Anticarcinogenic Effect of a Flavonoid Antioxidant, Silymarin, in Human Breast Cancer Cells MDA-MB 468: Induction of G1 Arrest through an Increase in Cipl/p21 Concomitant with a Decrease in Kinase Activity of Cyclin-dependent Kinases and Associated Cyclins1

Abstract

There is an increasing interest in identifying potent cancer preventive and therapeutic agents against breast cancer. Silymarin, a flavonoid antioxidant isolated from milk thistle, exerts exceptionably high to complete anticarcinogenic effects in tumorigenesis models of epithe!ial origin. In this study, we investigated the anticarcinogenic effect of silymarin and associated molecular mechanisms, using human breast carcinoma cells MDA-MB 468. Silymarin treatment resulted in a significantly high to complete inhibition of both anchorage-dependent and anchorage-independent cell growth in a doseand time-dependent manner. The inhibitory effects of silymarin on cell growth and proliferation were associated with a G1 arrest in cell cycle progression concomitant with an induction of up to 19-fold in the protein expression of cyclin-dependent kinase (CDK) inhibitor Cipl/p21. Following silymarin treatment oT cells, an incremental binding of Cipl/p21 with CDK2 and CDK6 paralleled a significant decrease in CDK2-, CDK6-, cyclin Dl-, and cyclin E-associated kinase activity with no change in CDK2 and CDK6 expression but a decrease in G1 cyclins Dl and E. Taken together, these results suggest that silymarim may exert a strong anticarcinogenic effect against breast cancer and that this effect possibly involves an induction of Cipl/p21 by silymarin, which inhibits the threshold kinase activities of CDKS and associated cyclins, leading to a G1 arrest in cell cycle progression. Received 11/18/97; revised 1/12/98; accepted 1/12/98. 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. C This work was supported in part by United States Public Health Service Grant CA 645 14 (to R. A.). 2 To whom requests for reprints should be addressed, at Center for Cancer Causation and Prevention, AMC Cancer Research Center, 1600 Pierce Street, Denver, CO 80214. Phone: (303) 239-3561; Fax: (303) 233-9562; E-mail: agarwalr@amc.org. INTRODUCTION Breast cancer is the most common nonskin malignancy and the second leading cause of cancer deaths in women in the United States (1). One approach to controlling breast cancer is chemopreventive intervention, a means of cancer control in which the disease is prevented, slowed, or reversed by the administration of one or a combination of naturally occurring or synthetic compounds. Several studies have shown that microchemicals present in the diet, several herbs, and plants are the most desirable class of agents for the prevention and/or intervention of various cancers (Refs. 2-8 and references therein). Among these, po]yphenolic antioxidants are receiving increasing attention in recent years (5-8). One such naturally occurring polyphenolic flavonoid antioxidant is silymarin, isolated from milk thistle (Silybum marianum (L.) Gaertn; Ref. 9). Silymarin is accepted for use in humans because it is used clinically in Europe and Asia for the treatment of alcoholic liver diseases (10). As a therapeutic agent, it is well tolerated and largely free from adverse effects, it is nontoxic in acute, subchronic, and chronic tests even at large doses, and there is no known LD50 for this agent (Ref. 1 1 and references therein). Silymarin protects against hepatotoxicity and lipid peroxidation induced by several xenobiotic agents (9-13) and has been shown to be a strong antioxidant capable of scavenging free radicals (Refs. 14-16 and references therein). Limited shortterm studies have suggested that sibymarin may be a potent anticarcinogenic agent (17, 18). In our recent studies, we showed that silymarin exerts exceptionally high to complete protective