Estrogenic and DNA-damaging activity of Red No. 3 in human breast cancer cells.

Exposure to pesticides, dyes, and pollutants that mimic the growth promoting effects of estrogen may cause breast cancer. The pesticide DDT and the food colorant Red No. 3 were found to increase the growth of HTB 133 but not estrogen receptor (ER) negative human breast cells (HTB 125) or rat liver epithelial cells (RLE). Red No. 3, beta-estradiol, and DDT increase ER site-specific DNA binding to the estrogen response element in HTB 133 cells and increase cyclin-dependent kinase 2 activity in MCF-7 breast cancer cells. Site-specific DNA binding by p53 in RLE, HTB 125, HTB 133, and MCF-7 cells was increased when they were treated with Red No. 3, which suggests that cellular DNA was damaged by this colorant. Red No. 3 increased binding of the ER from MCF-7 cells to the estrogen-responsive element. Consumption of Red No. 3, which has estrogenlike growth stimulatory properties and may be genotoxic, could be a significant risk factor in human breast carcinogenesis. ImagesFigure 4. AFigure 4. BFigure 5. AFigure 5. BFigure 6.Figure 7. AFigure 7. BFigure 7. C

In the United States, breast cancer is the most commonly diagnosed cancer and the second leading cause of cancer deaths (1). It has been estimated that one in eight U.S. women will develop breast cancer (2). Further, the incidence ofbreast cancer in the United States is increasing 1% per year in the 1990s (2). The personal and economic impacts of breast cancer make this disease a serious national health care concern. The total cost of breast cancer burdens the U.S. economy with direct and productivity losses of approximately $3.8 billion per year (3).
Great progress has been made recently in determining the molecular basis of familial forms of breast cancers (4). However, only 5 to 15% of the total number of breast cancers can be traced to an inherited familial defect (4). Although the etiology of the majority of breast cancers remains undetermined, epidemiologic studies have indicated that a dietary component contributes to the risk of developing the disease (5,6). High-fat diets may be a risk factor for breast cancer (5,6). However, the focus on fat as a major risk factor has shifted to the effects of hormones and hormone-mimicking chemicals (2).
Compounds that mimic the effects of estrogen are popularly referred to as xenoestrogens or environmental estrogens (7,8). Pesticides such as DDT, dyes (phenol red), and polychlorinated biphenyls are estrogenic (9)(10)(11). It has been suggested that exposure to xenoestrogens increases the risk of breast cancer for women in industrialized countries (8,14). However, the role of DDT and other environmental estrogens in the etiology of breast cancer in humans is controversial (7,(12)(13)(14). This paper was presented in part at the Workshop on Hormones, Hormone Metabolism, Environment, and Breast Cancer held 28-29 September 1995 in New Orleans, Louisiana. Manuscript received at EHP 6 June 1996; manuscript accepted 23 August 1996. The submitted manuscript has been authored by a contractor of the U.S. Government under contract DE-AC05-96OR22464. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes.
Address correspondence to Dr. C. Dees, PhotoGen L.L.C., 7327  Recently, it has been shown that phthalate plasticizers from plastic foodpackaging materials exhibit estrogenic activity (15). Butylated hydroxyanisole, a commonly used food preservative, also has weak estrogenic effects (15). Phenosulfothiazine, a red dye used in tissue culture media as a pH indicator, is a weak estrogen that stimulates the growth of human breast cells and binds to the estrogen receptor (ER) in MCF-7 human breast cancer cells (9). Because some food dyes are carcinogens (e.g., Red No. 4), we hypothesized that the dietary component that increases the risk of breast cancer in U.S. women might be xenoestrogenic food dyes. We examined the ability of food colorants to stimulate ER-positive and negative cells to enter the cell cycle. We also examined dyetreated cells for effects indicating damage to genetic material.

Proliferation Assays
Proliferation assays were performed by placing the cells in serum-free phenol red and methionine-free medium for 24 hr. Before adding xenoestrogens, cells were returned to media containing methionine but without serum or phenol red. Cells were incubated with xenoestrogens for 72 hr and then released by trypsinization. Cell counts were performed manually or with a Coulter model S cell counter.
The reaction was stopped using gel electrophoresis sample buffer, and the reaction products were separated on a 14% polyacrylamide gel (Novex, San Diego, CA).

