rb in the regulation of prostate growth hasbeen demonstrated; loss of Rb in mouse knockout models
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016
leads to the development of spontaneous prostate hyper-plasia.54
Substantial evidence, in both breast and PCa models invitro, of a clear ability to modulate the IGF pathwaysuggests a favorable outcome for a controlled clinical studyof the impact of a tomato rich, or lycopene supplementeddiet. The ability of lycopene to mimic the effect of PTEN andsuppress the constitutive hyperphosphorylation of AKT anddecrease Rb phosphorylation through decreased cyclin D1
induces quiescence and apoptosis in androgen-responsive and -/j.clnu.2007.01.002
Lycopene-induced cell cycle arrest and death of prostate cancer cells 11
availability in cancer cells55 and in these prostate cell lineslacking functional PTEN, further supports the mechanism oflycopene’s therapeutic potential in prostate cells being inpart through inhibition of AKTactivation; a compensation forthe absence of PTEN. This study provides mechanistic insightfor a potential therapeutic role for lycopene as well as itsuse in the prevention of PCa.
Acknowledgements
We would like to thank Catherine Wood and Lori Keddafor their expert technical assistance in HPLC analysisof lycopene samples and Dr. Venket Rao for supplyingLycopenTM.
Financial support: Canadian Micronutrients Network,CPC-Bionet, Michael Smith Foundation for Health Research,National Cancer Institute of Canada, Canadian ProstateCancer Research Initiative, Prostate Cancer Research Foun-dation of Canada.
References
1. Jones HA, Metz JM, Devine P, Hahn SM, Whittington R. Rates ofunconventional medical therapy use in patients with prostatecancer: standard history versus directed questions. Urology2002;59(2):272–6.
2. Di Mascio P, Kaiser S, Sies H. Lycopene as the most efficientbiological carotenoid singlet oxygen quencher. Arch BiochemBiophys 1989;274(2):532–8.
3. Agarwal A, Shen H, Agarwal S, Rao AV. Lycopene content oftomato products: its stability, bioavailability and in vivoantioxidant properties. J Med Food 2001;4(1):9–15.
4. Giovannucci E, Ascherio A, Rimm EB, Stampfer MJ, Colditz GA,Willett WC. Intake of carotenoids and retinol in relation to riskof prostate cancer. J Natl Cancer Inst 1995;87(23):1767–76.
5. Rao AV, Agarwal S. Role of lycopene as antioxidant carotenoid inthe prevention of chronic diseases: a review. Nutr Res 1999;19(2):305–23.
6. Giovannucci E, Rimm EB, Liu Y, Stampfer MJ, Willett WC. Aprospective study of tomato products, lycopene and prostate.J Natl Cancer Inst 2002;94(5):391–8.
7. Ansari MS, Gupta NP. A comparison of lycopene and orchidect-omy vs orchidectomy alone in the management of advancedprostate cancer. BJU Int 2005;95(3):453.
8. Ansari MS, Gupta NP. A comparison of lycopene and orchidect-omy vs orchidectomy alone in the management of advancedprostate cancer. BJU Int 2003;92(4):375–8.
9. Ansari MS, Gupta NP. A comparison of lycopene and orchidect-omy vs orchidectomy alone in the management of advancedprostate cancer. BJU Int 2004;94(4):678.
10. Clark PE, Hall MC, Borden Jr. LS, et al. Phase I-II prospectivedose-escalating trial of lycopene in patients with biochemicalrelapse of prostate cancer after definitive local therapy.Urology 2006;67(6):1257–61.
11. Guns ES, Cowell SP. Lycopene in prostate cancer prevention andtreatment. Nat Clin Pract: Urol 2005;2:38–43.
12. Shi J, Maguer ML. Lycopene in tomatoes: chemical and physicalproperties affected by food processing. Crit Rev Biotechnol2000;20(4):293–334.
13. Richards LR, Benghuzzi H, Tucci M, Hughes J. The synergisticeffect of conventional and sustained delivery of antioxidants onLNCaP prostate cancer cell line. Biomed Sci Instrum 2003;39:402–7.
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016/
14. Rao AV, Balachandran B. Role of oxidative stress and antiox-idants in neurogenerative diseases. Nutr Neurosci 2002;5(5):291–309.
15. Tang L, Jin T, Zeng X, Wang JS. Lycopene inhibits the growth ofhuman androgen-independent prostate cancer cells in vitro andin BALB/c nude mice. J Nutr 2005;135(2):287–90.
