Effects of Soy Isoflavones on Estrogen and Phytoestrogen Metabolism in Premenopausal Women’

Vol. 7. /101-1/08, December 1998 Cancer Epidemiology. Biomarkers & Prevention I/UI

Xia Xu, Alison M. Duncan, Barbara E. Merz, and Introduction Mindy S. Kurzer2 A role for estrogens in the development of breast cancer has Department of Food Science and Nutrition. University of Minnesota. St. Paul. been known for more than a century, since Beatson demon- Minnesota 55108 strated that oophorectomy induced tumor remissions in human

breast cancer ( 1 ). This role of estrogen in human breast carci- nogenesisis is supported by observations of estrogen-related Abstract risk factors, such as high serum and urine estrogen bevels (2. 3), Isoflavones and lignans are soy phytoestrogens that have postmenopausal obesity, early onset of menstruation, and late been suggested to be anticarcinogenic. The mechanisms menopause (4). It has been proposed that the total lifetime by which they exert cancer-preventive effects may involve exposure of a woman to estrogen is a determinant of her breast modulation of estrogen synthesis and metabolism. To cancer risk. evaluate this hypothesis, a randomized, cross-over soy In addition to the major endogenous estrogens, E23 and isoflavone feeding study was performed in 12 healthy specific estrogen metabobites may also influence breast cancer premenopausab women. The study consisted of three diet risk. E, and E1 are metabolized via three major pathways: the periods, each separated by a washout of -3 weeks. Each first pathway involves 16a-hydroxylation to form l6a-(OH)E diet period lasted for three menstrual cycles plus 9 days and E3; the second pathway leads to formation of 2-(OH)E2 (averaging -100 days), during which subjects consumed and 2-(OH)E1; and the third pathway forms 4-(OH)E, and their habitual diets supplemented with soy protein 4-(OH)E1 (Fig. I ). Both 2-hydroxybated and 4-hydroxylated E2 powder providing 0.16 (control diet), 1.01, or 2.01 mg of and E1 are catechol estrogens, which are further metabolized by total isoflavones per kg of body weight per day (10 ± 1.1, methylation (5, 6). Although the mechanisms by which estro- 65 ± 9.4, or 129 ± 16 mg/day, respectively). A 72-h urine gen increases breast carcinogenesis are not entirely clear, sub- sample was collected during the midfollicular phase (days stantiab evidence indicates that the key mechanisms are: (a) 7-9) of the fourth menstrual cycle in each diet period. mitogenic properties of the parent estrogens and their metabo- Urine samples were analyzed for 10 phytoestrogens and lites through classical estrogen receptor-mediated processes; 15 endogenous estrogens and their metabolites by a and (b) metabolic activation of E1 and E2 to genotoxic metab- capillary gas chromatography-mass spectrometry method. olites such as l6a-(OH)E1, 4-(OH)E2, and 4-(OH)E1 (7-10). Urinary excretion of isoflavonoids and bignans The main estrogen metabolite that has been proposed to be significantly increased with increased isoflavone a risk factor for breast cancer is l6cs-(OH)E1. 16a-(OH)E1 has consumption. Compared with the control diet, increased been shown to exhibit genotoxicity through induction of un- isoflavone consumption decreased urinary excretion of scheduled DNA synthesis and stimulation of anchorage-inde- estradiol, estrone, estriol, and total estrogens, as well as pendent growth of mammary epithelial cells ( 1 1 ). 16a-(OH)E1 excretion of the hypothesized genotoxic estrogen also irreversibly binds the estrogen receptor, resulting in long- metabolites, 16a-hydroxyestrone, 4-hydroxyestrone, and lasting effects such as persistent hyperpnoliferation and 4-hydroxyestradiol. Of importance are the observations of up-regulated expression of the c-rnvc oncogene, even allen a significant increase in the 2-hydroxyestrone/16a- withdrawal (12). Furthermore, the extent of estrogen b6a- hydroxyestrone ratio and a decrease in the genotoxic/total hydroxylation is significantly greater in strains of mice that estrogens ratio. These data suggest that soy isoflavone express mammary tumor virus and show high incidence of consumption may exert cancer-preventive effects by mammary cancer, when compared with those strains that do not decreasing estrogen synthesis and altering metabolism express mammary tumor virus (13). In humans. the relative away from genotoxic metabolites toward inactive extent of estrogen metabolism via the I 6cs-hydroxylation path- metabolites. way is significantly increased in patients with breast cancer ( I 3-17). Recent studies suggest that 4-hydroxylated catechol estrogens may be as harmful as 16a-(OH)E because their electrophilic quinone products react with DNA to form depuninating Received 6/9/98: revised 10/1/98: accepted 10/7/98. adducts known to generate mutations that initiate cancer both in The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in vitro and in vivo (10). In vito treatment with 4-(OH)E2 has been accordance with 18 U.S.C. Section 1734 solely to indicate this fact. I This research was supported by NIH Grant CA-66016 and General Clinical Research Center Grant MOl-RROO400 from the National Center for Research Resources. The soy powders were kindly donated by Protein Technologies 3 The abbreviations used are: E,. estradiol: E1, estrone: I6a-(OH)E. Iflo-hy- Intemational (St. Louis, MO). droxyestrone: E5, estriol: 2-0HE2. 2-hydroxyestradiol: 2-)OH)E1, 2-hy- 2 To whom requests for reprints should be addressed, at the Department of Food droxyestrone: 4-(OH )E2, 4-hydnoxyestradiol: 4-(OH )E, . 4-hydroxyestrone: CYP, Science and Nutrition. University of Minnesota. I 334 Eckles Avenue. St. Paul. cytochrome P-450: low- and high-iso. low- and high-isotlavone, respectively: MN 55108. E-mail: [email protected]. ODMA. O-desmethylangolensin.

