The effects of dietary phytoestrogens on aromatase activity in human endometrial stromal cells

– Dietary phytoestrogens have been reported to inhibit arom atase activity in placental microsomes, but the effects in the human endometrium are unknown. Aromatase, the rate-limiting enzyme in the conversion of androgens to estrogens, has recently been shown to be expressed in the endometrium of women with endometriosis and is thought to play a role in the pathophysiology of this disease. Therefore, the objective of this study was to sc reen dietary phytoestrogens for their ability to inhibit aromatase activity in human endometrial stromal cells (ESC) and identify potential novel therapeutic agents for the treatment of endometriosis. The inhibition of aromatase activity by direct interaction with the dietary phytoestrogens genistein, daidzein, chrysin, and naringenin was tested in a cell free assay. Furthermore, test compound effects on aromatase activity in ESC cultures were also examined. Genistein and daidzein were inactive in the human recombinant aromatase assay whereas naringenin and chrysin inhibited aromatase activity. However, genistein (1 nM to 1 mM) stimulated aromatase activity in ESC whereas other phytoestrogens had no effect. Immunopositive aromatase cells were demonstrated in genistein-treated ESC but not in untreated control cultures. Taken together, our data suggest th at genistein can increase aromat ase activity in ESC likely via increased enzyme expression. phytoestrogens / endometriosis / aromatase / genistein / endometrium 1. INTRODUCTION Cytochrome P450 aromatase (P450AROM) is the rate limiting enzyme that catalyzes the conversion of androstenedione and testo- sterone to estrone and 17 β-estradiol, respectively. While the ovaries are the pri- mary source of estrogen in the body, local production of estrogen by other tissues has also been demonstrated in estrogen depend- ent diseases such as breast cancer [1, 2] and endometriosis [3–5]. Endometriosis is a common gynecologic disorder that is char- acterized by the pres ence of endometrial glands and stroma outside of the uterine cavity. Endometriosis is an estrogen depend- ent disease [7, 8] that affects approximately 14% of all women of reproductive age, and 30–50% of infertile women [9]. Local pro- duction of estrogens by ectopic endometrial * Corresponding author: [email protected] Article published by EDP Sciences and available at http://www.edpsciences.org/rnd or http://dx.doi.org/10.1051/rnd:2005055 710 K.M. Edmunds et al. implants in women with endometriosis may explain treatment failures and the persist- ence of recalcitrant en dometriosis in post- menopausal women [6]. Therefore targeted inhibition of local estrogen production in endometriotic lesions by inhibition of aro- matase activity may have a place in the management of this disease. As endometri- osis is the leading cause of hospitalization for gynecologic surgery [10], thus novel, safe and effective treatment options are urgently needed. Phytoestrogens are a class of plant estro- gens that include isoflavones, flavones, fla- vonones and several mycotoxins such as coumestrol and zeralanone. Phytoestrogens are thought to have health benefits such as providing protection against breast cancer development [11, 12] and are potentially useful in the management of menopausal symptoms [13, 14]. Dietary factors such as phytoestrogens have been shown to inhibit aromatase activity [15, 16] without altering plasma estrogen concentrations [17]. There- fore, therapeutic use of phytoestrogens may be of benefit to women with endometriosis. Soy-based foods have high phytoestro- gen content of which genistein is the dom- inant isoflavone [18]. Chrysin, a flavone found in the plant Passiflora coerula and naringenin, a flavo none found in citrus fruits, have been shown to inhibit aromatase activity in hepatocytes and placental micro- somes in vitro [19, 20]. However, the effects of these comp ounds on aromatase expression are unknown. Furthermore, aro- matase expression is regulated via different promoter regions in a tissue specific man- ner [21] and thus the effects of phytoestro- gens on endometrial aromatase expression and activity are also unknown. Therefore, the objective of this study was to screen sev- eral phytoestrogens for their ability to directly inhibit arom atase activity and to determine the effect of dietary phytoestro- gens on aromatase expression and activity in human endometrial stromal cells. Genis- tein and daidzein, the dominant phytoestro- gens in the diet, together with chrysin and naringenin, two phytoestrogens previously shown to inhibit aromatase activity were selected as the test compounds for this study. We hypothesized that phytoestro- gens will inhibit aromatase activity in endometrial stromal cell cultures and thus potentially provide a novel therapeutic option that is both natural and effective in the management of endometriosis. 