NUTRITION AND CANCER, 58(2), 222–229Copyright C© 2007, Lawrence Erlbaum Associates, Inc.

The Effect of the Phytoestrogens Genistein, Daidzein, and Equol onthe Growth of Tamoxifen-Resistant T47D/PKCα

Debra A. Tonetti, Yiyun Zhang, Huiping Zhao, Sok-Bee Lim, and Andreas I. Constantinou

Abstract: Soy supplements are often consumed by womenfor alleviating menopausal symptoms or for the perceivedprotective effects against breast cancer. More concerning isthe concurrent consumption of soy isoflavones with tamox-ifen (TAM) for prevention or treatment of breast cancer. Wepreviously described a T47D:A18/protein kinase C (PKC)αTAM-resistant tumor model that exhibits autonomous growthand estradiol-induced tumor regression. We compared theestrogenicity of the isoflavones genistein, daidzein, and thedaidzein metabolite equol in the parental T47D:A18 andT47D:A18/PKCα cell lines in vitro and in vivo. Whereasequol exerts estrogenic effects on T47D:A18 cells in vitro,none of the isoflavones stimulated T47D:A18 tumor growth.T47D:A18/PKCα tumor growth was partially stimulated bygenistein, yet partially inhibited by daidzein. Interestingly,coadministration of TAM with either daidzein or genisteinproduced tumors of greater size than with TAM alone. Thesefindings suggest that simultaneous consumption of isoflavonesupplements with TAM may not be safe.

Introduction

The Hormone Therapy Trials of the Women’s Health Ini-tiative (WHI) study were halted early due to the multiple risksassociated with estrogen/progesterone combination hormonereplacement therapy (HRT) (1). Included among the risksare increased incidence of cardiovascular disease and the de-velopment of invasive breast cancer in women on the HRTarm. Since this finding, many women have turned to soy andphytoestrogen supplements as a perceived safer alternativeto HRT to alleviate the symptoms of menopause. However,there is not yet consensus regarding their efficacy for therelief of menopausal symptoms (2–5). There is a lower in-cidence of breast cancer in Asian women compared to theWestern world, often attributed to a soy-rich diet (6), yet ad-dition of soy to the Western diet has not yet been shown toreduce breast cancer risk (7,8).

D. Tonetti, Y. Zhang, H. Zhao, and S.-B. Lim are affiliated with the Department of Biopharmaceutical Sciences, College of Pharmacy, University of Illinoisat Chicago, Chicago, IL 60612. A. I. Constantinou is affiliated with the Department of Biological Sciences, University of Cyprus, Nicosia, Cyprus.

Tamoxifen (TAM) is currently approved for the treatmentof both premenopausal and postmenopausal patients withestrogen receptor alpha (ERα)-positive, early and advancedbreast cancer and is the only Federal Drug Administrationapproved drug for the chemoprevention of breast cancer inhigh risk premenopausal and postmenopausal women (9).The predominant mechanism of action of TAM is to com-pete with 17β-estradiol (E2) for binding to the ER, causinga conformational change and preventing transcriptional ac-tivation. Because hot flashes are a bothersome side effectof TAM therapy, many women may perceive phytoestrogensto be a natural and safe alternative to HRT. Alternatively,breast cancer patients may increase dietary soy intake eitherthrough food or supplements because soy has been touted topromote breast health. It was recently reported that 19–21%of breast cancer patients use some herbal supplements, ofwhich isoflavones are included (10, 11). However, the ef-fects of combining TAM and isoflavones in breast cancerpatients are not clear.

Genistein and daidzein are the most abundant phytoe-strogens found in soy and are classified as isoflavonecompounds. These compounds bind to ERα and ERβ; how-ever, both isoflavones have a much higher affinity for ERβ

(12). Both chemoprevention and therapeutic animal mod-els have been used to examine whether isoflavones aug-ment or negate the actions of TAM (13–15). We recentlyreported that daidzein cooperates with TAM to prevent 7,12-dimethylbenzanthracene-induced rat mammary tumors, per-haps acting through the metabolic conversion to equol (14).Using an MCF-7 xenograft model in athymic mice, Ju etal. (15) showed that dietary genistein blocked the inhibitoryeffect of TAM. These two models simulate breast cancerchemopreventive and therapeutic settings, respectively. Toaddress the effect of isoflavones on TAM-resistant breast can-cer, we tested the effects of combining TAM and isoflavonesin our TAM-resistant T47D:A18/protein kinase C alpha(PKCα) breast cancer model. We previously reported thatPKCα overexpression in T47D:A18 cells cause autonomousand TAM resistant growth in vitro and TAM-resistant, au-tonomous, and E2-inhibitory growth in vivo (16,17). Since

E2 inhibits tumor growth in this model, we wished to deter-mine whether the isoflavones genistein and daidzein wouldact in a similar fashion as E2 by inhibiting tumor growth orwhether combination with TAM would result in augmenta-tion or blockade of the TAM-stimulatory effects.

