Cross-talk between vitamin D receptor (VDR)- and peroxisome proliferator-activated receptor (PPAR)-signaling in melanoma cells

Abstract. The expression and signaling of the vitamin Dreceptor (VDR) and peroxisome proliferator-activatedreceptor (PPAR) α, δ, γ was investigated in the melanomacell line MeWo. Using real-time PCR, the mRNA of thenuclear receptors (NR) was detected. The strongestexpression was found for the VDR, approximately 3-foldhigher compared to the expression of PPARα or PPARδ, andthe weakest expression was for PPARγ. After treatment withcorresponding ligands, the expression of the VDR, PPARαand PPARδ was elevated up to 5-fold, while the PPARγexpression was not significantly affected. Treatment with1α,25-dihydroxyvitamin D3 (1,25(OH)2D3, calcitrol)resulted in 40% inhibition of MeWo cell proliferation, thatwas associated with a 5-fold increase in VDR mRNA.Interestingly, cell proliferation was differentially modulatedby treatment with the PPAR ligands. While docosahexaenoicacid (DHA) treatment resulted in a statistically significantincrease (approximately 10% ), the other PPAR ligandsinhibited MeWo cell proliferation. GW501516 (PPARδligand) and WY14643 (PPARα ligand) both had anantiproliferative effect of approximately 10% . Theseantiproliferative effects were not associated with modulationof PPARα or PPARδ expression. In contrast, stimulation ofMeWo proliferation by DHA was associated with a 3- and4-fold increase in the expression of PPARα and PPARδ,respectively. Analyzing the cross-talk between the VDR andPPAR signaling pathways, the 1,25(OH)2D3 treatmentresulted in an approximately 2-fold increase in expression ofPPARα and PPARδ, while the expression of PPARγ wasunaffected. Treatment with GW501516 and WY14643 resulted

in an increase in the VDR expression (2-fold after 120 h).The simultaneous treatment with 1,25(OH)2D3 partiallyantagonised the DHA- and alpha-linolenicacid (ALA)-induced up-regulation of PPAR expression. In contrast,treatment with the PPAR ligands had no pronounced effecton the 1,25(OH)2D3-induced increase in VDR expression.Simultaneous treatment with the PPAR ligands bezafibrateor ALA resulted in an up to 6-fold reduction of the1,25(OH)2D3-induced elevation of the 1α,25-dihydroxyvitamin D3-24-hydroxylase (CYP24A1) expression.Simultaneous treatment with the PPAR ligands and1,25(OH)2D3 resulted in only marginal modulation of1,25(OH)2D3-induced inhibition of cell proliferation.However, simultaneous treatment with bezafibrate and1,25(OH)2D3 resulted in a statistically significant partialantagonisation of the 1,25(OH)2D3-induced inhibition ofMeWo cell proliferation. In conclusion, PPAR and VDR havea role in growth regulation in melanoma cells andfunctionally relevant cross-talk between these nuclearsignaling pathways is indicated, but not at the level of cellproliferation, where 1,25(OH)2D3 has a dominant effect.

Malignant melanoma is the most aggressive form of skincancer, with a dramatic increase in incidence and co-morbidity, worldwide in recent years (1). Metastasizedmalignant melanoma has a very poor prognosis (2). It ishighly resistant to conventional chemotherapeutic treatments,with dacarbazine having the best single-agent efficacy (2).Furthermore, multiagent chemotherapeutic regimens showedno evidence of prolonged overall survival rates in melanomapatients compared to dacarbazine treatment alone (3, 4).Thus, there is a need to develop new therapeutic andpreventive approaches for melanoma, using synthetic ornatural substances to prevent or reverse the transition ofpremalignant lesions into invasive cancer (5).

Substances with a potential role in melanoma chemo-prevention or therapy include vitamin D analogs andperoxisome proliferator-activated receptor (PPAR) ligands,such as lipid-lowering drugs, fibrates and statins (6-13). It is

3647

Correspondence to: Professor Dr. med. J. Reichrath, Department ofDermatology, The Saarland University Hospital, Kirrbergerstr.66421 Homburg/Saar, Germany. Tel: +49 68411623802, Fax: +4968411623845, e-mail: [email protected]

Key Words: Nuclear receptors, malignant melanoma, peroxisomeproliferator-activated receptors, PPAR, vitamin D receptor, VDR.

ANTICANCER RESEARCH 29: 3647-3658 (2009)

Cross-talk between Vitamin D Receptor (VDR)- and Peroxisome Proliferator-activated Receptor

(PPAR)-signaling in Melanoma CellsPIT SERTZNIG1, TOM DUNLOP2, MARKUS SEIFERT1, WOLFGANG TILGEN1 and JÖRG REICHRATH1

1Department of Dermatology, The Saarland University Hospital, Homburg/Saar, Germany;2University of Kuopio, Department of Biosciences/Biochemistry, P.O.Box 1627, FI-70211 Kuopio, Finland