2. MATERIALS AND METHODS 2.1. Cell-free assay The ability of the test compounds to interact directly with the enzyme to alter aromatase activity was investigated in a cell free assay by modification of the fluores- cence assay described previously [22], using human recombinant aromatase expressed in insect cell microsomes (CYP19 suprasomes BD Gentest Biosciences, Woburn, USA) and 0.25 µM dibenzylfuo- rescein (BD Gentest Biosciences, Woburn, USA) as the substrate. The ability of test compounds (1 pM–100 µM in 0.1 M potas- sium phosphate buffer pH 7.4) to inhibit aromatase enzyme activ ity (0.4 pmol aro- matase/well was examined by incubation in the presence of cofactors (40 µM NADP, 100 µM Glucose-6-phosphate, 100 µM MgCl 2) and DMSO (1%). Assays were per- formed in a 96-well black walled culture plate (Becton Dickinso n, Franklin Lakes USA) in a total volume of 202 µL. Reac- tions were started by addition of 50 µL of prewarmed (37 °C) enzyme to the pre- warmed plates. Blank wells contained 50 µL of buffer in place of the enzyme. The plate was incubated at 37 ºC for 1 h and pre- liminary experiments showed that enzyme activity was linear up to 90 min. The reac- tion was stopped by the addition of 75 µL of 2 M NaOH to each well. Fluorescence was measured using a PerkinElmer HTS 7000 Bio Assay Reader at an excitation wavelength = 485 nm and emission = 535 nm. Fluorescein (Sigma Aldrich, Oakville, Canada) was used as the standard. Genistein induced increased aromatase activity 711 Non-linear least-squares regression analy- sis was used to fit inhibition curves to the equation: where E max and Emin are the maximum and minimum effects of the test compound, respectively. pIC 50 is the negative log of the molar concentration of the compound that produces 50% inhibition of enzyme and log C is the molar concentration of the com- pound that produces the effect E. IC 50 val- ues were converted to Kis using the Cheng Prusoff equation [23]: Ki = IC50/(1 + S/Km)) where S is the substrate concentration and Km is the Michaelis constant for the enzyme. The K m and maximum velocity (Vmax) of the enzyme reactions were deter- mined under the same conditions as described for inhibition experiments except that the substrate concentration varied between 0 and 0.4 µM. Data for these exper- iments were fit by non-linear least-squares regression to: V = (Vmax × S)/(Km + S) where V is the reaction velocity at substrate concentration S. 2.2. Endometrial stromal cell culture Endometrial biopsies were obtained from eighteen women aged 27–44 (mean (± SD) of 38.3 ± 6.0 years) undergoing benign gynecologic surgery at McMaster University Medical Centre. Informed con- sent was obtained from each patient by a research nurse and all procedures were car- ried out in accordance with approval of the McMaster University Research Ethics Board. Among the eighteen patients included in this study, eleven had a laparoscopic diagnosis of endometriosis and seven did not have any evidence of pelvic endometri- osis. None of the study subjects had received endocrine therapy in the previous six months before surgery. Endometrial tis- sue (1–2 g) obtained at hysterectomy was rinsed in Hanks’ balanced salt solution (HBSS) containing 200 units·mL –1 penicillin, 0.2 mg·mL–1 streptomycin and 0.5µg·mL–1 amphotericin B (Sigma Aldrich, Oakville Canada) to remove blood and debris. Sep- aration of the endometrial stromal cells was performed as previously described [24]. Briefly, the tissue was minced into 1 mm 3 fragments and digested for 2.5 h at 37 °C in medium containing collagenase type IA (2 mg·mL–1, Sigma-Aldrich, Oakville, Canada). After digestion, the remaining tis- sue fragments were mechanically dispersed and the dispersed cells were filtered through a 100 µm and subsequently a 40 µm cell strainer (Becton Dickson, Franklin Lakes, USA). Centrifugation (10 min, 725 × g) was used to pellet the cells after which time they were resuspended in 3 mL of plat- ing media [DMEM:F12, 4% FBS, 1% ITS+ and 1% antibiotic antimycotic solution (100 units·mL –1 penicillin, 0.1 mg·mL –1 streptomycin and 0.25 g·mL –1 amphoter- icin B (Sigma Aldrich, Oakville, Canada)]. Red blood cells were removed by layering the cell suspension over 3 mL of Ficoll- Paque PLUS (Amersham Biosciences, Uppsala, Sweden) in a sterile 15 mL poly- propylene tube. The solution was centri- fuged for 10 min at 400 × g. The media/ Ficoll interface layer containing the stromal cells was plated into 48 well Falcon tissue culture plates (Becton Dickson, Franklin Lakes, USA) at a density of 200 000 cells/ well/0.5 mL. Media was changed after 48 h and the cells were treated after 96 h in cul- ture, when the cells were near confluence. Purity of the cell preparation was confirmed by immunostaining for vimentin (mesen- chymal cell marker) and cytokeratin (epi- thelial cell marker) as described below. 