2-(OH)E2 2-MeOE2

H 4-MeOE2 4-(OH)E2 cH3o E2 HO 17-epiE3

HcV©IZ9 16a-(OH)E1 E3 16-ketoE2

HO E1 2-(OH)E1 2-MeOE1 16-epiE3

HO99 HO 4-(OH)E1 cH3o 4-MeOE1

Fig. I. Main pathways of estrogen metabolism.

shown to induce DNA single-strand breaks and other mutagenic significantly increases urinary excretion of 2-(OH)E, and products of oxidative damage in liver and kidney of Syrian 2-(OH)E1, both of which have been proposed to be benign and hamsters ( 1 8, 19). These effects are of particular significance weak estrogens in men and women (33). A low-fat, high-fiber because 4-(OH)E, is known to be a potent long-acting estrogen diet decreases plasma E1 sulfate concentrations (34) and urinary (20-22). Consistent with these observations are in vitro studies excretion of I 6a-hydroxylated estrogen metabolites and in- showing that microsomes prepared from human mammary ad- creases urinary excretion of 2-hydroxylated estrogens in enocarcinoma and fibroadenoma predominantly catalyze the women (35). 4-hydroxylation of E,, although this does not occur in micro- In addition to the traditional low-fat, high-fiber diet, soy- somes prepared from normal tissue (23). In the only human bean consumption has been suggested to contribute to the low study to report excretion of these metabolites, Aldercreutz et a!. incidence and mortality of breast cancer in East and Southeast (24) reported that the average amount of 24-h urinary 4-(OH)E1 Asia, also via effects on endogenous estrogens and estrogen in premenopausal Finnish women at high risk of breast cancer metabolism (36). Epidemiological studies have shown an in- was at least double that in premenopausal Asian women at low verse association between soybean consumption and risk of risk of breast cancer. breast cancer (37-40). Ingram et a!. (41) recently reported a Because only -5% of all breast cancer cases can be strong inverse association between the risk of both premeno- attributed to genetic predisposition, it has been postulated that pausal and postmenopausal breast cancer and urinary excretion diet may exert cancer-preventive effects through beneficial of specific isoflavonoids and lignans, phytoestrogens that are effects on endogenous estrogen concentrations and metabolism present in soy. Although numerous nonhormonal properties of (25-28). Vegetarian diets result in anovulation (29), lowered soy phytoestrogens have been reported, they have also been urinary (30) and plasma (29, 31, 32) estrogen levels and in- shown to inhibit key enzymes of estrogen synthesis (42-45) creased fecal (3 1 ) estrogens. Consumption of indole-3-carbinol, and specific CYP isoenzymes responsible for producing geno- which is abundant in cruciferous vegetables, significantly de- toxic estrogen metabolites (46, 47). Data in humans are incon- creases urinary excretion of E,, E, E3, and 16a-(OH)E1 and sistent, as there have been reports of both reduced (48, 49) and

Table I Subject characteristics” (10 ± 1 . 1, 65 ± 9.4, and 129 ± 16 mg of total isoflavones pen day for control, low-iso, and high-iso diets, respectively). The Prestudy Control Low-iso diet High-iso diet concentrations of all 12 isoflavone isomers (the aglycone, glu- Age (yr) 26.0 ± 4.9 coside, acetylglucoside, and malonylglucoside forms of daid-

Body weight (kg) 63.8 ± 7.6 65.1 ± 8.4 64.1 ± 8.1 64.3 ± 8.2 zein, genistein, and glycitein) were analyzed in the labonatory Body mass index 22.5 ± 1.9 23.0 ± 2.2 22.6 ± 2.1 22.7 ± 2.0 of Professor Pat Murphy at Iowa State University by a reversed- (kg/rn2) phase high-performance liquid chromatography method, as de- Body fat (%) 28.2 ± 5.2 28.4 ± 5.4 28.1 ± 5.4 27.2 ± 4.5 scnibed previously (53). On average, the proportions of daid- ‘, Values are means ± SD (ii 12). zein, genistein, and glycitein were 37, 55. and 8%, respectively, of the total isoflavones in the soy protein powders. Ninety- seven % of daidzein, 97% of genistein, and 91% of glycitein increased (50) serum E, concentrations in premenopausal were present as their glucoside conjugates, and the remainder women and reduced (5 1) serum E2 concentrations in postm- were present as the aglycones. The composition of the soy enopausal women after soy consumption. Petrakis et a!. (52) protein powders has been described previously (54). found that prolonged consumption of soy protein isolate caused Subjects were free living throughout the entire study. They breast cell hyperplasia, suggesting an estrogenic effect of soy were instructed to minimize phytoestrogen ingestion by avoid- on breast cells in premenopausal women. At this time, there ing whole grains, oilseeds such as flaxseed (rich in lignans). and have been no reports of the effect of soy isoflavone consump- legumes, including soybean foods and foods containing textur- tion on estrogen metabolism. For this study, we postulated that ized vegetable protein, hydrolyzed vegetable protein, or soy soy isoflavone consumption in women will decrease estrogen protein isolate. Food intake was monitored by two 3-day diet synthesis and modulate estrogen metabolism away from for- records kept during each menstrual cycle, one during the fol- mation of potentially carcinogenic metabolites. To evaluate this licular phase and one during the luteal phase. Energy, macro- hypothesis, a randomized, cross-over soy isoflavone feeding nutrient, and dietary fiber intakes were analyzed by a comput- study was performed in 12 healthy premenopausal women. erized nutrient analysis program (Nutritionist IV, Version 4.0: The Hearst Corporation, San Bruno, CA). Body density was Materials and Methods calculated from the sum of the four-skinfold thickness and a predictive equation was used to determine percentage body fat Subjects. This study was a substudy of a larger project performed to evaluate the effects of soy isoflavone consumption on (55). reproductive hormones in women. Fourteen subjects partici- Sample Collection and Analysis. Three continuous 24-h pated in the larger study. All subjects were recruited from the urine samples were collected during the midfolbicular phase Minneapolis-St. Paul metropolitan area. Exclusionary criteria (days 7-9) of the fourth menstrual cycle of each diet period. included athleticism; vegetarian diet; regular consumption of a Twenty-four-h urines were collected in 3-liter containers con- high-fiber, high-soy, or low-fat diet; cigarette smoking; regular taming 3 g of ascorbic acid. Urinary creatinine was measured to consumption of vitamin and mineral supplementation greater evaluate the completeness of urine collection using an enzy- than the Recommended Dietary Allowances; current pregnancy matic assay kit (Johnson & Johnson Clinical Diagnostics Inc., or lactation; regular use of medication including aspirin; use of Rochester, NY). After recording the 24-h urine volume, sodium hormones or antibiotics within 6 months of the start of the azide was added to achieve a 0. 1% (w/v) concentration. Urine study; history of chronic disorders including endocrine or gy- samples were stored at -20#{176}C until analysis. Immediately necological diseases; benign breast disease; irregular menstrual before analysis, the three 24-h urine aliquots from menstrual cycles; <90% or >120% ideal body weight; weight change of cycle days 7-9 were thawed and proportionally combined to > 10 lb within the previous year or weight change of >5 lb create a 72-h pooled sample. within the previous 2 months; consumption of more than two Ten-ml aliquots in duplicate from each 72-h pooled urine alcoholic beverages per day; and a history of food allergies. sample were extracted and analyzed for 10 phytoestrogens Health status of the subjects was verified by health history, (equol, ODMA, dihydrodaidzein, daidzein, genistein, glycitein, physical exam, and routine blood and urine screening. enterodiol, enterolactone, matairesinol, and coumestnob) and 15 Data from two of the fourteen subjects recruited into the endogenous estrogens and their metabolites [Fig. 1; E1, E2, E3. larger study were eliminated from this substudy due to anovu- l6a-(OH)E , 2-(OH)E , 2-(OH)E,, 4-(OH)E1 , 4-(OH)E2, lation and incorrectly timed urine collections. Thus, 12 subjects 2-methoxyestrone, 2-methoxyestradiob, 4-methoxyestrone, participated in this substudy (Table 1 ). Their prestudy averages 4-methoxyestradiol, 16-ketoestradiol, 16-epiestriol. and 17- for age, body weight, body mass index, and percentage body fat epiestriol] by an ion-exchange chromatography and capillary were 26.0 ± 4.9 years, 63.8 ± 7.6 kg, 22.5 ± 1.9 kg/m2, and gas chromatography-mass spectrometry method originally de- 28.2 ± 5.2%, respectively. veboped by Adlercreutz and colleagues (56-58). Briefly, after Study Design and Diet. Prior to the study, the protocol was ethoximation to protect the carbonyl functions, phytoestnogens approved by the University of Minnesota Institutional Review and estrogen metabolites were extracted on Bond Elut C18 Board Human Subjects Committee. The study was performed columns (Chrom Tech, Apple Valley, MN). Seven deuterated using a randomized, cross-over design following the individual phytoestrogen internal standards (synthesized by Drs. T. Hase menstrual cycles of each subject. The study consisted of three and K. W#{228}h#{228}l#{228},Department of Chemistry. University of Hel- diet periods, each separated by a washout period of -3 weeks. sinki, Helsinki, Finland; Ref. 58) and nine deuterated estrogen Each diet period lasted for three menstrual cycles plus 9 days, internal standards (Ref. 57; C/D/N Isotopes Inc., Pointe-Claire, during which subjects consumed their habitual diets supple- Quebec, Canada) were added to each sample for compounds in mented with a soy protein powder (Protein Technologies In- each of four analyte fractions described below. The aliquots ternational, St. Louis, MO), providing a daily dose of 0. 16 ± were subsequently hydrolyzed with He!ix potnatia enzyme ex- 0.01 (control), 1.01 ± 0.04 (low-iso diet), or 2.01 ± 0.03 tract (Sigma Chemical Co., St. Louis, MO). The isoflavonoid (high-iso diet) mg of total isoflavones per kg of body weight fraction was separated from the nest of the compounds on the