0250-7005/2009 $2.00+.40

well known that 1α,25-dihydroxyvitamin D3 (1α,25(OH)2D3,calcitriol), the most biologically active natural vitamin Dmetabolite, acts via binding to its corresponding intranuclearreceptor (VDR), present in target tissue cells (14). The VDRbelongs to the nuclear receptor (NR) superfamily of trans-acting transcriptional regulatory factors, which includes thesteroid and thyroid hormone receptors as well as the retinoid-X receptors (RXR) and retinoic acid receptors (RAR) (15-17). In various cell types, including normal and malignanthuman melanocytes, the effects of 1α,25(OH)2D3 and VDR-mediated genomic pathways include the regulation of cellgrowth and differentiation (reviewed in 6 and 7; 18-20). Non-genomic cytoplasmic signaling pathways are increasinglybeing recognized as well. Additionally, 1,25(OH)2D3-mediated effects on tumor invasion, angiogenesis andmetastastic behavior have been shown in variousmalignancies (reviewed in 6 and 7; 18-20). 1,25(OH)2D3exerts a significant inhibitory effect on the G1/S checkpointof the cell cycle by up-regulating the cyclin-dependent kinaseinhibitors p27 and p21, and by inhibiting cyclin D1 (reviewedin 7). Indirect mechanisms of 1,25(OH)2D3-mediated growthregulation include the up-regulation of transforming growthfactor-β and the down-regulation of the epidermal growthfactor receptor (reviewed in 7). 1,25(OH)2D3 has the capacityto induce apoptosis either indirectly through effects on theinsulin-like growth receptor and tumour necrosis factor-α ormore directly via the B-cell lymphoma 2 (Bcl-2) familysystem, the ceramide pathway, the death receptors (e.g. Fasreceptor) and the stress-activated protein kinase pathways(Jun N terminal kinase and p38) (reviewed in 7).Mechanisms involved in the 1,25(OH)2D3-induced inhibitionof tumor invasion and metastasis include inhibition of serineproteinases and metalloproteinases (reviewed in 7) as well asthe up-regulation of E-cadherin (21). This role is furthersuggested by the fact that the VDR gene is a target ofepithelial to mesenchymal transition (EMT) promoters suchas Snail1 (22). EMT is an organised process duringdevelopment and is under the control of master generegulators such as transcription factors Snail, Slug and Twist,amongst others (23). The discovery of the derivation ofcutaneous melanomas from the migratory neural crest cellpopulation supports the intrinsic predisposition hypothesis(24). In addition to this, elements of the neural crestmolecular circuitry responsible for migratory behaviourduring development might be reactivated during melanomapathogenesis. Little is known at present about thetranscriptional regulation of Slug and other melanocyte-specific factors that may be of importance for melanomapathogenesis and progression, by NRs.

In a variety of human carcinomas, PPAR ligands and otheragents influencing the PPAR signaling pathways have beenshown to have chemopreventive potential by mediatingtumor suppressive activities. Therefore modulating PPAR

signaling pathways represents a potential novel strategy forinhibiting tumor carcinogenesis and progression. Threegenetically and functionally distinct PPAR isotypes havebeen described: PPARα (NR1C1: nuclear receptor subfamily1, group C, member 1), PPARδ (NR1C2) and PPARγ(NR1C3). All three PPARs are adopted orphan nuclearhormone receptors (25-27), but exhibit distinct patterns oftissue distribution (28).

PPARδ is expressed in a wide range of tissues and cells,with relatively higher levels of expression noted in the skin,brain, adipose tissue, kidney, heart and digestive tract,suggesting possible developmental or physiological roles inthese tissues (29-33). PPARγ is expressed at high levels inadipose tissue and is an important regulator of adipocytedifferentiation and lipid metabolism (34-36). High levels ofPPARγ have also been observed in post-mitotic anddifferentiated colonic epithelial cells (37-39). PPARγ is alsofound in other cell types, such as hepatocytes, fibroblasts,and epithelial cells (31, 40). The highest PPARα expressionhas been shown in the liver (41) and in tissues with highfatty acid catabolism such as the kidney, heart, skeletalmuscle and brown fat (31, 42, 43).

PPARs are activated by a number of natural ligands such aslong-chain fatty acids, in particular, polyunsaturated fatty acids(including linoleic acid, linolenic acid and arachidonic acid)(44-46). In the last decade, a large number of synthetic ligandshave been identified. These include the thiazolidinediones(troglitazone, rosiglitazone, pioglitazone), a class of insulin-sensitizing agents and the fibrates (bezafibrate, clofibrate,fenofibrate), which are used as hypolipidaemic drugs. Interestin exploiting these NRs is testified to by the existence ofnewer ligands such as GW2331, GW9578, GW501516, L-165041, L-796449, L-805645 and WY14643 (47).

After activation through a ligand, the PPARs and VDRform a heterodimer with the RXR. This heterodimerpreferentially binds to specific peroxisome proliferator orvitamin D response elements in enhancer sites of theregulated target genes and regulates gene expression. Thecomplex regulatory network of PPAR and VDR signalingpathways, and the interaction between them, deservessystematic analysis. These pathways regulate a multitude ofgenes that are of importance for the regulation of variouscellular functions including cell proliferation, celldifferentiation, immune responses and apoptosis and maytherefore also play an important role in cancer treatment andprevention.

Recently, NRs have been sub-divided into several classes,depending upon tissue-specific expression patterns. Thisclassification may reflect shared responsibility incoordinating the transcriptional programs necessary toexecute idiotypic physiological pathways within any giventissue or organ (48). Consequently, the VDR and PPARswere assigned to a distinct cluster of NRs. Studying the


3648

expression profile of NRs in individual tissues offers asimple yet powerful way to obtain highly related informationabout their physiological functions as individual proteins(48). Considering the enormous number of direct andindirect target genes of the VDR and PPARs, it is tempting tospeculate that these NRs may modulate proliferation anddifferentiation of normal and malignant melanocytes at leastin part via direct or indirect regulation of melanocyte-specific factors. In this paper, the expression profile of theVDR and PPARs was analyzed in the melanoma cell lineMeWo to further elucidate the putative role of these NRsignaling pathways for the growth of melanoma cells.Additionally, putative cross-talk between the VDR andPPARs, a hypothesis that has been supported by recentfindings, was investigated. It has been shown that the humanPPARδ gene is a primary target of 1,25(OH)2D3 and its NR(49). Moreover, since both the VDR and PPAR compete fortheir predominant heterodimerisation partner, RXR complextranscriptional regulation of target genes may be expectedwhen both NR types are activated.