16. Hwang ES, Bowen PE. Cell cycle arrest and induction ofapoptosis by lycopene in LNCaP human prostate cancer cells.J Med Food 2004;7(3):284–9.
17. Levy J, Bosin BE, Feldman B, et al. Lycopene is a more potentinhibitor of human cancer cell proliferation than either alpha-carotene or beta-carotene. Nutr Cancer 1995;24:257–66.
18. Karas M, Amir H, Fishman D, et al. Lycopene interferes with cellcycle progression and insulin-like growth factor I signaling inmammary cancer cells. Nutr Cancer 2000;36(1):101–11.
19. Kim L, Rao AV, Rao LG. Effect of lycopene on prostate LNCaPcancer cells in culture. J Med Food 2002;5(4):181–7.
20. Kotake-Nara E, Kushiro M, Zhang H, Sugawara T, Miyashita K,Nagao A. Carotenoids affect proliferation of human prostatecancer cells. J Nutr 2001;131:3303–6.
21. Bose S, Crane A, Hibshoosh H, Mansukhani M, Sandweis L,Parsons R. Reduced expression of PTEN correlates with breastcancer progression. Hum Pathol 2002;33(4):405–9.
22. Dahia PL. PTEN, a unique tumor suppressor gene. Endocr RelatCancer 2000;7(2):115–29.
23. Cantley LC, Neel BG. New insights into tumor suppression: PTENsuppresses tumor formation by restraining the phosphoinositide3-kinase/AKT pathway. Proc Natl Acad Sci USA 1999;96(8):4240–5.
24. Horoszewicz JS, Leong SS, Kawinski E, et al. LNCaP model ofhuman prostatic carcinoma. Cancer Res 1983;43:1809–17.
25. Ahmed S, Johnson CS, Rueger RM, Trump DL. Calcitriol (1,25-dihydroxycholecalciferol) potentiates activity of mitoxantrone/dexamethasone in an androgen independent prostate cancermodel. J Urol 2002;168(2):756–61.
26. Portigal CL, Cowell SP, Fedoruk MN, Butler CM, Rennie PS,Nelson CC. Polychlorinated biphenyls interfere with androgen-induced transcriptional activation and hormone binding. ToxicolAppl Pharmacol 2002;179(3):185–94.
28. Nahum A, Hirsch K, Danilenko M, et al. Lycopene inhibition ofcell cycle progression in breast and endometrial cancer cells isassociated with reduction in cyclin D levels and retention ofp27(Kip1) in the cyclin E-cdk2 complexes. Oncogene 2001;20(26):3428–36.
29. Ben-Dor A, Nahum A, Danilenko M, et al. Effects of acyclo-retinoic acid and lycopene on activation of the retinoic acidreceptor and proliferation of mammary cancer cells. ArchBiochem Biophys 2001;391(2):295–302.
30. Obermuller-Jevic UC, Olano-Martin E, Corbacho AM, et al.Lycopene inhibits the growth of normal human prostateepithelial cells in vitro. J Nutr 2003;133(11):3356–60.
31. Shim M, Cohen P. IGFs and human cancer: implicationsregarding the risk of growth hormone therapy. Horm Res 1999;51(Suppl 3):42–51.
32. Mucci LA, Tamimi R, Lagiou P, et al. Are dietary influences onthe risk of prostate cancer mediated through the insulin-likegrowth factor system? BJU Int 2001;87(9):814–20.
33. Kiyama S, Morrison K, Zellweger T, et al. Castration-inducedincreases in insulin-like growth factor-binding protein 2promotes proliferation of androgen-independent human pros-tate LNCaP tumors. Cancer Res 2003;63(13):3575–84.
34. Scorilas A, Plebani M, Mazza S, et al. Serum human glandularkallikrein (hK2) and insulin-like growth factor 1 (IGF-1) improvethe discrimination between prostate cancer and benign
induces quiescence and apoptosis in androgen-responsive and -j.clnu.2007.01.002
prostatic hyperplasia in combination with total and %free PSA.Prostate 2003;54(3):220–9.
35. Chan JM, Stampfer MJ, Ma J, et al. Insulin-like growth factor-I(IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. J Natl Cancer Inst 2002;94(14):1099–106.
36. Oliver SE, Barrass B, Gunnell DJ, et al. Serum insulin-like growthfactor-I is positively associated with serum prostate-specificantigen in middle-aged men without evidence of prostatecancer. Cancer Epidemiol Biomarkers Prev 2004;13(1):163–5.
37. DiGiovanni J, Kiguchi K, Frijhoff A, et al. Deregulatedexpression of insulin-like growth factor 1 in prostate epitheliumleads to neoplasia in transgenic mice. Proc Natl Acad Sci USA2000;97(7):3455–60.