effects against carcinogenesis in different mouse tumorigenicity models of epithelial origin (19, 20). Defects in the regulation of cell cycle progression are thought to be one of the commonest features of transformed cells (2 1 ). Eukaroytic cell cycle progression is regulated by sequential activation and subsequent inactivation of a series of CDKs3 at different phases (22, 23). The activities of CDKs are positively regulated by cyclins (24); there are five major types of cyclins, namely A, B, C, D, and E, which act at different checkpoints of cell cycle (2 1 ). Cyclins D and E are required for progression through G1 (25). As cells enter G1, the CDK4and/or CDK6-cyclin D complex appears to be necessary for transition through early G (23), whereas CDK2-cyclin E com3 The abbreviations used are: CDK, cyclin-dependent kinase; CDKI, CDK inhibitor; RB, retinoblastoma; FACS, fluorescence-activated cell sorting. Research. on July 15, 2017. © 1998 American Association for Cancer clincancerres.aacrjournals.org Downloaded from 1056 Anticarcinogenic Effect of Si!ymarin in Breast Carcinoma Cells plex is involved in transition from late G1 into S phase (26, 27). Whereas CDK activity is negatively regulated by CDKJs (24, 28), CDKI proteins are often absent or inactive in cancerous cells (Refs. 29 and 30 and references therein). At least seven CDKIs have been identified in mammalian cells, and on the basis of homology, they fall into the p2! and p16 families, each of which comprises several members (24, 28). The p21 family includes p21 (also known as Cipl, WAF1, and so forth; Ref. 31), Kipl/p2’l (32, 33), and Kip2/pS7 (34, 35). These CDKIs preferentially bind to and inhibit most CDKcyclin complexes (21, 28, 30). Out of several CDKIs, Cipl/p2l is an important mediator of cell cycle arrest imposed by p53 in response to DNA damage (31, 36, 37). However, Cipl/p21 expression can also be induced by other factors in a p53independent fashion (38-40). Several in vitro studies have demonstrated that relative levels of Cipl/p21 may be critical in determining the threshold kinase activity of various CDK-cyclin complexes (41, 42), suggesting that appropriate levels of Cipl/ p21 may be critical in cell growth regulation. When these data are taken together, it can be appreciated that the agents that overcome the loss of tumor suppressor genes and/or the presence of aberrant endogenous cell cycle regulation that contributes to the tumorigenic process should be useful for the preventive intervention of cancer. Using MDA-MB 468 cells, which are estrogen receptor-negative breast carcinoma cells that overexpress a transcriptionally active mutant p53 protein (43, 44), in this study, we demonstrate the anticarcinogenic effect of silymarin and suggest the involvement of G1 phase of the cell cycle and regulatory proteins associated with it as a possible molecular mechanism of the effect of silymarmn. MATERIALS AND METHODS Cell Line and Other Reagents. The MDA-MB 468 cell line was obtained from the American Type Culture Collection (Rockville, MD). Cells were grown in improved minimum essential Eagle’s medium with 2 mr i glutamine (Biofluids, Rockville, MD), 10% fetal bovine serum, 100 units/mI penicillin, and 100 pg/mI streptomycin at 37#{176}C in a 5% CO2 atmosphere. Silymarin was purchased from Aldrich Chemical Co. (Milwaukee, WI) and dissolved in ethanol. The final concentration of ethanol in culture medium during silymarin treatment did not exceed 0.5% (vlv), and therefore, the same concentration of ethanol was present in control dishes. Anti-Cip!