Materials and Methods

Cell culture. Human melanoma cell line, MeWo (50), establishedfrom a metastatic melanoma lymph node, were obtained from thetumor and cell line bank of the Dermatology clinic of the SaarlandUniversity Hospital. The MeWo cells were cultivated in RPMI(supplemented with 10% FCS) at 37°C, in a 5% CO2 atmosphereusing culture dishes (diameter, 10 cm) or 24-well plates (Greiner,Frickenhausen, Germany). Semi-confluent cells were incubated with10% FCS (charcoal-treated), with or without 1,25(OH)2D3 andvarious PPAR ligands (alpha-linolenicacid (ALA), bezafibrate,docosahexaenoicacid (DHA), GW501516, or WY14643, or theindicated ligand combinations. Treatment of cells with vehicle(ethanol) alone served as the control treatment. 1,25(OH)2D3 andthe PPAR ligands bezafibrate, ALA, DHA and WY14643 werepurchased from Sigma (Taufkirchen, Germany). GW501516 waspurchased from Alexis/Axxora (Lörrach, Germany). Theconcentration of 1,25(OH)2D3 (10–8 M) was chosen because it waspreviously shown to be highly effective in inducing genomic effectsin MeWo cells which strongly express the VDR (51) withoutexerting toxic effects. With the exception of bezafibrate, PPAR

ligands were used at a concentration of 10–8 M as well. BecausePPAR ligands and 1,25(OH)2D3 are cytotoxic at higherconcentrations, we used low ligand concentrations in our studies. Itwas important to minimize toxic effects because the cells weretreated for over 120 h to show transcriptional effects on mRNAlevels. However, only fibrates activate PPARs at concentrations inthe high micromolar range (52). The final concentration of 200 μMfor bezafibrate was chosen because this concentration showedsignificant effects without exerting any cytotoxicity in other studies(53). All treatments were repeated every 48 h.

Proliferation assay. The cells were plated at a density of 1×103

cells/well into 24-well plates. Cell proliferation was quantified bycrystal violet (CV) dye staining. The cells were washed once withPBS and fixed with ethanol (70% ) for 30 min at room temperature.The cells were then incubated with a CV solution (1% w/v in 20%ethanol) for 30 min at room temperature and rinsed with waterthoroughly. After drying, the dye was extracted with 70% ethanoland its absorbance determined at 550 nm using a microplate reader(Titertek Multiskan Plus MK II, Labsystems, Helsinki, Finland).

RNA isolation. RNA isolation was carried out with a RNeasy Kit(Qiagen, Hilden, Germany) according to the manufacturer’s manual.The quantitative real-time (RTq) PCR was carried out with anOmniscript RT Kit (Qiagen), using Oligo-dT-primers and 2 μg RNAper reaction as a template.

Quantitative real-time PCR (RTqPCR) and analysis. The expressionof the human VDR, 1α,25-dihydroxyvitamin D3-24-hydroxylase(CYP24A1) and the PPAR isotype genes was analyzed in the MeWocells using RTqPCR (60 cycles in a LightCycler; Roche MolecularBiochemicals, Mannheim, Germany) and gene-specific primersfrom Qiagen and TIB Molbiol (Berlin, Germany) (Table I). In orderto calculate the normalized ratio, the relative amount of the targetgene (VDR, CYP24A1, PPAR) and a reference gene (β2-microglobulin) was determined for each sample and one calibrator,integrated into each LightCycler run. The relative ratio of target toreference for each sample and for the calibrator was first calculatedto correct for sample-to-sample variations caused by differences inthe initial quality and quantity of the nucleic acid (Roche, technicalnote LC 13/2001). The target/reference ratio of each sample wasthen divided by the target/reference ratio of the calibrator usingrelative quantification software (Roche Relquant). This second stepnormalized the different detection sensitivities of the target andreference amplicons. Thus the normalization to a calibrator provided

Sertznig et al: VDR and PPAR-signalling in Melanoma Cells

3649

Table I. Primers used in RTqPCR.

Primer Sequence Source

PPARα QuantiTect Primer Assay (QT00017451) QiagenPPARδ QuantiTect Primer Assay (QT00078064) QiagenPPARγ QuantiTect Primer Assay (QT00029841) QiagenVDR (forward) 5’-CCA GTT CGT GTG AAT GAT GG-3’ TIB MolbiolVDR (reverse) 5’-GTC GTC CAT GGT GAA GGA-3’CYP24A1 (forward) 5’-GCA GCC TAG TGC AGA TTT-3’ TIB MolbiolCYP24A1 (reverse) 5’-ATT CAC CCA GAA CTG TTG-3’β2-Μicroglobulin (forward) 5’-CCA GCA GAG AAT GGA AAG TC-3’ TIB Molbiolβ2-Μicroglobulin (reverse) 5’-GAT GCT GCT TAC ATG TCT CG-3’

a constant calibrator point between the PCR runs. Each sample wasanalyzed in quadruplicate; the final values were expressed as themedian of N-fold differences in the target gene expression in thetreated samples relative to the control samples.

