Previous results indicate that the polyphenol resveratrol inhibits cell growth of colon carcinoma cells via modulation of polyamine metabolic key enzymes. The aim of this work was to specify the underlying molecular mechanisms and to identify a possible role of transcription factor peroxisome proliferator–activated receptor γ (PPARγ). Cell growth was determined by bromodeoxyuridine incorporation and crystal violet staining. Protein levels were examined by Western blot analysis. Spermine/spermidine acetyltransferase (SSAT) activity was determined by a radiochemical assay. PPARγ ligand–dependent transcriptional activity was measured by a luciferase assay. A dominant-negative PPARγ mutant was transfected in Caco-2 cells to suppress PPARγ-mediated functions. Resveratrol inhibits cell growth of both Caco-2 and HCT-116 cells in a dose- and time-dependent manner (P < 0.001). In contrast to Caco-2-wild type cells (P < 0.05), resveratrol failed to increase SSAT activity in dominant-negative PPARγ cells. PPARγ involvement was further confirmed via ligand-dependent activation (P < 0.01) as well as by induction of cytokeratin 20 (P < 0.001) after resveratrol treatment. Coincubation with SB203580 abolished SSAT activation significantly in Caco-2 (P < 0.05) and HCT-116 (P < 0.01) cells. The involvement of p38 mitogen-activated protein kinase (MAPK) was further confirmed by a resveratrol-mediated phosphorylation of p38 protein in both cell lines. Resveratrol further increased the expression of PPARγ coactivator PGC-1α (P < 0.05) as well as SIRT1 (P < 0.01) in a dose-dependent manner after 24 hours of incubation. Based on our findings, p38 MAPK and transcription factor PPARγ can be considered as molecular targets of resveratrol in the regulation of cell proliferation and SSAT activity, respectively, in a cell culture model of colon cancer. (Cancer Res 2006; 66(14): 7348-54)
- colon cancer
- polyamine metabolism
Resveratrol is a naturally occurring polyphenol present in red wine, peanuts, and grapes ( 1, 2). It has been speculated that dietary resveratrol could be an explanation for the so-called “French paradox,” as it exhibits multiple cardioprotective properties ( 3, 4). Furthermore, we and others reported potent chemopreventive effects of resveratrol and its analogues in various carcinogenesis models ( 5– 8). The polyamines spermidine and spermine as well as their precursor putrescine are essential for normal cell growth, development, and tissue repair ( 9, 10). Correlation of excess polyamine levels with cancer was first reported in the late 1960s, when Russel and Snyder ( 11) reported high levels of ornithine decarboxylase activity, the pivotal enzyme of polyamine biosynthesis, in several human cancers. In colorectal cancer tissue, polyamine contents are increased 3- to 4-fold over that found in the equivalent normal colonic tissue ( 12, 13). Based on these findings, pharmacologic or natural inhibitors of polyamine metabolism have been studied in vitro ( 13, 14) and in vivo ( 15) as new potent therapeutic strategies in cancer treatment and prevention. Peroxisome proliferator–activated receptors (PPARs) are ligand-inducible transcription factors belonging to the nuclear hormone receptor superfamily ( 16, 17), which regulate transcription of target genes by heterodimerizing with the retinoid X receptor and binding to PPAR response elements ( 16, 18). PPARγ is expressed at high levels in colonic epithelial cells and colon cancer cells ( 19). Girnun and Spiegelman ( 20) hypothesize that PPARγ is exerting its effects early in the carcinogenic process by suppressing tumor formation. Activation of PPARγ therefore could function as an important molecular target of chemopreventive agents such as resveratrol. Recently, Babbar et al. ( 21) identified two PPAR response elements in the promoter of the spermine/spermidine acetyltransferase (SSAT) gene. Based on our former findings that resveratrol induces cell growth inhibition of colon cancer cells via induction of catabolic enzyme SSAT ( 13), the aim of this work was to specify the underlying molecular mechanisms and to identify a possible role of transcription factor PPARγ.