Table 2 Dietary intake”

Prestudy Control Low-iso diet High-iso diet

Length ofdiet periods (day)5 N/A 102 ± 22 99 ± 13 98 ± 12 Isotlavone (mg/day)” N/A 10 ± 1.1 65 ± 9.4 129 ± 16 Isotlavone (mg/kg/day)” N/A 0.16 ± 0.01 1.01 ± 0.04 2.01 ± 0.03 Energy’ kcal/day 2016 ± I29 2339 ± 82’ 2329 ± 80 2268 79t Mi/day 8.43 ± 0.54* 9.79 ± OW 9.74 ± 0.34t 949 ± 0.33k

Protein (g/day)’ 71.3 ± 59* 1 16.0 ± 43t I 5.0 ± 431 I 7.4 ± 4.2k Fat (5/day) 71.6 t 6.0 79.0 ± 3.3 76.2 ± 3.3 70.6 3.2 Carbohydrate (glday)’ 281 ± 17 302 ± 10 305 ± 10 298 ± 10 Dietary fiber (g/day)’ 12.2 ± 0.9* 8.9 ± 0.4k 9.2 ± 0.41 8.3 ± 0.4k

,‘ Prestudy diet data were based on one 3-day food record per subject: Control. low-iso. and high-iso data were based on seven 3-day food records per subject. a 12 subjects. N/A. not applicable. Values within rows with different symbols (* or t) were significantly different (P < 0.05).

I, Values are means ± SD.

‘ Least squares mean ± SE.

acetate form of QAE-Sephadex columns (Sigma). Estrogen ficient of variation of 5-6%. We, therefore, concluded that the metabobites with vicinab cis-hydroxyls (such as catechol estro- 24-h urine collections were complete and chose to present our gens) and lignan fractions were separated out on the borate and data as nmol of phytoestrogen per 24 h. When data were bicarbonate forms of QAE-Sephadex columns, respectively. analyzed relative to body weight (nmol of phytoestrogen per kg Neutral steroids were removed from the estrogens without of body weight per 24 h), to account for differences in isofla- vicinal cis-hydroxyls [such as E.,, E1, E3, and 16a-(OH)E1] by vone consumption due to body weight differences, or relative to the free base form of DEAE-Sephadex columns (Sigma). Tn- urinary creatinine (nmol of phytoestrogen pen mmob of creati- methylsilyb derivatives of the samples and standards were an- nine), the same results were obtained. alyzed by a Hewlett Packard (Wilmington, DE) 5890 and As a result of unequal variance, all phytoestrogen data 597 1A quadrupole gas chromatography-mass spectrometry in- were log-transformed before data analysis. Table 3 shows the strument operated with a Unix 59940A ChemStation and a HP geometric means and 95% confidence intervals for the urinary 7673 autosampler in the selective ion-monitoring mode. phytoestrogen data. Compared with the control diet, urinary All samples from each subject were analyzed in duplicate excretion of isoflavonoids (genistein, daidzein, dihydrodaid- in the same batch. Duplicate quality control urine samples from zein, and glycitein) and lignans (enterodiol and enterolactone) the midfollicular phase were also analyzed with each batch. For were significantly increased by isoflavone consumption in a the phytoestrogen analyses, intra-assay coefficients of variation dose-dependent manner (Table 3). Urinary isoflavone metabo- ranged from 0.4 to 6.9%, and interassay coefficients of varia- bites (ODMA and equol) were also significantly increased by tion ranged from 2.0 to 10. 1% . For the estrogen metabolite both the low-iso and high-iso diets, despite the high variability analyses. intna-assay coefficients of variation ranged from 1.5 in excretion (Table 3). Equol variability was particularly high: to 6.5%, and interassay coefficients of variation ranged from 4 of the 12 subjects were equol-producers who excreted high 1.3 to 11.3%. amounts of equol, and the other 8 subjects were predominant Statistics. The effects of diet on urinary excretion of phy- ODMA-producers who excreted minimal amounts of equol. toestrogens and estrogen metabolites were determined by During all three diet periods, urinary excretions of coumestrol ANOVA (GLM) using Statistical Analysis System, Version and matairesinol were extremely low. Nevertheless, urinary 6.12 (SAS Institute, Inc., Cary, NC). Subjects and diet periods coumestrol was significantly increased by both the low-iso and were treated as blocks. Data were examined for homogeneity of high-iso diets; urinary matairesinol was significantly increased variance before ANOVA (GLM). If necessary, bog-transforma- by high-iso diet only (Table 3). tion of the data was performed before analysis. A P value of Urinary Estrogens and Estrogen Metabobites. Urinary ex- <0.05 was considered to be significant. cretion of E,, E3, 4-(OH)E,, 4-(OH)E1, 2-(OH)E2, 2-methoxyestrone, 16-ketoestradiol, and 16-epiestriol were signifi- Results cantly decreased by both the low-iso and high-iso diets, Diet and Body Weight. The diet and bod’ weight and com- although in most cases, there was no significant difference position results for the entire group of I 4 subjects were reported between these two diets (Table 4). Urinary excretion of total separately (54). Body weight, body mass index, and percentage estrogens, E1, 16a-(OH)E1, and 4-methoxyestrone were signif- body fat of the I 2 subjects in this substudy are presented in icantly decreased by the high-iso diet when compared with the Table I . The mean consumption of energy, macronutnients, and control diet, but they were not decreased by the low-iso diet dietary fiber (including the contribution from the daily soy (Table 4). Although in most cases, we were unable to statisti- powder) for the 1 2 subjects in this substudy are shown in Table cabby differentiate the low-iso diet from one on both of the other 2. There were no significant differences in body weight, body diets, a trend toward a dose response is apparent. Urinary