Statistical analysis. All the data are presented as mean±SEM of atleast 3 experiments. Statistical significance was calculated by theStudent’s t-test. The mean differences were considered to besignificant when p<0.05.

Results

Proliferation of MeWo cells. Treatment with the VDR ligand1,25(OH)2D3 resulted in strong and statistically significantinhibition of cell proliferation (40% after 120 h, p<0.0005)(Figure 1). Treatment with the PPAR ligands differentiallymodulated the proliferation of the MeWo cells. Whiletreatment with several PPAR ligands resulted in a statisticallysignificant inhibition of cell proliferation (approximately

10% after 120 h with either GW501516 or WY14643,p<0.0005), treatment with DHA resulted in a smallstatistically significant increase of cell proliferation(approximately 10% after 120 h, p<0.005) (Figure 1).Treatment with ALA or bezafibrate resulted only in a weakand statistically insignificant inhibition of cell proliferation(4% , p=0.3936 and 5% , p=0.0705, respectively) (Figure 1).In combination with the PPAR ligands, 1,25(OH)2D3,appeared to have a dominant inhibitory effect on cellproliferation. Whilst combination with ALA, DHA andWY14643 appeared to have no effect on the cell numbersinhibited by 1,25(OH)2D3, GW501516 and bezafibrate had aslightly antagonistic effect by increasing the cell numbers to63.44% and 65.82% , relative to the control, respectively.

VDR, PPARα, δ and PPARγ expression. The RTqPCR analysisrevealed that the mRNAs for PPARα, δ, γ and the VDR wereexpressed in the MeWo melanoma cells (Figure 2a-e). The


3650

Figure 1. MeWo cell numbers as a percentage of the control (EtOH) after 120 h of culture. n.s., Not significantly different form the control;significantly different from the control at **p<0.005 and ***p<0.0005.

steady state level of the VDR gene expression wasapproximately 3 times higher compared to the expression ofPPARα and δ (p<0.0005) (Figure 2a), while the expression ofPPARγ was lowest and almost undetectable. No significantchange in the relative mRNA expression levels of any the NRsanalyzed was observed over the 120 h cultivation period(Figure 2b-e).

Effect of VDR and PPAR ligand treatment on receptor geneexpression. The steady state levels of the VDR, PPARα andPPARδ genes were increased up to 5-fold by treatment withtheir corresponding ligands, whilst the expression of PPARγwas not modulated (Figure 3a-d). The ligand-inducedincrease in expression was most pronounced and statistically


3651

Figure 2. a, MeWo cells untreated, cultured over 24, 48, 72, 96, 120 h(in total 20 values for each gene). b, Correlation between PPARαexpression and time of culture. c, Correlation between PPARδ expressionand time of culture. d, Correlation between PPARγ expression and timeof culture. e, Correlation between VDR expression and time of culture.


3652


3653

Figure 3. Effect of ligand treatment on VDR and PPAR expression. n.s., Not significantly different form the control; significantly different from thecontrol at *p<0.05, **p<0.005 and ***p<0.0005.

significant: for the VDR after treatment with 1,25(OH)2D3(3-fold after 24 h, p<0.0005; more than 4-fold after 120 h,p<0.0005) (Figure 3a); for PPARα after treatment witheither bezafibrate or ALA (approximately 4-fold after 120 h,p<0.05) (Figure 3b) and for PPARδ after treatment withbezafibrate, ALA or DHA (approximately 3-fold after 120h; bezafibrate: p<0.0005; ALA: p<0.005 and DHA: p<0.005)(Figure 3c). Treatment with WY14643 and GW501516,selective PPARα and PPARδ ligands, respectively, mostlydid not modulate the PPAR gene expression. The exceptionsto this were the modest increases in VDR, PPARα andPPARδ expression with WY14643, as well as VDRexpression after 120 h treatment with GW501516.

Cross-talk between VDR- and PPAR-signaling pathways. Atboth 24 h and 120 h, the 1,25(OH)2D3 treatment resulted in astatistically significant, 1.5-fold increase in expression of PPARα(p<0.05), whilst the expression of PPARγ was not significantly

affected (Figure 3b and 3d). PPARδ expression was slightlyincreased by 1,25(OH)2D3 treatment at 24 h (p<0.05), but by120 h the effect was not significant. Interestingly, treatment withseveral PPAR ligands resulted (after 120 h) in an increase inVDR expression (approximately 2- to 3-fold, GW501516 andWY14643, p<0.0005; approximately 1.5-fold, ALA and DHA,p<0.05) (Figure 3a). Moreover, simultaneous treatment with1,25(OH)2D3 and PPAR ligands led to a significant 1.5-foldincrease of PPARα expression compared to the treatment withPPAR ligands alone after 24 h of treatment (bezafibrate, DHAand GW501516, p<0.05, ALA and WY14643, p<0.005) (Figure4a). The PPARδ expression was only increased by thecombination of 1,25(OH)2D3 with bezafibrate (p<0.005) orDHA (p<0.05) after 24 h. The bezafibrate or ALA-induced up-regulation of the PPARα and PPARδ expression was after 120 hof treatment partially antagonised (PPARα: bezafibrate: p<0.005,ALA: p<0.05; PPARδ: bezafibrate: p<0.05, ALA: p<0.005)(Figure 4 a-b).


3654

Figure 4. PPAR expression in MeWo cells treated with 1,25(OH)2D3 plus ligand in comparison with the PPAR ligand alone.