2.3. Cell treatment and aromatase activity assay Cells were washed twice in HBSS and incubated for a minimum of 1 h in serum-free DMEM-F12 containing 100 units·mL –1 penicillin, 0.1 mg·mL –1 streptomycin and EE min += Emax Emin–() /1 1 0 pIC 50 Clog–– +() 712 K.M. Edmunds et al. 0.25µg·mL–1 amphotericin B (Sigma Aldrich, Oakville, Canada) prior to treatment for 24 h with increasing log concentrations (10 –9– 10–4 M) of genistein, daidzein, naringenin or chrysin (Sigma Aldrich, Oakville, Can- ada) diluted in serum free media. To exam- ine the role of estrogen receptor mediated effects, the cells were also treated with gen- istein in the presence of a non-selective estrogen receptor antagonist (ICI 182,780; Tocris, Ellisville, USA). After 24 h, the treatment media was removed and replaced with 500 µL of [1 β- 3H]-androstenedione [2.5 µCi·mL–1] (Perkin Elmer, Boston, USA) in DMEM-F12 (containing 100 units·mL–1 penicillin, 0.1 mg·mL –1 streptomycin and 0.25 µg·mL–1 amphotericin B) for 4 h at 37 °C. Aromatase activity was assayed using a radiometric technique that quantifies the incorporation of tritium from [1 β-3H]- androstenedione into 3H-labeled water as previously described [25]. Briefly, aro- matase activity was determined by transfer- ring 300 µL of the incubation medium to glass tubes, adding 300 µL of dextran coated activated charcoal (250 mg·mL –1, BD Biosciences, Oakville, Canada) to each tube and incubating for 2 h at 4° C. The samples were then centrifuged (15 min, 2500 × g) and the tritiated water content was determined by counting the supernatant in 5 mL of scintillation fluid (Aqueous Count- ing Scintillant, Amersham, England) in a liquid scintillation counter. To control for variation in the number of cells in each well, the aromatase activity was normalized to the cell protein content in each well as determined by the Bradford method. Due to variation in basal aromatase activity between patients, normalized aromatase activity was converted to a percentage of the control level for each culture. The aromatase assay is based on the release of tritiated water and the specificity of the assay was determined by co-incubation with 4-hydroxyandros- tendione an irreversible inhibitor of the cat- alytic activity of aromatase [26] to block the formation of tritiated water. 2.4. Immunocytochemistry Cells were seeded into 8 well Lab-Tek chamber slides (BD Biosciences, Oakville, Canada) at a density of 200 000 cells/well/ 0.5 mL. Media was changed after 48 h and the cells were treated with genistein (10 –6) after 96 h in culture. After 24 h of treatment, the cells were fixed in 10% neutral buffered formalin, washed in PBS, and endogenous peroxidase activity was quenched by incu- bating the cells in 3% hydrogen peroxide (in methanol) for 5 min. The cells were washed in PBS, incubated with the primary anti- bodies (Dako Diagnostics, Mississauga, Canada) for cytokeratin (1:50), and vimen- tin (1:50) for 1 h at room temperature and immunostaining was identified using EnVision (Dako Diagnostics, Mississauga, Canada) with diaminobenzidine (Sigma- Aldrich, Oakville, Canada) as the chro- mogen. The cells were counterstained with Carazzi hematoxylin. For negative controls, the cells were incubated with non-immune serum in place of the primary antibodies. To stain for the presence of aromatase in the genistein treated cultures and untreated controls, immunohistochemistry was per- formed on the chamberslides using a pri- mary monoclonal mouse antibody against human aromatase (1:50 Serotec, Raleigh, USA). Immunostaining was identified with the avidin-biotin-peroxidase technique using the Vectastain kit (Vector Laboratories, Burlington, Canada) with diaminobenza- dine as the chromogen and Carazzi hema- toxylin as a counter stain. 2.5. Statistical analyses Data were analyzed for equal variance and normal distribution. An effect of treat- ment on ESC aromatase activity was tested using a one-way analysis of variance (ANOVA) and differences between doses were determined using the Tukey multiple comparison method. A p value < 0.05 was considered to be statistically significant for all procedures used. Genistein induced increased aromatase activity 713 3. RESULTS 3.1. Recombinant human aromatase activity Aromatase activity in the presence of increasing substrate yielded a Km of 0.26µM (pKm = 6.6 ± 0.2) and a Vmax of 2.2 ± 1 pmol fluorescein released per mol enzyme per minute (Fig. 1). Naringenin (Ki = 0.3 µM) and chrysin (Ki = 1 µM) were potent inhib- itors of recombinan t human aromatase whereas genistein and daidzein were weak (Ki > 50 µM) inhibitors (Fig. 2). 3.2. Aromatase activity after phytoestrogen treatment of endometrial stromal cells Immunocytochemical staining for cells of mesenchymal origin and epithelial cells illustrated that our cultures consisted of Figure 1. Michaelis-Menten plot of the diben- zyfluorescein deakylase activity of recombinant human aromatase deter mined as described in

and methods. Points represent means and standard errors of triplicates within a single experiment. K m and Vmax values from this expe- riment were 0.22 µM and 1.4 pmol/pmol/min, respectively. Figure 2. Human recombinant aromatase activity as indicated by fluorimetrically quantified DBF dealkylase after treatment with naringenin, chrysin, genistein and dadizein. Each data point is the mean (± SEM) from three separated experiments. Naringenin and chrysin were effective inhibitors of the enzyme with a Ki = 0.3 and 1.0 µM, respectively, while genistein and daidzein were ineffective as shown by a Ki > 50 µM. 714 K.M. Edmunds et al. greater than 99% endometrial stromal cells (data not shown). Phytoestrogen treatment did not attenu- ate aromatase activity in ESC from women with endometriosis (n = 11) at any concen- tration tested (Fig. 3). However, genistein (10 –9–10–6 M) treatment of ESC from women without endometriosis (n = 7) resulted in a significant increase in aromatase activ- ity (P < 0.05) to approximately 150% above the activity observed in untreated ESC from the same patient (Fig. 4), whereas daidzein, naringenin and chrysin treatment had no effect. Furthermore, the genistein induced increase in aromatase activity was not atten- uated by co-treatment with the estrogen receptor antagonist ICI 182,780 ( P > 0.1, Fig. 5). 