The inhibition of the estrogenic effects of pesticides and environmental chemicals by curcumin and isoflavonoids.

Many environmental chemicals and pesticides have been found to be estrogenic and have been shown to stimulate the growth of estrogen receptor-positive (ER-positive) human breast cancer cells. Since it is difficult to avoid human exposure to environmental estrogens, a potentially important area of research is the development of dietary strategies to prevent the stimulated growth of breast tumors by environmental estrogens. In this context, the inhibitory action of curcumin and a combination of curcumin and isoflavonoids were studied in ER-positive human breast cancer cells (MCF-7 and T47D) and ER-negative MDA-MB-231 cells induced by the pesticide o,p'-DDT and the environmental pollutants 4-nonylphenol and 4-octylphenol. The median inhibitory concentration (IC50) for curcumin in T47D cells was 10 microM when measured at either a 48-hr or a 6-day incubation time. The IC50 value for curcumin was within the 8-10 microM range for inhibiting the growth of T47D cells induced by a 10- microM concentration each of 4-nonylphenol, 4-octylphenol, and o, p'-DDT. The IC50 for curcumin in MCF-7 cells induced by 10 microM of either o,p'-DDT, 4-octylphenol, or 4-nonylphenol were 9, 39, and >50 microM, respectively. A combination of curcumin and isoflavonoids was able to inhibit the induced growth of ER-positive cells up to 95%. For MDA-MB-231 cells, the IC50 for curcumin was 17 microM, which was reduced to 11 microM in the presence of 25 microM genistein. Curcumin and genistein induce drastic changes in the morphological shape of both ER-positive and ER-negative cells. Data presented here indicate that a mixture of curcumin and isoflavonoids is the most potent inhibitor against the growth of human breast tumor cells. These data suggest that combinations of natural plant compounds may have preventive and therapeutic applications against the growth of breast tumors induced by environmental estrogens.

Environmental chemicals may be involved in the etiology of a variety of human cancers. The cause for the majority of human tumors has been attributed to exposure to environmental carcinogens, pollutants, pesticides, drugs, UV-radiation, and tobacco products (1). A nutritional deficit or surplus or the absence of preventive micronutrients in the diet can further increase the susceptibility for developing cancers. In addition, tumors induced by environmental pollutants may be prevented by dietary strategies. It is therefore plausible that an imbalance between the exposure to cancer-causing environmental factors and the dietary intake of preventive nutrients facilitates the initiation and growth of tumors. The incidence of breast tumors in women is increasing, and environmental chemicals have been implicated in part for this increase (2). In vitro and in vivo data have shown that endogenous estrogens, such as estradiol, or environmental estrogens [e.g., DDT, polychlorinated biphenyls, 4-nonylphenol (4-NP), 4-octylphenol (4-OP), and many others] promote mammary carcinogenesis (3)(4)(5). A majority of the human breast tumor cells express estrogen receptors, and the growth of such tumors is generally estrogen dependent. The growth of estrogen receptor (ER)-positive tumors can be inhibited by antiestrogens blocking the estrogen receptor (6,7) or by inhibiting estrogen production (8).
Since it may not be possible to avoid human exposure to environmental chemicals, developing nutritional strategies for prevention and control of breast tumors induced by environmental estrogens is a highly relevant concept. In this context, natural plant-derived foods that modulate tumor growth also could form a framework for developing safe chemopreventive drugs. Two such dietary natural plantderived products of interest are curcumin and isoflavonoids. Curcumin (diferuloylmethane)-the major constituent of turmeric (a spice used in Indian dishes), which is isolated from the rhizomes of the Curcuma longa Linn plant-has shown anticarcinogenic properties in animal and cell models. It inhibits skin, forestomach, stomach, and colon tumors in animals (9)(10)(11)(12)(13)(14) and mammary tumorigenesis in rats (15 Curcumin has also been shown to have antistimulation properties toward human immunodeficiency virus (HIV) type I by inhibiting HIV-1 integrase and protease (23). The antiinflamatory and antioxidant properties of curcumin have been known for a long time (24). Most of the above studies on animal and cell models suggest curcumin, also found in many plant products other than turmeric, is an antiproliferative agent. Its effects on the growth of breast tumor cells are less well studied, and epidemiological or dinical studies of the risk for breast cancer in populations consuming curcumin in their daily diets have not been well documented.
Another class of plant-derived dietary products, isoflavonoids, have been shown to have many beneficial clinical and antitumorigenic properties . In addition to the beneficial properties, in vitro data have indicated that certain dietary compounds such as genistein and some metabolites of isoflavonoids, at concentrations below 10 pM, can enhance the growth of MCF-7 cells (28)(29)(30). At higher concentrations (>100 PM), isoflavonoids have been shown to cause nonbeneficial cytotoxicity or sublethal injury (29). The effect of low or high levels of isoflavonoids would yield undesirable results. Considering these limitations, studies of the effect of the combination of curcumin and isoflavonoids were initiated in our laboratory, where concentrations of individual compounds could be controlled to achieve the maximum beneficial effects against the growth ofbreast cells (31).