The simultaneous treatment with the PPAR ligands haddifferential effects on the 1,25(OH)2D3-induced up-regulationof VDR expression. While treatment for 120 h with DHA orGW501516 elevated (p<0.05) the 1,25(OH)2D3-induced up-regulation of VDR expression, treatment with other PPARligands (bezafibrate: p<0.0005, ALA: p<0.05) reduced the1,25(OH)2D3-induced up-regulation of VDR expression(Figure 3a).

As shown in Table II, the treatment with 1,25(OH)2D3increased the ratios of VDR/PPARα and VDR/PPARδ ascompared to the treatment with vehicle (ethanol) alone.Interestingly, treatment with the PPAR ligands GW501516and WY14643 increased the ratios of VDR/PPARα andVDR/PPARδ as well. The highest ratios of VDR/PPARαand VDR/PPARδ were observed after simultaneoustreatment with 1,25(OH)2D3 and GW501516 or WY14643.In contrast to treatment with GW501516 or WY14643,treatment with DHA, ALA and bezafibrate in generalreduced the ratios of VDR/PPARα and VDR/PPARδ. Aftersimultaneous treatment with 1,25(OH)2D3 and DHA, ALAor bezafibrate the ratios of VDR/PPARα and VDR/PPARδin general decreased as compared to the ratios aftertreatment with 1,25(OH)2D3 alone.

Effect of PPAR activation on 1,25(OH)2D3-mediated genetranscription. In order to provide insight into the functionalityof 1,25(OH)2D3-mediated signaling, the levels of a knowntarget gene, CYP24A1, were assessed and as expected, thesteady state mRNA was robustly up-regulated by 1,25(OH)2D3


3655

Table II. VDR/PPARα and VDR/PPARδ ratios after 120 h of treatmentwith different VDR and/or PPAR ligands

Treatment VDR/PPARα VDR/PPARδ

Without treatment 5.0 3.7EtOH 3.6 4.51,25(OH)2D3 11.5 12.3Bezafibrate 1.3 1.9ALA 1.2 1.6DHA 2.1 1.9WY14643 6.6 7.9GW501516 11.4 10.01,25(OH)2D3+bezafibrate 10.6 8.41,25(OH)2D3+ALA 6.0 8.91,25(OH)2D3+DHA 9.4 11.41,25(OH)2D3+WY14643 11.3 15.61,25(OH)2D3+GW501516 15.2 17.5

Figure 5. Effect of ligand treatment on CYP24A1 expression in MeWo cells compared with EtOH (vehicle, control). n.s., Not significantly differentform the control; significantly different from the control at *p<0.05, **p<0.005 and ***p<0.0005.

(Figure 5). Simultaneous treatment with the PPAR ligandsbezafibrate (p<0.0005), GW501516 (p<0.05) and ALA(p<0.05) resulted in an up to 6-fold reduction of the1,25(OH)2D3-induced stimulation of CYP24A1 geneexpression. Simultaneous treatment of DHA with 1,25(OH)2D3resulted in a statistically significant (p<0.005) increase inCYP24A1 gene expression compared to the treatment with1,25(OH)2D3.

Discussion

Earlier investigations showed that PPARα activation reducedthe metastatic potential both in B16F10 mouse melanomacells and in human SkMel188 cells in vitro via down-regulation of Akt signaling (5). In the present study, thePPAR mRNA expression in MeWo cells was presented forthe first time. LightCycler-based RTqPCR was chosen todetect and quantify NR expression because of its broaddynamic range of detection, quantitative reliability, andreproducibility. Interestingly, the expression of PPARγ wasvery weak (near the detection limit), questioning whether thepreviously reported antiproliferative effects of PPARγagonists on MeWo cells (4) were indeed mediated viasignaling pathways of this NR or via other unknownmechanisms. The VDR expression was approximately threetimes higher compared to the expression of PPARα and δ. Itis well accepted that the responsiveness to NR ligands varieswith the number of corresponding receptors in the targettissues. Therefore, these findings were well in line with ourobservation that 1,25(OH)2D3 exerts stronger effects onMeWo proliferation as compared to PPAR ligands. Nosignificant change in expression of any of the NRs wasobserved over the 120 h culture.

It was previously shown that NR expression may beinduced in target tissues by treatment with the correspondingligands, and that this up-regulation is in many cases of highimportance for mediating the biological effects of the NRligands (23). Interestingly, in the present study, theexpression of VDR, PPARα and PPARδ, but not of PPARγwas induced, up to 5-fold, by the ligand treatment. Theincrease in expression was most pronounced for VDR aftertreatment with 1,25(OH)2D3 (3-fold after 24 h, up to 5-foldafter 120 h), for PPARα after treatment with bezafibrate orALA and for PPARδ after treatment with bezafibrate, ALAor DHA (all approximately 4-fold after 120 h).

Treatment of the MeWo cells with the VDR-ligand1,25(OH)2D3 resulted in strong and statistically significantinhibition of cell proliferation which was in agreement withour previous results (51). Treatment with the PPAR ligandsdifferentially modulated proliferation, while GW501516 orWY14643 resulted in a statistically significant inhibition,DHA resulted in statistically significant increase and ALAor bezafibrate only in weak and not significant inhibition of

cell proliferation. These results were in line with otherfindings. The effects of PPAR were reported to vary stronglywith dosage, high concentrations (up to 30 μM ciglitazone,troglitazone, rosiglitazone, pioglitazone) exertingantiproliferative effects while low concentrations might evenstimulate cell proliferation (4). In the present study, aconcentration of 10–8 M was chosen both because significanteffects on cell growth were shown previously and becausethis concentration may be reached in clinical use. At higherdoses, the use of ciglitazone and other PPAR ligands islimited because of several side-effects (54).