Materials and Methods
Cell Culture and Materials
Caco-2 cells of passages 53 to 61 were kept in DMEM, supplemented with 10% FCS, 1% penicillin/streptomycin, 1% sodium pyruvate, and 1% nonessential amino acids. HCT-116 cells of passages 17 to 30 were cultured in McCoy's 5A supplemented with 10% FCS and 1% penicillin/streptomycin. Both cell lines were maintained at 37°C in an atmosphere of 95% air and 5% CO2. Cos7 cells were cultured in DMEM high-glucose supplemented with 10% FCS containing 100 units/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L glutamine, and 1 mmol/L sodium pyruvate at 37°C and 10% CO2. The cells were passaged weekly using Dulbecco's PBS containing 0.25% trypsin and 1% EDTA. The medium was changed thrice per week. Cells were screened for possible contamination with mycoplasma at monthly intervals. For experiments, the cells were seeded onto plastic cell culture wells in serum-containing medium and allowed to attach for 24 hours. For the SSAT activity assay, the cells were synchronized in medium containing 1% FCS 24 hours before treatment. WY-14643, pioglitazone HCl, L-165,041, GW7647, and SB203580 were obtained from Calbiochem (San Diego, CA); resveratrol was obtained from Sigma-Aldrich; FCS and DMSO were obtained from Sigma; DMEM and Optimem I from Life Technologies, Inc.; sodium pyruvate solution, glutamine, penicillin, and streptomycin stock solutions from PAA Laboratories GmbH; Lipofectamine 2000 from Invitrogen; and Dual-glo Luciferase Assay system from Promega.
SDS-PAGE and Immunoblot Analysis
Caco-2 cells were seeded in 80-cm2 flasks; 24 hours after plating, cells were incubated with substances for different time intervals. Cytosolic and nuclear extracts were obtained according to the instructions of the manufacturer (Active Motif, Rixensart, Belgium). Protein was quantified with the Bio-Rad protein colorimetric assay. After addition of sample buffer to the cellular extract and boiling samples at 95°C for 5 minutes, protein was separated on 10% SDS-polyacrylamide gel. Protein was transferred onto nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany) and the membrane was blocked for 1 hour at room temperature with 3% skim milk in Tris-buffered saline containing 0.05% Tween 20 (TBST). Next, blots were washed and incubated overnight at 4°C in TBST containing either 5% bovine serum albumin or 3% milk skim powder with a 1:1,000 or 1:500 dilution of primary antibodies for p38 and phospho-p38 (all from Cell Signaling, Beverly, MA), SIRT1, PGC-1α, and cytokeratin 20 (all from Santa Cruz Biotechnology, Santa Cruz, CA), and PPARγ (Calbiochem). The horseradish peroxidase–conjugated secondary antibody (Santa Cruz Biotechnology) was diluted at 1:2,000 and incubated with the membrane for another 30 minutes in skim milk. After chemiluminescence reaction (ECL; Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom), bands were detected after exposure to Hyperfilm-MP (Amersham International plc, Buckinghamshire, United Kingdom). Blots were reprobed with β-actin antibody (Santa Cruz Biotechnology). For quantitative analysis, bands were detected and evaluated densitometrically with ProViDoc system (Desaga, Wiesloch, Germany), normalized for β-actin density.
Cells were suspended and cultured on 96-well dishes at a density of 104 per well (0.28 cm2). Twenty-four hours after plating, cells were incubated for 24 to 72 hours with substances. At given time points following treatment, cell numbers were assessed by crystal violet staining. Medium was removed from the plates and cells were fixed with 5% formaldehyde for 5 minutes. After washing with PBS, cells were stained with 0.5% crystal violet for 10 minutes, washed again with PBS, and unstained with 33% acetic acid. Absorption, which correlates with the cell number, was measured at 620 nm.
The effects of resveratrol on DNA synthesis of cells were assessed using the cell proliferation ELISA kit (Roche Diagnostics, Tokyo, Japan). This assay is a colorimetric immunoassay for quantification of cell proliferation based on the measurement of bromodeoxyuridine (BrdUrd) incorporation during DNA synthesis and is a nonradioactive alternative to the [3H]thymidine incorporation assay. Cells were grown in 96-well culture dishes (104 per well), incubated with resveratrol for different time intervals, and then labeled with BrdUrd for a further 4 hours. Incorporated BrdUrd was measured colorimetrically.
SSAT Enzyme Activity Determination
Cells were washed twice with cold homogenizing buffer [10 mmol/L Tris-HCl (pH 7.5), 2.5 mmol/L DTT, 1 mmol/L EDTA], harvested by scraping, disrupted by sonification, and centrifuged at 15,000 × g at 4°C for 15 minutes. The radiochemical assay of the SSAT activity was done by the estimation of labeled N1-acetylspermidine synthesized from [14C]acetyl-CoA (Hartman Analytic, Braunschweig, Germany) and unlabeled spermidine (0.3 μmol/L) as described earlier ( 13).