38. Nickerson T, Chang F, Lorimer D, Smeekens SP, Sawyers CL,Pollak M. In vivo progression of LAPC-9 and LNCaP prostatecancer models to androgen independence is associated withincreased expression of insulin-like growth factor I (IGF-I) andIGF-I receptor (IGF-IR). Cancer Res 2001;61(16):6276–80.
39. Krueckl SL, Sikes RA, Edlund NM, et al. Increased insulin-likegrowth factor I receptor expression and signaling are compo-nents of androgen-independent progression in a lineage-derivedprostate cancer progression model. Cancer Res 2004;64(23):8620–9.
40. Knudsen ES, Wang JY. Differential regulation of retinoblastomaprotein function by specific Cdk phosphorylation sites. J BiolChem 1996;271(14):8313–20.
41. Chau BN, Wang JY. Coordinated regulation of life and death byRB. Nat Rev Cancer 2003;3(2):130–8.
43. Kitagawa M, Higashi H, Jung HK, et al. The consensus motif forphosphorylation by cyclin D1-Cdk4 is different from that forphosphorylation by cyclin A/E-Cdk2. Embo J 1996;15(24):7060–9.
44. Takaki T, Fukasawa K, Suzuki-Takahashi I, et al. Preferences forphosphorylation sites in the retinoblastoma protein of D-typecyclin-dependent kinases, Cdk4 and Cdk6, in vitro. J Biochem(Tokyo) 2005;137(3):381–6.
Please cite this article as: Ivanov NI, et al. Lycopene differentiallyindependent prostate cancer cell lines. Clin Nutr (2007), doi:10.1016
45. Keenan SM, Lents NH, Baldassare JJ. Expression of cyclin Erenders cyclin D-CDK4 dispensable for inactivation of theretinoblastoma tumor suppressor protein, activation of E2F,and G1-S phase progression. J Biol Chem 2004;279(7):5387–96.
46. Arata Y, Fujita M, Ohtani K, Kijima S, Kato JY. Cdk2-dependentand -independent pathways in E2F-mediated S phase induction.J Biol Chem 2000;275(9):6337–45.
47. Cheng L, Rossi F, Fang W, Mori T, Cobrinik D. Cdk2-dependentphosphorylation and functional inactivation of the pRB-relatedp130 protein in pRB(�), p16INK4A(+) tumor cells. J Biol Chem2000;275(39):30317–25.
48. Bertram J, Peacock JW, Fazli L, et al. Loss of PTEN is associatedwith progression to androgen independence. Prostate 2006;66(9):895–902.
49. Pollak MN, Schernhammer E, Hankinson S. Insulin-like growthfactors and neoplasia. Nat Rev Cancer 2004;4(7):505–18.
50. Yang Q, Fung KM, Day WV, Kropp BP, Lin HK. Androgen receptorsignaling is required for androgen-sensitive human prostatecancer cell proliferation and survival. Cancer Cell Int2005;5(1):8.
51. Tyagi A, Agarwal C, Agarwal R. Inhibition of retinoblastomaprotein (Rb) phosphorylation at serine sites and an increase inRb-E2F complex formation by silibinin in androgen-dependenthuman prostate carcinoma LNCaP cells: role in prostate cancerprevention. Mol Cancer Ther 2002;1(7):525–32.
52. Zi X FD, Agarwal R. Anticarcinogenic effect of flavonoidantioxidant, sliymarin, in human breast cancer cells MDA-MB468: Induction of G1 arrest through an increase in Cip1/p21concomitant with a decrease in kinase activity of cyclin-dependent kinases and associated cyclins. Clin Cancer Res1998;4:1055–64.
53. Zi X AG, Kung HJ, Agarwal R. A flaxonoid antioxidant, Silymarin,inhibits activation of erbB1 signalling and induces cyclin-dependent kinase inhibitors, G1 arrest, and anticarcinogeniceffects in human prostate carcinoma DU145 cells. Cancer Res1998;58:1920–9.
54. Abate-Shen C, Shen MM. Molecular genetics of prostate cancer.Genes Dev 2000;14(19):2410–34.
55. Radu A, Neubauer V, Akagi T, Hanafusa H, Georgescu MM. PTENinduces cell cycle arrest by decreasing the level and nuclearlocalization of cyclin D1. Mol Cell Biol 2003;23(17):6139–49.
induces quiescence and apoptosis in androgen-responsive and -/j.clnu.2007.01.002
