3 .2... ”‘1“ ’.‘ ' o v . . J... t I v. I v . > I . 0 I . . . . . . . .. . ... v . , . .. , . ‘ l in u v ‘ I. o it! .v. l I. n. p :O.HMVVO.L . . .n .u n. . . , u u May In . .3 .. "ma .. . .. . . V A. 'I1" 3%. ",. . . .45: “ I. . . . .. , . an»: a Var. .3? ézuw . . :zw . . . . . , . Saws : “U.u...n\uw..w.... -1 . .. . k3: . V Av... . . R 5.3....» t: ah. . hi. “Btu-r . 7 ‘5'... E . an. . . ‘Iv . ‘ q .. . . . 5.5;. .-.....i k “‘wfinwmmwv... . . 5%.... .. .. . . .321. ‘ .. . . .. . . so... 4... o . , . (it: :. Wt! v 1 .5 u 1...! t u. . " .L‘ .' ”(Ii'fiw ALA-g; '-' ," “‘1’ . ' ‘4‘: I. I . . . I III... . 2w. ... . . .. . . . . V” .. I .. . b..~....x...» .45.. Jib .. ...v¢..\ up. .. .. . . . . .. Z. a “I“... . .1»! ...:.. 3...»..3 . A .. . . . 45353.: five...“ wuwufiofi . . . . . 33! ‘ ‘ VOA ,nnli‘li u‘ll ‘I . A I.’..-.O.‘.’.ll: 0 .. . a»! .18....9Mntl ! J.Ei. . . . .5 not. . ‘1‘ 9 I .05- 4 I U.‘ -t {A‘l ‘ It! .w..u:.t.....u. .....W..u. . ‘ . . grirtvu Brawl!!! 1- s n 0 x v v ~ .11 .v 0.. . to" .. In . no nulqtt'l‘ XV: ctr ‘ cf fr NE. I Z ...n.utv1.....i~. .13.... y. ‘ >0l..n . . - ..... .nI-«P. 43‘}. 7 i‘ . .lqp‘. THEQS mlllllllllllllllllulllllll , 31293 01716 3530 This is to certify that the dissertation entitled THE BIOLOGICAL AND BIOCHEMICAL ROLES OF GENISTEIN AND ITS METABOLITE, DIHYDROGENISTEIN, IN HUMAN BREAST CARCINOGENESIS presented by Ching—Yi Hsieh has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science/Environmental Toxicology Adz Z/W Major “tofessor Date />,//e/? 3L MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771 LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE MTE DUE DATE DUE CBQT82372003 THE BlOLO MEI THE BIOLOGICAL AND BIOCHEMICAL ROLES OF GENISTEIN AND ITS METABOLITE, DIHYDROGENISTEIN, IN HUMAN BREAST CARCINOGENESIS By Ching-Yi Hsieh A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition and The Institute for Environmental Toxicology 1997 THE BlObOGlC METAB( Genistein (GE biochemical effect Inhibition of protei Three Objectives 0 Mental effect: “'45 found to cnha and tumor STOWIh hm cancer (HB Wing in the m ABSTRACT THE BIOLOGICAL AND BIOCHEMICAL ROLES OF GENISTEIN AND ITS METABOLITE, DIHYDROGENISTEIN, IN HUMAN BREAST CARCINOGENESIS By Ching—Yi Hsieh Genistein (GEN) is one of the isoflavones found in soy products. Several biochemical effects, including estrogenic effects of GEN have been reported, i.e. inhibition of protein tyrosine kinase, DNA topoisomerase and antioxidant activity. Three objectives of the present study aim to evaluate both the potential beneficial and detrimental effects of GEN related to breast cancer development. First, dietary GEN was found to enhance mammary gland development, and to show uterotrophic effects and tumor growth of implanted MCF-7 cells [estrogen receptor (ER)-positive human breast cancer (HBC)] in ovariectomized athyrnic mice. GEN acts as an estrogen agonist resulting in the proliferation of MCF-7 cells and in the induction of the expression of an estrogen responsive gene, p82, in MCF-7 cells in vitro. GEN produces a dose- dependent stimulatory effect on growth of MCF-7 cells at concentrations from 1 nM to 1 14M and a growth inhibitory effect at levels greater than 10 M in both MCF-7 and MDA- mB-23l (ER- negative HBC) cells. The effect of GEN on cell growth at lower concentration appears to be mediated by an ER-mediated pathway and the effect of GEN at higher concentration could be mediated by a different mechanism (e. g. tyrosine kinase -’ inhibition). Second, dihydrogenistein (DHG) is the only metabolite of GEN found in the urine of humans consuming soy. DHG increased cell growth in a dose-dependent manner in ER-positive MCF-7 cells, but not in ER-negative MDA-mB-231 cells. DHG differs from GEN in failure to inhibit cell growth at higher concentration (above 25 M) in either Ell-positive by DHG. Additio block the stimulatio competed With {3”} Increased after treat of genistein to inhit cells (Le. Type I cel basal epithelial cell llcells. The resul increased the differ culmres derived frc cell growth of Typ: comimitations higl CW Flow 1 moon of botl Wm blot 3113]). the GUS transition Significantly enhar HBEC. Since Ty in either ER-positive or ER-negative HBC cells. p82 mRNA expression was stimulated by DHG. Additionally, we demonstrated that the ER antagonist, ICI 164,3 84, is able to block the stimulation of p82 expression and cell growth induced by DHG. DHG competed with [3H]estradiol for binding to the ER. The level of p21 WAF ”C 1P 1 protein increased after treatment with GEN, but not with DHG. Lastly, we studied the ability of genistein to inhibit the growth of two types of normal human breast epithelial (HBEC) cells (i.e. Type I cells with lurninal and stem cell characteristics, and Type H cells with basal epithelial cell phenotypes) and to induce the differentiation of Type 1 cells to Type H cells. The results show that GEN, at concentrations lower than 1 uM, significantly increased the differentiation of Type I HBEC to Type H cells in one of two primary cultures derived from different human subjects. Significantly, GEN completely arrested cell growth of Type I HBEC at concentrations higher than 5 M and Type II HBEC at concentrations higher than 50 M in all of the six independent primary cultures examined Flow cytometric analysis revealed that GEN was able to arrest cell cycle progression of both Type I and Type II HBEC at both 61/8 and GZ/M checkpoints. Western blot analysis showed that the level of pZIWAFI’Cm, which negatively regulates the Gl/S transition and cdc2 protein, which positively regulates the GZ/M transition, are significantly enhanced and decreased respectively by GEN in both Type I and Type H HBEC. Since Type I HBEC have been shown to be more susceptible to neoplastic transformation, the inhibition of Type I cell growth by genistein at physiological dose could reduce the number of target cells for carcinogenesis, thereby providing a mechanism for chemoprevention of breast cancer. Depending on the dosage of GEN and the target cell type, our results provide evidence for both beneficial and detrimental effects of GEN on breast carcinogenesis. To Wen-Jen, Kevin and Karen iv I would encouragement lwould also 1 guidance and SI other members Steve Bursian,f The techni $318M)! ackno Brad Upham. I Emma. Saitoh, Finally, I v ACKNOWLEDGMENTS I would like to thank my major advisor, Dr. William Helferich, for his encouragement and help in accomplishing my research during the past four and half years. I would also like to express my heartfelt thanks to Dr. Chia-Cheng Chang for his guidance and support in the last year of my research. My sincere gratitude goes to the other members of my graduate committee, Dr. John Linz, Dr. Maurice Bennink and Dr. Steve Bursian, for their kindness and valuable advice. The technical and intellectual support and advice of numerous colleagues are gratefully acknowledged and appreciated to: Dr. James Trosko, Dr. Melinda Wilson, Dr. Brad Upharn, Dr. Kyung-Sun Kang, Dr. Ross Santell, Dr. Elizabeth Ship, Sherri Batterman, Dr. Wei Sun, Angela Cruz, Chia-Jen Liu, Nester DeoCampo, and Maki Saitoh. Finally, I would like to thank my husband, Wen-Jen Tsay, for his sacrifice, understanding and encouragement. I also want to thank my parents and sisters for their love and support. Without their help, it is impossible for me to finish this PhD program. My son and daughter, Kevin and Karen, who give me the most wonderful joy. I would like to thank them for their presence in my life during these four and half years. They are the best gifts that I have ever had. LIST OF TABLE LIST OF FIGUR LIST OF ABBRI CHAPTER 1 Introduction... CHAPTER 2 Lilemnlre Rev 1. Breast C II. Soybean H1- Genisteil IV. Biochm V- Bi010gic V1 Concent TABLE OF CONTENTS LIST OFABBREVIATIONS............ CHAPTER] Introduction.............. CHAPTER 2 LiteratureReview... . Breast Cancer Epidemiology and Etiology SoybeanandBreastCancer... Genistein Content in Soybeans and mSoy Products... Biochemical Effect of Genistein... Biological Effect of Genistein... s0.05). The uterine weight of the genistein treatment group was also increased in comparison with control group. These data suggest that dietary genistein has the potential to stimulate estrogenic responses imam. Effect of Gen'ntein on Tumor Growth in Athymic Nude Mice. Since genistein stimulated both mammary gland and uterine growth, it was of interest to determine the effects of an estradiol (2 mg) pellet or dietary genistein (750 ppm) on MCF-7 cell tumor growth. Ovariectomized athymic mice, implanted with MCF-7 cell tumors, were divided into three treatment groups after tumors reached an average cross-sectional area of approximate 55 m2. The three treatment groups were 1) negative controls (without estradiol pellets or genistein treatments), 2) positive controls (estradiol pellets) and 3) dietary genistein (750 ppm) treatment groups. Tumor growth was monitored weekly. Tumors in tht reorient. the oiestradiol trt morimatel) an addition: were remover I an average In the dietarj stitched to '. area in dicta “Eli's on ge; Weeks 0,, es. agonin m mimic mic Plum: (:1 geniStein 1e. W01 Wilding fie image of i l. {CR-7 Cells DISCl'ssr 34 Tumors in the positive control group grew rapidly and alter an additional two weeks on estrogen treatment, the average tumor cross-sectional area had reached 120 mm2(Figure 5). After two weeks of estradiol treatment the average tumor weight was 750 mg and total tumor weight per mouse was approximately 3 gram (approximately 10% of the body weight). Therefore, mice were killed after two additional weeks on estrogen treatment due to size of the tumors. After the estrogen pellets were removed in the negative control group, tumor growth stopped and tumor size was maintained at an average cross-sectional area of approximately 58 mm2 for an additional 12 weeks (Figure 5). In the dietary genistein treatment group, after the estrogen pellets were removed and mice were switched to 750 ppm genistein diets, tumor growth increased. The average tumor cross-sectional area in dietary genistein treatment group increased from 55 mm2 to 120 mm2 after 12 additional weeks on genistein treatment. These MCF-7 cell tumors were similar in size to tumors after two weeks on estradiol treatment. These results indicated that dietary genistein acts as an estrogen agonist in_yim to stimulate growth of estrogen-dependent MCF-7 tumor cells implanted into athymic mice. Plasma Genistein Concentration Analysis in Athymic Nude Mice Fed Genistein. Plasma genistein levels were measured by high performance liquid chromatography (HPLC). The concentration of free genistein in plasma was 0.24: 0.08 nM. The concentration of total genistein including he and conjugated forms in plasma was 2.1 $0.14 uM. These concentrations are within the range of concentrations which stimulated p82 mRNA expression (Figure 2) and the growth of MCF-7 cells (Figure 1). DISCUSSION Geniste stimulation ill-IQ At lou adosc-depen stimulation v commutation 25 ll.\l-100 l growth Stimi 05%ed a c 35 Genistein elicits a concentration-dependent dual threshold effect with regard to growth stimulation or inhibition on cultured estrogen-dependent human breast cancer (MCF-7) cells in yin. At low concentrations of genistein, in dextran charcoal-strip fetal bovine serum, we observed a dose-dependent increase in MCF-7 cell proliferation fiom 0.01-1 uM (Figure 1). Maximal growth stimulation was observed at 1 uM and this level of growth was sustained up to 10 nM. At higher concentrations, we observed a dose-dependent inhibition in cell growth when concentrations were 25 uM-IOO uM These results are consistent with those reported by Martin et al. (1978), in which growth stimulation in MCF-7 cells was observed a concentration of 200 nM. Wang et al. (1996) observed a concentration-dependent grth stimulation effect between 10 nM and 1 pM. Wang et al. (1996) also observed that growth inhibition at concentrations higher than 10 pM. These results suggest that there are at least two dose—dependent mechanisms by which genistein can alter cell proliferation. One mechanism stimulates proliferation at low concentrations of genistein (10 nM to 1 nM) and is likely mediated via the estrogen receptor. The other mechanism, which is anti- proliferative, is active at high concentrations of genistein (25-100 nM) and is likely mediated via anti-tyrosine phosphorylation and inhibition of cell cycle progression (Pagliacci et al., 1994). The threshold for the stimulation of proliferation was in the range of concentrations required to stimulate expression of p82 mRNA (Figure 2). Additionally, pS2 gene expression was enhanced at concentrations up to 50 pM, However, cell proliferation was inhibited at concentrations above 25 [AM These observations suggest that the phytoestrogen estrogen receptor-mediated mechanism is active at the higher concentrations however the growth inhibitory mechanism associated with genistein overrides the growth stimulatory effects from genistein. of CW he u’ar 36 Genistein binds to the estrogen receptor with an affinity approximately 100-fold less than that of estradiol (Wang et al., 1996 and Santell et al., 1997). To further confirm whether genistein is acting via an estrogen receptor mechanism, we evaluated expression of the p82 gene in cultured MCF-7 cells. The human pS2 gene was initially characterized as a gene whose expression is specifically controlled by estrogen in the breast cancer cell line MCF-7 (Brown et al., 1984). The increase in p82 mRNA after addition of estradiol to the culture medium is a primary transcriptional event, suggesting that control of pS2 gene promoter activity by the estrogen receptor is mediated by a cis-acting estrogen-responsive element that could be located in the 5'-flanking region of the pS2 gene (Jakowlew et al., 1984). pS2 stimulation by genistein (Figure 2) is consistent with the results observed on cell proliferation in that 10 nM (Figure 1) was effective of increasing proliferation and p82 gene expression. Additionally, Wang et al. reported that tamoxifen will inhibit genistein- stimulated p82 gene expression in MCF-7 cells. These data further suggest that genistein can act via the estrogen receptor mediated mechanism. Here we report that dietary genistein (750 ppm) fed to 28 day athymic ovairectomized mice will stimulate mammary gland growth (Figure 3, 4). This is the first report in which dietary genistein has been shown to enhance mammary gland growth imam. Recent reports (Lamartiniere et al., 1995 and Murrill er al., 1996) in which high dosages of genistein were administered subcutaneously have been reported to stimulate mammary gland differentiation, promoting lobuloalveolar development in immature Sprague-Dawley rats. No studies have been reported in which the effect of genistein was evaluated for its potential to stimulate growth of estrogen-dependent tumors imam. The studies presented here provide compelling evidence that MCF-7 estrogen-dependent tumors can be stimulated to grow in ovariectomized athymic [:7 E? 37 host mice supplemented with dietary genistein (750 ppm). It is important to point out that both estrogen and genistein have been shown to be chemopreventive in the rat DMBA animal model (Lamartiniere et al., 1995, Murrill er al., 1996 and Nagasawa et al., 1974). Estrogen administered to neonatal rats has been shown to reduce the incidence of DMBA-induced mammary tumors (Nagasawa et al., 1974). In the studies with genistein, three subcutaneous injections of genistein were given to young rats which were protective against DMBA initiated carcinogenesis. The authors suggest that genistein is acting as an estrogen agonist to stimulate mammary gland differentiation. This enhanced mammary gland differentiation may be the mechanism responsible for the chemopreventive effects of genistein in the rat DMBA mammary carcinogenesis model. Thus, there appears to be a paradox with regard to the action of genistein on tumor growth in rodent models. Our results suggest in the ovariectomized athymic mouse that genistein stimulates mouse mammary gland growth without maturation which is different than the effect of genistein in the immature rat in which genistein stimulate prolactin secretion (Santell et al., 1997) and ultimately results in lobuloalveolar development and mammary gland maturation (Lyons et al., 1958). Genistein may act similarly to estradiol because estradiol is known to reduce tumor incidence in some instances (Miller, 1990) whereas there are numerous reports (Henderson et al., 1988) that indicate that estrogen act as tumor promoters. Thus, whether genistein acts as a chemopreventative agent or as a tumor promoter will likely depend on the timing of administration of the genistein. - The present study focused on the estrogenic effect of genistein imam and imam systems. Genistein binds to the estrogen receptor, with an affinity approximately 100-fold lower than estradiol, resulting in enhanced proliferative activity. In culture cells, genistein stimulates MCF- 38 7 cell growth at concentrations as low as 100 nM achieving similar effects to that of estradiol at 1 nM (Figure 1). We observed a 240% increase in growth at 100 nM genistein treatment compared to control. Expression of the estrogen-responsive gene, pS2 was also induced in response to treatment with genistein at concentrations higher than 100 nM (Figure 2). Imam, using ovariectomized athymic nude mice implanted with estrogen-responsive MCF-7 cells, we have demonstrated that treatment with 750 ppm dietary geni stein was able to stimulate the growth of these estrogen-responsive MCF-7 cell tumors in the absence of estrogen (Figure 5). Plasma genistein was 240 nM (free form) which is above the concentration required for MCF-7 cell growth stimulation observed imam. From this concentration, one can predict from the in yitm cell growth studies that at this dosage of genistein will be stimulatory to implanted MCF-7 cell tumors. Thus, these results are consistent with imam cell culture studies. At high concentrations (above 25 nM) imam a dose—dependent decrease in MCF-7 cell growth is observed, however, it is unlikely that the concentrations required to inhibit MCF-7 cell growth can be achieved imam. In summary, genistein can act as an estrogen agonist resulting in proliferation of human breast cancer cells (MCF-7) imaim and enhances growth of MCF-7 cell tumors implanted into ovariectomized athymic nude mice imam. Thus, there is the potential of dietary genistein to stimulate growth of estrogen-dependent tumors in women with low circulating endogenous estrogen levels such as in postmenopausal women. ACKNOWLEDGMENTS 39 The authors would like to thank Dr. Y.C. Chang and Dr. M.G. Nair for providing genistein. Thanks are also due to Dr. P. Dickerson-Weber for her technical assistance and Dr. L. Bourquin for his statistical consultant. Sincere appreciation is extended to Dr. J. Linz for his critical review of the manuscript. Figure 1. Effects of estradiol and genistein on the growth of estrogen responsive MCF-7 cells. MCF-7 cells were cultured in the presence of various concentrations of genistein (10 nM- 100 uM) or control media for 96 hours, in IMEM media containing 5% fetal bovine serum (FBS), penicillin (100 units/ml) and streptomycin (100 ug/ml) at 37°C in a humidified atmosphere of 5% CO2 in air. Proliferation was assessed by DNA content as measured using HOEST reagent and fluorometric analysis (expressed as percent of control). Fluorescence was measured by excitation at 350 nm and emission at 455 um and was used to determine DNA content. The results (mean, n=8) are expressed relative to cells grown without genistein. C represents the vehicle control and E represented as treatment with 1 nM estrogen in media. 41 wig/Z???” _ z w m m o 1 65:00 me Eocene I I. 25 I. .01 .l 42 Figure 2. Concentration-dependent stimulation of p82 gene expression by genistein. MCF- 7 cells were cultured in IMEM or MEM media in the presence of various concentrations of genistein (1 nM-50 uM). RNA was isolated after 72 hours. Northern blot analysis and detection for p82 was performed using standard as described in methods. 43 Con , 5 E2 1nM " Gen 1nM Gen 10nM Gen 100nM Gen 500nM t germ Con 52 1 nM E2 200pM Gen 1 uM Gen 10uM Gen 25uM Gen 50uM . r, fr .C‘illflol'l- also a Figure 3. Mammary whole gland mounts from ovariectomized athymic mice fed various dietary treatments. Mice were ovariectomized on day 28, and treatments begun on day 35. Mice were fed A, control (AIN 93-G), B, genistein (750 ppm) or C, estradiol (1 ppm) containing diets. Mice were killed on day 40 and mammary glands were removed for whole gland mount preparation as described in Material and Methods. Note the increase in size and numbers of end buds indicated by arrows. Numerical data are presented in Figure 4. Estradiol (1 ppm) 46 Figure 4. Effect of genistein on end bud development in ovariectomized athymic female mice. Mice were ovariectomized on day 28, and treatments initiated on day 35. Mice were fed control (AIN 93-G), genistein (750 ppm) or estradiol (1 ppm) containing diets. Mice were killed on day 40 and mammary glands were removed for whole gland mount preparation as described in Material and Methods. Each bar represents the Mean: SEM of four mice per experimental group. The probability (*P=0.05) that estrogen and genistein treated groups had higher numbers of end buds than controls was determined by ANOVA. 47 15 u-flIu-H me 23:32 E2 lpprn GEN 750”- CON Figure 5. The effect of estrogen pellet (2 mg) and dietary genistein (750 ppm) on MCF-7 tumor growth in athymic nude mice. MCF-7 human breast cancer cells were injected subcutaneously into four sites on the flanks of mice at 1 x 10" cells per site. After tumors had formed, the mice were grouped to equalize tumor area and dietary treatment initiated Experimental groups included negative control AIN-93G (5 mice, 15 tumors=n), positive control implanted 2 mg estrogen pellet (5 mice, 17 tumors=n) and AIN-93G + genistein 750 ppm (5 mice, 17 tumors=n). Data are expressed as the change in tumor areas for each week of measurement. The treatnrenfiweek interaction is statistically significant (P<0.0001). Treatment means for each week are compared using the Least Significant Difference method. 49 Tl m ._ ... «lie. 1 .1. If m T» n \m n T T 1.1 r T . a... a p a yul 5 8 1 4 7 o 42l m 9 7 5 4 no.2 3:283...er 3E3... ens—e: 12 10 on Treatment Weeks CHAPTER 4 Effects of Dihydrogenistein, a Metabolite of Genistein, on Human Breast Cancer Cells In Vitrg ABSTRACT Dihydrogenistein (DHG) is the only metabolite of genistein (GEN) found in the urine of human volunteers consuming soy. DHG differs from GEN by the saturation of one double bond (at C2 and C3). In previous studies from our laboratory, we have shown that GEN produces a dose-dependent stimulatory effect on growth of MCF-7 cells [an estrogen receptor (ER)-positive human breast cancer (HBC) cell line] at concentrations from 10 nM to 1 M and a grth inhibitory effect at levels greater than 10 uM in both MCF-7 and MDA-mB-231 (ER- negative HBC) cells. The effect of GEN on cell growth at lower concentrations is likely mediated by an ER-mediated pathway and the effect of GEN at higher concentration is mediated by a different mechanism. We evaluated the potential of DHG to act by similar mechanisms to GEN on growth of HBC cells. We observed that DHG increased cell grth in a dose-dependent manner from 1 nM to 80 M in MCF-7 cells, but not in MDA-mB-231 cells. However, DHG did not inhibit cell growth at higher concentrations (above 25 uM) in either ER-positive or ER-negative HBC cells. Furthermore, stimulation of p82 mRNA expression by DHG occurred in a dose-dependent fashion from 10 nM to 80 M in MCF-7 cells. Experiments were carried out to confirm that DHG acts as an estrogen (E) agonist and 50 Inna" 51 elicits estrogenic effect by ER-mediated mechanism. The results showed that the ER antagonist, ICI 164,384 (100nM), blocks the stimulation of p82 expression and cell growth induced by DHG. Furthermore, DHG was found to compete with [3H]estradiol for binding to the estrogen receptor with 50% inhibition at 0.6 nM. Although we failed to detect a difference between GEN and DHG treatment in protein tyrosine phosphorylation as originally suspected, we did find that the level of p21 “PW“ protein was increased after treatment with GEN (50 nM), but not with DHG (50 uM). The level of p53 protein expression, however, was not changed by GEN treatment, indicating that p21 cm’w‘“ could be induced by GEN by a p53-independent mechanism. These results suggest that minor modification in the chemical structure of GEN results in the loss of its growth inhibitory effect at high (>25uM) concentrations with no significant change in estrogenic activity. INTRODUCTION Dihydrogenistein is the only metabolite of genistein found in the urine of human volunteers (Joannou, G.E. et al. 1995). When volunteers were fed with soy (Kelly, G.E. et al. 1993), urinary dihydrogenistein was detected and identified by profile capillary gas chromatography (GC) and electron ionization mass spectrometry (GC-EIMS) analysis of the trimethylsilyl ether (TMS) derivatives. However, no quantitation was reported in this study for the concentration of dihydrogenistein in urine. Additionally, dihydrogenistein was found when pure genistein was fermented with human fecal bacteria under anaerobic conditions (Chang Yu-Chen and Nair M.G. 1995a). 52 Dihydrogenistein is a reduction product of genistein by hydrogenation at the C2 and C3 position. Chang and Nair (1995a) were able to synthesize dihydrogenistein from genistein by hydrogenation to provide sufficient quantities for imam studies. (Chang Yu-Chen and Nair MG. 1995b). Chang and Nair (1995a) conducted various bioassays with dihydrogenistein to evaluate the antifungal, antibacterial, mosquitocidal , nematocidal and topoisomerase effects. To date, no other characterization of dihydrogenistein has been conducted in human cells. Furthermore, it is not clear whether dihydrogenistein, formed after bacterial biotransformation, has a different biological effect from that of genistein. The major known effects of genistein in human breast cancer cells are ER-mediated estrogenic effect (Wang, T.Y. et al. 1996) and non-ER-mediated protein tyrosine kinase inhibition effect (Akiyama, T. et al. 1987). We have shown that genistein produces a concentration-dependent effect on growth stimulation of MCF-7 cells at lower concentrations (10 nM to 1 uM). Additionally, we have shown that, at higher concentrations (>10 uM), genistein inhibited cell growth (see chapter 3). The effect of genistein on cell grth at lower concentration appears to be mediated by an estrogen receptor pathway, while the effects at higher concentrations were independent of estrogen receptor since these inhibitory effects were observed in both estrogen receptor positive and estrogen receptor negative cells. Hence, the effect of genistein at higher concentrations is likely to be mediated by the non-ER-mediated protein tyrosine kinase inhibition effect (Akiyama, T. et al. 1987). Genistein is known to inhibit the tyrosine kinase activity of a number of kinases including both receptor and cytosolic forms 53 (Akiyama, T. et al. 1987, Nakafutu et al. 1992). Genistein at 100 M inhibited epidermal growth factor (EGF)-, nerve growth factor (N GF)—, fibroblast growth factor (FGF)- and insulin-induced RaszGTP complex formation in rat pheochromocytoma PC-12 cells ( Nakafutu et al. 1992), and erythropoietin-induced RaszGTP formation in human erythroleukemia cells (Torti et al. 1992). Furthermore, genistein at >500 uM significantly inhibits the total protein tyrosine phosphorylation (Koroma and DE Juan, 1994). Since DHG has been found in human urine, it is important to evaluate the potential estrogenic and growth inhibitory effects of dihydrogenistein in human breast cancer cells. Ogawara et a1. (1989) reported that a slight change in the structure of genistein could cause a significant decrease in its ability to inhibit protein tyrosine kinase activity. In order to clarify the structure-activity relationship, the tyrosine kinase inhibitory activities of synthetic flavonoids, isoflavonoids and genistein derivatives (PKI-l to PK]- 24) were investigated. The results indicates that a hydroxyl group at position C5 was essential for inhibitory activity and that hydroxyl group at C7 and C4' positions was necessary for full expression of the inhibitory activity (fig. 6). Results fiom this study show that dihydrogenistein does not inhibit cell proliferation in MCF—7 or MBA-231 cells. The results of Akiyama et al. (1989) and our studies suggest that the hydrogenation at C2 and C3 position of dihydrogenistein may cause the loss of tyrosine kinase inhibitory activity as compared to its parent compound, genistein. This needs to be further examined. 54 p21CIPl/wm, a 21 kDa protein, is a major inhibitor of cyclin/CDK catalytic activity. Each member of the cyclin kinase family is inhibited by p21 mm“, but their relative affinities vary with each enzyme (Gu et al., 1993, Xiong et al., 1993 and Harper et al., 1993). As a potent inhibitor of G1 cyclin-dependent kinase, p21 “Pl/WA“ has been found to cause cell cycle arrest specifically in Gl/S transition (Harper et al., 1993). On the other hand, p21 CPI/WA}? 