Material and Methods

Cell Lines and Culture Conditions

T47D:A18 is a hormone-dependent human breast cancercell clone that was previously described (18). T47D:A18/neoand T47D:A18/PKCα20 clones have been described (16).T47D:A18, T47D:A18/neo and T47D:A18/PKCα20 cloneswere maintained in RPMI 1640 phenol-red medium supple-mented with 10% fetal bovine serum (FBS). T47D:A18/neoand T47D:A18/PKCα20 clones were supplemented withG418 (500 µg/ml). Prior to cell proliferation and luciferaseassays, all cell lines were maintained in phenol red-free RPMI1640 supplemented with 10% 3X dextran-coated charcoal-treated FBS (estrogen-depleted media) for 3 days. Prior tocell injection into mice, all cell lines were grown in completemedium containing 10% FBS.

Proliferation Assays

The cell clones T47D:A18 and T47D:A18/PKCα20 wereseeded at 4 × 104 cells/ml estrogen-depleted media into T25tissue culture flasks. The following day (Day 1), mediumcontaining either ethanol (control), 17β-estradiol (Sigma-Aldrich, St. Louis, MO) (E2, 10−9 M), daidzein (10−7 M),genistein (10−7 M), (R, S)-equol (10−7 M) (LC Laborato-ries, Woburn, MA), or 4-hydroxytamoxifen (4-OHT; Sigma-Aldrich) (10−7 M) was added. All compounds were dissolvedin 100% ethanol and added to the medium at a 1:1000 dilu-tion. Cells were counted on Days 2–10.

Transient Transfection and Luciferase Assays

Estrogen response element (ERE)-tk-Luc plasmid con-tains 3 tandem copies of the vitellogenin ERE sequenceinserted into pT109luc(19). ERE-tk-luc plasmid was tran-siently cotransfected with the β-galactosidase expressionplasmid pCMVβ (for transfection efficiency normalization)into the T47D:A1(8 cells by electroporation.

Establishment of T47D:A18 and T47D:A18/PKCα

Tumors in Athymic Mice

T47D:A18 cells were combined with Matrigel in a1:1 (vol/vol) ratio and injected subcutaneously (1 × 107

cells/site) into the axillary mammary fat pads of ovariec-tomized 5–6-wk-old BALB/c athymic mice (Harlan SpragueDawley, Madison, WI). T47D:A18/PKCα cells were injectedin a similar fashion except without the addition of Ma-trigel. Mice were divided into treatment groups consisting of10 mice/group: control (no treatment), E2 (E2 capsule), TAM(Sigma-Aldrich), genistein, genistein + TAM, daidzein, anddaidzein + TAM. E2 was administered via silastic capsules(1.0 cm) implanted subcutaneously between the scapules,and the capsules were replaced every 8 wk. TAM was ad-ministered po at a dose of 1.5 mg/animal daily for 5 daysas previously described (17). Custom diets (Harlan-Teklad,Madison, WI) were formulated to contain genistein (0.206g/kg) or daidzein (0.144 g/kg). A casein control diet was fedto the control, E2, and TAM treatment groups. Tumor cross-sectional area was determined weekly by Vernier calipers andcalculated using the formula length/2 × width/2 × π . Meantumor area was plotted against time in weeks to monitor tu-mor growth. The mice were sacrificed by CO2 inhalation andcervical dislocation.

The composition of all of the diets used is summa-rized in Table 1. TD 02481 (Harlan-Teklad) was used asa base for the daidzein (TD 02483, 144 mg daidzein/kg) andgenistein (TD 02482, 206 mg genistein/kg) diets. Daidzein

Table 1. Diet Composition

TD-7912 (Control DietChemical TD 02481 (Casein Control) TD 02482 (Genistein) TD 02483 (Daidzein) Containing Soy)

Casein(%) 20.8% 20.8% 20.8% Crude Protein (19%)Corn starch 37.6% 37.6% 37.6% Crude Fat (5%)Maltodextrin 13.2% 13.2% 13.2% Crude Fiber (5%)Sucrose 10.0% 10.0% 10.0% Ground cornCorn Oil 0.7% 0.7% 0.7% Dehulled soybean mealCellulose 0.5% 0.5% 0.5% Ground oatsL-Cystine 0.35% 0.35% 0.35% Wheat middlingsMineral mix 0.3% 0.3% 0.3% Dehydrated alfalfa mealVitamin mix 0.15% 0.15% 0.15% Soybean oil

1 Corn gluten mealMineralsVitaminsAmino acids

Genistein None 206 mg/kg None NoneDaidzein None None 144 mg/kg None

Vol. 58, No. 2 223

and genistein were purchased from Indofine Chemical Co.(Hillsborough, NJ), and the selected concentrations werebased on the amounts of each isoflavone present in the well-tolerated and effective 16% (wt/wt) soy protein isolate dietused in previous studies (20,21). The composition of thesoy-containing Teklad Diet 7912 is shown in Table 1.