/p21 antibody was from Calbiochem (Cambridge, MA). Anti-Kipl/p27 and anti-CDK4 antibodies were from Neomarkers, Inc. (Fremont, CA). Antibodies to cycbin Dl, cyclin E, CDK2, and CDK6; rabbit antimouse immunoglobulin-horseradish peroxidase-conjugated and goat antirabbit immunoglobulin-horseradish peroxidase-conjugated secondary antibodies; and RB-glutathione Stransferase fusion protein were purchased from Santa Cruz Biotechnology, Inc (Santa Cruz, CA). Histone Hl was from Boehringer Mannheim Corp. (Indianapolis, IN). [ y-32P]ATP (specific activity, 3000 Ci/mmo!) was from New England Nuclear (Boston, MA). ECL detection system was from Amersham Corp. (Arlington Heights, IL). Cell Growth Assay. For the anchorage-dependent cell growth assay, MDA-MB 468 cells were plated at 0.5 X i0 cells/60-mm plate under the culture conditions detailed above. On day 2, cells were fed with fresh medium and either left untreated or treated with ethanol alone or silymarin at a dosage of 10, 25, 50, or 75 pg/mI of medium dissolved in ethanol. The cultures were fed with fresh medium with or without same concentrations of silymarin every alternate day up to the end of the experiment; each treatment and time point had four plates. At days 1-5 of these treatments, cells were trypsinized and collected in counting vials. Each plate was washed thoroughly with isotonic buffer containing 0.1% formalin, and washings were collected in the original vials with trypsinized cells. Each vial was counted in a Coulter counter to determine the total number of cells. In a parallel experiment, at various experimental intervals, the number of viable cells was also determined by the trypan blue dye exclusion test. Soft Agar Colony Formation Assay. MDA-MB 468 cells were cultured under the conditions detailed above. For anchorage-independent cell growth, a soft agar colony formation assay was performed using 6-well plates. Each well contamed 2 ml of 0.5% agar in medium as the bottom layer, 1 ml of 0.38% agar in medium and 2000 cells as the feeder layer, and 1 ml of 0.38% agar in medium either with vehicle ethanol or with different doses of silymarin in ethanol as the top layer. Each treatment had three wells. Cultures were maintained at 37#{176}Cin a humidified 5% CO2 atmosphere. The number of colonies was determined by counting them under an inverted phase-contrast microscope at X 100 magnification; a group of more than 10 cells was counted as a colony. The rate of colony growth and the optimum time for scoring colonies was assessed by counting colonies at 5, 7, 10, 15, 20, and 30 days and was found to be optimal at day 10. The wells were also examined on day 1 to eliminate possible artifacts caused by any clumps of cells. Cell Cycle Analysis. MDA-MB 468 cells at 80-90% confluency were either untreated or treated with ethanol alone or varying concentrations of silymarin (25-75 pg/mi of medium) in ethanol. Twenty-four, 48, and 72 h after these treatments, medium was aspirated, monolayers were quickly washed twice with cold PBS, cells were trypsinized, and cell pellets were collected. The cells were washed twice with PBS, fixed in cold methanol, and rewashed with PBS to remove methanol. After being suspended in 500 p1 of PBS, cells were digested with 20 pglml RNase at 37#{176}C for 30 mm and chilled on ice for 10 mm, and then cellular DNA was stained with propidium iodide (50 pg/ml) by incubation for 1 h at room temperature in the dark. Cell cycle distribution was analyzed by flow cytometry using the Becton Dickinson FACS system. Silymarin Treatment of Cells for Cell Cycle Regulatory Molecules. MDA-MB 468 cells were grown as detailed above, and 80-90% confluent cultures were treated with either ethanol alone or varying concentrations of silymarin (10-75 pg/mi of medium) in ethanol. Sixteen h after these treatments, medium was aspirated, cells were quickly washed twice with cold PBS, and 0.5 ml lysis buffer (10 m Tris-HC1, pH 7.4, 150 mM NaC!, 1% Triton X-100, 1 rnr,i EDTA, 1 nm EGTA, 0.2 mi i sodium vanadate, 0.2 m t phenylmethylsulfonyl fluoride, 0.5% NP4O, 1 pg/rn] pepstatin, 0.2 units/mi aprotinin) was added to each plate. After 15 