We previously reported that the combination of curcumin and genistein is the most sensitive inhibitor against estrogenic pesticide-induced growth of ER-positive MCF-7 human breast cancer cells (31). Since the inhibitory potency of a single compound is limited and higher concentrations of inhibitory compounds may cause toxicity to normal cells or induce sublethal effects that may cause cellular transformation or tumor stimulation, the use of a combination of natural compounds with different mechanisms of action constitutes a Growth of cells in control medium without curcumin was normalized to 100%.  Methods for details. Cell growth was measured on day 6 using the tetrazolium (MTT) assay. Induced cell growth was normalized to 100% in each case. reasonable approach for preventing the harmful effects of environmental estrogens. In this study, the ER-positive (MCF-7 and T47D) and ER-negative (MDA-MB-231) cell lines were used as models for human breast cancer. In this paper, we describe the inhibitory and cell killing effects of curcumin and curcumin plus isoflavonoids on the growth of these cells and on the growth of ER-positive cells induced by the estrogenic environmental chemicals o,p'-DDT, 4-NP, and 4-OP. These data support the condusion that a combination of curcumin and isoflavonoids is the most effective means for preventing the growth of both ER-positive and ER-negative cells.

Materials and Methods
Chemicals. Stock solutions of o,p'-DDT from Ultra Scientific, Kinston, Rhode Island; 4-NP and 4-OP from Aldrich Chemical Company, Milwaukee, Wisconsin; 17-,B estradiol and curcumin from Sigma Chemical Company, St. Louis, Missouri; and genistein, daidzein, biochanin A, formononetin, and equol from Indofine, Sommerville, New Jersey, were prepared by dissolving the compounds in DMSO. All stock solutions of compounds were diluted with phenol redfree tissue culture media supplemented with 5% dextrane-coated charcoal-treated serum (DCC; Sigma Chemical Co.) and 100 units/ml of penicillin/streptomycin to an appropriate concentration (test media). The control media did not contain more than 0.5% DMSO. Cell culture and growth procedure. Estrogen receptor-positive MCF-7 and T47D and ER-negative MDA-MB-231 human breast cancer cells were grown in phenol red RPMI 1640 media (Gibco BRL, Grand Island, NY) supplemented with 5% fetal calf serum (FCS) and 5 ml 10,000 U/ml of penicillin/streptomycin in an incubator maintained at 5% CO2 95% air and 100% humidity at 37°C.
Prior to each experiment with MCF-7 and T47D cells, we replaced phenol red maintenance media with phenol red-free media for 24 hr. The phenol red-free media was then replaced with phenol red-free RPMI 1640 media supplemented with 5% DCC-stripped bovine serum for 48 hr. Cells were detached by adding 2-3 ml 2.5% trypsin solution. After 15 min, cells were harvested and washed with Ca-Mg free phosphate buffered saline (PBS). Washed cells were suspended in phenol red-free RPMI 1640 media supplemented with 5% DCC and antibiotics and were gently agitated by passing up and down in a pipet to get a single cell suspension. Cells were counted by a Coulter Counter. Cells (2-3 x 105/ml) were plated in a 12-well culture plate (2 ml/well) and were allowed to attach for 48 hr.
After 48 hr, media was removed and replaced by test media (prepared as above) containing the appropriate concentrations Table 2  of environmental chemicals, curcumin, and/or isoflavones. Control cells were incubated in phenol red-free media containing 5% DCC and an equivalent amount of the solvent without the test material. The solvent concentration did not exceed 1%, and this concentration does not appreciably alter the cell growth. The cell growth was estimated on day 6 using the tetrazolium assay (MTT), as reported previously (31). The absorbance at 540 nm was used as a measure of cell density of live cells. The percent proliferation was calculated using the following equation: Percent proliferation = Cell Density with Test Compound X 100 Cell Density of Control Estrogen receptor-negative MDA-MB-231 cells (2 x 105 in each well) were plated in 12-well plates and were allowed to attach for 24 hr. We replaced the media with the media containing appropriate concentrations of curcumin and isoflavones. Cell densities were estimated on day 4 using the MTT assay. Growth ofcontrol and treated cells was calculated using the above equation, and data were normalized to 100%. In order to see the time-dependent inhibition, cells plated in 12-well plates were treated with phenol red-free media containing varying concentrations of curcumin for 4 hr. Media containing curcumin was removed. After washing cells with PBS, curcumin and phenol red-free complete medium was added to each well and cells were allowed to grow in the incubator for 5 days.