Statistical Analysis

When comparing to one group, data were analyzed usingunpaired t-test. When comparing groups, data were analyzedusing 1-way analysis of variance followed by the DunnettMultiple Comparison test. All statistics were performed us-ing GraphPad InStat version 3.00 for Windows 95 (GraphPadSoftware, San Diego, CA). Significant differences were in-dicated when P < 0.05.

Results

Effects of Daidzein and Equol on T47D:A18/neoand T47D:A18/PKCα Cells in Culture

We previously reported the stable transfection of PKCα

into the T47D:A18 cell line results in TAM-resistant andautonomous growth in vitro and in vivo (16). An additionalgrowth characteristic is apparent in vivo; E2 inhibits tumorformation and causes complete tumor regression (17). Weexamined whether the phytoestrogens genistein, daidzein,or equol exert differential growth effects in the parentalT47D:A18/neo versus the T47D:A18/PKCα cell clones. Thehormone-dependent T47D:A18/neo cell line is growth stim-ulated by E2 and to a lesser extent by equol but not by genis-tein, daidzein, or 4-OHT (Fig. 1A). As previously reported,T47D:A18/PKCα cells grow both in the absence and pres-ence of E2 and are TAM resistant as indicated by growth inthe presence of 4-OHT (16). Addition of genistein, daidzein,or equol does not inhibit T47D:A18/PKCα growth. There-

fore, equol has an estrogenic effect on T47D:A18/neo cells,but it is difficult to determine whether equol, genistein, ordaidzein are estrogenic in the T47D:A18/PKCα cell linebecause these cells exhibit autonomous growth. To deter-mine whether these ligands exert an estrogenic effect onERE transcriptional activation, an ERE-luciferase constructwas transiently transfected, and luciferase activity was mea-sured in each cell line (Fig. 2). In T47D:A18/neo cells, E2induced ERE-luciferase transactivation by 22-fold over con-trol, whereas equol caused a 12-fold induction. The pureantiestrogen, ICI 182, 780, reversed the estrogenic effectsof equol, suggesting that equol is acting through the ER.Daidzein and genistein did not result in increased ERE-luciferase activity. ERE-luciferase transactivation is consis-tent with the growth effects produced by these ligands in theT47D:A18/neo cell line (Fig. 1A). As previously reported,basal ERE-luciferase activity in T47D:A18/PKCα cells ishigher than the parental T47D:A18/neo, and the fold induc-tion produced by E2 relative to control is less (threefold)(16).Equol exhibits ERE-luciferase activity that is similar to E2(2.7-fold) and is reversed with ICI 182,780, again suggest-ing that equol exerts estrogenic effects through the ER (Fig.2B). Daidzein is less estrogenic and shows about one half thelevel of induction (1.5-fold) compared to E2, and genisteinis similar to the untreated control. Therefore, both equol anddaidzein exhibit estrogenic effects in the T47D:A18/PKCα

TAM-resistant cell line and is reversed by the pure antiestro-gen ICI. This is compared to weaker estrogenic effects ofequol in the parental T47D:A18/neo cells.

The Effects of Genistein and Daidzein on the Growthof T47D:A18 Cells In Vivo

To determine the in vivo effects of genistein and daidzein,T47D:A18 cells were injected into ovariectomized, athymic

Figure 1. Proliferation rate of clones T47D:A18/neo (A) and T47D:A18/ PKCα (B), untreated (control) or treated with 17β-estradiol (E2) (10–9 M), 4-hydroxytamoxifen (4-OHT) (10–7M), daidzein (10–7M), genistein (10–7M), or equol (10–7M). Proliferation rate was assessed by cell counting as described inMaterials and Methods. The results are expressed as total cell number ± SE for each time point. These results are representative of 3 independent experiments.

224 Nutrition and Cancer 2007

Figure 2. Luciferase assay measuring estrogen response element (ERE)-luciferase transcriptional activation in T47D:A18 (A) and T47D:A18/PKCα (B) cellclones. The ERE-luciferase and β-galactosidase plasmids were transiently cotransfected into each cell line and were untreated (control; C) or treated with17β-estradiol (E2) (10–9 M), 4-hydroxytamoxifen (OHT, 10−7 M), antiestrogen ICI 182, 780 (I; 10−7 M), daidzein (D; 10–7 M), equol (Eq; 10–7 M), orgenistein (G; 10–7 M) or combinations of D + I, Eq + I, or G + I for 24 h as described in Materials and Methods. Fold mean normalized luciferase activityrelative to control ± SE is shown. * Indicates significant difference (P < 0.05) when compared to untreated control.

mice. As previously reported (17), T47D:A18 cells form tu-mors in response to E2 but not in the absence of E2 or inthe presence of TAM (Fig. 3A). Daidzein treatment aloneresulted in transient growth the 1st wk followed by declin-ing tumor size throughout the experimental period (Fig. 3B).Treatment with a combination of daidzein and TAM showedno difference in tumor size compared to the untreated control(Fig. 3A). Genistein treatment alone or in combination withTAM did not result in significant tumor growth comparedwith TAM treatment alone (Fig. 3C). Treatment of T47D:A18tumors with 1.5 mg TAM is known to produce TAM-stimulated tumors after 10 wk (22). These results suggest thatgenistein and daidzein treatment either alone or in combina-tion with TAM do not stimulate T47D:A18 tumor growth.