It has previously been reported that the human PPARδgene was a direct transcriptional target of 1,25(OH)2D3 (49),but this effect was not observed in the present study. Thiswas possibly because different time-points and a differentcell type was used. However, in combination with bezafibrateand DHA, this hormone was able to induce PPARδexpression, thereby indicating that interactions do occur. Itis possible that endogenous PPAR ligands produced by othercell types might allow for gene activation events.1,25(OH)2D3 was also responsible for an up-regulation inthe expression of PPARα, both after one and five days oftreatment. However, it is difficult to determine whether thiswas a direct effect because of the long time lapse betweenhormone addition and the first measurement by RT-PCR ofthe expression of the NR gene levels. This needs to bedetermined in future studies. 1,25(OH)2D3 also had aninhibitory effect on the gene activiation by the PPAR ligands.Bezafibrate-, ALA- and DHA-induced PPARδ expressionwas suppressed in the presence of 1,25(OH)2D3 at 120 h.Again, it is difficult to suggest that these were primaryeffects and further clarification is needed.

Mild antiproliferative effects were observed withGW501516 and WY14643 treatment, which correlated withan increase in VDR expression afforded by these compounds,at the same time-point (120 h, Figure 3a). It is thereforetempting to speculate that these compounds augment theVDR signaling system to achieve their antiproliferativeaction. However, the capability of MeWo cells to synthesise1,25(OH)2D3 de novo should be verified. Knowing thattreatment with the PPAR ligands did not increase theCYP24A1 gene expression (Figure 5), it is alternativelypossible that the VDR protein itself exerts an antiproliferativeeffect since it was unregulated by the hormone wherever anantiproliferative effect was observed.

In summary, PPARs and the VDR are expressed inmelanoma cells and, besides 1,25(OH)2D3, some of thePPAR ligands investigated inhibit the proliferation of MeWocells. Additionally, the biological effects of PPAR ligands onmelanoma cells may, at least in part, be mediated via vitaminD-signaling pathways. Stimulation of MeWo cells withPPAR ligands modulates the expression of the VDR, whichmay increase the sensitivity of these cells for the action of


3656

endogeneously produced 1,25(OH)2D3. Treatment withPPAR ligands, at least in part, antagonizes theantiproliferative effects of 1,25(OH)2D3. These effects maybe due to cross-talk between PPAR- and VDR-mediatednuclear signaling pathways that is at least in part explainedby the fact that both types of NR compete for the sameheterodimerisation partner, RXR. Further investigations arerequired to show the importance of this effect. In conclusion,PPARs may be of importance for growth regulation innormal and malignant melanocytes and may represent atherapeutical target in malignant melanoma.

References

1 Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward Eet al: Cancer statistics, 2004. CA Cancer J Clin 54: 8-29, 2004.

2 Eigentler TK, Caroli UM, Radny P and Garbe C: Palliativetherapy of disseminated malignant melanoma: a systematic reviewof 41 randomised clinical trials. Lancet Oncol 4: 748-759, 2003.

3 Hauschild A, Weichenthal M, Balda BR, Becker JC, Wolff HH,Tilgen W et al: Prospective randomized trial of interferon α-2band interleukin-2 as adjuvant treatment for resectedintermediate- and high-risk primary melanoma without clinicallydetectable node metastasis. J Clin Oncol 21: 2883-2888, 2003.

4 Freudlsperger C, Moll I, Schumacher U and Thies A: Anti-proliferative effect of peroxisome proliferator-activated receptorγ agonists on human malignant melanoma cells in vitro.Anticancer Drugs 17: 325-323, 2006.

5 Grabacka M, Plonka PM, Urbanska K and Reiss K: Peroxisomeproliferator receptor α activation decreases metastatic potentialof melanoma cells in vitro via down-regulation of Akt. ClinCancer Res 12(10): 3028-3036, 2006.

6 Osborne JE and Hutchinson PE: Vitamin D and systemic cancer:Is this relevant to malignant melanoma? Br J Dermatol 147: 197-213, 2002.

7 Hansen CM, Binderup L, Hamberg KJ and Carlberg C: VitaminD and cancer: effects of 1,25(OH)2D3 and its analogs on growthcontrol and tumorigenesis. Front Biosci 6: 820-848, 2001.

8 Demierre MF and Nathanson L: Chemoprevention of melanoma:an unexplored strategy. J Clin Oncol 21: 158-165, 2003.

9 Dellavalle RP, Nicholas MK and Schilling LM: Melanomachemoprevention: a role for statins or fibrates? Am J Ther 10:203-210, 2003.

10 Rubins HB, Robins SJ, Collins D et al: Gemfibrozil for thesecondary prevention of coronary heart disease in men with lowlevels of high-density lipoprotein cholesterol. Veterans AffairsHigh-Density Lipoprotein Cholesterol Intervention Trial StudyGroup. N Engl J Med 341: 410-418, 1999.

11 Grabacka M, Placha W, Plonka PM et al: Inhibition ofmelanoma metastases by fenofibrate. Arch Dermatol Res 296:54-58, 2004.