mass index, percentage body fat, or mean daily consumption of excretion of , 2-methoxyestradiol, 4-methoxyestra- energy, macnonutnients. or dietary fiber among the three diet diob and 17-epiestriol were not significantly affected by diet, periods, although prestudy energy. protein, and dietary fiber although a trend toward decreased excretion with isoflavone consumption were significantly different from those during the intake is apparent (Table 4). study (Tables I and 2). The ratio of genotoxic estrogen metabolites [l6a-(OH)E,, Urinary Phytoestrogens. Subjects’ urinary creatinine was 4-(OH)E,, and 4-(OH)E] to total estrogens was significantly quite consistent, as shown by an average within-subject coef- decreased by both the low- and high-iso diets (Table 5). The

Table 3 Urinary phytoestrogens (nmol/24 hY’

Control (,i = I 1 )h Low-iso diet (n = 1 1 )‘ High-iso diet (ii = 12)

Isoflavonoids Genistein 997 (807_l,232)* 6,529 (5.378_7,925)t 14,200 ( I 1.75 I-l7,I59)

Daidzein 995 (719-l,378) 4,9f4 (3,634-6.781 )‘ 9,528 (7,1 l()-l2,768) Dmhydrodaidzein 364 (249-532) 1,538 ( l.057_2,236)t 2.799 ( l,986-3.946) ODMA 1,155 (805_l,657)* 8,656(6,007_12,474)t 20.292 ( 14.65I_28.l06)t Equol 1 17 (62-222) 270 (l42_5I2)t 438 (246-780)’

Glycitemn 262 (205-334) 1.4 19 ( 1 , I 06- 1.822)’ 2,5 10 (2.0 I 4-3, I 28)0 Coumestrol 26.9 (22.3_32.6)* 38.6 (3l.946.8)t 36.8 (31.1-43.6)’ Lignans Enterodmol 171 (l43_204)* 251 (211-298)’ 296 (252-347)’ Enterolactone 1.132 (97l_l,320)* 1,874 ( 1,600-2,196)’ 2,782 (2.422_3,l96)0

Matairesinol 8.9 (7.1-1 1 .1 )* 10.9 (8.7-l3.6)’ I 3.9 ( I I .4-16.9)’ Total 6,438 (5,533_7,49l)* 31,132 (26,696-36,306) 64,666 (56,41874.121)

‘, Due to unequal variance. all data were log-transformed before data analysis. Geometric means (95% confidence intervals) are presented. When data were analyzed relative to urinary creatinine (nmol of phytoestrogens/mmol of creatinine) or relative to body weight (nmol of phytoestrogens/kg of body weight/24 h). the same results were obtained. Values within rows with different symbols (8, t, or ) were significantly different. (P < 0.05). b One subject did not collect urine samples at the end of the control diet period. ‘. One subject used antibiotics at the end of the low-iso diet period. and her data were not included.

Table 4 Urinary estrogens and estrogen metabolites (nmol/24 h)” Table 5 Urinary estrogen metabolite ratios”

Control Low-iso diet High-iso diet Control Low-iso diet High-iso diet

(n II)” (n 11)’ (n 12) (n 11)” (ii = II)’ (‘1 = 12)

E, 7.71 ± 0.31* 6.38 ± 0.31’ 5.90 ± 0.28” Genotoxic/total 0.10 ± 0.01 * 0.08 ± 0.01 0.08 ± 0.01’ E 17.49 ± 0.65 16.15 ± 0.64*” 14.95 ± 0.59’ 2-(OH)E1/l6a-(OH)E 18.00 ± 3.92k 31.37 ± 3.97’ 28.33 ± 3.58*t E3 1 1.29 ± 0.39* 9.80 ± o,39t 8.19 ± 0.36 2E-tota1/l6a-totaI” 2.91 ± 0.40 3.92 0.41 3.76 ± 0.37 l6a-(OH)E” 3.67 0.25 3.36 ± 0.19 2.84 ± 0.22 2.72 (2.34-3.l6) 335 (2.87-3.91)’ 3.49 (3.05-4(X))’ 2.67 (2.24-3.l9) 2.16 (l.82-2.56)’ 1.75 (1.49-2.06)’ 2E,-total/4E1-total 8.51 ± 0.79k I 1.91 ± 0.76’ 10.68 ± t).72” 4-(OH)E, 1.75 ± 0.I7 1.04 ± 0.l7t 0.92 ± 0.16’ 2E,-total/4E,-total 4.42 ± 0.69 5.42 ± 0.7 I 6.44 ± 0.63 4-(OH)E, 5.44 ± 0.51 3.87 ± 0.51 3.53 ± 0.46 2-total/4-total 7.17 ± 0.59* 10.17 ± 0.58’ 9.45 ± 0.54t 4.81 (4.03_5.74)* 3.46 (2.91-4.10)’ 3.39 (2.89-3.97)’ “ Least squares means ± SE. Genotoxic/total = (l6a-(OHE1 +4-(OH)E2+4- 2-(OH)E, 4.86 ± 0.18* 4.06 ± 0.19t 3.71 ± 0.17t (OH)E, )/total estrogens: 2E -totalllfla-total = + 2-MeOE, )/( l6a- 2-(OH)E1 36.05 ± 1.72 38.27 ± 1.74 34.32 ± 1.57 (OH)E1 +E3+ l7-epiE3): 2E1-total/4E1-total = (2-(OH)E, +2-MeOE1 )/(4-(OH)E1 4-MeOE2 0.14 ± 0.01 0.16 ± 0.02 0.12 ± 0.01 +4-MeOE1): 2E-totaV4E,-total = (2-(OH)E+ 2-MeOE,)/(4-(OH)E,+4-MeOE2:

4-MeOE1” 0.47 ± 0.08 0.30 ± 0.08 0.28 ± 0.07 2-total/4-total = (2-(OH)E + 2-(OH)E,+ 2-MeOE1 + 2-MeOE,)/(4-(OH)E1 +4- 0.40 (0.310.5l)* 0.29 (0.230.37)*t 0.27 (0.2l0.33)t (OH)E,+4-MeOE1 +4-MeOE). 2-MeOE1. 2-methoxyestrone: 2-MeOE2. 2-me- 2-MeOE, 2.45 ± 0.1 1 2.34 ± 0.1 1 2.15 ± 0.10 thoxyestradiol; 4-MeOE , 4-methoxyestrone: 4-MeOE. 4-methoxyestradiol. Values within rows with different symbols (8 or t) were significantly different (P < 0.05). 2-MeOE1” 7.77 ± 0.27 6.78 ± 0.27 6.23 ± 0.24 /‘ One subject did not collect urine samples at the end of the control diet period. 7.31 (6.85_7.80)* 6.48 (6.066.92)t 5.94 (5.6O-6.3l) ‘. One subject used antibiotics at the end of the low-iso diet period. and her data l6-ketoE2 2.00 ± 0.10’ 2.70 ± 0.1 1* 2.27 ± 0.1 1” were not included. ± ± ± l7-epiE 0.88 0.08 0.72 0.08 0.79 0.08 ,! Due to unequal variance. data were log-transformed before ANOVA (GLM). 16-epiE3 2.59 ± 0.16* 2.01 ± 0.16’ 2.02 ± O.lS Geometric means (95% confidence intervals) are presented below the least Total estrogens 105.26 ± 2.60k 97.50 ± 2.67k’ 87.93 ± 2.38” squares means.

‘, Least squares means ± SE. When data were analyzed as nmol of estrogen metabolite/mmol of urinary creatinine, the same results were obtained. 4-MeOE2, 4-methoxyestradiol: 4-MeOE1, 4-methoxyestrone; 2-MeOE2, 2-methoxyestradiol; 2-MeOE1, 2-methoxyestrone; l6-ketoE2, 16-ketoestradiol; l7-epiE3, 17- The ratios of 2E1-total to 4E1-total (which reflects the ratio epiestriol; 16-epiE3. 16-epiestriol. Values within rows with different symbols (*, of 2-hydroxylation to 4-hydroxylation of E1), and 2-total to t. or ) were significantly different (P < 0.05). 5 One subject did not collect urine samples at the end of the control diet period. 4-total (which reflects the ratio of 2-hydroxylation to 4-hy- ‘- One subject used antibiotics at the end of the low-iso diet period, and her data droxybation for both E1 and E2), were significantly increased by were not included. both the low- and high-iso diets (Table 5). There was also a ‘I Due to unequal variance, data were log-transformed before ANOVA (GLM). trend toward an increase in the ratio of 2E2-total to 4E,-total Geometric means (95% confidence intervals) are presented below the least squares means. (which reflects the ratio of 2-hydroxylation to 4-hydroxybation of E2) by the low- and high-iso diets, although this was not statistically significant. genotoxic estrogen metabolites made up -8% of the total estrogens in subjects consuming the high- or low-iso diets, compared with - 10% in subjects consuming the control diet. Discussion The ratio of 2-(OH)E1 to 16a-(OH)E1 was significantly in- As expected, increased urinary excretion of isoflavonoids creased in subjects who consumed the low-iso diet compared (daidzein, genistein, glycitein, dihydrodaidzein. ODMA, and with the control diet, although the high-iso diet was not statis- equol) was observed in our subjects after both low and high soy tically distinguishable from the other diets. The ratio of 2E1- isoflavone consumption for more than three menstrual cycles total to l6a-total (which reflects the ratio of 2-hydroxylation of (an average of - 100 days), in a clear dose-dependent manner. E1 to 16a-hydroxylation of E1; Ref. 59), was significantly These results are consistent with those reported from short-term increased by both the low- and high-iso diets (Table 5). (1- or 9-day) soy feeding studies (60-62), although the short-

term studies found daidzein to be the major urinary isofla- 4-hydroxybated catechol estrogens (70). It has been shown that vonoid, and we observed the daidzein metabobite ODMA to be a-naphthoflavone, a synthetic flavone and inhibitor ofCYPlAl