The Effects of Genistein and Daidzein on the Growthof TAM-Resistant T47D:A18/PKCα Cells In Vivo

As previously reported, T47D:A18/PKCα tumors exhibitautonomous and TAM-resistant growth and are inhibitedby E2 (Fig. 4A). Furthermore, E2 causes complete tumorregression when administered to growing T47D:A18/PKCα

tumors (17). Therefore, because daidzein and equol exert es-trogenic effects in the T47D:A18/PKCα cells in vitro (Fig. 2),we hypothesized that perhaps these isoflavones may inhibittumor growth in a similar fashion as E2. Both daidzein andgenistein when given in combination with TAM stimulateT47D:A18/PKCα tumors to grow faster than mice treatedwith TAM alone or either isoflavone by itself (Fig. 4B).Interestingly, daidzein treatment initially stimulatedT47D:A18/PKCα tumors (Weeks 1–7), followed by a periodof growth inhibition (Weeks 8–12), and eventual regrowth(Weeks 12–14). Genistein also stimulated T47D:18/PKCα

tumor growth through Week 10 followed by a period oftumor stabilization (Weeks 11–14). These results suggestthat both daidzein and genistein exhibit partial attenuation

of T47D:A18/PKCα tumor growth. The growth inhibitoryeffects of daidzein appear to be greater than genistein,perhaps in part due to the metabolic conversion to equol.Therefore, daidzein and genistein exhibit weaker growthinhibitory effects than E2 in this TAM-resistant tumormodel. However, in combination with TAM, both genisteinand daidzein stimulate T47D:18/PKCα tumor growth to agreater extent than TAM or either isoflavone alone.

The Addition of Soy as a Protein Source in the DietEffects the Ability of T47D:A18/PKCα Tumorsto Grow Autonomously

We noticed that the control diet (TD 02481) formu-lated to be devoid of soy products and used as a basecontrol diet seemed to prevent the autonomous growth ofT47D:A18/PKCα tumors previously observed using themouse chow supplied from the University of Illinois atChicago (UIC) Animal Facility (TD 7912). In fact, boththe untreated control and E2-treated groups produced dis-tinct tumor growth curves with the custom diet (soy free)from what we routinely observed using the UIC-supplieddiet. Tumors from mice fed the soy-free control diet initiallygrew but at Week 6 began to regress (Fig. 4B) in contrastto the robust tumor growth observed in mice fed the UICdiet containing soy products (Fig. 4A). Mice fed the soy-freecontrol diet exhibited E2-induced tumor growth inhibition;however, by Week 8, these tumors began to grow. This isin contrast to the complete suppression of tumor growth inthe E2-treated group fed the UIC soy-containing diet (Figs.4A, 4B). We speculate that perhaps the inhibitory effects ofE2 can be overridden by a component in the soy-based diet.No apparent differences in growth characteristics were ob-served between the soy-free and soy-containing diets fed tomice bearing the parental T47D:18 tumors. To directly com-pare the 2 control diets on the ability of T47D:A18/PKCα

Vol. 58, No. 2 225

Figure 3. Initiation and growth of T47D:A18 tumors in athymic mice. A: 40 mice received injections of the T47D:A18 cell line and were treated(10 mice/group) with a 1-cm 17β-estradiol (E2) capsule, tamoxifen (TAM) (1.5 mg TAM /day po), TAM + daidzein (Dz) (Diet TD02483), or left un-treated (Control). The E2, TAM, and untreated groups were fed the soy-free control diet (TD02481). B: 30 mice received injections of the T47D:A18 cell lineand treated with (10 mice/group) TAM (1.5 mg TAM/day po, fed the control diet TD02481), genistein (Gen) (Diet TD02482), or TAM + Gen (Diet TD02482).C: 20 mice received injections of the T47D:A18 cell line and treated (10 mice/group) with either a 1-cm E2 capsule (Diet TD02481) or Dz (Diet TD02483).