mm in lysis buffer at 4#{176}C, the cell lysate was scraped from the plate, collected in microcentrifuge tubes and left on ice for additional 15 ruin. The lysates were cleared by centrifugation for 5 mm in a table-top centrifuge at 4#{176}C, the Research. on July 15, 2017. © 1998 American Association for Cancer clincancerres.aacrjournals.org Downloaded from Clinical Cancer Research 1057 clear supernatants were collected, and protein concentration was determined. For the studies assessing the time-dependent effect of silymarin, cultures were treated with either ethanol or silymain at a concentration of 75 ji.g/ml of medium, cells were harvested 4, 8, 16, 24, and 48 h later, and cell lysates prepared as described above. Western hnmunob!otting and Immunoprecipitation. For Western immunoblotting, 40-100 ig of protein lysate per sample were denatured with 2 X SDS sample buffer, samples were subjected to 12% SDS-PAGE, and separated proteins were transferred on to membrane. The bevels of Cipl/p21, Kipl/p2’7, CDK2, CDK4, CDK6, cyclin Dl, and cyclin E were determined using specific primary antibodies followed by peroxidase-conjugated appropriate secondary antibody and visualization by the ECL detection system. For studies evaluating the binding of CDKs with cyclins, 50 p.g of protein lysate was mixed with 0.5 i.g of anti-CDK2 or anti-CDK6 antibody, and then this mixture was rotated at 4#{176}C for 4 h. Thereafter, 5 p.] of protein A-agarose beads were added, and this mixture was incubated overnight at 4#{176}C. The next day, beads were collected by centrifugation and washed four times with lysis buffer, and the immunoprecipitated CDK2 or CDK6 was denatured with 30 i.l of 1 X SDS sample buffer. The proteins were subjected to SDS-PAGE on 12% gel, and the separated proteins were transferred on membrane by Western blotting. The membrane was probed with anti-cyclin E antibody in the case of CDK2 or anti-cyclin Dl antibody in the case of CDK6 followed by peroxidase-conjugated appropriate secondary antibody and visualization by ECL detection system. Similarly, the levels of CDK2 and CDK6 bound to Cipl/p2l were determined by immunoprecipitating Cipl/p21 using antiCipl/p21 antibody and 5 i.l of protein G-agarose beads, and following SDS-PAGE and Western blotting, the membranes were probed with specific primary antibody to CDK2 or CDK6 followed by peroxidase-conjugated appropriate secondary antibody and visualization by the ECL detection system. Kinase Activity Assays. CDK2and cyclin E-associated Hi histone kinase activity was determined as described by Wu et a!. (45). Briefly, using anti-CDK2 or anti-cyclin E antibody (2 J.g) and protein A-agarose beads (20 pA), CDK2 and cyclin E were immunoprecipitated from 200 p.g of protein lysate per sample as detailed above. Beads were washed three times with lysis buffer and then once with kinase assay buffer (50 mrvi Tris-HC1, pH 7.4, 10 ms i MgCl2, and 1 ms t DY!’). Phosphorylation of histone Hi was measured by incubating the beads with 40 p.1 of “hot” kinase solution [0.25 i.l (2.5 p.g) of histone Hi, 0.5 il of y-32P]ATP, 0.5 p.! of 0. 1 m i Al?, and 38.75 a! of kinase buffer] for 30 mm at 37#{176}C. The reaction was stopped by boiling the samples in SDS sample buffer for 5 mm. The samples were analyzed by 12% SDS-PAGE, and the gel was dried and subjected to autoradiography. Similarly, CDK6and cyclin Dl-associated RB kinase activity was determined as described by Wu et a!. (45) and detailed above with some modifications. Briefly, vehicleor silymarin-treated cells were lysed in RB lysis buffer (50 mi t HEPES-KOH, pH 7.5, containing 150 mM NaC1, 1 mi i EDTA, 2.5 mn EGTA, 1 nmi DY!’, 0.1% Tween 20, 10% glycerol, 80 mtvi 3-glycerophosphate, 1 mM sodium fluoride, 0. 