the predominant urinary isoflavonoid. This difference suggests and CYP1 B I , completely suppresses E,-induced tumorigenesis that the length of soy consumption may influence the activities through inhibition of the formation of 4-(OH)E, and its elec- of gut microfboral enzymes responsible for ODMA formation. trophilic quinone (70). At high concentrations, genistein also In addition to providing isoflavones, soy likely contains weakly inhibits CYPIA2 (46), an enzyme that catalyzes the small amounts of plant lignans such as secoisobaniciresinol and 2-hydroxybation of E, (71) and E2 (72) in human liver. Con- matairesinol, which can be converted to the mammalian big- sumption of the synthetic isoflavone, ipniflavone by rats, results nans. enterodiol, and entenolactone by human gut bacteria (63). in a strong suppression of liver CYP3A4 (47), the enzyme that Although the quantities are low, they are significant, as shown catalyzes the 4- and 16a-hydroxylation of estrogens in humans by significantly increased lignan excretion after soy consump- (71). tion. It is also possible that consumption of soy isoflavones This is the first study to report that soy isoflavone con- alters intestinal microflora toward increased bignan production. sumption in humans lowers urinary estrogen excretion and The between-subject variability in urinary phytoestrogen leads estrogen metabolism in a favorable direction by prefer- excretion within a diet was great, even after adjusting for entialby decreasing production of genotoxic 16a- and 4-hy- differences in consumption due to provision of isoflavones droxylated estrogen metabobites. This mechanism may contrib- relative to body weight. Within a diet, excretion of genistein, ute to the observed inverse associations between breast cancer daidzein, ODMA. and equol varied by 3-. 4-, 7-, and 800-fold. and soy consumption. Of additional significance is the possi- respectively. This high variability in phytoestrogen excretion bility that, in some studies of dietary effects on endogenous after soy consumption is consistent with previous reports (61, estrogen, urinary estrogen bevels may be more sensitive and 62, 64) and is likely due to the dependence of phytoestrogen significant end points than plasma concentrations of reproduc- formation and absorption on the composition of gut microflora tive hormones or functional end points such as the menstrual (65), which varies substantially among individuals (66). This cycle. variability was particularly high for equol. because only 4 of the I 2 subjects excreted significant quantities, consistent with pre- Acknowledgments vious reports that 30-40% of human subjects produce large We are grateful to Professor Herman Adlercreutz, Department of Clinical Chem- quantities of equol after soy consumption (67). istry. University of Helsinki, for his advice on our analytical work, and to Dr. Will Our data show that overall urinary estrogen excretion was Thomas. Division of Biostatistics, School of Public Health, and Dr. Gary Oehlert. significantly decreased by soy phytoestrogen consumption, as School of Statistics, University of Minnesota, for their advice on our data reflected by reduced excretion ofE,, E1, E3, and total estrogens. analysis. We would also like to thank the subjects and staff at the General Clinical Research Center, University of Minnesota, for their effort in carrying out this Because we did not observe significant effects on plasma es- human feeding study. trogen concentrations (54), reduced urinary estrogen excretion is likely due to decreased estrogen synthesis rather than decreased clearance. This explanation is consistent with in vitro References studies showing that soy phytoestrogens reduce estrogen syn- I . Beatson, G. On the treatment of inoperable cases of breast cancer by oophorectomy. Lancet, ii: 104-107, 1898. thesis through inhibition of key steroidogenic enzymes such as 2. Toniolo, P. G.. Levitz, M., Zeleniuch-Jacquotte, A., Banerjee. S., Koenig. K., aromatase (42-44), 3/3-hydroxysteroid dehydrogenase, and Shore, R. E., Strax, P., and Pastemack, B. S. A prospective study of endogenous I 7j3-hydroxysteroid dehydrogenase (45, 68, 69). estrogens and breast cancer in post-menopausal women. J. NatI. Cancer Inst. Excretion of the genotoxic estrogen metabolites l6a- (Bethesda), 87: 190-197, 1995. (OH)E, 4-(OH)E,, and 4-(OH)E1 was also significantly re- 3. Key. T. J. A.. Wang. D. Y.. Brown. I. B.. Hermon, C., Allen, D. S.. Moore, duced by soy phytoestrogen consumption. Evaluation of the I. W.. Bulbrook, R. D., Fentiman. I. S., and Pike, M. C. A prospective study of urinary oestrogen excretion and breast cancer risk. Br. J. Cancer, 73: 1615-1619, estrogen metabolite ratios suggests that, in addition to the 1996. absolute amounts, the relative amounts of both 16a-hydroxy- 4. Henderson, B. E., Ross, R.. and Bemstein, L. Estrogens as a cause of human lation and 4-hydnoxylation of estrogens (pathways producing cancer: the Richard and Hinda Rosenthal Foundation Award Lecture. Cancer the genotoxic estrogen metabolites) were significantly de- Res., 48: 246-253. 1988. creased after soy phytoestrogen consumption. At the same time, 5. BalI, P.. and Knuppen. R. Catecholestrogens: chemistry, biogenesis. metabo- the proportion of estrogen metabolized by 2-hydroxylation (a lism, occurrence and physiological significance. Acta Endocrinol., 93 (Suppl. 232): 1-127, 1980. pathway producing benign and weak estrogen metabolites) was 6. Mantucci, C. P., and Fishman, J. P450 enzymes of estrogen metabolism. significantly increased when compared with the proportion Pharmacol. Ther., 57: 237-257, 1993. metabolized by 16a-hydroxybation, although the absolute 7. Fishman, J.. Osbome. M. P., and Telang. N. T. The role of estrogen in amount of 2-(OH)E1, the most abundant estrogen metabolite mammary carcinogenesis. Ann. N. Y. Acad. Sci.. 768: 91-100. 1995. found in urine, was not significantly changed. These data sug- 8. Bradlow, L., Telang. N. T.. and Osbom, M. P. Estrogen metabolites as gest that it is possible to modulate production of estrogens and bioreactive modulators of tumor initiators and promoters. Biol. React. lntermed., their metabolites through consumption of soy phytoestrogens. V: 285-296, 1996. In addition, because we did not observe significant effects on 9. Yager. J. 0.. and Liehr, J. G. Molecular mechanisms of estrogen carcinogen- concentrations of plasma estrogens or the menstrual cycle (54), esis. Annu. Rev. Pharmacol. Toxicol., 36: 203-232, 1996. it is likely that urinary estrogen excretion is a more sensitive 10. Cavalieri. E. L., Stack. D. E.. Devanesan, P. D., Todorovic, R., Dwivedy, I., Higginbotham. S.. iohansson, S. L.. Patil, K. D.. Gross, M. L.. Gooden, J. K., indicator of the effects of isoflavones on endogenous estrogen. Ramanathan, R., Cemy. R. L., and Rogan, E. G. Molecular origin of cancer: Although few data are available. the likely mechanism by catechol estrogen-3,4-quinones as endogenous tumor initiators. Proc. NatI. Acad. which soy isoflavones modulate estrogen metabolism is via Sci. USA, 94: 10937-10942, 1997. effects on the activity of specific CYP isoenzymes responsible II. Telang. N. T.. Suto, A., Wong. G. Y., Osborne, M. P., and Bradlow, H. L. for estrogen hydroxylations. Genistein has been shown to in- Induction by estrogen metabolite l6a-hydroxyestrone of genotoxic damage and aberrant proliferation in mouse mammary epithelial cells. J. NatI. Cancer Inst. hibit rat liver CYP1AI (46), an enzyme that catalyzes the (Bethesda), 84: 634-638. 1992. conversion of catechol estrogens to their electrophilic quinones, I 2. Swaneck, G. E.. and Fishman. J. Covalent binding of the endogenous estro- compounds that may be responsible for the genotoxicity of gen lbcs-hydroxyestrone to estradiol receptor in human breast cancer cells:

characterization and intranuclear localization. Proc. NatI. Acad. Sci. USA. 85: 37. Nomura, A., Henderson, B. E., and Lee, J. Breast cancer and diet among the 7831-7835. 1988. Japanese in Hawaii. Am. I. Clin. Nutr.. 3/: 2020-2025. 1978. 13. Bradlow. H. L.. Hershcope, R. 3., Martucci, C. P., and Fishman, J. l6a- 38. Hirayama, T. A large-scale cohort study on cancer risks by diet. with special hydroxylation of estradiol: a possible risk marker for breast cancer. Ann. N. Y. reference to the risk reducing effects of green-yellow vegetable consumption. Acad. Sci., 464: 138-151, 1986. Princess Takamatsu Symp.. 160: 41-53. 1985. 14. Schneider, J., Bradlow, H. L., Strain, G., Levin, J., Anderson, K., and 39. Lee, H. P., Gourley. L.. Duffy. S. W., Esteve, I., Lee, I., and Day. N. E. Fishman, J. Effects of obesity on estradiol metabolism: decreased formation of Dietary effects on breast cancer risk in Singapore. Lancet. 337: I 197-12(X). 1991. non-uterotropic metabolites. J. Clin. Endocrinol. Metab., 56: 973-978, 1983. 40. Wu, A. H., Ziegler, R. G., Horn-Ross, P. L., Nomura, A. M., West, D. W.. 15. Fishman, J., Schneider, J., Hershcopf, R. J.. and Bradlow. H. L. Increased Kolonel, L. N.. Rosenthal, I. F., Hoover, R. N.. and Pike, M. C. Tofu and risk of estrogen- l6a-hydroxylase activity in women with breast and endometrial cancer. breast cancer in Asian-Americans. Cancer Epidemiol. Biomark. Prey., 5: 901- J. Steroid Biochem., 20: 1077-1081, 1984. 906, 1996. 16. Kabat, G. C., Chang, C. J.. Sparano, J. A., Sepkovic, D. W., Hu, X., Khalil, 41. Ingram, D.. Sanders, K., Kolybaba. M., and Lopez. D. Case-control study of A., Rosenblatt, R., and Bradlow, H. L. Urinary estrogen metabolites and breast phyto-oestrogens and breast cancer. Lancet. 350: 990-994. 1997. cancer: a case-control study. Cancer Epidemiol. Biomark. Prey., 6: 505-509, 42. Adlercreutz, H.. Bannwart, C., W#{228}hiilS,K., MSkeIS, T., Brunow. G., Hase, T.. I997. Arosemena, P. J., Kellis. J. T., Jr., and Vickery. E. L. Inhibition of human 17. Ursin. G.. London, S., Stanczyk. F. Z.. Gentzschein. E., Paganini-Hill. A., aromatase by mammalian lignans and isoflavonoid phytoestrogens. I. Steroid Ross, R. K.. and Pike, M. C. A pilot study of urinary estrogen metabolites Biochem. Mol. Biol., 44: 147-153, 1993. ( l6a-OHE1 and 2-OHE, ) in postmenopausal women with and without breast 43. Campbell. D. R., and Kurzer. M. S. Flavonoid inhibition of arornatase cancer. Environ. Health Perspect.. 105 (Suppl. 3): 601-605, 1997. enzyme activity in human preadipocytes. J. Steroid Biochem. Mo). Biol.. 46: 18. Han, X., and Liehr, J. G. DNA single strand breaks in kidneys of Syrian 381-388, 1993. hamsters treated with steroidal estrogens. Hormone-induced free radical damage 44. Wang. C.. MSkelS. T., Hase. T.. Adlercreutz, H.. and Kurzer. M. S. Lignans preceding renal malignancy. Carcinogenesis (Lond.), 15: 977-1000, 1994. and flavonoids inhibit aromatase enzyme in human preadipocytes. I. Steroid 19. Han, X.. and Liehr, J. G. 8-Hydroxylation of guanine bases in kidney and Biochem. Mol. Biol., 50: 205-212, 1994. liver DNA of hamsters treated with estradiol: role of free radicals in estrogen- 45. Keung, W-M. Dietary estrogenic isoflavones are potent inhibitors of -hy- induced carcinogenesis. Cancer Res., 54: 55 15-55 17, 1994. droxysteroid dehydrogenase of P. teswsteronii. Biochem. Biophy. Res. Corn- 20. Bamea, E. R., MacLusky, N. J., and Naftolin, F. Kinetics of catecholestro- mun., 3: 1137-1144, 1995. gen-estrogen receptor dissociation: a possible factor underlying differences in 46. Lee. H., Wang. H-W.. Su, H-Y., and Hao, N. I. The structure-activity catecholestrogen biological activity. Steroids, 41: 643-656, 1983. relationships of flavonoids as inhibitors of cytochrome P-450 enzymes in rat liver 21. van Aswegen. C. H.. Purdy. R. H., and Wittliff, J. L. Binding of 2-hy- microsomes and the mutagenicity of 2-amino-3-methyl-imidazol4,5-Jlquinoline. droxyestradiol and 4-hydroxyestradiol to estrogen receptors from human breast Mutagenesis. 9: 101-106, 1994. cancers. J. Steroid Biochem., 32: 485-492, 1989. 47. Monostory, K.. and Vereczkey. L. Interaction of theophylline and ipnflavone 22. Schutze. N., Vollmer, G., Tiemann, I., Geiger. M., and Knuppen, R. Cat- at the cytochrome P450 level. Eur. J. Drug Metab. Pharmacokinet.. 20: 43-47. echolestrogens are MCF-7 cell estrogen agonists. J. Steroid Biochern. Mol. Biol., I995. 46: 781-789. 1993. 48. Lu, L. J. W., Anderson, K. E., Grady. I. I.. and Nagamani. M. Effects of soya 23. Liehr. J. G.. and Ricci, M. J. 4-Hydroxylation of estrogens as marker of consumption for one month on steroid hormones in premenopausal women: human mammary tumors. Proc. NatI. Acad. Sci. USA, 93: 3294-3296, 1996. implications for breast cancer risk reduction. Cancer Epidemiol. Biomark. Prey.. 24. Aldercreutz, H.. Gorbach, S. L., Goldin, B. R., Woods, M. N., Dwyer. J. T.. 5: 63-70, 1995. and HSmSlSinen, E. Estrogen metabolism and excretion in Oriental and Caucasian 49. Nagata, C., Kabuto. M., Kurisu. Y.. and Shimizu. H. Decreased serum women. J. NatI. Cancer Inst. (Bethesda), 86: 1076-1082, 1994. estradiol concentration associated with high dietary intake of soy products in 25. Goldin, B. R.. Adlercreutz, H., Gorbach, S. L., Woods, M. N., Dwyer, J. T., premenopausal Japanese women. Nutr. Cancer. 29: 228-233. 1997. Conlon, T., Bohn, E.. and Gershoff, S. N. The relationship between estrogen 50. Cassidy. A.. Bingham. S.. and Setchell, K. D. R. Biological effects of a diet levels and diets of Caucasian American and Oriental immigrant women. Am. J. of soy protein rich in isoflavones on the menstrual cycle of premenopausal Clin. Nutr., 44: 945-953, 1986. women. Am. J. Clin. Nutr., 60: 333-340, 1994. 26. Adlercreutz, H. Diet and sex hormone metabolism. In: Rowland. I. ted.). 51. Baird. D. D., Umbach. D. M.. Lansdell, L., Hughes. C. L.. Setchell. K. D. R.. Nutrition, Toxicity and Cancer, pp. 137-195. Boca Raton. FL: CRC Press, 1991. Weinberg, C. R. Haney. A. F.. Wilcox, A. I.. and Mclachlan. I. A. Dietary 27. Bernstein. L.. and Ross, R. K. Endogenous hormones and breast cancer risk. intervention study to assess estrogenicity of dietary soy among postmenopausal Epidemiol. Rev., 15: 48-65, 1993. women. J. Clin. Endocrinol. Metab., 80: 1685-1690, 1995. 28. Ziegler, R. G., Hoover, R. N., Pike, M. C. Hildesheim, A., Nomura, A. M., 52. Petrakis, N. L.. Barnes, S., King. E. B.. Lowenstein. I.. Wiencke. I.. Lee. West, D. W.. Wu-Williams, A. H., Kolonel, L. N., Horn-Ross, P. L., Rosenthal, M. M.. Miike, R.. and Kirk, M., and Coward, L. Stimulatory influence of soy J. F.. and Hyer, M. B. Migration patterns and breast cancer risk in Asian- protein isolate on breast secretion in pre- and postmenopausal women. Cancer American women. J. NatI. Cancer Inst. (Bethesda), 85: 1819-1827, 1993. Epidemiol. Biomarkers Prey., 5: 785-794. 1996. 29. Pirke, K. M., Schweiger, U.. Laessle, R., Dickhaut, B., Schweiger. M., and 53. Wang, H-J., and Murphy, P. A. Isoflavone content in commercial soybean Waechtler, M. Dieting influences the menstrual cycle: vegetarian i’S. nonvegetar- foods. J. Agric. Food Chem., 42: 1666-1673, 1994. ian diet. Fertil. Steril., 46: 1083-1088, 1986. 54. Duncan, A. M., Mer,o, B. E., Xu, X.. Nagel. T. C.. Phipps. W. R.. and Kurzer. 30. Armstrong. B. K.. Brown, i. B.. Clarke, H. T., Crooke, D. K., Hahnel, R., M. S. Soy isotlavones exert modest hormonal effects in premenopausal women. Masarei, J. R.. and Ratajczak. T. Diet and reproductive hormones: a study of J. Clin. Endocrinol. Metab.. in press. 1999. vegetarian and nonvegetarian postmenopausal women. J. NatI. Cancer Inst. (Be- 55. Dumin, J., and Wornersley, I. Body fat assessed from total body density and thesda), 67: 761-767, 1981. its estimation from skinfold thickness: measurement on 481 men and women aged 31 . Goldin. B. R., Adlercreutz, H., Gorbach, S. L., Warram, J. H., Dwyer. J. T., 16 to 72 years. Br. J. Nutr.. 32: 77-97, 1974. Swenson, L., and Woods, M. N. Estrogen excretion patterns and plasma levels in 56. Fotsis. T., and Adlercreutz, H. The multicomponent analysis of estrogens in vegetarian and omnivorous women. N. EngI. J. Med., 307: 1542-1547, 1982. urine by ion exchange chromatography and GC-MS-I. Quantitation of estrogens 32. Schultz. T. D., and Leklem, J. E. Nutrient intake and hormonal status of after initial hydrolysis of conjugates. J. Steroid Biochem.. 28: 203-213, 1987. premenopausal vegetarian Seventh-Day Adventists and premenopausal nonveg- 57. Bannwart, C., Adlercreutz, H., Whh#{228}l#{228},K.,Brunow, G.. and Hase, T. Deu- etanians. Nutr, Cancer, 4: 247-259, 1983. terium labelled ethoximes as stable isotope internal standards in the GC-MS-SIM 33. Michnovitz, J. J., Adlercreutz, H., and Bradlow, H. L. Changes in levels of determination of oxo-estrogens in human urine extracts: preliminary results. In: urinary estrogen metabolites after oral indole-3-carbinol treatment in humans. S. Gorog and E. Heftmann (eds.), Advances in Steroid Analysis. pp. 283-286. J. NatI. Cancer Inst. (Bethesda), 89: 718-723, 1997. Budapest: Akademiai Kiado, 1988. 34. Woods, M. N.. Gorbach, S. L., Longcope. C., Goldin, B. R., Dwyer. J. T., and 58. Adlercreutz, H., Fotsis, T., Bannwart. C., WahhIS, K.. Brunow. G.. and Hase, Momll-LaBrode, A. Low-fat, high-fiber diet and serum estrone sulfate in pre- T. Isotope dilution gas chromatographic-mass spectrometric method for the menopausal women. Am. J. Clin. Nun’., 49: 1 179-1183, 1989. determination of lignans and isoflavonoids in human urine. including identifica- 35. Longcope. C.. Gorbach. S.. Goldin. B.. Woods, M., Dwyer. J., MomIl. A., tion of genistein. Clin. Chim. Acta, 199: 263-278. 1991. and Warram, J. The effect of a low fat diet on estrogen metabolism. J. Clin. 59. Ziegler. R. G., Rossi, S. C.. Thomas. R. F., Bradlow, H. L., Adlercreutz, H.. Endocrinol. Metab.. 64: 1246-1250, 1987. Sepkovic. D., Kiuru, P.. WahSIa. K.. Vaught, J. B.. Donaldson. J. L.. Falk. R. T.. 36. Adlercreutz, H., and Mazur, M. Phyto-oestrogens and Western diseases. Ann. Fillmore, C., Siiteri, P. K., Hoover, R. N., and Gail, M. H. Quantifying estrogen Med., 29: 95-120, 1997. metabolism: an evaluation of the reproducibility and validity of enzyme immu-