Figure 4. Initiation and growth of T47D:A18/PKCα tumors in athymic mice. A: 30 mice received injections of the T47D:A18/PKCα cell line and randomizedto 3 treatment groups (10 mice/group): 17β-estradiol (E2) (1-cm E2 capsule), tamoxifen (TAM) (1.5 mg TAM/day po), or Control (untreated). All mice werefed the UIC supplied soy-containing Teklad 7912 diet. B: 70 mice received injections of the T47D:A18/PKCα cell line and randomized to 7 treatment groups(10 mice/group): E2 (1-cm E2 capsule, fed soy-free TD02481 diet), TAM (1.5 mg TAM/day po, fed soy-free TD02481 diet), daidzein (Dz) (TD02483 diet), Dz+ TAM (TD02483 diet), genistein (Gen) (TD02482 diet), Gen + TAM (TD02482 diet) or Control (untreated, TD02481 diet). *, indicates significant difference(P < 0.05) in tumor size when comparing Dz + TAM vs. TAM groups and Dz + TAM vs. Dz; †, indicates significant difference (P < 0.05) in tumor sizewhen comparing Gen + TAM vs. TAM and Gen + TAM vs. Gen groups.

226 Nutrition and Cancer 2007

Figure 5. Effect of soy-free and soy-containing control diets on the growthof T47D:A18/PKCα tumors. T47D:A18/PKCα cells were injected into 20athymic mice and randomized to be fed either the soy-containing Teklad7912 diet or the soy-free Teklad 02481 diet.

tumors to grow autonomously, T47D:A18/PKCα tumorswere initiated in 20 athymic ovariectomized mice andrandomized to 2 groups fed either the soy-free control diet(TD 02481) or the soy-based control diet (TD 7912). Micefed the soy-based diet consistently exhibit T47D:A18/PKCα

autonomous tumor growth, whereas mice fed the soy-freecontrol diet show initial autonomous growth followed by tu-mor regression after 4 wk on the diet (Fig. 5) These resultssuggest that a component in the soy-based diet may be stim-ulating T47D:A18/PKCα tumor growth.

Discussion

Theaddition of soy or purified isoflavones to the diet ispopular among women interested in alleviating menopausalsymptoms as well as preventing breast cancer. While theefficacy of this approach is currently not clear, a more wor-risome application is the combination of soy or isoflavonesupplements with TAM in women who are either at high riskof developing breast cancer or in those that are being treatedfor breast cancer. We examined the effects of isoflavones onthe growth characteristics of a TAM-resistant breast cancertumor model developed in our laboratory, T47D:A18/PKCα

(16,17,23). This model is interesting in that E2 inhibits tumorgrowth and is capable of causing tumor regression, a charac-teristic that may be relevant in patients (24–27). Our primaryobjective was to determine whether the isoflavones genisteinand daidzein or the daidzein metabolite equol would produceestrogenic effects in this model in a similar fashion as E2, thuscausing inhibition of tumor growth. A secondary objectivewas to determine whether combining TAM with either genis-tein or daidzein would augment or attenuate tumor growth.

In vitro, genistein, daidzein, and equol do not inhibitT47D:A18/PKCα autonomous cell proliferation, whereasequol, but not genistein and diadzein, stimulates T47D:A18cell proliferation. Only equol induced ERE-luciferase activ-ity in the T47D:A18 cells, while equol and daidzein both in-duced ERE-luciferase activity in the T47D:A18/PKCα cells.None of the isoflavone compounds either alone or in combi-nation with TAM stimulated T47D:18 tumor growth. Similar

to our results, it was reported that both daidzein and equolexhibit estrogenic effects in MCF-7 cells in vitro (28). Highconcentrations of daidzein caused a slight stimulatory effecton MCF-7 tumors; however, equol did not stimulate MCF-7 tumor growth (28). In contrast to our results with T47D:A18 tumors, genistein was reported to stimulate the growthof MCF-7 tumors both alone and in combination with post-menopausal levels of E2 (29–31). We did not determine thecombination of E2 and genistein in the T47D:A18 tumormodel; however, both daidzein and genistein in combinationwith TAM produced T47D:A18/PKCα tumors of greater sizethan TAM alone. Interestingly, daidzein alone initially stim-ulated T47D:A18/PKCα tumor growth followed by transientgrowth inhibition and eventual stimulation. Genistein alsocaused initial growth stimulation followed by a period of tu-mor stabilization. These in vivo results suggest that dietarysupplementation in conjunction with TAM in the therapeuticsetting is not beneficial and may in fact be harmful.

We have previously reported that the combination ofdaidzein with TAM produces increased protection againstrat mammary carcinogenesis, perhaps due to the antioxi-dant effects of equol, the main product of intestinal bacterialmetabolism of daidzein (14). In that study, the isoflavoneswere initiated prior to the induction of tumors and con-tinued to be provided throughout the study, representinga classic model of cancer chemoprevention. In the presentstudy, the diets were initiated after the injection of PKCα-overexpressing, TAM-resistant breast cancer cells. There-fore, in the present study, the tumor cells (T47D:A18/PKCα)were already present prior to the introduction of the diets. Inthe current animal model, effective agents simply attenuateor inhibit the growth of existing tumors, and consequently,these agents have a therapeutic potential. We report herethat daidzein and genistein do not cause regression of TAM-resistant tumors in mice and at best cause transient tumorstabilization. In fact, mice fed diets containing genistein ordaidzein in combination with TAM developed larger tumorsthan those given TAM alone. The combined data from the2 studies suggest that although daidzein provides additionalprotection than TAM alone in the prevention of mammarytumors (14), it is ineffective and perhaps harmful in the treat-ment of existing TAM-resistant tumors.