ER-ERE Mobiity Shift Prcedures
Cells were cultured in 175-cm2 flasks in DMEM/F12 (phenol red free) supplemented with 10% FBS. The medium was replaced with fresh medium without serum 24 to 48 hr prior to adding compounds for growth promoting effects. Cells were then incubated for 1 hr with xenoestrogens or exposed to electromagnetic fields (EMF) for 1 hr. Cell extracts were prepared by removing the medium and washing the monolayers three times with PBS, pH 7.4. Cells were lysed by the addition of binding buffer (20% glycerol, 0.4 mM KCI, 2 mM DTT, 1 mM PMSF, in 20 mM Tris-HCl buffer, pH 7.5) using a glass Dounce homogenizer. The lysate was centrifuged at 10,000xg for 15 min and the supernatants retained for testing. Total protein content of the extracts was determined using BCA protein assays (Pierce Biochemicals, Rockford, IL). Protein content for all samples was equalized prior to performing the binding assay. The estrogen responsive element (ERE) (GTCCAAAGTCAGGTCA CAGTGACCGATCAAGTT) as described by Kumar and Chambon (18) and the complementary strand were synthesized, prepared in double-stranded form, and end-labeled with [32P]-ATP using T4 kinase. Binding reactions consisted of 5 pl of protein (approximately 5 ng), 0.5 ng 32P-labeled oligonucleotide, 1 pl of a 1.9-pg/ml Poly dl, dC solution (Sigma Chemical, St. Louis, MO), and 25 pl binding buffer. Binding reactions were incubated at room temperature for 20 min. The entire reaction mixture was then separated on 6% nondenaturing polyacrylamide gels (Novex, San Diego, CA) and visualized by autoradiography.
Competitive binding studies using the ER from MCF-7 were performed as described previously (11) with modifications to accommodate the use of cultured cells. To confirm visual interpretation of ER-ERE mobility shift increases stimulated by xenoestrogens, autoradiographs were scanned using a Hewlett Packard ScanJet IIcx. Densitometric evaluation was performed using SigmaScan software.

p53 Mobility Shift Procedures
Procedures for p53 mobility shift assays were similar to those described previously (16,19,20). RLE cells were cultured in 175 cm2 flasks in Richter's medium supplemented with 0.5% newborn calf serum. The medium was replaced prior to adding compounds for testing with fresh medium containing test compounds without serum. Cells were then incubated for 2 hr with DNA-damaging agents or exposed to EMF. Untreated control cells were also examined. S9 homogenate (Molecular Toxicology, Annapolis, MD) was prepared from rats treated with Aroclor 1254. The S9 mix components were 8 mM MgCl2, 33 mM KCl, 5 mM glucose 6-phosphate, 4 mM NADP, 100 mM sodium phosphate, pH 7.4, and S9 at 10% (v/v) of mix. Food colorants were added to 1-ml S9 mix and then added to cultured cells.
Nuclear extracts from the cells were prepared as described (13). Briefly, the medium was removed from the cells and the monolayers washed with PBS, pH 7.4. Cells were lysed by the addition of 2.5 ml buffer (20% glycerol, 10 mM NaCl, 1.5 mM MgCI2, 0.2 mM EDTA, 1 mM DTT, 1 mM PMSF, and 0.1% Triton X-100 in 20 mM HEPES buffer, pH 7.6). The lysate was centrifuged at 800xg for 4 min; the resulting pellet was diluted with three volumes of 500 mM NaCl in buffer (see previous description), then incubated at 4°C for 30 min with agitation. The mixture was centrifuged at 35,000xg for 10 min and the supernatants containing p53 were removed for immediate analysis. The total protein content of the extracts was determined using BCA protein assays (Pierce Biochemicals). Protein content for all samples was equalized prior to performing the binding assay. The consensus p53 binding sequence (GGACATGCCCGGG CATGTCC) was synthesized, prepared in double-stranded form, and end-labeled with [32P]-ATP). A 21mer randomized at each base was synthesized and used as a nonspecific competitor control. Binding reactions consisted of 20 pg nuclear protein, 0.5 ng 32P-labeled oligonucleotide, and 0.5 pg salmon sperm DNA (Sigma Chemical) with buffer (without Triton) in a final volume of 25 pl. Binding reactions were incubated at room temperature for 20 min; 8 pl of the reaction mixture was separated on 6% nondenaturing polyacrylamide gels (Novex) and visualized by autoradiography.