1 also has been reported to induce differentiation in a number of cell types in vitro, such as the myelomonocytic cell line U937 (Liu etal., 1996). In this regard, GEN has been well documented as both a cell cycle inhibitor and a differentiation inducer in several cancer cell lines(Constantinou and Huberrnan 1995 and Pagliacci et al., 1994). A model has been presented by Liu et al., (1996) depicting a key 1CIP ”W AF 1 in facilitating the differentiation of cancer cells in response to role for p2 inducers such as hormonal ligands in a p53-independent manner. Accordingly, it seems possible that p21 CIPWAF 1 could be one of the target proteins of GEN to induce cell growth arrest and/or differentiation in cancer cells. In summary, the biological effect and mechanisms of function of DHG are largely unknown. There are three major potential mechanisms of DHG which were tested in this paper. First, the estrogen receptor (ER)-mediated pathway was analyzed by cell proliferation, ER competitive binding assay and ER-dependent p82 gene expression. Second, the protein tyrosine kinase-mediated pathway was studied by comparing the total protein tyrosine phosphorylation level in western blots. Thirdly, the p21 CPI/w” 1 protein mediated pathway was analyzed by western blots to detect changes in p21 55 /w . . cm AF 1 protern expressron. MATERIALS AND METHODS Chemicals. Genistein and dihydrogenistein were synthesized by Chang and Nair (Chang et al., 1994 and Chang and Nair, 1995b). ICI 164,384 was a gift from Dr. T. Zacharewski of the Department of Biochemistry, Michigan State University. The other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO). The effect of dihydrogenistein on the proliferation of estrogen receptor positive MCF-7 and estrogen receptor negative MDA-mB-23l human breast cancer cells in culture. MCF-7 (estrogen receptor positive) and MDA-mB-231(estrogen receptor negative) cells (American Type Culture Collection) were maintained in the IMEM medium (Biofluids Inc., Rockville, MD) containing 5% fetal bovine serum (FBS), penicillin (100 units/ml) and streptomycin (100 pig/ml). Cells were incubated at 37°C in a humidified atmosphere of 5% C02 in air. MCF-7 and MDA-mB-231 cells (American Type Culture Collection) were harvested, and then were plated at 1.5x103 cells per well in 24 well tissue culture polystyrene plates in medium containing 5% FBS for one day. After removing the medium, cells were washed with phosphate buffered saline (PBS) and then were incubated with medium containing 5% charcoal dextran-treated fetal bovine serum (Hyclone Inc., Logan, UT) in phenol-red free media for an additional two days. The medium was renewed using the same medium mentioned above plus dihydrogenistein at 1 nM, 10 nM, 100 nM, 200 nM, 500 nM, 1 nM, 10 nM, 25 nM, 50 56 nM, 80 M or estrogen at 1 nM. Proliferation was assessed by fluorometric analysis of DNA content (as an indicator of cell number) daily for 4 days (West et al. 1985, Labarca, C. & K. Paigen. 1980). Cells were lysed i_n__s_i1a with 10 mM EDTA, pH 12.3 at 37°C for 30 minutes and neutralized with KH2P04 and the resulting DNA stained by using Hoechst 33258 reagent. Fluorescence was measured by excitation at 350 nm and emission at 455 nm and compared to standard salmon sperm DNA to determine DNA content. The effect of dihydrogenistein on the expression of the estrogen responsive p82 gene in estrogen receptor positive MCF-7 human breast cancer cells in vitro. MCF- 7 cells were cultured in 100 mm x 20 mm polystyrene tissue culture plates under the condition described above. Twenty-four hours after initial plating of the MCF-7 cells, estradiol or dihydrogenistein were added. The final concentration of estradiol was 1 nM and the final concentrations of dihydrogenistein were 1 nM, 10 nM, 100 nM, 1 pM or 10 nM. Total RNA were isolated 72 hours after the addition of estradiol or dihydrogenistein using the method of Chomczynski and Sacchi (1987) as modified by Xie and Rothblum (1991). Briefly, the cells were lysed with 1.7 ml of lysis buffer which contained 4 M guanidium thiocyanate, 25 mM sodium citrate, 100 mM 2- mercaptoethanol and 0.5% sodium sarcosyl (pH 7 .0). The cell lysate was transferred from the tissue culture plate to a 15 m1 Corex tube, and mixed with water-saturated phenol containing 0.04% (w/w) hydroxyquinoline and 2 M sodium acetate (pH 5.0) in a ratio of 10:10: 1. This was followed by the addition of 0.36 ml of chloroformzisoamyl alcohol (24:1). Tubes were vortexed for 10-20 seconds, placed on ice for 30 minutes, and then centrifuged at 12,000 x g for 30 minutes at 4°C. The upper layer containing 57 the RNA was then transferred to a new tube containing an equal volume of isopropanol at -20°C. Samples were vortexed and RNA was precipitated by placing the mixture at - 20°C for at least two hours. The RNA precipitate was recovered by centrifugation at 12,000 x g for 30 minutes. The resultant pellet was washed twice with 70% ethanol, and dried under vacuum. RNA was resuspended in 100 pl diethyl pyrocarbonate-treated (DEPC) water. Ten ug of RNA were used for Northern blot analysis. Northern blot analysis was performed to determine pSZ mRN A content using pS2 cDNA probe ( Jakowlew et al. 1984). The detailed procedure is described below. Probing for p82. A plasmid containing a 559 base pair cDNA fragment encoding the full length p82 (Jakowlew et al. 1984) was obtained from the American Type Culture Collection. 2511 restriction endonuclease digestion of the plasmid (Jakowlew et al. 1984) produced DNA fragments of 4.4, 0.32 and 0.24 kb. These fragments were isolated from the agarose gel slice using the methods adapted from Moore et al. (1995). Twenty-five ng of p52 cDNA (0.32 and 0.24 kb) and so ncr [or -32P]dCTP were used to label the DNA by the random primed labeling method using a commercial kit and following the manufacturer's instructions (Gibco BRL, Gaithersburg, MD). Northern Blot Hybridization and Detection. For detection of pS2 expression, 10 pg of total RNA were separated on 1.2% formaldehyde agarose gels and transferred to a Hybond-N Nylon membrane (Amersham, Arlington Heights, IL). The RNA was cross- linked onto the membrane by UV light for 3 minutes using a 25 watt transilluminator (Hoefer Scientific Instruments, San Francisco, CA). The RNA on the membrane was then hybridized with the 32P-labeled p82 DNA probe and detected by autoradiography. Northern blots were also probed for human glucose-3-phosphopate-dehydrogenase 58 (G3PDH) cDNA (Clontech Laboratories, Inc., Palo Alto, CA), a house keeping gene, to confirm equal loading of RNA among treatment and control groups. The effect of the pure estrogen receptor antagonist (ICI 164,384) on dihydrogenistein stimulated cell growth and p82 expression in estrogen receptor positive MCF-7 human breast cancer cells cultured in vitro. MCF-7 cells were treated with ICI 164,384 athO nM which has been reported as a maximal inhibition concentration (Colin et al. 1994) in conjunction with various concentrations of dihydrogenistein. As mentioned above, proliferation was assessed by fluorometric analysis of DNA content daily for 4 days (West et al. 1985, Labarca, C. & K. Paigen. 1980). Hoechst 33258 reagent was added and fluorescence was measured by excitation at 350 nm and emission at 455 nm to determine the DNA content. Total RNA was isolated 3 days afier the addition of ICI 164,384 and dihydrogenistein using the method of Chomczynski et al. (1989) as modified by Xie and Rothblum (1991 ). Northern blot analysis was performed as described above. Determination of receptor binding affinity of dihydrogenistein to human estrogen receptor (hER) isolated from estrogen receptor positive MCF—7 human breast cancer cells. Competitive binding assays were conducted using a modification of the hydroxyapatite (HAP) binding procedure of Murdoch et a1, 1990. A human breast cancer cell line, MCF—7, was the source of human estrogen receptor (hER) for the binding assays. Four days before harvest, the medium was changed to 5% DCC-FBS medium in which the serum supplement was treated with dextran-coated charcoal 59 (Hyclone Laboratories) to remove steroid hormones. Twenty four hours before harvesting, the 5% DCC-FBS medium was replaced with serum-free medium to minimize the level of 17 B-estradiol in the cells and to improve the recovery of unliganded hER. Cells at near confluence were gently suspended by a 30 min incubation in 1 mM EDTA in Ca- Mg— free phosphate buffered saline at 37°C. Cell suspensions were combined and centrifuged at 800 g for 10 min at room temperature to pellet cells. The cells were washed once with homogenization buffer [10 mM Tris HCl pH 7.5, 1.5 mM NazEDTA, 1 mM dithiothreitol, 1 mM sodium molybdate, and 10% (v/v) glycerol], resuspended in 2 ml ice cold homogenization buffer, disrupted with 60 passes of a Dounce homogenizer on ice, and centrifuged at 800 xg, at 4°C for 10 min to pellet cellular debris. The hER preparation was obtained by centrifugation of the supernatant at 100,000 xg at 4°C for 30 min. The hER preparation was stored in 200 ill aliquots at - 80°C until use. Protein was determined using the Bradford dye binding assay in 1 ml homogenization buffer (Bradford, 1976). Inhibition of 10 nM [3H]-17B-estradiol (3H-E2) binding to hER was measured by incubation at 4 C for 2 hr of 40 pM hER in 1 ml Assay Buffer [10 mM Tris HCl pH 7.5, 1.5 mM NazEDTA, 1 mM dithiothreitol, and 10% (v/v) glycerol]. Triplicate analyses were conducted at concentrations ranging from 0.001 to 1 HM for genistein and dihydrogenistein. Total binding of 3H-E2 was estimated in the absence of competitor and was used to define the 100% binding level. Following a two hr incubation at 4°C, 0.5 ml of HAP was added. Samples were incubated at 4°C for 15 min to allow proteins in the supernatant to bind the HAP and were mixed at five min intervals during this 60 incubation. Samples were then centrifuged at 3,000 xg in a Beckrnan swinging bucket centrifuge for five min at 4°C. The supernatant was removed, and pellets were rinsed three times with 2.5 ml aliquots of ice-cold assay buffer. After the third rinse, the pellets were transferred to scintillation vials with 4 ml Safety Solve per vial (of a liquid scintillation fluid). The radioactivities were then measured using a Beckrnan LS-100 liquid scintillation counter (Beckman Instruments, Indianapolis, IN). Tritium efficiency standards were prepared fresh each time the assay was performed, and were used to determine the disintegrations per min (DPM) from the counts per min (CPM) reported by the detector. The resulting data was utilized to determine the ICSO of genistein and dihydrogenistein. Determination of the effect of dihydrogenistein on overall protein tyrosine phosphorylation of human breast cancer cells in culture. Total protein tyrosine phosphorylation content was assessed on the control, positive control (stimulated with 100 nM EGF) and treatment (genistein 300 uM or dihydrogenistein 300 uM alone or plus 100 nM EGF) cultures in both MCF-7 and MDA-mB-231 cells which were plated in 60 mm dishes. There are three steps included in this method: protein isolation, SDS-PAGE and electroblotting, hybridization and autoradiography (Kang et al., 1996). First, for protein isolation, 300 pl lysis buffer were added to a cell monolayer after discarding the culture medium. The lysis buffer was made up of 20% SDS lysis solution containing several protease and phosphatase inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 uM antipain, 0.1 uM anprotinin, 1 pM leupeptin, 0.] uM sodium orthovanadate and 5 mM sodium fluoride). The lysate was carefully 61 harvested by scraping into 1.5 ml microfuge tubes. Cell debris were removed by centrifugation at 800 xg, and then equal volumes of 2 x sample loading buffer (160 mM Tris-HCl, pH 6.8, 4% SDS, 30% glycerol, 5% B-mercaptoethanol, 10 mM dithiothreitol, 0.01% bromophenol blue) were added to the lysate which was boiled for 5 min at 100°C. Protein concentrations were determined by the DC protein assay kit (Bio-Rad Co., Richmond, CA) and samples were stored at -20°C, until ready for electrophoretic analysis. For SDS-PAGE, equal amounts of protein samples were run at constant voltage (200 V) for 45 min on 12%, 1 mm thick, precast, polyacrylarnide minigels (Bio-Rad Laboratories, Richmond, CA), using a miniprotean apparatus (Bio-Rad). Broad range, prestained (10 to 220 kDa) molecular weight protein markers (GIBCO ) were run simultaneously with the samples. For electroblotting, hybridization and autoradiography, gels were equilibrated in transfer buffer and then electroblotted onto 0.4 um Immobilon-P membranes (Millipore, Bedford, MA) using the trans-Blot SD (Bio-Rad) apparatus. Membranes were blocked in 5% nonfat dried skim milk in PBS containing 0.1% Tween 20 for 1 hr and probed with either antiphosphotyrosine antibody (Oncogene Science, Cambridge, MA) at a 1:1000 dilution or anti-EGF receptor antibody (Oncogene Science, Cambridge, MA) at a 1:50 dilution in the blocking solution, for a minimum of 1 hr at room temperature. Blots were washed in PBS and 0.1% Tween 20 and incubated for 1 hr at room temperature in a double secondary antibody matrix composed of IgG-HRP (horseradish peroxidase) for detecting protein samples (diluted 1:1000) in the blocking solution. Blots were washed (three times, 5 min each), immersed in chemoluminescent (ECL) immunodetection 62 reagents, and then exposed to Kodak X-ray film from 15 sec to 5 min, as described in Amersham ECL protocols (Amersham Co., Arlington Heights, IL). [W lCIPl U1 and Determination of the effect of dihydrogenistein on the level of p2 p53 protein expression in MCF-7 cells. MCF—7 cells were plated in 60 mm dishes, in the absence or presence of genistein and dihydrogenistein at the concentration of 50 nM. The methods of protein isolation, SDS-PAGE and electroblotting, hybridization and autoradiography which are used for sample analysis were described as above. The only difference is that membranes were blocked in 5% nonfat dried skim milk in PBS containing 0.1% Tween 20 for 1 hr and probed with different primary antibodies which 1 CP'IWAF' polyclonal antibody (Oncogene Science, Cambridge, MA) at a were anti-p2 1:500 dilution, anti-p53 monoclonal antibody (Oncogene Science, Cambridge, MA) at a 1:1000 dilution, or anti-actin polyclonal antibody (Oncogene Science, Cambridge, MA) at a 1:1000 dilution in the blocking solution, for a minimum of 1 hr at room temperature. The expression of actin served as an internal control of total protein levels in samples. RESULTS The effect of dihydrogenistein on the proliferation of estrogen receptor positive MCF-7 and estrogen receptor negative MDA-mB-23l human breast cancer cells in culture. Dihydrogenistein increased MCF-7 cell (ER positive) growth from 1 nM up to 25 M in a dose-dependent manner, then reached a maximum effect which is 63 approximately 2 times higher than the control. Estrogen treatment at concentrations of 1 nM also showed a 2 fold stimulatory effect on MCF-7 cell growth Although the cell growth showed a slight decrease at dihydrogenistein concentrations higher than 25 nM, it was not significantly different to the effect of genistein treatment at the same concentration. (Fig. 7) In contrast, dihydrogenistein did not stimulate or inhibit cell growth in MDA-mB-231 (ER negative) cells at both low and high concentrations (Fig. 8). Thus, ER appears to play a key role to cause this different response to dihydrogenistein between MCF-7 and MDA-mB-231 cells. Dihydrogenistein, however, did not inhibit cell proliferation at higher concentrations (about 25 uM) as shown by genistein in both MCF-7 and MDA-mB-231 cells. The effect of dihydrogenistein on the expression of estrogen responsive p82 