12 Staels B, Van Tol A, Fruchart JC and Auwerx J: Effects ofhypolipidemic drugs on the expression of genes involved in highdensity lipoprotein metabolism in the rat. Ir J Med Sci 32: 490-498, 1996.

13 Staels B, Dallongeville J, Auwerx J, Schoonjans K, Leitersdorf Eand Fruchart JC: Mechanism of action of fibrates on lipid andlipoprotein metabolism. Circulation 98: 2088-2093, 1998.

14 Stumpf WE, Sar M, Reid FA, Tanaka Y and DeLuca HF: Targetcells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach,kidney, skin, pituitary and parathyroid. Science 209: 1189-1190,1979.

15 Baker AR, McDonnell DP and Hughes M: Cloning andexpression of full-length cDNA encoding human vitamin Dreceptor. Proc Natl Acad Sci USA 85: 3294-3298, 1988.

16 Kliewer SA, Umesono K, Mangelsdorf DJ and Evans RM:Retinoic X receptor interacts with nuclear receptors in retinoicacid, thyroid hormone and vitamin D3 signalling. Nature 355:446-449, 1992.

17 Evans RM: The steroid and thyroid hormone receptorsuperfamily. Science 240: 889-895, 1988.

18 Seifert M, Rech M, Meineke V, Tilgen W and Reichrath J:Differential biological effects of 1,25-dihydroxyvitamin D3 onmelanoma cell lines in vitro. J Steroid Biochem Mol Biol 89-90:375-379, 2004.

19 Danielsson C, Fehsel K, Polly P and Carlberg C: Differentialapoptotic response of human melanoma cells to 1 α,25-dihydroxyvitamin D3 and its analogues. Cell Death Differ 5:946-952, 1998.

20 Danielsson C, Torma H, Vahlquist A and Carlberg C: Positiveand negative interaction of 1,25-dihydroxyvitamin D3 and theretinoid CD437 in the induction of human melanoma cellapoptosis. Int J Cancer 81(3): 467-470, 1999.

21 Palmer HG, Gonzalez-Sancho JM, Espada J, Berciano MT, PuigI, Baulida J, Quintanilla M, Cano A, de Herreros AG, Lafarga Mand Munoz A: Vitamin D(3) promotes the differentiation of coloncarcinoma cells by the induction of E-cadherin and the inhibitionof beta-catenin signaling. J Cell Biol 154(2): 369-387, 2001.

22 Palmer HG, Larriba MJ, Garcia JM, Ordonez-Moran P, Pena C,Peiro S, Puig I, Rodriguez R, de la Fuente R, Bernad A, PollanM, Bonilla F, Gamallo C, de Herreros AG and Munoz A: Thetranscription factor SNAIL represses vitamin D receptorexpression and responsiveness in human colon cancer. Nat Med10(9): 917-919, 2004.

23 Peinado H, Olmeda D and Cano A: Snail, Zeb and bHLH factorsin tumour progression: an alliance against the epithelialphenotype? Nat Rev Cancer 7(6): 415-428, 2007.

24 Bronner-Fraser M: Neural crest cell migration in the developingembryo. Trends Cell Biol 3: 392-397, 1993.

25 Wahli W, Braissant O and Desvergne B: Peroxisome proliferator-activated receptors: transcriptional regulators of adipogenesis,lipid metabolism and more. Chem Biol 2: 261-266, 1995.

26 Desvergne B and Wahli W: PPAR: a key nuclear factor innutrient/gene interactions? In: Inducible Transcription. BauerleP (ed.), Birkhäuser, Boston vol 1 pp. 142-176, 1995.

27 Chawla A, Repa JJ, Evans RM and Mangelsdorf DJ: Nuclearreceptors and lipid physiology: opening the X-files. Science 294:1866-1870, 2001.

28 Escher P and Wahli W: Peroxisome proliferator-activatedreceptors: insight into multiple cellular functions. Mutat Res448: 121-138, 2000.

29 Amri EZ, Bonino F, Ailhaud G, Abumrad NA and Grimaldi PA:Cloning of a protein that mediates transcriptional effects of fattyacids in preadipocytes. Homology to peroxisome proliferator-activated receptors. J Biol Chem 270: 2367-2371, 1995.

30 Desvergne B and Wahli W: Peroxisome proliferator-activatedreceptors: nuclear control of metabolism. Endocr Rev 20: 649-688, 1999.


3657

31 Braissant O, Foufelle F, Scotto C, Dauca M and Wahli W:Differential expression of peroxisome proliferator-activatedreceptors (PPARs): tissue distribution of PPAR-α, -β, and -γ inthe adult rat. Endocrinology 137: 354-366, 1996.

32 Escher P, Braissant O, Basu-Modak S, Michalik L, Wahli W andDesvergne B: Rat PPARs: quantitative analysis in adult rattissues and regulation in fasting and refeeding. Endocrinology142: 4195-4202, 2001.

33 Michalik L, Desvergne B, Basu-Modak S, Tan NS and Wahli W:Nuclear hormone receptors and mouse skin homeostasis:implication of PPARβ. Horm Res 54: 263-268, 2000.

34 Tontonoz P, Hu E, Graves RA, Budavari AI and Spiegelman BM:mPPAR γ 2: tissue-specific regulator of an adipocyte enhancer.Genes Dev 8: 1224-1234, 1994.

35 Chawla A, Schwarz EJ, Dimaculangan DD and Lazar MA:Peroxisome proliferator-activated receptor (PPAR) γ: adipose-predominant expression and induction early in adipocytedifferentiation. Endocrinology 135: 798-800, 1994.