Cell morphology. For morphological studies, MDA-MB-23 1 and MCF-7 cells were plated in tissue culture dishes containing glass coverslips. After reaching near confluence, cells were treated with curcumin and genistein as mentioned above and were incubated for additional 48 hr. Treated cells were washed with PBS, fixed with methanol, stained with Giemsa, and photographed.
MDA-MB-23 1 control and curcumintreated cells (48 hr) were also stained by the terminal deoxynucleotidyl transferasemediated nick labeling (TUNEL) reaction for determining apoptosis using the Boehringer Mannheim kit and kit protocol (Boehringer Mannheim, Indianapolis, IN). Cells were permeabilized and stained with propidium iodide (PI). Apoptotic cells were quantitated using a Zeiss fluorescence microscope (Zeiss, Thornwood, NY)

Effect of Curcumin on the Growth of T47D Cells
The T47D cells used in these studies respond to 17p-estradiol (i.e., ER-positive).
The effect of curcumin on the growth of these cells as a function of the concentration and time of incubation is shown in Figure 1. The cell growth in 2% FCS has been normalized to 100% for data shown in Figure 1. The cell growth is inhibited as a function of curcumin concentration when measured on day 6 while the inhibition levels off at approximately 40% when measured after 48 hr (Fig. 1). However, the concentration of curcumin required to inhibit the growth of T47D cells by 50% (IC50) is approximately the same (10 ,uM) when measured after 48 hr or on day 6. Effect of Curcumin on Induced Proliferation of ER-positive Cells MCF-7 cells. It has been reported that the pesticide o,p'-DDT and the environmental chemicals 4-NP and 4-OP have estrogenic properties. These compounds can induce the proliferation of MCF-7 cells (Table 1) (32). 4-NP was found to be more potent at inducing growth at the 5 pM concentration. Higher concentrations (>10 pM) of 4-NP and 4-OP are found to be toxic to MCF-7 cells.
The effect of curcumin on the proliferation of MCF-7 cells induced by o,p'-DDT, 4-NP, and 4-OP is shown in Figure 2. For the sake of calculating the IC50 values for individual antagonistic compounds, the percent growth of cells induced by environmental chemicals has been normalized to 100%. The curcumin IC50s for inhibition of growth induced by o,p'-DDT, 4-OP, and 4-NP are 9, 39, and >50 pM, respectively. Curcumin appears to be a potent inhibitor of o,p'-DDT-induced cell proliferation and a less effective inhibitor for 4-NP-induced proliferation. At 50 pM ofcurcumin, the cell growth induced by o,p'-DDT is reduced to 12% of that noted in the absence ofcurcumin. Table 2 shows the inhibitory effects of various combinations of curcumin and isoflavonoids on the growth of MCF-7 cells induced by environmental pollutants. The induced growth by pollutants has been normalized to 100%. The percent growth in the presence ofthe combination of curcumin and isoflavonoids was then calculated with respect to this normalized value. Data presented in Table 2 show a synergistic inhibitory effect in the presence of the combination of curcumin and isoflavonoids. As shown in Table 2, the 4-NP-induced cell growth is reduced to 48% (52% inhibition) in the presence of a mixture of curcumin (10 pM) and genistein (25 FM). A mixture of curcumin and equol at 25 pM each reduces the growth of MCF-7 cells induced by 4-NP and o,p'-DDT to 18% and 1 1%, respectively. Tamoxifen, a well-studied antiestrogen can reduce the growth of MCF-7 cells induced by 4-NP to 70-80%, while the mixtures of tamoxifen (2.5 pM) and curcumin (10-75 pM) can substantially reduce the induced growth ( Table 2). The combination of curcumin and genistein at 50 pM each is able to reduce the growth of MCF-7 cells induced by 10 pM 4-OP to 6% (94% inhibition). T47D cells. The done ofT47D cells used here is estrogen responsive and shows induced proliferation in the presence of 17p-estradiol.
The growth of T47D cells in the presence of 5 nM 17p-estradiol was normalized to 100%; relative normalized values for the induced percent proliferation by o,p'-DDT, 4-NP, and 4-OP are 83, 79, and 76%, respectively. In Figure 3, the induced growth by individual environmental chemicals has been normalized to 100% and each data point represents the percent growth inhibition as a function of curcumin concentration. The IC50 values of curcumin are within the 8-10 pM range for inhibiting the induced growth by 10 pM concentrations of 4-NP, 4-OP, and o,p'-DDT. Unlike the inhibitory effects seen in MCF-7 cells, curcumin seems to be an equally effective growth inhibitor of T47D cells in the presence ofeach ofthese growth inducers.