3.3. Immunocytochemistry Immunopositive aromatase staining was evident as a diffuse brown cytoplasmic pre- cipitate (Fig. 6) that was absent in control cultures where the primary antibody was substituted with non-immune serum. Immu- nopositive staining was focally present in some but not all genistein (10 –6 M) treated ESC from eutopic endometrium of women without endometriosis. Moreover, no immu- noreactive aromatase staining was visible in the untreated ESC taken from the same patient. 4. DISCUSSION The objective of the current study was to screen dietary phytoestrogens for their abil- ity to inhibit human recombinant aromatase activity and to determine the effect of die- tary phytoestrogens on endometrial stromal cell aromatase activity in culture. Although naringenin and chrysin inhibited aromatase in our cell-free assay, they were ineffective in endometrial stromal cell cultures from Figure 3. Aromatase activity was unchanged in genistein treated endometrial stromal cell cultures from women with endometriosis ( n = 11). The control bar represen ts the aromatase activity from the vehicle treated cells from each of the patients and the data bars represent the aromatase activity of the cells following treatment with genistein represented as percent of control. The control value has arbitrarily been set to 100% and data are presented as the mean ± SEM. Genistein induced increased aromatase activity 715 Figure 4. The effects of genistein treatment for 24 h on aromatase activity in endometrial stromal cells obtained from eutopic endometrium of women without endometriosis (n = 7). The control bar represents the aromatase activity from the untreated cells from each of the patients and the data bars represent the aromatase activity of the cells following treatment with genistein represented as percent of control. The results are the mean (± SEM) fro m seven different cultures. Values with different superscripts are significantly (P < 0.05) different. Figure 5. The effects of 10 –6 M genistein (GEN) alone and in combination with 10–6 M ICI 182 780 (ICI), and non-selective estrogen receptor antagonist, on aromatase activity obtained from the euto- pic endometrium of women without endometriosis (n = 3). The results are the mean (± SEM) from three different cultures. Means identified with a different letter were si gnificantly different ( P = 0.008). 716 K.M. Edmunds et al. women with and without endometriosis and thus are unlikely to have any potential ther- apeutic benefit in the management of endometriosis. In contrast, genistein, the dominant isoflavone found in soy-based foods, was inactive in the cell-free assay but to our surprise increased aromatase activity in endometrial stromal cells of women without endometriosis. These data suggest that the observed effects of genistein are not mediated through direct effects of genistein on enzyme activity but indirectly via enhanced aromatase expression in endome- trial stromal cells or via intermediates on aromatase activity. This point is supported by evidence of immunocytochemical stain- ing for aromatase in genistein treated but not untreated cells. In our study, the con- centrations of genistein that were used to treat the human endometrial stromal cells (1 nM to 10 mM) correspond to the serum concentrations of both Asian and Cauca- sian women who are consuming soy-based foods [27–29] and thus are considered to be physiologically relevant. Taken together, our results suggest that while phytoestro- gens may have health benefits such as the proposed protection against breast cancer development [11, 12], genistein is unlikely to have any therapeutic value in the man- agement of endometriosis and more impor- tantly may increase aromatase activity in the endometrium and thus could be an important factor in the pathobiology of this enigmatic disease. In the present study, aromatase was not detected by immunohistochemistry in con- trol cultures of endometrial stromal cells from women without endometriosis. In addition, aromatase activity of vehicle treated endometrial cells was at background levels for the assay and thus supports the view that aromatase is either absent or inhibited in the endometrium from women without endometriosis. Our findings are in agreement with prior studies in which aro- matase cytochrome P450 has been reported to be expressed in the endometrium of women with endometriosis but is either absent [4], or expressed at low levels in the endometrium of women without endome- triosis [30]. Therefore, the patients in the current study were grouped into two cate- gories: endometriotic and non-endometri- otic. None of the phytoestrogens tested inhibited aromatase activity of the ESC from women with endometriosis. However, in the current study, genistein-treatment of Figure 6. Immunocytochemical staining for aromatase in untreated cells (A) and cells treated with 10–6 M genistein (B) reveals positive staining in the treated cells (arrows). Genistein induced increased aromatase activity 717 ESC from women without endometriosis