Cell Proliferation Studies
We examined a number of synthetic food dyes to determine their potential for growth-promoting activity on ER-positive growth-arrested human breast cancer cells. The effects of Red No. 3 were also tested on ER-negative HTB 125 breast cells and on a RLE cell line. One synthetic food dye (Food Drug and Cosmetics Red No. 3) was found to stimulate the growth of ERpositive human breast cancer cells in proportion to the applied dose ( Figure IA). ER-negative cultured breast cells did not respond to Red No. 3 ( Figure 1B), nor did Environmental Health Perspectives * Vol 105, Supplement 3 * April 1997 RLE cells ( Figure 1C). However, HTB 125 cells may be myoepithelial in origin, whereas HTB 133 cells are transformed and are probably derived from secretory epithelia. Therefore, HTB 125 cells may differ from HTB 133 cells in many other aspects besides being ER-negative. ER-positive breast cancer cells also respond to DDT in a dose dependent manner (Figure 2A), whereas RLE did not respond ( Figure 2B). These two studies suggest that Red No. 3 could stimulate the growth of human breast cells and that effects are mediated through the ER. Therefore, we examined the effects of the steroidal antiestrogen ICI 182,780 on breast cells treated with Red  Competitive binding studies (11) were used to confirm that Red No. 3 has estrogenic activity. Red No. 3 successfully competed for the ER from the MCF-7 cells (Figure 3). Red No. 3 (25 pg/ml) increased ER-ERE binding approximately 2.5-fold over that produced by the control (1 times protein control). Doubling the protein in the control reaction mixture (2 times protein control) increased the intensity of the ER-ERE complex approximately 3-fold higher than produced by the 1 times protein control ( Figure  4A). The lowest concentration of estradiol that increased ER-ERE binding was 100 pM, which increased ER-ERE intensities approximately 1.3-fold over the 1 times protein control. DDT (300 nM) increased ER-ERE binding nearly 2-fold over the 1 times protein control ( Figure 4A).
Red No. 3 increased ER-ERE binding 1.5-fold over the protein control ( Figure  4B). Low concentrations of ICI 182,780 (10 nM), when added to the medium of HTB 133 cells containing Red No. 3 (25 pg/ml), completely inhibited increased ER binding to the ERE but only partially inhibited the response stimulated by DDT (0.3 pM) (0.5-fold of the control) and P-estradiol (10 nM) (0.5-fold of the control) ( Figure 4B). Complete inhibition of estradiol-induced ER-ERE binding by ICI 182,780 requires 10 to 15 M excess of the inhibitor (17). ICI 182,780, when used at a concentration 10 to 15 times that of estradiol, can inhibit the ER-ERE binding to levels below that of the untreated controls (17). Since ICI 182,780 can inhibit Red No. 3 ER-ERE complex formation ( Figure 4B) and stimulation of cell growth (Figure 2), this confirms that the actions of Red No. 3 are mediated through the ER. Figure 4B shows that phenol red (50 pg/ml) simulates the ER-ERE binding as well (1.3-fold increase over control).

Cycfin Kinase Assays
Entrance into the cell cycle requires acivation of Cdk2, which is a late event in G1/S transition (21,22). The cyclin E-Cdk2 complex then phosphorylates the tumor suppressor protein pRb1O5 (21)(22)(23). We reasoned that if Red No. 3 were capable of stimulating breast cell growth, the activity of Cdk2 must be increased in treated cells. Increased activity of Cdk2 can be demonstrated in ER-positive breast cancer cells using immune complex assays with histone H I as the target ( Figure 5) (17). DDT stimulates Cdk2-associated kinase activity in MCF-7 breast cancer cells, but increased Cdk2 activity requires larger amounts of DDT when compared with estradiol ( Figure 5A). Increased Cdk2 activity was also stimulated by adding Red No. 3 to MCF-7 cells ( Figure 5B). Between 3 to 10 pg/ml of Red No. 3 were required to achieve the same phosphorylation of the histone target as that produced by 2% (v/v) FBS ( Figure 5B). . DDT and DNA-damaging agents including actinomycin, mitomycin C, and 5-fluorouracil (5-FU), also were tested for effects on p53-DNA site-specific binding. Figure 6 shows that p53 extracts prepared for DNA binding studies, which were isolated from MCF-7 cells treated with Red No. 3 Figure 6). Therefore, Red No. 3 appears to penetrate human breast cancer cells in vitro and has access to the nuclear compartment including the genetic material and the enzymes that modify it.