gene expression in estrogen receptor positive MCF-7 human breast cancer cells in vitro. As has been noted above, dihydrogenistein can be an estrogen agonist and elicits estrogenic effect by an estrogen receptor-mediated mechanism. This is supported by the stimulation of p82 mRNA expression by dihydrogenistein which occurred in a concentration-dependent fashion from 10 nM to 80 M in MCF-7 cells (Fig. 9). The ER-mediated estrogenic effect of dihydrogenistein is thus confirmed by these experiments. The effect of the pure estrogen receptor antagonist (ICI 164,384) on dihydrogenistein-stimulated cell growth and p52 expression in estrogen receptor positive MCF-7 human breast cancer cells cultured in vitro. As it has been noted 64 above, dihydrogenistein might act as an estrogen agonist and elicits estrogenic effect through an estrogen receptor-mediated mechanism. If so, the enhanced p82 mRNA expression and cell proliferation induced by dihydrogenistein are expected to be inhibited by ICI 164,384, the estrogen receptor antagonist, at 100 nM. Indeed, the prediction is confirmed by results presented in Figs. 10 and 1 1. As shown, ICI 164,384 blocked the cell proliferation effect of dihydrogenistein at a concentration of 1 M as well as estrogen at a concentration of 1 nM and genistein at a concentration of 1 M in MCF-7 cells (Fig. 10). ICI 164,384 also blocked the effect of dihydrogenistein in the stimulation of p82 expression at a concentration of 1 uM in MCF-7 cells (Fig. 11). Determination of the receptor binding affinity of dihydrogenistein to the human estrogen receptor (hER) isolated from estrogen receptor positive MCF-7 human breast cancer cells. To further characterize the interaction between dihydrogenistein and the ER pathway, we examined the ability of dihydrogenistein to compete with [3H]estradiol (1 nM) for binding to the ER. Wong et al. (1996) showed that genistein competed with 3H-E2 for binding to the estrogen receptor with 50% inhibition at 500 nM. Dihydrogenistein competed with 3H-E2 for binding to the estrogen receptor with 50% inhibition at 600 nM (Fig. 12). Here, we clearly show that the estrogen receptor binding affinity for dihydrogenistein is similar to the estrogen receptor binding affinity for genistein. These results, together with the results from the cell proliferation, p82 gene expression assays and studies of estrogen antagonist, ICI 164,384, supported the interpretation that dihydrogenistein induces an ER-dependent estrogenic response. 65 The effect of dihydrogenistein on overall protein tyrosine phosphorylation of human breast cancer cells in culture. In order to investigate the effect of DHG on overall protein tyrosine phosphorylation, we evaluated the effect of DHG and GEN at various dosages on protein tyrosine phophorylation in MCF-7 cells and MDA-mB-231 cells. The results failed to show a significant difference in protein tyrosine phosphorylation among control, EGF (100 nM), genistein (300 M ) and dihydrogenistein (300 M ) treated MCF-7 cells (fig. 13). One explanation could account for these results. In MCF-7 cells (the transformed tumor cells), the receptor tyrosine kinases or cellular tyrosine kinases might be already overexpressed. Therefore, any effect of weak stimulators (EGF) or inhibitors (genistein) may not be detected. In contrast, EGF (100 nM) pie-treated cells showed an increase in number and intensity of tyrosine phosphorylated proteins compared with those (control) treated with the solvent DMSO in MDA-mB-231 cells. Notably, genistein at 300 M plus EGF (100 nM) in preheated cells inhibited the expression of a major phosphotyrosine protein (180 kDa) compared with the positive control, i.e. EGF pretreated MDA-mB-231 cells. Similarly, cells pre-treated with dihydrogenistein (300 uM) plus EGF (100 nM) also showed the same inhibition of the expression of this protein band. (Fig. 14b) The same western blot reprobed with anti-EGF receptor antibody showed a single band at the same position where the major band was located. (Fig. 14a) The data indicate that this 180 kDa protein could be the EGF receptor. However, this study failed to show any differences in protein tyrosine phosphorylation between genistein (300 M ) and dihydrogenistein (300 uM ) treatment in MDA-mB-231 cells. 1 ”WA“ and p53 protein The effect of dihydrogenistein on the level of p2 expression in MCF-7 cells. We have observed that the level of p21 CPI/w“ 1 protein increased after treatment with genistein at 50 nM, but not with dihydrogenistein at 50 nM. The relative ratio of p21 CIPl/WAFI protein levels are 1: 2.5: 1, for control: genistein (50 M ): dihydrogenistein (50 M ), respectively. There are no significant differences in the level of p53 protein expression among these treatments. (Fig. 15) These data indicate that the mechanism of p21 CIPWAF ' induction could be p53-independent. It is known that p21 CIR/WA“ is a very important cell cycle regulatory protein especially at the Gl/S checkpoint. Thus, our results showed that genistein, at a concentration of 50 M, which is a cytostatic, but not cytotoxic concentration, can increase the p21 CIPI’WAF] protein expression level to cause cell cycle arrest. However, dihydrogenistein at 50 M does not have this effect. The differential effect on the level of p21 cm’w‘“ protein expression between dihydrogenistein and genistein could be the major reason that dihydrogenistein failed to inhibit cell grth at concentrations higher than 25 11M compared to the effect of genistein at the same concentrations. Our data showing that the induction of p21 “PW/AF] by genistein could be p53-independent imply that the effect of genistein might not be due to its effect on DNA damage in MCF-7 cells (El-Deiry et al., 1995). DISCUSSION Genistein is a principle isoflavone present in soy beans and soy products and is 67 suspected to be a chemoprevective agent for breast cancer. Dihydrogenistein is the only metabolite of genistein found in the urine of human volunteers consuming the soy (Joannou, G.E. et al. 1995). The major objectives of this study were to investigate the biological effect of DHG and its biochemical mechanisms of action in human breast cancer cells. Our data revealed that there are at least two important properties of dihydrogenistein in MCF-7 cells. First, dihydrogenistein is able to bind to the estrogen receptor as a ligand, and induces several estrogenic responses which include the increase in proliferation of MCF-7 cells and the induction of p82 gene expression. Second, unlike genistein, dihydrogenistein appears to lose its ability to induce the expression of p21 CIPI’WAFI at concentrations higher than 25 M. This may explain its inability to inhibit cell proliferation at concentrations higher than 25 M as compared to genistein at the same concentration. Akiyama, et al. (1987) showed that genistein is able to inhibit tyrosine kinase activity in vitro and proposed that it could be the mechanism that genistein inhibits tumor cell growth. Our results however fail to reveal a significant difference in protein tyrosine phosphorylation among control, EGF (100 nM), genistein (300 nM) and dihydrogenistein (300 M ) treated MCF-7 cells (Fig. 13). Since genistein (300 M) has been reported to inhibit protein tyrosine kinase activities in several cancer cell lines (Koroma et al., 1994 and Clark et al., 1996). Several explanations could account for these results. First, in MCF-7 cells (the transformed tumor cells), the receptor tyrosine kinases or cellular tyrosine kinases might be already overexpressed Therefore, any effect of weak stimulators (EGF) or inhibitors (genistein) may not be detected. Second, 68 genistein may specifically inhibit tyrosine phosphorylation of certain proteins whose expression may be low and not detectable by assessing the overall levels of tyrosine phosphorylation. Third, cellular metabolism or drug exclusion mechanisms may reduce the intracellular genistein concentrations to levels below that needed to inhibit tyrosine phosphorylation in cells. In contrast to the study of overall tyrosine phosphorylation, the specific study of EGF-R tyrosine phosphorylation revealed that both genistein and dihydrogenistein were capable of inhibiting the tyrosine phosphorylation. Therefore, the differential response of MCF—7 cells to higher concentrations of genistein and dihydrogenistein in cell growth could not be attributed to the differential ability of the two compounds to inhibit protein phosphorylation. Therefore other mechanisms differentially exerted by higher concentrations of genistein and dihydrogenistein must be considered. 1 CPI/WA“ protein expression could be a The ability of genistein to induce the p2 mechanism that genistein inhibits cell proliferation of MCF—7 cells at cytostatic concentrations from 25 to 50 nM. Genistein appears not to affect the expression of other cell cycle regulatory proteins in MCF-7 cells, such as p16 (another family of cyclin kinase inhibitors), cyclin D1, cdc2 (a G2/M cyclin dependent kinase) and pr (retinoblastoma protein). The functions of p21 CPI’WAF 1 could contribute to the following three different mechanisms. First, p21 CI? 1m” 1 could form a complex with cyclin D ,and CDK 4 or 6 (in Gl/S transition) and inhibit CDK-dependent phosphoryaltion and inactivation of pr (Harper et al., 1995) resulting in G1 arrest. Second, p21 CIPWAFI could bind to PCNA and inhibit DNA synthesis in S phase (Baylin, 69 1997). In this manner, p21 CPI/WA“ would work as a tumor suppressor gene inhibiting tumor cell growth. Third, p21 CPI/WA“ could bind to cyclin A or B and cdc2 kinase complex (in G2/M transition) to inhibit cell cycle progression from G2 to mitosis (M) phase resulting in tumor cell arrest at G2 phase. The mechanism by which genistein induces the p21 CIPWAF‘ protein expression in MCF-7 cells, however, is not clear. In this regard, dihydrogenistein differs from its parent compound, genistein, by the loss its ability to induce the p21 cpl/WA}: 1 protein expression at cytostatic concentrations (25 to 50 M). This mechanism provides potential explanation for the difference in cell growth inhibition at higher concentrations between dihydrogenistein and genistein. In summary, our results indicate that a chemical modification of genistein results in the loss of growth inhibitory effect at high (>25 uM) concentrations of the isoflavone and no significant change in estrogenic activity. At the biological level, dihydrogenistein, similar to genistein, is expected to promote breast cancer growth at physiological concentrations. 70 Figure 6. The chemical structures of genistein and dihydrogenistein. 71 Genistein Dihyd rogenistein 72 Figure 7. The effect of dihydrogenistein on the proliferation of estrogen receptor- positive MCF-7 human breast cancer cells in culture. MCF-7 cells were cultured in the presence of various concentrations of dihydrogenistein (DHG)(1 nM-80 nM) or estrogen (E2) at 1 nM for 96 hr, in IMEM medium containing 5% fetal bovine serum (FBS), penicillin (100 units/ml) and streptomycin (100 jig/ml) at 37°C in a humidified atmosphere of 5% CO; in air. Proliferation was assessed by DNA content as measured using Hoechst reagent and fluorometric analysis (expressed as percent of control). Fluorescence was measured under excitation at 350 nm and emission at 455 nm and was used to determine the DNA content. The results (mean of 8 duplicates) are expressed relative to cells grown without dihydrogenistein. Bars represent stande error of mean (MeaniSEM). C represents vehicle control containing 1 to 1000 dilution of DMSO in grth medium. II“ 08 09 93 0|- l |-' IO' I-OO' l-OO' Relative Growth (% of Control) d d N 092 o 's A OIOIUODOJPMIIQ / 74 Figure 8. The effect of dihydrogenistein on the proliferation of estrogen receptor - negative MDA-mB-231 human breast cancer cells in culture. MDA-mB-231 cells were cultured in the presence of various concentrations of dihydrogenistein (25, 50 and 80 nM) or estrogen (E2) at 1 nM for 96 hr, in IMEM medium containing 5% fetal bovine serum (FBS), penicillin (100 units/ml) and streptomycin (100 jig/ml) at 37°C in a humidified atmosphere of 5% CO; in air. Proliferation was assessed by DNA content as measured using Hoechst reagent and fluorometric analysis (expressed as percent of control). Fluorescence was measured under excitation at 350 nm and emission at 455 nm and was used to determine the DNA content. The results (mean of 8 duplicates) are expressed relative to cells grown without dihydrogenistein. Bars represent standard error of mean (MeaniSEM). C represents the vehicle control containing 1 to 1000 dilution of DMSO in growth medium. recepto 131"! a l 35. Illa" ifetrllrr 16.115 ‘7 W“SZ Wul NIEILSINEIDOHIIAHIG wnog Wn08 ZH Relative Growth (% of Control) I» O l 03 O l 19 O l — OZl 091 76 Figure 9. The effect of dihydrogenistein (DHG) on the expression of the estrogen responsive p82 gene in estrogen receptor positive MCF-7 human breast cancer cells i_n_vit;o. MCF-7 cells were cultured in IMEM medium in the presence of various concentrations of dihydrogenistein (DHG) ( 10 nM-lO nM) or estrogen (E2) (1 nM). RNA was isolated after 72 hr treatment. Northern blot analysis and detection of p82 was performed as described in methods. The control represents the vehicle control containing a 1 : 1000 dilution of DMSO in growth media. Northern blots were also probed with human glucose-3-phosphopate-dehydrogenase (G3PDH) cDNA (Clontech Laboratories, Inc., Palo Alto, CA) as an internal control, to confirm equal loading of RNA among treatment and control groups. 77 a Q 5% =N Control E2 1nM DHG 10nM DHG 100nM DHG 500nM DHG luM DHG 10uM 78 Figure 10. The effect of estrogen receptor antagonist (ICI 164,384) (ICI) (100 nM ) on dihydrogenistein (DHG) stimulated cell growth in estrogen receptor positive MCF—7 human breast cancer cells cultured in_vitm. MCF-7 cells were cultured in IMEM medium as described above in the presence of DHG (1 uM), genistein (GEN) ( 1 uM), estrogen (E2) (1 nM) with or without ICI (100 nM). Proliferation was assessed by DNA content as measured using Hoechst reagent and fluorometric analysis (expressed as percent of control). Fluorescence was measured under excitation at 350 nm and emission at 455 nm and was used to determine the DNA content. The results (mean of 8 duplicates) are expressed relative to cells grown without dihydrogenistein. Bars represent standard error of mean (MeaniSEM). C represents the vehicle control containing a l : 1000 dilution of DMSO in growth medium. Relative Growth (96 of Control) ON 6 G 021 - 081 7 00C U '- H a U :- II F} a. _ G ' l- 3 80 Figure 11. The effect of the estrogen antagonist (ICI 164,384) (ICI) (100 nM) on dihydrogenistein (DHG) stimulated p82 expression in estrogen receptor positive MCF-7 human breast cancer cells cultured i_r_l__vi§m. MCF-7 cells were cultured in IMEM medium in the presence of DHG (1 uM), genistein (GEN) ( 0.1 nM) or estrogen (E2) (1 nM) with or without ICI (100 nM). RNA was isolated after 72 hr of treatment. Northern blot analysis and detection of p82 was performed as described in methods. C represents the vehicle control containing a 1 : 1000 dilution of DMSO in growth medium. Northern blots were also probed with human glucose-3-phosphopate-dehydrogenase (G3PDH) cDNA (Clontech Laboratories, Inc., Palo Alto, CA) as an internal control, to confirm equal loading of RNA among treatment and control groups. 