1 mi i sodium orthovanadate, 1 m phenylmethybsulfonyl fluoride, 10 g/mi each of beupeptin and aprotinin), and using anti-CDK6 or anti-cyclin Dl antibody (2 IJ.g) and protein G-agarose beads (20 p.1), specific proteins were immunoprecipitated from 200 p.g of protein lysate per sample as detailed above. Beads were washed three times with RB lysis buffer and then once with RB kinase assay buffer (50 mr i HEPES-KOH, pH 7.5, containing 2.5 nmi EGTA, 10 nmi 3-glycerophosphate, 1 miii sodium fluoride, 0.1 mM sodium orthovanadate, 10 mivi MgCl2, 1 mi i DY!’). Phosphorylation of RB was measured by incubating the beads with 40 pJ of hot RB kinase solution [0.25 ii.! (2 .g) of RB-glutathione S-transferase fusion protein, 0.5 ii.! of [-y-32P]ATP, 0.5 p.1 of 0.1 nmi ATP, 38.75 il of RB kinase buffer] for 30 mm at 37#{176}C. The reaction was stopped by boiling the samples in SDS sample buffer for 5 mm. The samples were analyzed by 12% SDS-PAGE, and the gel was dried and subjected to autoradiography. Densitometric and Statistical Analysis. Autoradiograms of the Western immunoblots were scanned with a Microtek MSF-300GS image scanner (Microtek, Torrance, CA) that was linked to a Macintosh Ilsi computer. The image generated by the Microtek gray-scale scanner was captured by Image Studio and then analyzed for intensity of gray relative to background by Scan Analysis. The quantitation of image intensities on film was carried out at less than maximal densities. Radiolabeled bands for kinase activities were quantitated with a Phosphorlmager and analyzed with ImageQuant software (Molecular Dynamics). As needed, the twotailed Student’s t test was used to assess statistical significance of the difference between vehicleand silymarin-treated samples. Unless specified otherwise, the results shown in each case are representative of three independent experiments with similar findings. RESULTS Silymann Exerts both Antiproliferative and Anticarcinogenic Effects in MDA-MB 468 Cells. Using an anchoragedependent growth assay, we first assessed whether sibymarmn affords an antiproliferative effect in human breast carcinoma cells MDA-MB 468. As shown in Fig. 1, silymarmn treatment significantly inhibited the cell proliferation in a doseand timedependent manner. Whereas lowest concentration of silymarin tested (10 i.g/ml) showed almost no inhibition, a concentration of 25 of silymarin/ml showed a time-dependent inhibition in cell proliferation, with almost 50% inhibition (P < 0.001 ) at day 5 (Fig. 1 ). Treatment of cells with silymarin at higher doses of 50 and 75 g/ml showed a complete inhibition (P < 0.0001) of anchorage-dependent cell growth beginning at day 2 and onward at all time points studied, up to 5 days of treatment (Fig. 1). A 75 i.g/ml dose of silymarin also showed a reduction in initial cell number. Inhibition of soft agar colony formation potential of malignant cells is used extensively as a short-term in vitro assay to assess the anticarcinogenic effects of agents undergoing testing. Using this assay, we next assessed the anticarcinogenic effect of sibymarin in breast carcinoma cells MDA-MB 468. First, we determined the optimal time period and number of cells per plate, as well as number of cells per colony. In these studies, we found that 2000 cells grown in soft agar gave rise to optimum colonies of more than 10 cells per colony after 10 days of seeding (data not shown). As many as 97 ± 7 (mean ± SE of Research. on July 15, 2017. © 1998 American Association for Cancer clincancerres.aacrjournals.org Downloaded from

7 Figures and Tables

Cite this paper

@inproceedings{Zi2005AnticarcinogenicEO, title={Anticarcinogenic Effect of a Flavonoid Antioxidant, Silymarin, in Human Breast Cancer Cells MDA-MB 468: Induction of G1 Arrest through an Increase in Cipl/p21 Concomitant with a Decrease in Kinase Activity of Cyclin-dependent Kinases and Associated Cyclins1}, author={Xiaolin Zi and Denise K. Feyes and Rajesh Agarwal}, year={2005} }