induced an increase in aromatase activity to 150% of the untreated controls similar to the findings using adrenocortical carci- noma cell lines treated with herbicides [31, 32]. Furthermore, our results are harmoni- ous with the previous finding that genistein (30 µM) increased aromatase activity 3 fold in the H295R human adrenocortical carci- noma cell line [33]. Hence, genistein treat- ment-induced changes in aromatase activity could lead to increased local levels of estro- gens in the endometrium. However, the functional significance of genistein induced changes in aromatase activity is unknown. A previous study has demonstrated that genistein treatment increased cell prolifer- ation and was weak ly estrogenic in endometrial stromal cell and Ishikawa cell cultures [34]. However, genistein treatment antagonized the effects of estradiol in these cultures suggesting that genistein is a com- petitive antagonist of estradiol. Therefore, a genistein induced increase in aromatase activity and local estrogen production in the endometrium could be relevant in hypoes- trogenic states such as menopause. While genistein treatment was without effect on the endometrium of macaque monkeys with surgically induced menopause [35], our proposal is supported by the observa- tion that endometrial hyperplasia was sig- nificantly more prevalent in postmenopausal women receiving soy tablets vs. a reference group that received a placebo [36]. Moreo- ver, a recent study [37] has also shown that high dose phytoestrogens can reverse the antiestrogenic effects of clomiphene citrate on the endometrium. Hence, we propose that the effects of soy isoflavones, including genistein on the endometrium is complex and requires further study. The mechanism through which genistein treatment increased aromatase activity in the endometrium remains unknown. Although estradiol has been shown to increase aro- matase activity in ESC cultures [38], sev- eral distinct lines of evidence lead us to suggest that genistein is not acting through an estrogen receptor mediated pathway in our cultures to increase aromatase activity. Genistein is a preferential estrogen receptor (ER)-β agonist and has been shown to have estrogenic actions in a variety of tissues in the rat [39, 40]. However, ER-α , not ER-β is the dominant ER sub-type expressed in the endometrium [41]. Furthermore, it is unlikely that genistein causes stimulation of aromatase in the endometrium by acting through a functional estrogen receptor path- way because we have shown that the stim- ulation of aromatase in ESC by genistein is not attenuated by co-t reatment of the cells with ICI 182,780 which is a non-selective estrogen receptor antagonist. We therefore propose that it is unlikely that genistein stimulates aromatase activity in endome- trial stromal cell cultures by acting directly via the ER. Alternatively, we propose that genistein can stimul ate aromatase activity in the endometrium through inhibition of phosphodiesterase activity and result in increased levels of cAMP. Support for this proposal comes from evidence that genis- tein inhibits cAMP-phosphodiesterase activity in a variety of cell types [42–44]. In addition, aromatase expression in the endometrium is regulated through cAMP- induced promoter II [45] and cAMP-treat- ment has previously been shown to result in a 26–60 fold increase in endometrial aro- matase activity [46]. Soy products are widely believed by the public to provide health benefits. The Food and Drug Administration has released an approval for foods that contain at least 6.25 g of soy protein/serving to contain a cardiovascular health claim (November 10, 1999; No. 279) and this has led to a plethora of soy-based and fortified foods as well as soy supplements to emerge on to the market [47]. Phytoestrogens are efficiently absorbed after ingestion and their bioavailability is high enough to have biological effects [48]. Furthermore, contemporary studies reveal that non-Asian women are ingesting increas- ing amounts of phytoestrogens in their diet as part of a trend towards a healthier lifestyle [49, 50]. Despite potential health benefits for women of some age groups, we speculate 718 K.M. Edmunds et al. that genistein consumption by women of reproductive age may have associated health risks. Moreover, epidemiological evidence demonstrates that Oriental women have a higher incidence of endometriosis than Caucasian women suggesting a link between endometriosis and dietary phytoestrogens, as Asian diets are high in soy isoflavones [51, 52]. Hence, consumption of soy prod- ucts by women of reproductive age may not be without conseque nce for endometrial aromatase activity and potentially endome- triosis. In summary, the resu lts of this study demonstrate that dietary compounds, such as genistein which is present in foods including soy milk and tofu that the general public views as healthy alternatives to tra- ditional foods in the North American diet, can increase the local production of estro- gen in the ESC. Genistein-induced changes in endometrial aromatase activity may have detrimental effects which could lead to increased risk for estrogen-dependent dis- eases which involve the dysregulation of aromatase such as endometriosis, adenom- yosis and uterine leiomyomas.