81 C) ‘u-‘é E c N m E2+ICI GEN 0.1uM GEN +ICI DHG luM DHG+ICI 82 Figure 12. Effects of dihydrogenistein (DHG) and genistein (GEN) on the binding of [’Hlestradiol to the ER. ER binding assays were performed as described in Materials and Methods. Competitors [estradiol (E2), GEN, DHG] were added at the indicated concentrations with 1 nM [3H]estradiol. The results (meaniSE, n = 3) are expressed as a percentage of control. % Control gnu _ _ _ 98. PS 9. 363.352.. :2. ‘ lllmn l¢l Omz lbl DID 84 Figure 13. Effects of dihydrogenistein (DHG) and genistein (GEN) on overall protein tyrosine phosphorylation in EGF-preteated MCF—7 cells. MCF-7 cells were cultured in the presence of dihydrogenistein (DHG) (300 uM) or genistein (GEN) (300 nM) with or without EGF (100 nM) which was added 15 min before the start of treatment. Total protein was collected after 24 hr. Western blot and detection for protein tyrosine phosphorylation was performed using procedures described in Materials and Methods. C represented the vehicle control which contained a 1 : 1000 dilution of DMSO in grth media. C EGF GEN+EGF GEN DHG+EGF DHG 86 Figure 14. Effects of dihydrogenistein (DHG) and genistein (GEN) on EGF receptor protein expression (A) and overall protein tyrosine phosphorylation (B) in EGF- preteated MDA-mB-231 cells. MDA-mB-23l cells were cultured in the presence of dihydrogenistein (DHG) (300 nM) or genistein (GEN) (300 nM) with or without EGF (100 nM) which was added 15 min before the start of treatments. Total protein was collected after 24 hr. Western blot analysis and detection of protein tyrosine phosphorylation (B) or EGF receptor (A) was performed using procedures described in Materials and Methods. A431 cell lysate was used as a positive control for EGF receptor (A). C represents the vehicle control containing a 1 : 1000 dilution of DMSO in growth media. HJIDEI ML (1 87 F! 0 lg fl A-431 C EGF GEN+EGF GEN DHG+EGF DHG C EGF GEN+EGF GEN DHG+EGF DHG 88 Figure 15. Effects of dihydrogenistein (DHG) and genistein (GEN) on p21 crrrrwxrr and p53 protein expression in MCF—7 cells. MCF-7 cells were cultured in the presence of dihydrogenistein (DHG) (50 uM) or genistein (GEN) (50 nM). Total protein was collected after 48 hr of treatment. Western blot analysis and detection of 1 CIPW AF 1, p53 and actin was performed using procedures described in Materials and p2 Methods. C represents the vehicle control which containing a l : 1000 dilution of DMSO in growth medium. Western blot was also probed for anti-human actin as an internal control to confirm equal loading of total protein among treatment and control groups. 89 mun wmu >02: Con GEN DHG Ii! CHAPTER 5 Genistein Suppresses the Proliferation and Induces the Differentiation of a Normal Human Breast Epithelial Cell Type With Stem Cell Characteristics in Vitro: a Possible Chemopreventive Mechanism of Genistein for Human Breast Cancer ABSTRACT Genistein, a natural isoflavonoid phyto-estrogen found in soy and other plant foods, has been shown to inhibit the growth of some breast cancer cell lines m at higher concentrations. The low incidence of breast cancer in countries with a flavonoid-rich soy-based diet and the suppression of experimental mammary tumors by prepubertal genistein treatment in rats suggest that genistein may exert a chemopreventive effect on human breast cancer. Two types of morphologically distinguishable normal human breast epithelial cells (HBEC) have been derived from reduction mammoplasty. Type I HBEC have luminal epithelial and stem cell characteristics; i.e. the ability to differentiate into Type II cells with basal epithelial cell phenotypes and the ability to form budding/ductal structures on Matrigel. These cell cultures were to assess the potential effect of genistein on chemoprevention of human breast cancer. We have analyzed the effects of genistein on cell proliferation, cell differentiation and cell cycle progression (flow cytometric analysis with propidiurn iodide-stained cells) in both Type I and Type II HBEC. Genistein, at concentrations lower than 1 uM, significantly increased the differentiation of Type I HBEC to Type D cells in one of two primary cultures derived 9O 91 from different human subjects. Genistein completely arrested cell growth of Type I HBEC at concentrations higher than 5 M and Type II HBEC at concentrations higher than 50 MA after 72 h treatment in all the 6 independent primary cultures examined. Flow cytometric analysis revealed that genistein was able to arrest cell cycle progression of both Type I and Type H HBEC at both Gl/S and G2/M checkpoints. Western blot analysis showed that the level of pZIWAFI’Cm , which negatively regulates the G1/S transition, and cdc2 protein, which positively regulates the GZ/M transition, are significant enhanced and decreased respectively after 72 h genistein treatment (50 nM) in both Type I and Type II HBEC. Type I HBEC have been shown in previous studies to be more susceptible to neoplastic transformation than Type II cells and Type I cells have also been shown in this study to be more sensitive to grth inhibition. Therefore, the reduction in target stem cells for neoplastic transformation might be a chemopreventive mechanism for genistein. INTRODUCTION Epidemiological data suggest that one explanation for the low incidence of breast cancer in oriental women is the consumption of a soy-rich diet (Messina et al., 1991). Genistein, which is one of the principle isoflavonoids in human soy-rich diet (tofu, soy- milk, soy-flour etc), could be the chemical responsible for the chemopreventive activity of soy products. Indeed, genistein has been shown to suppress mammary tumors induced by dirnethylbenz[a]anthracene (DMBA) and to enhance mammary gland differentiation in female Sprague-Dawley SD rats which were exposed to genistein at the 92 prepubertal stage (Murrill et al. 1996 and Lamartiniere et al. 1995). These data imply that genistein may play a role in human breast cancer prevention. Although the studies of Murrill et al., 1996 and Lamartiniere et al., 1995 demonstrate a chemopreventive effect of genistein for rat mammary tumors and imply that the induction of mammary gland differentiation is a possible chemopreventive mechanism of genistein, the biological effects and mechanisms of function of genistein in human mammary gland and normal human breast epithilial cells are largely unknown. Recently, a culture method has been developed to grow two morphologically distinguishable cell types of normal HBEC (Type I and Type 11), derived from reduction mammoplasty tissues (Kao et al., 1995). Type H cells express basal epithelial cell markers (i.e. cytokeratin 14 and or 6 integrin) whereas Type I HBEC express luminal epithelial cell phenotypes (i.e. cytokeratin 18 and epithelial membrane antigen, EMA) (Kao et al., 1995). Furthermore, Type 1 cells are deficient in gap junctional intercellular communication and possess stem cell characteristics (i.e. the ability to differentiate into Type H cells by cyclic AMP enhancing agents and the ability to form budding/ ductal structures on Matrigel) (Kao et al., 1995; Chang et al., 1996). Significantly, Type I HBEC express estrogen receptors (Kang et al., 1997) and have been found to be more susceptible to neoplastic transformation. After SV40 large T-antigen transformation, the Type I cells acquire anchorage independent growth and become immortal spontaneously at high frequency (Kao et al., 1995). 93 This HBEC system appears to be a relevant system to examine the effect of genistein on normal HBEC in relation to carcinogenesis. It provides an opportunity to test 1) if genistein is capable of inducing differentiation of Type [I HBEC into Type cells, 2) if Type I and Type II HBEC proliferation is differentially affected by genistein, and 3) if the expression of cell cycle regulating genes is affected by genistein. These are the objectives of this study. MATERIALS AND METHODS Culture Media The culture medium used in these studies to culture Type I and Type II HBEC is the MSU-l medium which is a 1:1 mixture (vol/vol) of a modified Eagle’s MEM (GIBCO BRL Life Technologies, Grand Island, NY) (D medium) (Chang et al., 1981) and a modified MCDB 153 (M-7403, Sigma) (Pittelkow et al., 1986) supplemented with EGF (0.5 ng/ml) (E-1264, Sigma Chemical Co., St. Louis, MO), insulin (5 rig/ml) (1-1882, Sigma), hydrocortisone (5 ug/ml) (H-0888, Sigma), human transferin (5 rig/ml) (T-7786, Sigma), and 17 B-estradiol (132) (1 x 10“ M) (5.2257, Sigma). The modified Eagle’s MEM (D medium) contains Earle’s balanced salt solution with 1 mg/ml sodium bicarbonate and 7.64 mg/ml sodium chloride, a 50% increase in all vitamins and essential amino acids (except glutamine), a 100% increase in all nonessential amino acids, and 1 mM sodium pyruvate (pH adjusted to 6.5 before the addition of sodium bicarbonate). 94 The Modified Eagle’s MEM (D medium) with 5% fetal bovine serum (FBS) (GIBCO) and penicillin (100 units/ml) and streptomycin (100 rig/ml) was used to grow MCF-7 and MDA-mB-231 cells (American Type Culture Collection). Acquisition, Processing, and Culturing of Human Breast Epithelial Cells (HBEC) Reduction mammoplasty tissues were obtained from six female patients of 18, 56, 23, 37, 38 and 18 year of age. The HBEC obtained from these reduction mammoplasty tissue specimens were designated as HME-21, HME-22, HME-23, HME-24, HME-25 and HME-27 in order. The tissue specimens were minced into small pieces with scalpels, then digested in collagenase-Type IA (C-2674, Sigma) solution (1 g tissue per 5,000 units of collagenase in 10 ml medium) at 37°C in a waterbath overnight (16-18 hours). The next morning, the solution containing the digested tissues was centrifuged to remove the collagenase solution. The cellular pellet was washed once with MSU-l medium before being suspended in the MSU-l medium supplemented with 5% fetal bovine serum (FBS) (GIBCO). Subsequently, the cells were plated in two flasks (150 cmz). After a 2 hour incubation, the cells (or cell aggregates) that remained in suspension were transferred to four to six flasks (75 cm2) for the purpose of reducing the number of attached fibroblasts. After an overnight incubation, the medium was changed to the FBS-free MSU-l medium. The MSU-l medium was changed once every 2 days for 1 week. Subsequently, the cells were removed with a solution of trypsin (0.01%) (Sigma) and ethylenediaminetetraacetic acid (EDTA) (0.02%) (Sigma) and stored in solution [phosphate buffered saline (PBS) containing 10% dimethylsulfoxide (DMSO)] 95 in liquid nitrogen. During this 1 week period, almost all of the fibroblasts can be removed by treatment (one to two times) with diluted trypsin (0.002%) solution. To start a culture from stored frozen cells, the frozen cells in liquid nitrogen were thawed and placed in MSU-l medium supplemented with 5% FBS for 4 hours for the attachment of residual fibroblasts. The epithelial cells in suspension were transferred to new plates and cultured in the FBS-free MSU-l medium. All cultures were incubated at 37°C in incubators supplied with humidified air and 5% C02. Separation of Type I and Type H HBEC The first passage of HBEC, recovered from liquid nitrogen storage, was plated in the MSU-l medium supplemented with 5% FBS as described above. After overnight culture, the cells that remained in suspension were transferred to new plates. Continued culture of these suspended cells, which later attached, in the FBS-containing medium gave rise to Type I cells. The attached cells, in the overnight culture, incubated in the FBS-free MSU-l medium supplemented with 0.4% BPE (Pel-Freez, Rogers, AR) gave rise to Type II cells. Assessment of HBEC Differentiation inflim To determine the effect of genistein on Type I HBEC differentiation, Type I HBEC were grown in several different treatment groups containing MSU-l medium and various concentrations of genistein (0.01, 0.1 and 1 nM) or DMSO at a 1:1000 dilution (solvent 96 control). Cholera toxin (CT) (1 ng/ml) (Sigma) was used as a positive control (Kao et al., 1995). Starting from single cell plating of pure Type I cells, the differentiation of Type I HEBC was measured by counting the number of Type II HEBC colonies and colonies of Type I surrounded by Type D cells. The percentage of these colonies among total colonies (Type 1, Type H and Type I surrounded by Type 11 colonies) indicates the differentiation potential of Type I cells under different treatments. Briefly, Type I cells(1 x 10‘) were plated in 60 mm plates in triplicate in the 5% FBS-containing MSU-l medium (which promotes the growth of Type I and inhibits Type II cell growth). The next day, the medium was replaced by the FBS-free MSU-l medium (which supports the growth of both Type I and Type H cells) for one day. Then chemicals for various treatments as described above were added to the FBS-free MSU-l medium. All cells were incubated for 7 days at 37°C (media changed twice). At day 7 after chemical treatments, the cells were washed twice with PBS and fixed with 2 m1 of 70% ethanol containing 0.1% acetic acid 2 ml crystal violet solution per plate was added to stain the colonies. Then, Type 1, Type I surrounded by Type D and Type II colonies were visually identified under a microscope and quantitated (Fig. 16). The identity of treatment for each plate was unknown (blind) during counting to ensure objectivity. Assessment of HBEC proliferation iayitm To determine the effect of genistein on growth of both Type I and Type II HBEC, HBEC were grown in MSU-l medium containing various concentrations of genistein (0.1 to 100 uM). The growth of I-IEBC was measured by quantitation of total nucleic acid 97 extracted from the culture (Li et al., 1990). Briefly, Type 1 (1 x 104 cells) and Type 11 (1 x 104 cells) HBEC (passage 2) were plated in 35 mm plates in triplicate in the 5% FBS- containing MSU-l medium for Type I HBEC and in the 0.4% BPE-containing MSU-l medium for Type H HBEC. The next day, the medium was replaced by the FBS- and BPE-free MSU-l medium for one day, then various concentrations (0.1 to 100 uM) of genistein were added to the FBS and BPE-free MSU-l medium. Cells were incubated for 7 days at 37°C (media changed twice) for measuring the dose—dependent growth at day 7. For measuring the growth curves with various concentrations of genistein treatments, the cells were harvested at the day the treatments started (day 0), and at day 1, 3, 5, 8 after the treatments. To harvest cells, the cells were washed twice with PBS and lysed with 2 ml of 0.1 N sodium hydroxide. The lysate was transferred into a 2.2 ml Eppendorf tube and centrifuged at 14,000 rpm for 2-3 min. The absorbance of the clear lysate at 260 nm was measured using a Beckrnan DU-7400 spectrophotometer (Schaumburg, IL). Each treatment was done in triplicate plates and the Lorentzian wave form curve fitting formula was used to determine the IC50 and the ICloo of genistein treatments in both Type I and Type II HBEC cells. Flow Cytometric Measurement and Cell Cycle Analysis A quantitative measure of cell cycle distribution was obtained by flow cytometric analysis of DNA content - cell number fi'equency histograms, as described in Fraker et al. (1995). Type I and Type II HBEC were incubated for 72 hours then the medium was replaced with fresh maintenance medium (MSU-l) containing genistein at 0, 5, 10, 25, 98 50, and 100 MA and collected after 3 days of treatment. Type II HBEC were also collected after incubation for 6 and 13 hours in MSU-l medium following the 72 hours treatment with genistein at 50 M for cell cycle recovery analysis. Briefly, there were two steps in this procedure. 