After the treatments, p53-DNA sitespecific binding increased in cultured HTB 125 breast cells after the application of DNA-damaging chemotherapeutic agents and Red No. 3 ( Figure 7A). Titration of Red No. 3 on RLE cells showed that treatment of the cells for 2 hr with 25 pg/ml was sufficient to increase p53-DNA binding. As little as 100 ng/ml of Red No. 3 was sufficient to significantly increase p53 binding after a 4-hr exposure (data not shown). For unknown reasons, DDT failed to increase p53-DNA binding when added to the medium of RLE cells ( Figure 7A).
We also examined other commonly used food dyes to determine their effects on p53-DNA binding and if metabolic activation or inactivation might occur by treatment with S9 liver extract. Blue No. 1, Green No. 3, and Yellow No. 5 slightly increased p53 binding in MCF-7 cells after treatment with S9 but not without treatment ( Figure 7B). Red No. 3 stimulated the largest increase in p53-DNA binding without S9 treatment ( Figure 7B). Although S9 treatment of Red No. 3 had no additional effect, p53-DNA site-specific binding was increased in HTB 133 cells treated with Red No. 3. The effects of Red No. 3 and other dyes on HTB 133 cells were similar to our studies using MCF-7 cells (Figures 7B, 7C). Preliminary studies using Red No. 40 on MCF-7 cells have shown no effect on p53-DNA binding (not shown). Some nongenotoxic stresses increase p53-DNA binding, including hypoxia and heat shock (24,25)   p53-DNA binding stimulated in RLE and human breast cells may occur by some mechanism other than DNA damage. However, it has been demonstrated that p53 recognizes damaged DNA (25,26) and it has not been reported that treatment of cultured cells with nongenotoxic agents such as estrogen will stimulate p53-DNA binding. Preliminary studies suggest that treatment of DNA with Red No modification of the DNA using endonucleases alters the results by action of the endonuclease unpublished data). Preliminary,, suggest that Red No. 3 decreases in the Ames II test (C Dees, u results). These results are cons two previous studies that suggest interacts directly with DNA or affects DNA-modifying enzymes (genointeractive) (28,29). Few other studies have examined the risks of DNA damage that might be associated with food colorants. No studies have specifically investigated the effects of synthetic food colorants on human breast cells. Red No. 3 has been shown to be mutagenic in the Bacillus subtilus sporulation assay (28), genointeractive in the Salmonella typhimurium assay (Ames test) (29), and carcinogenic in B6C3F1 mice (increased pheochromocytomas in males) (30). In the B. subtilus sporulation assay, S9 treatment appeared to decrease the effect of Red No. 3 (28), whereas in our study S9 treatment appears to have no effect. We believe that our results using a p53-gel mobility shift assay are consistent with a conclusion that Red No. 3 is capable HH1 of damaging DNA. Discussion noprecipitated Estrogen increases the risk of breast cancer d MCF-7 cells. (31), but the importance of xenoestrogens histone target, such as DDT in the etiology of breast cancer remains controversial (7,(12)(13)(14). It is interesting to note that the major route of exposure to a number of xenoestrogens is through the food chain (32). However, while the incidence of breast cancer in the United States is increasing, exposure levels to several estrogenic pesticides and pollutants are decreasing (32).