The authors gratefully acknowledge the tech- nical assistance of Ma ry Louise Beecroft, RN Scott Phillips and Vas ko (Bill) Georgievski. This study was supported by funding from the Canadian Network of Toxicology Centres (WGF) and the Canadian Institutes of Health Research (MOP 62681; WGF and ACH).

[1] Maggiolini M, Carpino A, Bonofiglio D, Pezzi V, Rago V, Marsico S, Picard D, Ando S. The direct proliferative stimulus of dehy- droepiandrosterone on MCF7 breast cancer cells is potentiated by overexpression of aro- matase. Mol Cell Endocrinol 2001, 184: 163– 171. [2] Sasano H, Harada N. Intratumoral aromatase in human breast, endometrial, and ovarian malignancies. Endocr Rev 1998, 19: 593–607. [3] Noble LS, Simpson ER, Johns A, Bulun SE. Aromatase expression in endometriosis. J Clin Endocrinol Metab 1996, 81: 174–179. [4] Kitawaki J, Noguchi T, Amatsu T, Maeda K, Tsukamoto K, Yamamoto T, Fushiki S, Osawa Y, Honjo H. Expression of aromatase cytochrome P450 protein and messenger ribonucleic acid in human endometriotic and adenomyotic tissues but not in normal endometrium. Biol Reprod 1997, 57: 514–519. [5] Kitawaki J, Kusuki I, Koshiba H, Tsukamoto K, Fushiki S, Honjo H. Detection of aromatase cytochrome P-450 in endometrial biopsy specimens as a diagnostic test for endometri- osis. Fertil Steril 1999, 72: 1100–1106. [6] Takayama K, Zeitoun K, Gunby RT, Sasano H, Carr BR, Bulun SE. Treatment of severe post menopausal endometriosis with an aro- matase inhibitor. Fertil Steril 1998, 69: 709–713. [7] Dizerega GS, Barber DL, Hodgen GD. Endometriosis: role of ovarian steroids in ini- tiation, maintenance, and suppression. Fertil Steril 1980, 33: 649–653. [8] Kitawaki J, Kado N, Ishihara H, Koshiba H, Kitaoka Y, Honjo H. Endometriosis: the pathophysiology as an estrogen-dependent disease. J Steroid Biochem Mol Biol 2003, 83: 149–155. [9] Rice VA. Conventional medical therapies for endometriosis. Ann N Y Acad Sci 2002, 955: 343–352. [10] Ajossa S, Mais V, Guerrierero S, Paoletti AM, Caffiero A, Murgia C, Melis GB. The preva- lence of endometriosis in premenopausal women undergoing gynecological surgery. Clin Exp Obstet Gynecol 1994, 21: 195–197. [11] Hilakivi-Clarke L, Ch o E, De Assis S, Olivo S, Ealley E, Bouker KB, Welch JN, Kan G, Clarke R, Cabanes A. Maternal and prepuber- tal diet, mammary development and breast cancer risk. J Nutr 2001, 131: 154S–157S. [12] Katadre M, Osborne M, Telang NT. Soy iso- flavone genistein modulates cell cycle pro- gression and induces apoptosis in HER-2/neu oncogene expressing human breast epithelial cells. Int J Oncol 2002, 21: 809–815. [13] Hale G, Bievre M, Hughes C. Exploring the role of progestins and phytoestrogens in men- opause. Integr Med. 1999, 2: 133–141. [14] Vincent A, Fitzpatrick LA. Soy isoflavones: are they useful in menopause? Mayo Clin Proc 2000, 75: 1174–1184. [15] Campbell DR, Kurzer MS. Flavonoid inhibi- tion of aromatase enzyme activity in human preadipocytes. J Steroid Biochem Mol Biol 1993, 46: 381–388. Genistein induced increased aromatase activity 719 [16] Adlercreutz H, Bannwart C, Wahala K, Makela T, Brunow G, Hase T, Arosemena PJ, Kellis Jr. JT, Vickery LE. Inhibition of human aromatase by mammalian lignans and isofla- vonoid phytoestrogens. J Steroid Biochem Mol Biol 1993, 44: 147–153. [17] Celec P, Ostatnikova D, Caganova M, Zuchova S, Hodosy J, Putz Z, Bernadic M, Kudela M. Endocrine and cognitive effects of short-time soybean consumption in women. Gynecol Obstet Invest 2004, 59: 62–66. [18] Tham DM, Gardner CD, Haskell