1) Fixation of cells: HBEC (second passage) were collected from T-25 flasks at 80% confluence by washing two times with PBS followed by trypsinization. Cell number was determined using a hemocytometer and the cell suspension diluted to approximately 1 x 106 cells/ml in MSU-l medium. For analysis of cell viability, a 50 ul aliquot from each sample was collected into a 0.5 ml microfuge tube. After the addition of 25 ul of 10% trypan blue, the sample was incubated at room temperature for 5 min, and cell viability was determined by the exclusion of trypan blue. For flow-cytometry analysis, the remaining samples were centrifuged at 350 x g for 5 min. and the supernatant removed. Then, the cells were resuspended in 5 ml of PBS and transferred to a Falcon 2056 tube. The cells were pelleted by low speed centrifugation (350 x g) for 5 min. The supernatant was aspirated and the pellets were washed with PBS. The cell pellet was resuspended, at a density of 1 x 106 cells/ml, in ice cold 70% ethanol with rapid but gentle mixing. After the cells were fixed in ethanol for 1 to 3 hours at 4°C, the sample could be stored at -20 °C until analysis. 2) Cell staining for the flow activated cell sorter (FACS): the cells were centrifuged at 400 x g for 5 minutes to remove the ethanol, the cellular pellet was washed with PBS, and then flow cytometric DNA staining reagents: 0.1 mM EDTA (pH 7.4), 0.1% of Triton X-100, 0.05 mg/ml RNase A (50 units/mg), and 50 ug/ml propidium iodide (P1) in PBS (pH 7.4). The tubes were gently vortexed and placed in the dark at 4°C overnight until reading on the flow activated cell sorter (FACS). Fluorescence was assessed on a FACS Vantage 99 (Beckton Dickinson) by excitation with an Argon laser at 488 nm and the emission measured at 620 to 700 nm. Data were collected with Lysis II software and the percent of cells in each phase of the cell cycle calculated with MPLUS software (Phoenix Flow). SDS-PAGE and Western blot analysis Proteins were extracted from normal HBEC and from MCF-7 cells grown in 60 mm dishes by treatment with 20% SDS lysis solution containing several protease and phosphatase inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 uM leupeptin, 1 uM antipain, 0.1 uM aprotinin, 0.] M sodium orthovanadate, 5 mM sodium fluoride). Afier sonication at three 10-s pulses from a probe sonicator, the cell lysates were stored at -20°C until use (Kang et al., 1996). The protein amounts were determined by the DC protein assay kit (Bio-Rad Co., Richmond, CA). Proteins were separated on 12.5% SDS polyacrylamide gels and transferred to Irnmobilon PVDF membranes (Millipore Co., Bedford, MA) at 90 V for 2 hours. The blots were blocked with 5 % dried skim milk in PBS containing 0.1 % Tween 20. P21 ”WA“ was detected by the anti-p21 C‘P'M" polyclonal antibody [(C-19)-G], (Santa Cruz Biotechnology, Inc. Santa Cruz, CA) which recognizes amino acids 146-164 mapping at the carboxyl terminus of human p21 CI? ”w AF 1. P53 was detected by the anti-p53 monoclonal antibody (Ab-2, Oncogene Science, Cambridge, MA) which recognizes amino acids 46-55 of human p53 and reacts with human wild-type and mutant p5 3. Cdc2 kinase was detected by the anti-cdc2 polyclonal antibody (Ab-1, Oncogene Science, Cambridge, MA) which recognizes the carboxyl terminal 8 amino acids of human cdc2. The anti-actin antibody (Oncogene Science, 100 Cambridge, MA) was utilized to detect actin, an internal indicator of total protein level in samples. Western blots were incubated with Horseradish peroxidase-conj ugated secondary antibody and detected with ECL chemiluminescent detection reagent (Amersham Co., Arlington Heights, IL). The membranes were exposed to X-ray film for 15 sec to 1 min. RESULTS The effect of genistein on the differentiation of Type I HBEC into Type II HBEC. Two types of normal HBEC can be developed in pure culture from reduction mammoplasty tissues. Only Type I HBEC at the second passage were used for induction of differentiation into Type H cells by genistein. As shown in Table l, genistein at concentrations from 0.01 M to 1 uM significantly increased the frequency of colonies containing Type H cells in HME 21. The effect of genistein was not as potent as cholera toxin, a positive control known to induce the differentiation (Kao et al. , 1995). The results from the experiment using a different HBEC culture derived from a different woman (HME 23), however, failure to show the differentiation ability of genistein (Table 1). The induction of differentiation of Type I to Type II HBEC by low concentrations of genistein is usually accompanied by enhanced total cell proliferation in the population. This was seen with HME 21 and HME 23 responded differently to low concentrations of genistein. The effects of different concentrations (5-100 uM) of genistein on growth of 101 Type I and Type I] HBEC in a 8-day time-course. Pure cultures of Type I and Type II HBEC at passage 2 were used to study the growth response to different concentrations of genistein. The results of these studies are presented in Figure 17, 18 and Table 2. In Figure 17, growth response curves in a 8-day time course were obtained for Type I and Type II HBEC of HME 21 exposed to different concentrations of genistein. The results show that 5 11M or higher concentrations of genistein completely inhibited the growth of Type I cells (Figure 17a). For Type 11 cells, 5-25 M of genistein showed partial inhibition of cell grth in a dose-dependent manner. A complete inhibition of cell grth was observed when the cells were exposed to 50 11M or higher concentrations of genistein. It is clear that Type I cells were much more sensitive to growth inhibition by genistein than Type 11 cells. The effect of increasing concentrations of genistein on the growth of Type I and Type II HBEC: dose-dependent inhibition. Results in Figure 18 showed that Type I cells of HME 21 (Fig. 18a) and HME 23 (Fig. 18b) responded differently to low doses of genistein. Genistein at 0.01— 1 uM stimulated cell growth in HME 21 Type I cells, but not HME 23 Type I cells. The stimulation of cell growth by low doses of genistein in HME 21 cells is believed to be due to the induction of Type I to Type D cell differentiation by genistein. The newly converted Type 11 cells are known to be highly proliferative compared to Type I cells. Genistein at 1 to 5 uM could completely arrest cell growth in HME 21 Type I cells and could completely arrest cell growth at concentration around 50 M in Type H cells (Fig. 18a). Genistein at concentration 102 around 10 M could completely arrest cell growth in HME 23 Type I cells and at concentration around 50 1.1M could completely arrest cell growth in Type H cells (Fig. 18b). Other different HBEC cultures have also been examined. Table 2 list the concentrations of genistein for complete (ICloo) and 50% (ICso) arrest of cell growth among cancer cells, Type I and Type H HBEC of different women after 7 day treatment. Consistently, the results indicate that Type 1 human breast epithelial cells derived from different women were more sensitive to growth inhibition by genistein than Type H cells (Fig. 3 and Table 2). Furthermore, normal human breast epithelial cells were found to be much more sensitive to grth inhibition by genistein than the breast carcinoma, MCF-7, cell line(Table 2). The effect of different concentrations of genistein on cell cycle progression of both Type I and Type II HBEC. The flow cytometric analysis revealed that genistein at 5-25 uM did not significantly change the distribution of cells at different phases of cell cycle in HME 21 Type I cells. From Fig. 17a, it is known that 5-25 uM genistein completely arrested the cell growth. It seems that these concentrations of genistein stop the cell cycle progression resulting in no accumulation of cells in a particular phase. For Type H cells, genistein at 50-100 uM significantly arrested the cells at G2. This effect seems to be reversible when genistein was removed from the medium (Fig. 19 and Table 3). WAFl/CIPI The effect of genistein on p21 and cdc2 kinase protein expression in 103 both Type I and Type II HBEC. Western blot analysis showed that the expression of pzl‘m‘m , which negatively regulates the 01/8 transition, and cdc2 kinase, which regulates the G2/M transition, were significantly enhanced or decreased respectively after 72 h of 50 M genistein (or 25 11M, data not shown) treatment in both Type I and Type H ' HBEC (Fig. 20) of I-IME 24 (or HME 21, data not shown). There were no change in the level of p53 and p16 protein expression (data not shown). Therefore, the cell cycle progression arrest could be mediated through a p53, pl6-independent mechanism. Thus, our results showed that genistein at 50 M, which is a cytostatic concentration, can WAFl/CIPI increase the p21 protein expression and decrease the cdc2 kinase protein expression to cause cell cycle arrest. DISCUSSION Genistein, a component of soy, is suspected to be a chemopreventive agent for breast cancer (Barnes et al., 1990). This is supported by animal experiments showing that genistein treatment during the prepubertal period can suppress the development of chemically-induced mammary tumors in rats (Murrill at al., 1996). Although the effect of genistein has been studied mm in breast cancer cell lines as described previously, its effect on normal human breast epithelial cells has not been reported. We believe the study of genistein on normal HBEC is more relevant to evidence concerning its chemopreventive potential. Fmthermore, since we have developed two types of normal HBEC from reduction mammoplasty tissues and showed that one of the cell types possessed stem cell characteristics and was more susceptible to neoplastic transformation, 104 we have a very relevant cell culture system to study the effect of genistein. The major result fiom this study is the finding that Type I HBEC were more sensitive to growth inhibition by genistein than Type H cells in all six different HBEC cultures examined. The doses of genistein that inhibited Type I cell growth are at low concentrations (0.1 to 1 M) which can be easily attained by some vegetarian women (Adlercreutz et al., 1993). In contrast, Type H HBEC and breast cancer cells (MCF-7) were inhibited by higher concentrations of genistein (> 50 M) which are beyond the physiological dose. The implication of this differential response is that the grth of basal or myoepithelial cells and cancer cells may not be inhibited by physiological doses of genistein in the body. On the other hand, the inhibition of Type I cell growth by physiological doses of genistein could reduce the number of Type I cells in the mammary gland Since Type I HBEC have stem cell characteristics and have been found to be more susceptible to neoplastic transformation, genistein could effectively reduce the number of target cells for neoplastic transformation and thereby function as a chemopreventive agent. Our study also revealed a possible mechanism by which genistein inhibited the growth of normal HBEC, i.e. by affecting the expression of cell-cycle regulating genes that arrested the cell cycle progression Specifically, the expression of p21 “W ”cm was enhanced while the expresson of cdc2 was decreased by genistein. The modulation of the expression of these genes could effectively arrest cells at G1 and G2 respectively. Cell cycle analysis confirmed that in Type I HBEC the distribution of cells at G1 and GZ 105 was not affected by concentrations of genistein that completely stopped cell growth. For Type H cells, there was a clear accumulation of cells at GZ by genistein that inhibit cell growth. The difference between the response of Type I and Type H cells to genistein could reflect the fact that there were more cells at S phase in the initial population of Type H cells than Type I cells. The study of the ability of genistein to induce differentiation of Type I to Type H cells resulted in the finding of positive response in one of two independent HBEC cultures studied The positive response in HME 21 appears to be real, since it coincided with the increase in cell proliferation by the low doses of genistein. The latter effect is interpreted as due to the induction of differentiation of Type I cells to Type H cells. These newly differentiated cells are highly proliferative. The variable response of different cell cultures derived from different individuals to genistein-induced differentiation is not clear. It could be due to developmental or genetic factors and should be investigated in the future. Since Type I cells in conjunction with Type H HBEC are able to form mammary organoid (budding/ductal structures) on Matrigel, it is possible to study the effect of genistein on the development of human mammary gland in the m system. The effect and mechanism of firnction of genistein revealed by this study should be confirmed by using this system and other studies in the future. 106 Figure 16. Normal human breast epithelial cells (HBEC) cultured in MSU-l medium contained three types 0 f epithelial cell colonies: Type I cells (A), Type [I cells (B), and Type I cells surrounded by Type 11 cells (C) (Kao et al., 1995). In pure Type I cell culture, the appearance of the latter two colonies indicates the differentiation of Type I to Type H cells. F... ru. r wu hT-l pill 108 Table 1. The effect of genistein on differentiation of Type I HBEC into Type II HBEC. Treatments No. of Type H and Type I surrounded by Type H colonies/ total colony no. (% of Type H and Type I/H colonies) HME 21 Control“ 57/159 (36.0%) Cholera toxin (1 ng/ml) 214/245 (87.4%)" Genistein (0.01 uM) 65/118 (55.1%)" Genistein (0.1 uM) 68/108 (63.0%)" Genistein (1 M) 49/73 (67.0%)" HME 23 Control“ 22/42 (52.8%) Cholera toxin (1 ng/ml) 43/54 (78.5%)" Genistein (0.01 nM) 20/36 (56.1%) Genistein (0.1 uM) 20/34 (58.4%) Genistein (1 M) 12/25 (48.0%) ’Vehicle control contains 1 to 1000 dilution of DMSO in MSU-l medium. *" highly significantly different from the control (P<0.01). 