In contrast, the diet of women in the United States includes processed foods that are increasingly likely to contain food colorants and additives (27). Industrialized countries rely heavily on processed foods; over the last several decades about 80% of their food supply has been processed by the food industry (27,33). Further, the use of food additives continues to increase at a rate of 4 to 5% annually (34). Processed foods are also more likely to be packaged in materials that may contaminate the food with estrogenic plasticizers (15). Therefore, ated with Red as the diet of industrialized countries ther colorants, becomes more reliant on processed foods, the exposure to estrogenic dyes, preservatives, and contaminants from packaging materials increases. Estrogenic pesticides 3 prior to and pollutants also contribute to the total restriction xenoestrogen exposure via foods. s produced It is difficult to determine the total s (C Dees Perhaps more important, the reported intake from young childhood through puberty was actually higher than for the total population: approximately 1.6 to 2.5 mg/kg/day at the 90th percentile (adjusted to 1995 estimates as above). Thus, during growth and development the intake of Red No. 3 may actually be higher than that necessary to induce p53 binding for greater than 10% of the population at a time when developing breast tissues may be most susceptible to xenobiotic challenge. In addition to inducing p53 binding to DNA, Red No. 3 also causes an induction of cell proliferation (inhibitable by antiestrogen) at a concentration of 10 pg/ml. This concentration is only approximately 3 times greater than the predicted concentration for the top 1% of the population or the top 10% of young children consuming dye-containing foods. The predicted physiological concentration of Red No. 3 is approximately 33% of the level necessary to enhance cell proliferation in vitro. However, it should be noted that the full interaction of growth factors on signal transduction in relation to steroid binding to ER, and the role of growth factors in inducing ER-positive cells to enter and complete the cell cycle, remains to be fully elucidated. For example, the ability of xenoestrogens to induce breast cancer cells to enter the cell cycle is enhanced when other growth factors (e.g., insulin) are present (36). Growth factors such as insulin, insulinlike growth factor, and epidermal growth factor are able to phosphorylate Environmental Health Perspectives * Vol 105, Supplement 3 * April 1997 B and activate the ER via kinase activated signaling pathways without the presence of compounds bound to the ER ligand binding site (37,38). Phorbol esters are also able to induce the phosphorylation of the ER (39). When the ER is phosphorylated it can bind to the estrogen responsive elements without a compound bound to the ligand binding site (38). In addition, other investigators have documented that the activity of one xenoestrogen may be synergistically enhanced when another one is present (40). Other factors may greatly enhance the ability of xenoestrogens such as Red No. 3 to induce ER-positive cells to enter the cell cycle. Therefore, a concentration of Red No. 3 in vivo that is 33% of that necessary to cause proliferation in vitro may be sufficient to trigger ER-positive cells to progress to the cell cycle.
Cancer risk from Red No. 3 may be further increased if developing reproductive tissues are exposed. The effects of extra estrogen on developing reproductive tissue has been demonstrated. In laboratory animals the mammary glands of female mouse pups exposed to inappropriate levels of estrogen during development are larger than those of control animals (41,42). Furthermore, the increase in terminal end bud formation observed in these mice increases the likelihood that they will develop breast cancer (41,42). In addition, exposure to estrogens during the development of the mouse reproductive tract permanently estrogenizes cells (43). Two genes that respond to estrogen (lactoferrin and epidermal growth factor) are persistently expressed after exposure to estrogen during development (43). Thus, the greatest risk associated with exposure to xenoestrogens may occur during the period from early childhood through puberty, a period in which the highest consumption of Red No. 3 occurs.
While the role of diet in increasing breast cancer risk of U.S. women is generally accepted, the particular component of the diet that confers the risk is not. Most of the current studies on the factors in the diet of U.S. women that contribute to breast cancer have focused on total fat content. Processed foods contain the highest levels of added fat also and are foods most likely to contain the highest levels of food colorants. Therefore, it is possible that the correlation of high fat foods to increased risk of breast cancer noted in previous studies is actually caused by the presence of xenoestrogenic food additives such as Red No. 3. Xenoestrogenic food additives and other xenoestrogens that are found in foods, including pollutants (e.g., dioxins, polychlorinated biphenyls), packaging contaminants (phthalates), and pesticides, may be in total the dietary factor that contributes to the high breast cancer risk of women in the United States and other industrialized countries. However, many other factors are likely to increase the risk of breast cancer for women living in industrialized countries. Additional risk factors may include high fat diets, poor exercise habits, and high total body fat (6). The age of menarche or menopause, alcohol use, and parity may also increase risk. While our studies suggest that a xenoestrogenic food dye may increase the risk of breast cancer, further studies are required to determine if the estrogenic and genotoxic effects of Red No. 3 on cells in vitro also occur in vivo.