Of particular interest is our finding that T47D:A18/PKCα

tumors exhibit hormone-independent and E2-inhibitedgrowth in mice fed a soy-containing diet; however, thesegrowth characteristics were different in mice fed a soy-freediet (Figs. 4B and 5). We can only speculate that a componentin the soy-containing diet (TD 7912) stimulates the growthof T47D:A18/PKCα tumors, whereas in combination withE2 results in growth suppression. This phenomenon is notobserved in the parental T47D:A18 tumors, but is unique tothe PKCα-overexpressing tumors. Because daidzein, and toa lesser extent genistein, causes partial tumor stabilization(Fig. 4B), it is possible that these isoflavones are responsi-ble for the complete growth suppression observed in com-bination with E2 in the soy-containing diet (Fig. 4A). Al-ternatively, metabolic conversion of daidzein to equol may

Vol. 58, No. 2 227

contribute to the “estrogenic” effects in vivo (growth sup-pression) because equol is nearly as potent as E2 on in vitrocell proliferation and ERE-luciferase activity (Figs. 1 and 2)and has a 100-fold higher affinity for ERα than daidzein (32).Although intestinal conversion of daidzein to equol occursin rodents, the extent of conversion can vary among mousestrains just as it does among humans (33,34). Finally, in ad-dition to soy, this diet also contains dehydrated alfalfa meal(Table 1); therefore, it is possible that the phytoestrogeniccoumestans may contribute to this growth profile.

Other studies have examined the effect of combining soyisoflavones with TAM in vitro (35,36) and in vivo (13,15),but this is the first report to examine the growth effects ofisoflavones on a TAM-resistant tumor model. It should alsobe noted that the TAM resistance in our model is due toPKCα overexpression, and we have previously reported thatPKCα is overexpressed at a much higher frequency in tumorsfrom patients that exhibit disease recurrence compared withpatients remaining disease free (37). However the observedeffect of isoflavones might be specific to only this model andmay not be generalized to all TAM-resistant tumors. Moreexperiments with different animal models must be performedto determine if the soy isoflavones are harmful to all TAMresistant tumors and even more importantly, if the soy com-ponents are harmful once mammary tumors have passed acritical stage of development. We chose to administer theisoflavones in the aglycone form (genistein and daidzein) asopposed to the glycone forms (genistin and daidzin), despitethe fact that the glycones are more abundant in soybeans andmost soy foods (38). Numerous animal studies have been con-ducted using the aglycone forms (28,30,39,40) because theglycones are quickly hydrolyzed by intestinal bacterial en-zymes to the aglycone forms. However, because there is evi-dence of metabolic variation between species (41) and amongindividuals (42), the plasma concentrations achieved in thisstudy may be more closely related to isoflavone supplementa-tion, and there is likely to be individual metabolic variability.

In summary, we report for the first time the growth effectsof combining the isoflavones genistein and daidzein withTAM in the TAM-resistant tumor model T47D:A18/PKCα.Two features of this model indicate its clinical relevance: 1)PKCα overexpression correlates with TAM-resistance in pa-tients (37) and 2) E2-induced regression is a potential featureof many breast cancers (24–27). Although our findings arespecific to this model, these results lend support to the notionthat concurrent consumption of dietary isoflavones with TAMadministration within a treatment setting may be unsafe.

Acknowledgments and Notes

This work was supported by Grant RO1 CA96517 awarded jointly by theNational Cancer Institute and the Centre for Complimentary and AlternativeMedicine. Address correspondence to Dr. Debra A. Tonetti, Department ofBiopharmaceutical Sciences, College of Pharmacy, University of Illinois atChicago, 833 S Wood St (M/C 865), Chicago, IL 60612. Phone: 312-413-1169. FAX: 312-996-1698. E-mail: dtonetti@uic.edu.

Submitted 11 November 2006; accepted in final form 20 February 2007.

References

1. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C,et al.: Risks and benefits of estrogen plus progestin in healthy post-menopausal women: principal results From the Women’s Health Ini-tiative randomized controlled trial. JAMA 288, 321–333, 2002.

2. Van Patten CL, Olivotto IA, Chambers GK, Gelmon KA, Hislop TG, etal.: Effect of soy phytoestrogens on hot flashes in postmenopausalwomen with breast cancer: a randomized, controlled clinical trial.J Clin Oncol 20, 1449–1455, 2002.