The following plasmids were used for transfection: pcDNA3 (Invitrogen), as an empty vector for control transfection, and plasmid pcDNA3-PPARγL468A/E471A, a dominant-negative double mutant, which was kindly provided by V.K. Chatterjee (Department of Medicine, University of Cambridge, Addenbrooke's Hospital, Cambridge, United Kingdom; ref. 22). These constructs were transfected into subconfluent Caco-2 cells with Lipofectamine 2000 (Invitrogen). After 6 hours, the cells were fed with fresh medium containing 10% FCS. Twenty-four hours later, the cells were fed with medium containing G418 (400 μg/mL) and culture medium was replaced twice a week. G418-resistant colonies were collected and used for further analysis.
Gal4-PPARγ Transactivation Assay
Plasmids. The Gal4-fusion receptor plasmid pFA-CMV-PPARγ-LBD, containing hinge region and the LBD of PPARγ, was constructed by integrating cDNA fragment obtained from PCR amplification of human monocytes into the SmaI/XbaI (Promega) sites of the pFA-CMV vector (Stratagene). The cDNA fragment contained bps 610-1,518 (NM_015869) of the PPARγ coding sequence. Frame and sequence of the fusion receptors were verified by sequencing. As reporter plasmids, we used pFR-Luc (Stratagene). For normalizing of transfection efficacy, we used pRL-SV40 (Promega).
Transfection. Cos7 cells were seeded at 30,000 per well in a 96-well plate. After 24 hours, transfection was carried out using Lipofectamine 2000 according to the protocol of the manufacturer. Transfection mixes contained 0.8 μL LF2000, 280 ng pFR-Luc, 2 ng pRL-SV40, and 14 ng of the PPARγ fusion receptor plasmid for each well. Four hours after transfection, medium was changed to DMEM without phenol red–containing 100 units/mL penicillin, 100 μg/mL streptomycin, 2 mmol/L glutamine, 1 mmol/L sodium pyruvate, appropriate concentration of test substance, and 0.1% DMSO, testing each concentration in triplicate wells. Cells were incubated overnight and assayed for reporter gene activity with the Dual-Glo Luciferase Assay system. Luminescence of both luciferases was measured in GENiosPro Luminometer (Tecan). Each assay was repeated at least thrice.
Calculations. Luciferase activity for all assays was corrected by subtracting background activity obtained from nontransfected controls. Relative light units were calculated by dividing firefly light units by renilla light units. Activation factors are gained by dividing mean values of relative light units for each concentration of agonist by mean relative light unit values of the DMSO control. Relative activation is calculated by dividing activation factors by the maximum activation factor. Calculation of EC50 values was done using the four-parameter logistic regression function of SigmaPlot2001 (SPSS, Inc.) using the mean of relative activation for each tested concentration of at least three determinations.
The data are expressed as means ± SE of at least three independent experiments. ANOVA was done when more than two groups were compared, and when significant (P < 0.05), multiple comparisons were done with the Turkey test. P < 0.05 was considered to be significant.
Effects of resveratrol on cell proliferation and cell counts. Caco-2 and HCT-116-cells were incubated with increasing concentrations of resveratrol (30-200 μmol/L) for 24, 48, and 72 hours. After each time interval, both cell proliferation ELISA (BrdUrd) and crystal violet staining were done. In HCT-116 cells, a significant time- and dose-dependent decrease in cell proliferation and cell counts could be measured ( Fig. 1 ). The same effects could be observed in Caco-2 cells, which is in accordance to our earlier studies (ref. 6; data not shown).
The role of PPARγ in resveratrol-induced activation of SSAT. Next, we examined the effects of resveratrol (50-100 μmol/L) on SSAT activity in Caco-2-wild type cells compared with Caco-2-empty vector and Caco-2-dnPPARγ mutant cells to investigate effects mediated by PPARγ. Resveratrol (100 μmol/L) leads to a significant increase of SSAT activity (P < 0.05 versus control) in Caco-2-wild type cells after 24 hours of incubation, which is in agreement with our previous data ( 13). In Caco-2-empty vector cells, resveratrol (100 μmol/L) also significantly increases SSAT-activity (P < 0.05 versus control) whereas no effects could be observed when PPARγ-mediated functions are suppressed in Caco-2-dnPPARγ mutant cells ( Fig. 2 ).