109 Figure 17. The effects of different concentrations (5-100 nM) of genistein on HME 21 Type I (a) and Type II (b) HBEC growth in a 8-day time-course experiment. Type I and II HBEC were cultured in the presence of various concentrations of genistein(5 uM-IOO nM) or DMSO as control for 1,3, 5 and 8 days, in MSU—l medium at 37°C in a humidified atmosphere of 5% C02 in air. Proliferation was assessed by DNA content as measured by lysis of cells with 0.1 N NaOH and reading the absorbance of the clear lysate at 260 nm. The units on the Y axis are expressed as equivalent cell number per well (12 well dishes) determined by DNA content. Results are expressed as average of triplicate plates i SE. (standard error). control “0" gonSuM —*— control "0' gen 5 all -t—° gen 10uM 110 2 3 4 5 s 7 s 9 Days after cell plating 2 3 4 5 s 7 s 9 J ‘l 1 O 953.8 83.: =2; .8 .2. =8 953.8 Sod: =2, .8. .2. :8 (b) (t) Days after cell platlng lll Figure 18. Dose-dependent inhibition of Type I and Type II HBEC growth by genistein treatment for 7 days. Two different cell cultures were used: (a) results of HME 21 and (b) results of HME 23 (open circles, Type I cells; open triangles, Type H cells). Type I and H HBEC were cultured in the presence of various concentrations of genistein (0.1, 1, 10, 25 and 50 nM) or DMSO (solvent control) for 7 days, in MSU-l medium at 37°C in a humidified atmosphere of 5% CO; in air. Proliferation was assessed by DNA content as measured by lysis of cells with 0.1 N NaOH and reading the absorbance of the clear lysate at 260 nm. Results shown are mean i S.E. (standard error) from triplicate experiments relative to cell growth without genistein treatment. 1.12 (a) 200 175 ' 150 0 125 Relative Growth genistein conc. (uM) (b) 125 100 75 Relative Growth genistein conc. (uM) 113 Table 2. The concentrations of genistein for complete (ICm) and 50% (1C5...) arrest of cell growth among cancer cells, Type I and Type II HBEC after 7 day treatment. Cell lines The concentrations of genistein (IC50) for 50% arrest of cell grth The concentrations of genistein(IC100) for complete arrest of cell growth Cancer cell lines MCF—7 25 M > 100 W MDA-mB-231 30 M > 100 M HME 21 Type I HBEC 1 11M 1 uM Type II HBEC 40 11M 50 11M HME 22 Type I HBEC 1 11M 10 11M Type H HBEC 7 1.1M 25 M HME 23 Type I HBEC 5 11M 10 M Type H HBEC 7 M 50 M HME 24 Type I HBEC 5 M 10 M Type H HBEC 7 M 50 M HME 25 Type I I-IBEC 5 M 10 M Type II HBEC 16 M 50 M HME 27 Type I HBEC 2 11M 10 M Type H HBEC 19M 50 M " HBEC designates normal human breast epithelial cells. 114 Figure 19. Flow-cytometric analysis of cell cycle distribution of HME 21 Type I (a) and Type 11(h) HBEC exposed to different concentrations of genistein. Type I and H HBEC were cultured in the presence of various concentrations of genistein(5 uM-IOO 11M) or DMSO (solvent control) for 3 days, in MSU-l medium at 37°C in a humidified atmosphere of 5% CO; in air. The DNA content-cell number frequency histograms display the dose-dependent effect of a 72-h treatment with increasing genistein doses in a representative experiment (S-phase reduction, Gl/S and G2/M arrest). 115 3 Cell Number AS 02:3. 0233:. m ES v. P III .41‘ I IrJ 62.38.: 3 ES @258... nm ES Cell Number J 0256. noimnams um ES II. 1.41 4‘ 02.38... 3 ES . 02>. 02.3.: 4 02.38.: :5 ES 116 Table 3. The effect of different genistein concentrations on cell-cycle distribution of Type I and Type II HBEC after 72-h treatment. Treatments G1 phase (%) S phase ("/3 G2 phase (%) HMEZI type H: Control 61.6 19.4 19.0 Genistein (10 uM) 64.0 19.8 16.2 Genistein (25 M) 65.2 16.5 18.3 Genistein (50 M) 47.5 2.4 50.1 6 h recovery" 51.9 5.3 42.8 13 h recovery“ 66.8 8.1 25.1 Genistein (100 nM) 39.6 14.5 45.9 I-IME21 type 1: Control 80.0 8.3 11.7 Genistein (5 uM) 76.4 7.5 16.1 Genistein (10 M) 79.3 6.1 14.5 Genistein (251,1M) 73.7 2.9 23.4 Data are expressed as percent of the cells in each phase of the cell cycle. Viabilities assessed by trypan blue exclusion were greater than 93% for all concentrations and time points. Vehicle control contained 1 to 1000 dilution of DMSO in MSU-l medium. ’Cells were allowed to recover for 6 hr or 13 hr in fresh growth medium after the 72 hr treatment. 117 Figure 20. The effect of genistein on the expression of p21 w“ 11cm and cdc2 kinase protein in Type I and Type II HBEC. HME 24 Type I and Type H HBEC were cultured as described above in the presence genistein (GEN) (50 nM) or DMSO (solvent control). Total proteins were collected after 48 hours of treatment. Western blot and / ~ , ‘ . W” ‘ c1131, cdc2 krnase and actin were performed usrng procedures detection of p21 described in Materials and Methods. Vehicle control contained a 1 : 1000 dilution of DMSO in grth medium. Western blot was also probed for actin using an anti-human actin as an internal control, to confirm equal loading of total proteins among treatment and control groups. ...-i, ‘ -’ “93V -.—-—~ [zd 19133 118 Type I Type I+GEN Type II g; Type II+GEN CHAPTER 6 CONCLUSIONS AND IMPLICATIONS OF THE STUDY This study attempted to elucidate the biological effect and biochemical functions of genistein and dihydrogenistein related to breast carcinogenesis. The following conclusions and implications may be drawn from the results obtained: 1. Genistein at concentrations lower than 10 M, which is within the physiological range, is able to promote the growth of tumors formed by ER-positive breast cancer cells in vivo through its estrogenic action. Therefore genistein could be detrimental to postmenopausal women who have low levels of circulating estrogen and develop mostly ER-positive breast cancer. 2. The IC50 of genistein to inhibit cancer cell growth is always higher than 15-30 uM in several cancer cell lines studied m (Barnes, 1995 and this study). The maximum physiological concentration of genistein has been estimated to be no more than 15 M by dietary administration (Barnes, 1995). Therefore, it is unlikely that cancer can be cured by eating soy or soy products. On the other hand, if genistein is administered as a pill to reach a high level (> 50 nM) in the body, it could kill normal breast cells as well as cancer cells according to this study. Therefore, it may not be feasible to use genistein as a chemotherapeutic drug. However, genistein has been conjugated to an antibody to target CD 19 cell surface receptor present in B-cell precursor leukemia to selectively inhibit CD 19-associated transmembrane receptor tyrosine kinases and triggered rapid apoptotic cell death (Uckurn et al., 1995). The 119 120 potential use of this strategy in solid tumors is not clear. 3. Similar to genistein, dihydrogenistein promotes breast cancer cell growth in a dose-dependent manner from 1 nM to 80 1.1M in ER-positive but not in ER-negative breast cancer cells. The grth promoting effect can be blocked by the estrogen antagonist and coincides with its induction of the expression of the ER-responsive, p82 protein. Dihydrogenistein differs from genistein in failure to inhibit cell growth at higher concentrations (about 25 nM) in either ER-positive or ER-negative breast cancer cells and in the failure to induce p21WAF ”cm expression at 50 M. This indicates that dihydrogenistein retains the estrogenic activity of the parental compound, genistein, but loses its ability to block cell-cycle progression and to inhibit cancer cell growth. Therefore, dihydrogenistein may be as harmful as genistein and probably no better than genistein as a chemopreventic agent. 4. Genistein preferentially inhibits the growth of a normal HBEC type with stem cell characteristics (Type I HBEC) at physiological concentrations (0.1 to 1 uM). Since Type I HBEC have been shown to be more susceptible to neoplastic transformation i_n \_/l_t_l'_Q, genistein could reduce the number of target cells for breast carcinogenesis, thereby functions as a chemopreventive agent. The mechanism appears to be mediated by the WAFl/CIPl ability of genistein to induce the p21 and to inhibit the cdc2 cell-cycle regulation protein expression. The conclusions from these studies using in vitro or animal models have their limitations. The potential mechanism that mediates the chemopreventive function of genistein as suggested by this study is consistent with evidence that genistein enhances 121 rat mammary gland differentiation (Murrill et al., 1996). The hypothesis needs to be substantiated by other epidemiological, clinical trial or in vitro human mammary organoid studies. Genistein was found to induce the differentiation of Type I to Type H HBEC in one of two cell cultures derived from different human subjects. The extent, mechanism and significance of this variable response are not known. The amount of dihydrogenistein in the human body as a result of metabloism, absorption and elimination is not known. These are some of the issues need to be investigated in future studies. LIST OF REFERENCES LIST OF REFERENCES Adlercreutz, H., P.I. Musey, T. Fotsis, C. Bannwart, K. Wahala, T. Wakela, G. Brunow, and T. Hase. Identification of lignans and phytoestrogens in urine of chimpanzees. Clinica Chimica Acta 158, 147-154, 1986a . Adlercreutz, H., T. Fotsis, C. Bannwart, K. Wahala, T. Wakela, G. Brunow, and T. Hase. Determination of urinary lignans and phytoestrogen metabolites, potential antiestrogens and anticarcinogens, in urine of women on various habitual diets. J. Steroid Biochem. 25, 791-797, 1986‘” . Adlercreutz, H. Western diet and western diseases: some hormonal and biochemical mechanisms an dissociations. Scand. J. Clin. Lab. Invest. 50:3-23, 1991. Adlercreutz, H., T. Fotsis, J. Lampe, C. Bannwart, K. Wahala, T. Wakela, G. Brunow, and T. Hase. Quantitative determination of lignans and isoflavonoids in plasma of omnivorous and vegetarian women by isotope dilution gas chromatography-mass spectrometry. Scand. J. Clin. Lab. Invest. 53 (suppl. 215):5-18, 1993. Adlercreutz, H. , C.I-I.T. Goldin, B.R. Gorbach, T. Fotsis, C. Bannwart, K. Wahala, T. Wakela, G. Bnlnow, and T. Hase. Soybean phytoestrogen intake and cancer risk. J. Nutr 125:757s-7705, 1995. Akimura, K., T. Utsumi, E.F. Sato, J. Klostergaard, M. Inoue and K. Utsumi. Role of tyrosyl phosphorylation in neutrophil priming by tumor necrosis factor-a and granulocyte colony stimulating factor. Arch. Biochem. Biophys. 298:703-709, 1992. Akiyama,T., J. Ishida, H.Ogawara, S.-I. Watanabe and K.-I. Suzuki. Genistein, a specific inhibitor of tyrosine-specific protein kinase. J. Biol. Chemistry 262: 5592-5595, 1987. Banerjee, M.R., B.G. Wood, F .K. Lin, and LR. Crump. Organ culture of whole mammary glands of the mouse. In: Tissue Culture Association Manual, Vol. 2, no. 4: 457-462, 1976. Rockville, MD. 122 123 Barnes, 8., C. Grubbs, and K.D.R. Setchell. Chemoprevetion by powdered soybean chips (PSC) of mammary tumors in Rats. Breast Cancer Res. Treat. 12, 128, 1988. Barnes, 8., C. Grubbs, K.D.R. Setchell, and J. Carlson. Soybeans inhibit mammary tumors in models of breast cancer. Prog. Clin. Biol. Res, 347: 239-253, 1990. Barnes, 8., M. Kirk and L. Coward. Isoflavones and their conjugates in soy foods: extraction conditions and analysis by I-IPLC-mass spectrometry. J. Agric. Food Chem., 42:2466-2474, 1994. Barnes, S. and T. G. Peterson. Biochemical targets of the isoflavone genistein in tumor cell lines. The Society for Experimental Biology and Medicine, 208: 103-108, 1995. Barnes, 8., T. G. Peterson, L. Coward. Rationale for the use of genistein-containing soy matrices in chemoprevention trials for breast and prostate cancer. Supplement 22: 1 81- 187, 1995. Batterham, T.J., D.A. Shutt, N.K. Hart, A.W.I-1. Braden, and H. J. Tweeddale. Metabolism of intraruminally administered [4- ”c1 fonnononetin and [4- “c1 biochanin A in Sheep. Aust. J. Agric. Res. 22:131-138, 1971. Baylin, S.B. Tying it all together: epigenetics, genetics, cell cycle, and cancer. Science 277: 1948-1949, 1997. Brown, A.M.C., J .M. Jeltsch, M. Roberts, and P. Chambon. Activation of pS2 gene transcription is a primary response to estrogen in the human breast cancer cell line MCF- 7. Proc. Natl. Acad Sci. USA 81: 6344-6348, 1984. Cai, Q.Y., R.O. Rahn and R.W. Zhang. Dietary flavonoids, quercetin, luteolin and genistein, reduce oxidative DNA damage and lipid peroxidation and quench free radicals. _. Cancer Letters 119: 99-107, 1997. Cairns, J. Mutation selection and the natural history of cancer. Nature (London) 255:197-200, 1975. Carmer, S.G., and MR. Swanson. An evaluation of ten pairwise multiple comparison 124 procedures by Monte Carlo methods. J. Am. Stat. Assoc. 68: 66-71, 1973. Carter, M.W., G. Matrone, and W.W.G. Smart, Jr. Effect of genistein on reproduction of the mouse. J. Nutr. 55:639—645, 1955. Cassidy, A., S. Bingharn & KDR. Setchell. Biological effects of a diet of soy protein rich in isoflavones on the menstrual cycle of premenopausal women. Am. J. Clin. Nutr. 60:333-40, 1994. Chang, C.C., J .A. Boezi, S.T. Warren, et al. Isolation and characterization of a UV- sensitive hypennutable aphidicolin-resistant Chinese hamster cell line. Somatic Cell Genet. 7:235-253, 1981. Chang, C.C., KS. Kang, C.Y. Kao, C.J. Liu and J .E. Trosko. A normal human breast epithelial cell type with stem cell characteristics and estrogen receptor expression is more susceptible to neoplastic transformation. World Congress on In Vitro Biology. In Vitro Cell Develop. Biol. 32: 14A-15A, 1996. Chang, Y.C., M.G. Nair, R.C. Santell, and W.G. Helferich. Microwave mediated synthesis of genistein. J. Agric. Food Chem. 42:1869-71, 1994. Chang, Y.C. and MG. Nair. Metabolites of daidzein and genistein and their biological activities. J. Natural Products 58: 1901-1905, 1995b. Chang, Y.C. and MG. Nair. Metabolites of daidzein and genistein by intestinal bacteria. J. Natural Products 58: 1892-1896, 1995a. Cheng, B, CD. Story, L. Yoder, W.H. Hale, and W. Burroughs. Estrogenic activity of isoflavone derivatives extracted and prepared from soybean oil meal. Science 118: 164- 165, 1980. Chomczynski, P. and N. Sacchi. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroforrn extraction. Anal. Biochem. 162:156-159, 1987. Clark, J.W., A. Santos-Moore, E.S. Stevenson and A. R Frackelton, Jr. Effects of 125 tyrosine kinase inhibitors on the proliferation of human breast cancer cell lines and proteins important in the ras signaling pathway. Int. J. Cancer 65: 1 86-191, 1996. Colin K.W., W. Kimberley, J .E. Sweeney, A. Warlters, E. A. Musgrove and Robert L. Sutherland Antiestrogen regulation of cell cycle progression and cyclin D1 gene expression in MCF-7 human breast cancer cells. Breast Cancer Research and Treatment 31295-105, 1994. Constantinou, A., K. Kiguchi, and E. Huberrnan. Induction of differentiation and strand breakage in human 1-H.-60 and K-562 leukemia cells by genistein. Cancer Res. 5022618- 2624, 1990. Davis, D.L., H. L. Bradlow, M. Wolff, T. Woodruff, D. G. Hoel and H Anton-Culver. Medical hypothesis: xenoestrogens as preventable causes of breast cancer. Environt. Health Presp. 101: 372-377, 1993. Dees, C., J .S. Foster, S. Ahamed, and J. Wimalasena. Dietary estrogens stimulate human breast cells to enter the cell cycle. Environt. Health Presp. 