WL. Poten- tial health benefits of dietary phytoestrogens: a review of the clinical, epidemiological, and mechanistic evidence. J Clin Endocrinol Metab 1998, 83: 2223–2235. [19] Kao Y, Zhou C, Sherman M, Laughton CA, Chen S. Molecular basis of the inhibition of human aromatase (estrogen synthetase) by flavone and isoflavones phytoestrogens: a site-directed mutagenesis study. Environ Health Perspect 1998, 106: 85–92. [20] Jeong HJ, Shin YG, Kim IH, Pezzuto JM. Inhibition of aromatase activity by flavonoids. Arch Pharm Res 1999, 22: 309–312. [21] Simpson ER, Zhao Y, Agarwal VR, Michael MD, Bulun SE, Hinshelwood MM, Graham- Lorence S, Sun T, Fisher CR, Qin K, Mendel- son CR. Aromatase Expression in Health and Disease. Recent Prog Horm Res 1997, 52: 185–213. [22] Stresser DM, Turner SD, McNamara J, Stocker P, Miller VP, Crespi CL, Patten CJ. A high-throughput screen to identify inhibi- tors of aromatase (C YP19). Anal Biochem 2000, 284: 427–430. [23] Cheng Y, Prusoff WH. Relationship between the inhibition constant (K I) and the concentra- tion of inhibitor which causes 50 per cent inhi- bition (I50) of an enzymatic reaction. Biochem Pharmacol 1973, 22: 3099–3108. [24] Arnold JT, Kaufman DG, Seppälä M, Lessey BA. Endometrial stromal cells regulate epi- thelial cell growth in vitro: a new co-culture model. Hum Reprod 2001, 16: 836–845. [25] Garzo VG, Dorrington JH. Aromatase activity in human granulosa cells during follicular development and the modulation by follicle- stimulating hormone and insulin. Am J Obstet Gynecol 1984, 148: 657–662. [26] Brodie AM, Shwarzel WC, Shaikh AA, Brodie HJ. The effect of an aromatase inhib- itor, 4-hydroxy-4-androstene-3,17-dione, on estrogen-dependent processes in reproduction and breast cancer. Endocrinology 1977, 100: 1684–1695. [27] Valentin-Blasini L, Blount BC, Caudill SP, Needham LL. Urinary and serum concentra- tions of seven phytoestrogens in a human ref- erence population subset. J Exp Anal Environ Epidemiol 2003, 13: 276–282. [28] Fanti P, Stephenson TJ, Kaariainen IM, Rezkalla B, Tsukam oto Y, Morishita T, Nomura M, Kitiyakara C, Custer LJ, Franke AA. Serum isoflavones and soya food intake in Japanese, Thai and American end-stage renal disease patients on chronic haemodialy- sis. Nephrol Dial Transplant 2003, 18: 1862– 1868. [29] Grace PB, Taylor JI, Botting NP, Fryatt T, Oldfield MF, Al-Mah arik N, Bingham SA. Quantification of isoflavones and lignans in serum using isotope di lution liquid chroma- tography/tandem mass spectrometry. Rapid Commun Mass Spectrom 2003, 17: 1350–1357. [30] Taga S, Yoshida N, Sekiba K. Distribution and cyclic change of aromatase cytochrome P- 450 activity in human uteri. J Steroid Biochem Mol Biol 1990, 37: 741–745. [31] Sanderson JT, Seinen W, Giesy JP, van den Berg M. 2-chloro-s-triazine herbicides induce aromatase (CYP19) activity in H295R human adrenocortical carcinoma cells: A novel mechanism for estrogenicity. Toxicol Sci 2000, 54: 121–127. [32] Sanderson JT, Boerma J, Lansbergen GW, den Berg M. Induction and inhibition of aro- matase (CYP19) activity by various classes of pesticides in H295R human adrenocortical carcinoma cells. Toxicol Appl Pharmacol 2002, 182: 44–54. [33] Sanderson JT, Hordijk J, Denison MS, Springsteel MF, Nantz MH, van den Berg M. Induction and inhibition of aromatase (CYP19) activity by natural and synthetic fla- vonoid compounds in H295R human adreno- cortical carcinoma cells. Toxicol Sci 2004, 82: 70–79. [34] Kayisli UA, Aksu CAH, Berkkanoglu M, Arici A. Estrogenici ty of isoflavones on human