3. Nikander E, Kilkkinen A, Metsa-Heikkila M, Adlercreutz H, PietinenP, et al.: A randomized placebo-controlled crossover trial with phytoe-strogens in treatment of menopause in breast cancer patients. ObstetGynecol 101, 1213–1220, 2003.

4. Tice JA., Ettinger B, Ensrud K, Wallace R, Blackwell T, et al.: Phy-toestrogen supplements for the treatment of hot flashes: the IsoflavoneClover Extract (ICE) Study: a randomized controlled trial. JAMA 290,207–214, 2003.

5. van de Weijer PH and Barentsen R: Isoflavones from red clover(Promensil) significantly reduce menopausal hot flush symptoms com-pared with placebo. Maturitas 42, 187–193, 2002.

6. Adlercreutz H and Mazur W: Phyto-oestrogens and Western diseases.Ann Med 29, 95–120, 1997.

7. Atkinson C, Warren RM, Sala E, Dowsett M, Dunning AM,et al.: A. Red-clover-derived isoflavones and mammographic breastdensity: a double-blind, randomized, placebo-controlled trial [IS-RCTN42940165]. Breast Cancer Res 6, R170–179, 2004.

8. Gikas PD and Mokbel K: Phytoestrogens and the risk of breast can-cer: a review of the literature. Int J Fertil Womens Med 50, 250–258,2005.

9. Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M,et al.: Tamoxifen for prevention of breast cancer: report of the NationalSurgical Adjuvant Breast and Bowel Project P-1 Study. J Natl CancerInst 90, 1371–1388, 1998.

10. Buettner C, Kroenke CH, Phillips RS, Davis RB, Eisenberg DM, et al.:Correlates of use of different types of complementary and alternativemedicine by breast cancer survivors in the nurses’ health study. BreastCancer Res Treat 100, 219–227, 2006.

11. Ashikaga T, Bosompra K, O’Brien P, and Nelson L: Use of compli-mentary and alternative medicine by breast cancer patients: prevalence,patterns and communication with physicians. Support Care Cancer 10,542–548, 2002.

12. Liu J, Burdette JE, Xu H, Gu C, van Breemen RB, et al.: Evaluationof estrogenic activity of plant extracts for the potential treatment ofmenopausal symptoms. J Agric Food Chem 49, 2472–2479, 2001.

13. Liu B, Edgerton S, Yang X, Kim A, Ordonez-Ercan D, et al.: Low-dose dietary phytoestrogen abrogates tamoxifen-associated mammarytumor prevention. Cancer Res 65, 879–886, 2005.

14. Constantinou AI, White BE, Tonetti D, Yang Y, Liang W, et al.: Thesoy isoflavone daidzein improves the capacity of tamoxifen to preventmammary tumours. Eur J Cancer 41, 647–654, 2005.

15. Ju YH, Doerge DR, Allred KF, Allred CD, and Helferich WG: Di-etary genistein negates the inhibitory effect of tamoxifen on growth ofestrogen-dependent human breast cancer (MCF-7) cells implanted inathymic mice. Cancer Res 62, 2474–2477, 2002.

16. Tonetti DA, Chisamore MJ, Grdina W, Schurz H, and Jordan VC: Stabletransfection of protein kinase C alpha cDNA in hormone-dependentbreast cancer cell lines. Br J Cancer 83, 782–791, 2000.

17. Chisamore MJ, Ahmed Y, Bentrem DJ, Jordan VC, and Tonetti DA:Novel antitumor effect of estradiol in athymic mice injected with aT47D breast cancer cell line overexpressing protein kinase c alpha.Clin Cancer Res 7, 3156–3165, 2001.

18. Pink JJ and Jordan VC: Models of estrogen receptor regulation byestrogens and antiestrogens in breast cancer cell lines. Cancer Res 56,2321–2330, 1996.

19. Catherino WH and Jordan VC: Increasing the number of tandem estro-gen response elements increases the estrogenic activity of a tamoxifenanalogue. Cancer Lett 92, 39–47, 1995.

228 Nutrition and Cancer 2007

20. Constantinou AI, Lantvit D, Hawthorne M, Xu X, van Breemen RB, etal.: Chemopreventive effects of soy protein and purified soy isoflavoneson DMBA-induced mammary tumors in female Sprague-Dawley rats.Nutr Cancer 41, 75–81, 2001.

21. Constantinou AI, Lucas LM, Lantvit D, Hawthorne M, Xu X, et al.:Soy protein isolate prevents chemically-induced rat mammary tumors.Pharm Biol 40, 24–34, 2002.

22. Schafer JM, Lee ES, O’Regan RM, Yao K, and Jordan VC:Rapid development of tamoxifen-stimulated mutant p53 breast tu-mors (T47D) in athymic mice. Clin Cancer Res 6, 4373–4380,2000.