The combination of curcumin and isoflavonoids is more potent in T47D cells induced by environmental pollutants (Table  1). Curcumin alone, at the concentration of 15 pM, could reduce the growth of T47D cells induced by o,p'-DDT (10 pM) to approximately 35% (Fig. 3), while a mixture of curcumin (15 pM) and genistein (10 and 25 pM) is able to reduce the induced growth to about 15% ( Table 1) Figure 4. Curcumin was found to be cytotoxic for these cells. The IC50 of curcumin is 17 pM (Fig. 4). In the presence of 25 pM genistein, the IC50 of curcumin is reduced to 11 pM (Fig. 4).  (Fig. 6B).  in phenol red-free media and 5% DCC serum responded to curcumin plus genistein in a similar manner as MDA-MB-231 cells in standard medium treated with curcumin (data not shown).
The TUNEL labeling for apoptosis demonstrated that curcumin treatment increased apoptosis, but the absolute number was small. Apoptosis data showed a 10% increase in the apoptotic death of cells when measured after 48 hr of treatment with 10 pM of curcumin.

Discussion
The data presented in this paper demonstrate several major points: 1) curcumin can inhibit the growth of human breast cancer cells independent of the expression of estrogen receptors; 2) curcumin more effectively inhibits the growth of ER-nega-

X
Articles * Curcumin and isoflavonoids inhibit estroqenic effect curcumin can inhibit the growth of both ER-positive MCF-7 and T47D cells induced by estrogenic environmental chemicals and pesticides; 4) the combination of curcumin and genistein is the most potent inhibitor against the growth of both ER-positive and ER-negative breast cancer cells; 5) tamoxifen, a well studied antiestrogen, could only reduce the growth of MCF-7 cells induced by 4-NP to about 70-80%, while a combination of tamoxifen and curcumin can further reduce the induced growth; and 6) the inhibitory effects of curcumin and isoflavonoids are synergistic and may involve separate pathways of action. The exact mechanism of action of curcumin for inhibiting the growth of tumor cells induced by estrogenic environmental chemicals (ER-positive cells) or ER-negative cells growing in the normal tissue culture media has not been fully explored. Many biological and medicinal properties of curcumin are now well recognized. The inhibitory action of curcumin against the growth of tumor cells should be associated with some of its biological properties. The anti-inflamatory properties of curcumin have been known for a long time. Curcumin has been shown to influence the activities of enzymes such as cyclooxigenase (COX), lipoxygenase (LOX), phospholipase A2 (PLA2), and phospholipase C-gama-1 (PLCgl) (12,13). Several metabolites of COX and LOX are involved in growth signaling (33)(34)(35). Increased activities of PLA2 and PLCgl have been observed in breast tumors (36). PLCgl may produce diacyl glycerol and inositol 1,4,5-triphosphate; these signaling compounds can stimulate protein kinase C, which has been implicated in modulating cell growth. Curcumin may be inhibiting the growth of both ERpositive and ER-negative cells by directly modulating the activities of COX, LOX, PLA2, and PLCgl enzymes, including the activity of protein kinase C, as has been reported in colon tumors (12,13).
The cell morphology data presented in this paper show that cells become round after 48 hr of treatment with curcumin or curcumin plus genistein. These data suggest the possibility of a disruption in cell matrix and a reduction in cytoplasmic contents of curcumin and curcumin plus genistein-treated cells. These initial structural disruptions could direct cells to initiate the apoptotic or programmed cell death process. There is some indication by TUNEL data (10% increase in apoptosis) that this process may actually be triggered. However, more data are needed for determining the exact mechanism by which curcumin inhibits the induced growth ofbreast tumor cells.
Genistein is well recognized for its inhibition of various kinds of protein kinases. There is a possibility that genistein inhibits cell growth by its direct action on protein kinases (e.g., protein kinase C, tyrosine kinase, MAP-kinase). Protein kinases are involved in the phosphorylation of estrogen and progesterone receptors. Phosphorylated receptors are involved in activating transcription factors. Since curcumin and genistein have been shown to have a synergistic effect against the growth of both ER-positive and ER-negative cells, the action of these compounds may be through separate pathways. The concentration and time-dependent growth inhibition data suggest that curcumin and isoflavonoids may not be competing for the same target in order to block the growth. The exact mechanism of action of these compounds remains to be elucidated.
These results support the concept that a combination of natural chemopreventive nutrients is more potent than individual compounds against suppressing the growth of breast tumors induced either by environmental chemicals or other carcinogens (31). Also, the in vitro data presented in this paper constitute the framework for further studies in animal models and clinical trials. Data on natural plant compounds may further be helpful in developing new therapeutic agents.