Effects of resveratrol on PPARγ transcriptional activity. To investigate the effects of resveratrol (50-100 μmol/L) on PPARγ ligand–dependent activity, we did a chimeric Gal4-PPARγ transactivation assay. Because the chimeric receptor contained only hinge region and ligand binding domain of the PPARγ, any effect of resveratrol affecting kinase-sensitive AF1 domain was ruled out. After incubation with resveratrol (100 μmol/L), we could generate similar effects of PPARγ agonist pioglitazone on PPARγ activity (P < 0.01; Fig. 3A ). To show evidence of resveratrol ability to increase PPARγ activity, we measured, after resveratrol treatment, the expression of cytokeratin 20, which is described to be a specific target gene of PPARγ activity in colorectal cancer cells ( 23). Incubation with resveratrol (30-100 μmol/L) led to an ∼40% increase of cytokeratin 20 expression at 100 μmol/L after 72 hours (P < 0.001; Fig. 3B).
Effects of resveratrol on PPARγ protein levels. We further did Western blot analysis to determine possible effects of resveratrol on translational level. However, no significant changes in PPARγ protein expression could be detected (data not shown).
Involvement of mitogen-activated protein kinase p38 in resveratrol-induced inhibition of cell proliferation and SSAT activation. There are several lines of evidence that resveratrol mediates its chemopreventive actions via modulation of mitogen-activated protein kinase (MAPK) pathways. To examine p38 MAPK–mediated actions, we used the specific inhibitor SB203580. This anti-inflammatory drug inhibits the catalytic activity of p38 MAPK by competitive binding in the ATP pocket ( 24). As shown in Fig. 4 , incubation with resveratrol (50-200 μmol/L) augmented phosphorylated p38 in a time- and dose-dependent manner, both in Caco-2 and HCT-116 cells (∼300% at 200 μmol/L after 16 hours; P < 0.01), whereas p38 MAPK concentration remained unaffected. To characterize the role of p38 activation in resveratrol-mediated induction of SSAT, we pretreated Caco-2 and HCT-116 with p38 inhibitor SB203580 (10-20 μmol/L) for 1 hour and then added resveratrol (100 μmol/L) for another 24 hours. Both in Caco-2-( Fig. 5A ) and HCT-116-cells ( Fig. 5B), coincubation with SB203580 significantly diminished resveratrol-induced SSAT activation [P < 0.05 versus resveratrol (100 μmol/L) in Caco-2; P < 0.01 versus resveratrol (100 μmol/L) in HCT-116].
Effects of resveratrol on PGC-1α and SIRT1 expression. Western blot analysis was done to determine possible effects of resveratrol on the expression of PPARγ coactivator 1α and sirtuin homologue SIRT1, which exhibits PPARγ-suppressive effects in white adipocyte tissue. Resveratrol (50-200 μmol/L) led to a significant dose-dependent increase in both PGC-1α (∼60% at 200 μmol/L; P < 0.05; Fig. 6A ) and SIRT1 (∼140% at 200 μmol/L; P < 0.01; Fig. 6B) expression after 24 hours of incubation.