endometrial stromal and glandular cells. J Clin Endocrinol Metab 2002, 87: 5539–5544. [35] Foth D, Cline JM. Effects of mammalian and plant estrogens on mammary glands and uteri of macaques. Am J Clin Nutr 1998, 68: 1413S–1417S. [36] Unfer V, Casini ML, Costabile L, Mignosa M, Gerli S, Di Renzo GC. Endometrial effects of long-term treatment with phytoestrogens a randomized, double-blind, placebo-control- led study. Fertil Steril 2004, 82: 145–148. [37] Unfer V, Casini ML, Costabile L, Mignosa M, Gerli S, Di Renzo GC. High dose of phytoes- trogens can reverse the antiestrogenic effects of clomiphene citrate on the endometrium in 720 K.M. Edmunds et al. To access this journal online: www.edpsciences.org patients undergoing intrauterine insemina- tion: A randomized trial. J Soc Gynecol Inves- tig 2004, 11: 322–328. [38] Bulun SE, Yang S, Fang Z, Gurates B, Tamura M, Sebastian S. Estrogen production and metabolism in endometriosis. Ann N Y Acad Sci 2002, 955: 75–85. [39] Santell RC, Chang YC, Nair MG, Helferich WG. Dietary genistein exerts estrogenic effects upon the uterus, mammary gland and the hypothalamic/pituitary axis in rats. J Nutr 1997, 127: 263–269. [40] Diel P, Smolnikar K, Schulz T, Laudenbach- Leschowski U, Michna H, Vollmer G. Phy- toestrogens and carcinogenesis-differential effects of genistein in experimental models of normal and malignant rat endometrium. Hum Reprod 2001, 16: 997–1006. [41] Weihua Z, Saji S, Makinen S, Cheng G, Jenson EV, Warner M, Gustafsson J. Estrogen receptor (ER) β, a modulator of ER α in the uterus. PNAS 2000, 97: 5936–5941. [42] Kuppusamy UR, Das NP. Effects of flavo- noids on cyclic AMP phosphodiesterase and lipid mobilization in rat adipocytes. Biochem Pharmacol 1992, 44: 1307–1315. [43] Nichols MR, Morimoto BH. Tyrosine kinase- independent inhibition of cyclic-AMP phos- phodiesterase by genistein and tyrphostin 51. Arch Biochem Biophys 1999, 366: 224–230. [44] Burvall KM, Palmberg L, Larsson K. The tyrosine kinase inhibitor genistein increases basal cAMP and poten tiates forskolin- induced cAMP accumulation in A549 human airway epithelial cells. Mol Cell Biochem 2002, 240: 131–133. [45] Sofi M, Young MJ, Papamakarios T, Simpson ER, Clyne CD. Role of CRE-binding protein (CREB) in aromatase expression in breast adi- pose. Breast Cancer Res Treat 2003, 79: 399– 407. [46] Noble LS, Takayama K, Zeitoun KM, Putman JM, Johns DA, Hinshelwood MM, Agarwal VR, Zhao Y, Carr BR, Bulun SE. Prostaglan- din E 2 stimulates aromatase expression in endometriosis-derived stromal cells. J Clin Endocrinol Metab 1997, 82: 600–606. [47] Setchell KD, Brown NM, Desai P, Zimmer- Nechemias L, Wolfe BE, Brashear WT, Kir- schner AS, Cassidy A, Heubi JE. Bioavaila- bility of pure isoflavones in healthy humans and analysis of commercial soy isoflavone supplements. J Nutr 2001, 131 (Suppl): 1362S–1375S. [48] Ross JA, Kasum CM. Dietary flavonoids: bio- availability, metabolic effects, and safety. Annu Rev Nutr 2002, 22: 19–34. [49] Goodman MT, Wilkens LR, Hankin JH, Lyu LC, Wu AH, Kolonel LN. Association of soy and fiber consumption with the risk of endometrial cancer. Am J Epidemiol 1997, 146: 294–306. [50] Guthrie JR, Ball M, Murkies A, Dennerstein L. Dietary phytoestrogen intake in mid-life Australian-born women: relationship to health variables. Climacteric 2000, 3: 254–261. [51] Miyazawa K. Incidence of endometriosis among Japanese women. Obstet Gynecol 1976, 48: 407–409. [52] Arumugam K, Templeton AA. Endometriosis and race. Aust N Z J Obstet Gynaecol 1992, 32: 164–165.