36 Tontonoz P, Hu E and Spiegelman BM: Stimulation ofadipogenesis in fibroblasts by PPAR γ 2, a lipid-activatedtranscription factor. Cell 79: 1147-1156, 1994.

37 Fajas L, Auboeuf D, Raspe E, Schoonjans K, Lefebvre AM,Saladin R, Najib J, Laville M, Fruchart JC, Deeb S, Vidal-PuigA, Flier J, Briggs MR, Steals B, Vidal H and Auwerx J: Theorganization, promoter analysis, and expression of the humanPPARγ gene. J Biol Chem 272: 18779-18789, 1997.

38 Mansen A, Guardiola-Diaz H, Rafter J, Branting C andGustafsson JA: Expression of the peroxisome proliferator-activated receptor (PPAR) in the mouse colonic mucosa.Biochem Biophys Res Commun 222: 844-851, 1996.

39 Lefebvre AM, Paulweber B, Fajas L, Woods J, McCary C,Colombel JF, Najib J, Fruchart JC, Datz C, Vidal H, DesreumauxP and Auwerx J: Peroxisome proliferator-activated receptor γ isinduced during differentiation of colon epithelium cells. JEndocrinol 162: 331-340, 1999.

40 Greene ME, Blumberg B, McBride OW, Yi HF, Kronquist K,Kwan K, Hsieh L, Greene G and Nimer SD: Isolation of humanperoxisome proliferator-activated receptor γ cDNA: expressionin hematopoietic cells and chromosomal mapping. Gene Expr 4:281-299, 1995.

41 Palmer CN, Hsu MH, Griffin KJ, Raucy JL and Johnson EF:Peroxisome proliferator activated receptor-α expression inhuman liver. Mol Pharmacol 53: 14-22, 1998.

42 Auboeuf D, Rieusset J, Fajas L, Vallier P, Frering V, Riou JP,Steals B, Auwerx J, Laville M and Vidal H: Tissue distributionand quantification of the expression of mRNAs of peroxisomeproliferator-activated receptors and liver X receptor-α inhumans: no alteration in adipose tissue of obese and NIDDMpatients. Diabetes 46: 1319-1327, 1997.

43 Gonzalez FJ: The role of peroxisome proliferator-activatedreceptor alpha in peroxisome proliferation, physiologicalhomeostasis, and chemical carcinogenesis. Adv Exp Med Biol422: 109-125, 1997.

44 Gottlicher M, Widmark E, Li Q and Gustafsson JA: Fatty acidsactivate a chimera of the clofibric acid-activated receptor and theglucocorticoid receptor. Proc Natl Acad Sci USA 89: 4653-4657,1992.

45 Forman BM, Chen J and Evans RM: Hypolipidemic drugs,polyunsaturated fatty acids, and eicosanoids are ligands forperoxisome proliferator-activated receptors α and δ. Proc NatlAcad Sci USA 94: 4312-4317, 1997.

46 Xu HE, Lambert MH, Montana VG, Parks DJ, Blanchard SG,Brown PJ, Sternbach DD, Lehmann JM, Wisely GB, WillsonTM, Kliewer SA and Milburn MV: Molecular recognition offatty acids by peroxisome proliferator-activated receptors. MolCell 3: 397-403, 1999.

47 Berger J and Moller DE: The mechanisms of action of PPARs.Annu Rev Med 53: 409-435, 2002.

48 Bookout AL, Jeong Y, Downes M, Yu RT, Evans RM andMangelsdorf DJ: Anatomical profiling of nuclear receptorexpression reveals a hierarchical transcriptional network. Cell126: 789-799, 2006.

49 Dunlop TW, Vaisanen S, Frank C, Molnar F, Sinkkonen L andCarlberg C: The human peroxisome proliferator-activated receptorδ gene is a primary target of 1alpha,25-dihydroxyvitamin D3 andits nuclear receptor. J Mol Biol 349(2): 248-260, 2005.

50 Carey TE, Takahashi T, Resnick LA, Oettgen HF and Old LJ:Cell surface antigens of human malignant melanoma; mixedhemadsorption assays for humoral immunity to culturedautologous melanoma cells. Proc Natl Acad Sci USA 73: 3278-3282, 1976.

51 Reichrath J, Rech M, Moeini M, Meese E, Tilgen W and SeifertM: In vitro comparison of the vitamin D endocrine system in1,25(OH)2D3-responsive and -resistant melanoma cells. CancerBiol Ther 6(1): 48-55, 2007.

52 Kliewer SA, Xu HE, Lambert MH and Willson TM: Peroxisomeproliferator-activated receptors: From genes to physiology.Recent Prog Horm Res 56: 239-263, 2001.

53 Kippenberger S, Loitsch SM, Grundmann-Kollmann M, SimonS, Dang TA, Hardt-Weinelt K, Kaufmann R and Bernd A:Activators of peroxisome proliferator-activated receptors protecthuman skin from ultraviolet-B-light-induced inflammation.J Invest Dermatol 117: 1430-1436, 2001.

54 Matthaei S, Stumvoll M and Häring HU: Thiazolidinedione(Insulinsensitizer): Neue Aspekte in der Therapie des Diabetesmellitus Typ 2. Dt Ärztebl 98: 912-918, 2001.

Received January 29, 2009Revised May 26, 2009

Accepted June 25, 2009


3658

Documents

Cross-talk between vitamin D receptor (VDR)- and peroxisome proliferator-activated receptor (PPAR)-signaling in melanoma cells