The present study clearly shows that resveratrol mediates growth inhibitory effects in colorectal cancer cell lines, at least partly, via polyamine degradation, whereas activation of transcription factor PPARγ seems to play a pivotal role. The plant polyphenol resveratrol (3,4′,5-trihydroxystilbene) exhibits multiple chemopreventive effects comprising cell growth inhibition ( 6, 25), induction of apoptosis ( 26), and prevention of angiogenesis ( 27), whereby the underlying molecular mechanisms are only partly understood ( 28, 29). Intracellular polyamine levels are maintained within very narrow limits because decreases of polyamine concentrations interfere with cell growth whereas an excess seems to be toxic ( 30). The three key enzymes of polyamine metabolism are ornithine decarboxylase and S-adenosylmethionine decarboxylase, the rate-limiting enzymes of polyamine biosynthesis, and SSAT, which controls polyamine catabolism ( 31). Wolter et al. showed that resveratrol-induced growth arrest of Caco-2 cells is accompanied by inhibition of polyamine biosynthesis as well as activation of polyamine catabolism ( 13). The peroxisome proliferator–activated receptor γ (PPARγ) is a nuclear receptor that acts as a transcription factor controlling the expression of multiple genes involved in cell growth, differentiation, and apoptosis of several malignant cell lines, and therefore seems to play a crucial role in carcinogenesis ( 32, 33). To abolish PPARγ-mediated functions, we transfected a dominant-negative mutant in Caco-2 cells. In this PPARγ mutant, two charged amino acid residues (Leu468 and Glu471) in helix 12 of the ligand binding domain are mutated to alanine, whereupon the mutant retains ligand and DNA binding but exhibits markedly reduced transactivation due to impaired coactivator recruitment ( 22). According to the current findings, an essential role for PPARγ in enhancing SSAT enzyme activity is assumed ( 21). In fact, we could show that, in contrast to Caco-2-wild type and Caco-2-empty vector cells, resveratrol failed to increase SSAT activity in Caco-2-dnPPARγ cells. The mitogen-activated protein kinase (MAPK) pathways have been recognized as a major signaling pathway by which cells transduce extracellular signals into an intracellular response. In several studies, resveratrol was shown to mediate multiple functions by modulating MAPK pathways ( 26, 34, 35). We could show as well that incubation with resveratrol causes phosphorylation, and thus activation, of p38 MAPK in colon cancer cells. Furthermore, combination of resveratrol with an inhibitor of p38 MAPK leads to an inhibition of resveratrol-induced SSAT activation both in Caco-2 and HCT-116 cells. Consequently, an activation of MAPK cascade by resveratrol can be assumed. As previously described, our results point out an important role of PPARγ in resveratrol-mediated actions whereas resveratrol-dependent PPARγ activation seems to be mediated at least partly by an activation of the ligand binding domain (LBD/AF2) because a Gal4-PPARγ chimeric receptor was activated by resveratrol at a concentration of 100 μmol/L, and this concentration was sufficient to induce SSAT as well. In addition, our results suggest that activation of PPARγ by resveratrol is due to kinase activation, leading to phosphorylation-dependent activation of PPARγ coactivators like PGC-1α ( 36). Coactivators all interact with a similar surface of the activated ligand binding domain of the receptors and have been suggested to mediate their transcriptional activity ( 37). It is well established that, in addition to transcription factors, coactivators can also be targets of multiple signal transduction pathways in response to different stimuli ( 38). Puigserver et al. ( 39) could show that PGC-1α is activated through p38 MAPK. The mechanism by which p38 activates PGC-1α is not yet clear, but it is suggested that p38 MAPK–mediated phosphorylation counteracts repressor effects, possibly by encouraging the release of a repressor from PGC-1α ( 40). On activation, PGC-1α docks on PPARγ and thus can modulate its transcriptional activity ( 41). In addition to PGC-1α, resveratrol further leads to an activation of SIRT1, a member of the silent information regulator 2 (Sir2) family of proteins (sirtuins; ref. 42). SIRT1 is mainly linked to negative regulation of gene expression as a cofactor through protein deacetylation ( 43). However, there is evidence that SIRT1 can act positively and negatively to control gene expression as a cofactor for PGC-1α. These opposite effects could possibly be due to the recruitment of a different set of coactivators and corepressors through PGC-1α/SIRT1 ( 44). This could further be an explanation for the repressive effects of SIRT1 on PPARγ in white fat where PGC-1α is very low ( 45). In summary, our data confirm our earlier studies showing that resveratrol-mediated growth inhibition of colorectal cancer cells seems to involve SSAT-induced polyamine catabolism. Here we further show that transcription factor PPARγ acts as a p38-dependent target in resveratrol-induced molecular mechanisms (see Fig. 7 ). Besides a decrease in cell proliferation, the observed reduction of BrdUrd incorporation is probably due to an induction of apoptosis, as Wolter et al. ( 6) showed an obvious increase of caspase-3-activity in resveratrol-treated cells. Recent studies further indicate that the activation of catabolic SSAT is related to an induction of programmed cell death (46). Further projects directing towards these aspects of resveratrol action are going to proceed.
Grant support: Deutsche Forschungsgemeinschaft grant GRK 757 (S. Ulrich) and the Else-Kröner Fresenius Foundation, Bad Homburg, Germany.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Received August 15, 2005.
- Revision received April 3, 2006.
- Accepted April 28, 2006.
- ©2006 American Association for Cancer Research.