105: 633-636, 1997. El-Deiry, W.S., V.E. Tokino, and B. Vogelstein. WAF 1, a potential mediator of p53 tumor suppression. Cancer Res. 55:2910-2919, 1995. Eldridge, AC, and W.F. Kwokek. Soybean isoflavones: effect of environment and variety on composition. J. Agric. Food Chem. 31:394-396, 1983. Ek, B., B. Westerrnark, A. Wasteson, and CH. Heldin. Stimulation of tyrosine-specific phosphorylation by platelet-derived growth factor. Nature 295 :419-420, 1982. Fraker, P.J., et al. Quantification of apoptotic events in pure and heterogeneous populations of cells using the flow cytometer. Method in Cell Biology, 46:57, 1995. Gewirtz, D.A. Does bulk damage to DNA explain the cytostatic effects of topoisomerase H inhibitors. Biochem. Biophys. Res. Commun. 42:2253-2258, 1991. Gottardis, M.M., S.Y. Jiang, M.H. Jeng, and V.C. Jordan. Inhibition of tarnoxifen- stimulated growth of an MCF-7 tumor variant in athymic mice by novel steroidal 126 antiestrogens. Cancer Res. 49: 4090-4093, 1989. Graham BK, C. Nelson, M.A. Waring, G. E. Joannou and A. Y. Reeder. Metabolites of dietary (soya) isoflavones in human urine. Clinica Chimica Acta 22329-22, 1993. Gu, Y., C.W. Truck, and DO. Morgan. Inhibition of CDK2 activity in vivo by an associated 20 K regulatory subunit. Nature 366:707, 1993. Harper, J .W., G.R. Adami, N. Wei, and SJ. Elledge. The p21 cdk-interacting protein Cipl is potent inhibitor of G1 cyclin-dependent kinases. Mol. Biol. Cell 6:387-400, 1995. Haslarn, S.Z. Progesterone effects on deoxyribonucleic acid synthesis in normal mouse mammary glands. Endocrinology 122: 464-470, 1988. Henderson, B.E., R. Ross, and L. Bernstein. Estrogen as a cause of human cancer. Cancer Res. 48: 246-253, 1988. Hildreth, N.G., R.E. Shore, PM. Dvoretsky. The risk of breast cancer after irradiation of the thymus in infancy. New Engl. J. Med. 321: 1281-1284, 1989. Hill, T.D., N.M. Dean, L.J. Mordan, A.F. Lau, M.Y. Kanernitsu and AL Boynton. PDGF-induced activation of phospholipase C is not required for induction of DNA synthesis. Science 248: 1660-1663, 1990. Hunter, T. and J .A. Cooper. Protein-tyrosine kinases. Annu. Rev. Biochem. 54:897- 930, 1985. Jakowlew, S.B., R. Breathnach, J. Jeltsch, P. Masiakowski, and P. chambon. Sequence of the p82 mRNA induced by estrogen in the human breast cancer cell line MCF-7. Nucleic Acids Res. 12:2861-78, 1984. Joannou, GE, GE. Kelly, A.Y. Reeder, M. Waring and C. Nelson. A urinary profile study of dietary phytoestrogens. The identification and mode of metabolism of new isoflavonoids. J. Steroid Biochem. Molec. Biol. 54:167-184, 1995. 127 Jordan, V.C., MM. Gottardis, S.-Y. Jiang and M-H. J eng. Inhibition of tamoxifen- stimulated grth of an MCF-7 tumor variant in athymic mice by novel steroidal antiestrogens. Cancer Res. 49: 4090-4093, 1989. Kang, KS, 1. Monita, A. Cruz, Y.J. Jeon, J .E. Trosko and CC. Chang. Expression of estrogen receptors in a normal human breast epithelial cell type with luminal and stem cell characteristics and its neoplastically transformed cell lines. Carcinogenesis 18:251- 257, 1997. Kao, C.Y., K. Nomata, C.S. Oakley, C.W. Welsch and CC. Chang. Two types of normal hmnan breast epithelial cells derived from reduction mammoplasty: phenotypic characterization and response to SV40 transformation. Carcinogenesis 16:53 1-53 8, 1995. Kasuga, M, Y. Zick, D.L. Blithe, M. Crettaz, and CR. Kahn. Insulin stimulates tyrosine phosphorylation of the insulin receptor in a cell-fiee system. Nature 298:667- 669, 1982. King, KL, and J .A. Cidlowski. Cell cycle and apoptosis: common pathway to life and death. J. of Cellular Biochem. 58:175-180, 1995. Koroma, B.M and E. De Juan, Jr. Phosphotyrosine inhibition and control of vascular endothelial cell proliferation by genistein. Biochem. Pharmacology 48: 809-818, 1994. Krygier, A., J .S. Murley, R.R. Mehta and A. Constantinou. Induction of human adenocarcinoma cell differentiation by the phytoestrogen genistein is independent of its antiestrogenic ftmction International J. of Oncology 10:753-757, 1997. Labarca, C. & K. Paigen. A simple, rapid, and sensitive DNA assay procedure. Anal. Bioch 102:344-352, 1980. Lamartiniere, C.A., J .B. Moore, N.M Brown, R. Thompson, M. J. Hardin and S. Barnes. 1995a. Genistein suppresses mammary cancer in rats. Carcinogenesis, 16:2833-2840. Lamartiniere, C.A., J. Moore, M Holland, and S. Barnes. Neonatal genistein chemoprevents mammary cancer. Proc. Soc. Exp. Biol. Med, 208: 120-123, 1995b. 128 Land, C.E., M. Tokunaga and S. Tukuoka. Studies of breast cancer risks among atomic bomb survivors. Gerber G.B., Tayler D.M., Cardis E. and Thiessen J .W. (eds) Brist. Inst. Radiol. Report 22:49, 1991. Leopold, A.S., M. Erwin, J. Oh, and B. Browning. Phytoestogens: adverse effects on reproduction in California quail. Science 191: 98-100, 1976. Levitzki, A., A. Grazit. Tyrosine kinase inhibition: An approach to drug development. Science 267: 1782-1788, 1995. Li, I.C., C.C. Chang, and J .E. Trosko. Thymidylate synthetase gene as a quantitative mutation marker in Chinese hamster cells. Mutat. Res. 243:233-239, 1990. Linassier, C.M, M. Pierre, J .B.L. Pecq and J. Pierre. Mechanisms of action in NIH 3T3 cells of genistein, an inhibitor of EGF receptor tyrosine kinase activity. Biochem. Pharrnacol. 39:187-193, 1990. Liu, M., M.H. Lee, M. Cohen, M. Bommakanti, and LP. Freedman. Transcriptional activation of the Cdk inhibitor p21 by Vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev. 10:142-153, 1996. Lyons, W.R., C.H. Li, and RB. Johnson. The hormonal control of mammary growth and lactation. Recent Progress in Hormone Research. 14:219-254,1958. Mathieson, RA, and W.D. Kitts. Binding of phyto-oestrogen and oestradiol-17B by cytoplasmic receptors in the pituitary gland and hypothalamus of the ewe. J. Endocr. 85:317-325, 1980. Matsukawa, Y., N. Msrui, T. Sakai, Y. Satomi, M. Yoshida, M. Matsumoto, H. Nishino, and A. Aoike. Genistein arrests cell cycle progression at G2-M. Cancer Res. 53:1328- 1331, 1993. Markovitz, J ., C. Linassier, P. Fosse, J. Couprie. J. Pierre, A. Jaquemin—Sablon, J .M. Saucier, J .8. Le Pecq and AK. Larson. Inhibitory effects of the tyrosine kinase inhibitor genistein on mammalian DNA topoisomerase H. Cancer Res. 49:5111-5117, 129 1989. Markert, C.L. Neoplasia: a disease of cell differentiation. Cancer Res. 28: 1908-1914, 1968. Marshall, B. Epidemiology. Search for a killer: focus shifts from fat to hormones. Science 259: 618-621, 1993. Martin, P.M., K.B. Horwitz, D.S. Ryan & W.L. Mcguire. Phytoestrogen interaction with estrogen receptors in human breast cancer cells. Endocrinology 103: 1 860-7, 1978. McManus, M.J. and CW. Welsch. Honnone-induced ductal DNA synthesis of human breast tissues maintained in the athymic nude mouse. Cancer Res. 41: 3300-3305, 1981. Messina, M.J., and V. Messina. Increasing use of soyfoods and their potential role in cancer prevention. J. Amer. Diet. Assoc. 91, 836-840, 1991. Messina, M.J., V. Peraky, K. D. R. Setchell, and S. Barnes. Soy intake and cancer risk: a review of the in vitro and in vivo data. Nutr. Cancer 21: 1 13-31, 1994. Miller, W.R. Endocrine treatment for breast cancers: biological rationale and current progress. J. Steroid Biochem. Molec. Biol. 37: 467-480, 1990. Moore, D.D., J. Chory, and R.K. Ribaudo, Current Protocols in Molecular Biology, Ausubel, F.M. et al., Ed. 27 Vol. 1: 261-262, 1995. Murakami, M.S., MC. Strobel and GP. Vande Woude. Cell cycle regulation, oncogenes, and antineoplastic drugs. Chapter lealignant transformation, 3-17, 1995. Murdoch, F.E., D.A. Meier, J .D. Furlow, KA.A. Grunwald, and J. Gorski. Estrogen receptor binding to a DNA response element in vitro is not dependent upon estradiol. Biochem. 29: 8377-8385, 1990. Murrill, W. B., N.M. Brown, J-X Zhang, P. A. Manzolillo, S. Barnes and CA. Lamartiniere. Prepubertal geni stein exposure suppresses mammary cancer and enhances gland differentiation in rats. Carcinogenesis 17 :1451-1457, 1996. 130 Naim, M., B. Gestetner, S. Zikah, Y. Birk, and A. Bondi. Soybean isoflavones characterization, determination, and antifungal activity. J. Agric. Food Chem. 22:806- 810, 1974. Nakafutu, M., T. Satoh, and Y. Kaziro. Differentiation factors, including nerve growth factor, fibroblast growth factor, and interleukin-6, induce an accumulation of an active raszGTP complex in rat pheochromocytoma pc12 cells. J. Biol. Chem. 267: 19448- 19454, 1992. Nagasawa, H., R. Yanai, M. Shodono, T. Nakamura, and Y. Tanabe. Effect of neonatally administered estrogen or prolactin on normal and neoplastic mammary growth and serum estradiol-17 B level in rats. Cancer Res., 34: 2643-2646, 1974. Ogawara, H., T. Akiyama, J. Ishida, S.-I. Watanabe and K.-I. Suzuki. A specific inhibitor for tyrosine protein kinase from pseudomonas. J. Antibiotics 39:606-608, 1986. Ogawara, H., T. Akiyama, J. Ishida, S.-I. Watanabe and K.-I. Suzuki. Inhibition of tyrosine protein kinase activity by synthetic isoflavones and flavones. J. Antibiotics 42:340-343, 1989. Pagliacci, M.C., M. Smacchia, G. Migliorati, F. Grignani, C. Riccardi, and I. Nicoletti. Growth-inhibitory effects of the natural phytoestrogen genistein in MCF-7 human breast cancer cells. Eur. J. Cancer 30: 1675-1682, 1994. Peterson, G. Evaluation of the biochemical targets of genistein in trmor cells. J. Nutr. 12527845-7895, 1995. Peterson, G., L. Coward, M. Kirk, C.N. Falany and S. Barnes. The role of metabolism in mammary epithelial cell growth inhibition by the isoflavones genistein and biochanin A. Carcinogenesis 17:1861-1869, 1996. Pollner, F. A holistic approach to breast cancer research. Environt. Health Presp. 101: 116-120, 1993. 131 Potter, V.R. Phenotypic diversity on experimental hepatomas: The concept of partially blocked ontogeny. Br. J. Cancer 3821-23, 1978. Reeves, PG, PH. Nielsen, and GO Fahey, Jr. AIN-93 purified diets for laboratory rodents: final report of the American institute of nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J .Nutr. 123: 1939-1951, 1993. Robinson, M.J., A.H. Corbett, and N. Osheroff. Effects of topoisomerase H-targeted drugs on enzyme-mediated DNA cleavage and ATP hydrolysis: Evidence for distinct drug interaction domains on topoisomerase H. Biochemistry 32:3638-3643, 1993. Rubin J .B., MA. Shia, and P. F. Pilch. Stimulation of tyrosine-specific phosphorylation in vitro by insulin-like growth factor 1. Nature 305 2438-440, 1983. Russo, I.H., M. Koszalka, J. Russo. Chorionic gonadotropin (hCG) inhibits initiation and progression of mammary carcinogenesis. Pro. Am. Assoc. Cancer Res. 31 :906a, 1990a. Russo, I.H., M. Koszalka, J. Russo. Human chorionic gonadotropin and rat mammary cancer prevention. J. National Cancer Institute 82:1286-1289, 1990b. Russo, I.H. and J. Russo. Hormone prevention of mammary carcinogenesis: a new approach in anticancer research. Anticancer Res 8: 1247-1264, 1988. Russo, I.H. and J. Russo. Mammary gland neoplasia in long-term rodent studies. Envir. Heal. Persp. 104:938-967, 1996. SAS Institute Inc. SAS User Guide: Statistics, Version 5 Ed, 1985. SAS Institute, Cary, NC. Santell, R.C., Y.C. Chang, M.G. Nair, and W.G. Helferich. Dietary genistein exerts estrogenic effects upon the uterus, mammary gland and the hypothalamic/pituitary axis in rats. J. Nutr. 127: 263-269, 1997. Setchell, K.D.R., S.P. Boniello, P. Hulme, D.N. Kirk, and M. Axelson. Nonsteroidal estrogens of dietary origin: possible roles in horrnone-dependent disease. Am. J. Clin. 132 Nutr. 40:569-578, 1984. Shore, R.E., L.H. Hempelmann, E. Kowaluk et al. Breast neoplasm in women treated with X-ray for acute postpartum mastitis. J. Natl. Cancer Inst. 59: 813-822, 1977. Shutt, D.A., R.H. Weston, and J .P. Hogan. Quantitative aspects of phyto-oestrogen metabolism in sheep fed on subterranean clover (Trifolium subterraneum cultivar clare) or red clover (Trifolium Pratense). Aust. J. Agric. Res. 21 :713-722, 1971. Shutt, DA, and RI. Cox. Steroid and phyto-oestrogen binding to sheep uterine receptors in vitro. J. Endo. 52:299-310, 1972. Sit, K..,H K.P. Wong and EH. Bay. Genistein inhibits DNA synthesis but has no effect on levels of DAG and IP3, cell rounding and alkalinization in sulphate-treated Chang liver cells. Biol. Signals 1:228-36, 1992. Soyatech Survey and Estimates, 1991. PO. Box 84, Bar Harbor ME and Soyfoods Center Survey, PO. Box 234, Lafayette, CA. The American Cancer Society, 1994. Cancer Facts and Figures. Atlanta, GA. Cancer facts and figures, ©1997, American Cancer Society, Inc., 1997. Tanimura, M, H. Kobuchi, T. Utsumi, T. Yoshioka, S. Kataoka, Y. Fujita and K. Utsumi. Neutrophil priming by granulocyte colony stimulating factor and its modulation by protein kinase inhibitors. Biochem. Pharmacol. 44:1045-1052, 1992. Torti, M., K.B. Marti, D. Altschuler, K. Yamarnoto, and EG. Lapetine. Erythropoietin induces p21“Ila activation and p120GAP tyrosine phosphorylation in human erythroleukemia cells. J. Biol. Chem. 267: 8293-8298, 1992. Troll, W., R Weisner, J. Shellabarger, S. Holtzrnan, and J .P. Stone. Soybean diet lowers breast tumor incidence in irradiated rats. Carcinogenesis 1: 469-472, 1980. Uckun, P.M., W.E. Evans, C.J. Forsyth, K.G. Waddick, L.T. Ahlgren, L.M Chekstrom, A. Burkhardt, J. Bolen, and DE. Myers. Biotherapy of B-cell precursor leukemia by 133 targeting genistein to CD 19-associated tyrosine kinases. Science 267 2886-891, 1995. Ushiro, H, S. Cohen. Identification of phosphotyrosine as a product of epidermal growth factor-activated protein kinase in A-431 cell membranes. J. Biol. Chem. 255:8363-8365, 1980. Varmur, H. and RA. Weinberg. Gene and biology of cancer. W.H. Freeman and Company, Chapter 2:38-39, 1993. Wang, H.-J. and PA. Murphy. Isoflavone content in commercial soybean foods. J. of Agric. and Food Chem. 42:1666-1673, 1994. Wang, T.T.Y., N. Sathyamoorthy and J. M. Phang. Molecular effects of genistein on estrogen receptor mediated pathways. Carcinogenesis 17: 271-275, 1996. West, DC, A. Sattar and S. Kumar. A Simplified in situ solubilization procedure for the determination of DNA and cell number in tissue cultured mammalian cells. Anal. Bch. 1472289-295, 1985. Wei, H., L. Wei, K. Frenkel, R. Bowen and S. Barnes. Inhibition of turnor-promoter induced hydrogen peroxide formation by genistein in vitro and in vivo. Nutr. Cancer 2021-12, 1993. Wei, H., R. Bowen, Q. Cai, S. Barnes and Y. Wang. Antioxidant and antipromotional effects of the soybean isoflavone genistein. Proc. Soc. Exptl. Biol. Med. 208: 124-130, 1995. Xie, W., and LI. Rothblum. Rapid, small-scale RNA isolation from tissue culture cells. Biotechniques 1 11325-27, 1991. Xiong, Y., G.J. Harmon, H. Zhang, D. Casso, R. Kobayashi, and D. Beach. P21 is a universal inhibitor of cyclin kinases. Nature 3662701, 1993. Xu, X., Harris, KS, Wang, H.J., Murphy, PA. and Hendrich, S. Bioavailability of soybean isoflavones depends upon gut microflora in women. J. Nutr. 125:2307-15, 1995. 134 Yamashita, Y., S. Kawada, and H. Nakano. Induction of mammalian topisomerase H dependent DNA cleavage by non-intercalative flavonoids genistein and orobol. FEBS Lett 288:46-50, 1991.