23. Lin X, Yu Y, Zhao H, Zhang Y, Manela J, et al.: Overexpression of PKCalpha is required to impart estradiol inhibition and tamoxifen-resistancein a T47D human breast cancer tumor model. Carcinogenesis 27, 1538–1546, 2006.

24. Osipo C, Gajdos C, Cheng D, Jordan VC, Liu H, et al.: Reversalof tamoxifen resistant breast cancer by low dose estrogen therapy.J Steroid Biochem Mol Biol 93, 249–256, 2005.

25. Ingle JN. Estrogen as therapy for breast cancer. Breast Cancer Res 4,133–136, 2002.

26. Peethambaram PP, Ingle JN, Suman VJ, Hartmann LC, and LoprinziCL: Randomized trial of diethylstilbestrol vs. tamoxifen in post-menopausal women with metastatic breast cancer: an updated analysis.Breast Cancer Res Treat 54, 117–122, 1999.

27. Lonning PE, Taylor PD, Anker G, Iddon J, Wie L, et al.: High-doseestrogen treatment in postmenopausal breast cancer patients heavilyexposed to endocrine therapy. Breast Cancer Res Treat 67, 111–116,2001.

28. Ju YH, Fultz J, Allred KF, Doerge DR, and Helferich WG: Effects of di-etary daidzein and its metabolite, equol, at physiological concentrationson the growth of estrogen-dependent human breast cancer (MCF-7) tu-mors implanted in ovariectomized athymic mice. Carcinogenesis 27,856–863, 2006.

29. Allred CD, Allred KF, Ju YH, Virant SM, and Helferich WG: Soy dietscontaining varying amounts of genistein stimulate growth of estrogen-dependent (MCF-7) tumors in a dose-dependent manner. Cancer Res61, 5045–5050, 2001.

30. Ju YH, Allred KF, Allred CD, and Helferich WG: Genistein stimulatesgrowth of human breast cancer cells in a novel, postmenopausal animalmodel, with low plasma estradiol concentrations. Carcinogenesis 27,1292–1299, 2006.

31. Ju YH, Allred CD, Allred KF, Karko KL, Doerge DR, et al.: Phys-iological concentrations of dietary genistein dose-dependently stim-ulate growth of estrogen-dependent human breast cancer (MCF-7)tumors implanted in athymic nude mice. J Nutr 131, 2957–2962,2001.

32. Sathyamoorthy N and Wang TT: Differential effects of dietary phyto-oestrogens daidzein and equol on human breast cancer MCF-7 cells.Eur J Cancer 33, 2384–2389, 1997.

33. Ward WE, Kim S, Chan D, and Fonseca D: Serum equol, bone mineraldensity and biomechanical bone strength differ among four mousestrains. J Nutr Biochem 16, 743–749, 2005.

34. Setchell KD, Brown NM, and Lydeking-Olsen E: The clinical impor-tance of the metabolite equol-a clue to the effectiveness of soy and itsisoflavones. J Nutr 132, 3577–3584, 2002.

35. Jones JL, Daley BJ, Enderson BL, Zhou JR, and Karlstad MD: Genis-tein inhibits tamoxifen effects on cell proliferation and cell cycle arrestin T47D breast cancer cells. Am Surg 68, 575–578, 2002.

36. Limer JL, Parkes AT, and Speirs V: Differential response to phytoestro-gens in endocrine sensitive and resistant breast cancer cells in vitro. IntJ Cancer 119, 515–521, 2006.

37. Tonetti DA, Morrow M, Kidwai N, Gupta A, and Badve S: Elevatedprotein kinase C alpha expression may be predictive of tamoxifen treat-ment failure. Br J Cancer 88, 1400–1402, 2003.

38. Coward L, Barnes N, Setchell K, and Barnes S: Genistein and daidzein,and their β-glycosides conjugates: anti-tumor isoflavones in soybeanfoods from American and Asian diets. J Agric Food Chem 41, 1961–1967, 1993.

39. Hsieh CY, Santell RC, Haslam SZ, and Helferich WG: Estrogeniceffects of genistein on the growth of estrogen receptor-positive humanbreast cancer (MCF-7) cells in vitro and in vivo. Cancer Res 58, 3833–3838, 1998.

40. Lamartiniere CA, Cotroneo MS, Fritz WA, Wang J, Mentor-Marcel R,et al.: Genistein chemoprevention: timing and mechanisms of action inmurine mammary and prostate. J Nutr 132, 552S–558S, 2002.

41. Gu L, House SE, Prior RL, Fang N, Ronis MJ, et al.: Metabolic phe-notype of isoflavones differ among female rats, pigs, monkeys, andwomen. J Nutr 136, 1215–1221, 2006.

42. Wiseman H, Casey K, Bowey EA, Duffy R, Davies M, et al.: Influenceof 10 wk of soy consumption on plasma concentrations and excretionof isoflavonoids and on gut microflora metabolism in healthy adults.Am J Clin Nutr 80, 692–699, 2004.

Vol. 58, No. 2 229