noassays for 2-hydroxyestrone and l6a-hydroxyestrone in urine. Environ. Health 66. Mallett, A. K., and Rowland, I. R. Factors affecting the gut microflora. In: Perspect.. 105 (Suppl. 3): 607-614, 1997. I. R. Rowland (ed), Role of the Gut Flora in Toxicity and Cancer, pp. 347-382. 60. Xu, X. Wang. H. I.. Murphy. P. A. Cook. L.. and Hendrich, S. Daidzein is London: Academic Press Limited, 1988. a more bioavailable soymilk isoflavone than is genistein in adult women. J. Nutr., 67. Kelly, G. E., Nelson, C., Waring, M. A., Joannou, G. E., and Reeder, A. Y. /24: 825-832, 1994. Metabolites of dietary (soya) isoflavones in human urine. Clin. Chim. Acta, 223: 61. Xu, X., Harris, K. S., Wang. H. J., Murphy, P. A., and Hendnich, S. 9-22, 1993. Bioavailability of soybean isoflavones depends upon gut microflora in women. I. 68. M#{228}kelS,5,, Poutanen, M., LehtimSki, J., Kostian, M-L., Santti, R., and Nutr., 125: 2307-2315. 1995. Vihko, R. Estrogen-specific 17(3-hydroxysteroid oxidoreductase type I (E.C. 62. Karr, S. C.. Lampe. J. W.. Hutchins, A. M.. and Slavin, J. L. Urinary 1 . I . 1 .62) as a possible target for the action of phytoestrogens. Proc. Soc. Exp. isoflavonoid excretion in humans is dose dependent at low to moderate levels of Biol. Med., 208: 51-59, 1995. soy-protein consumption. Am. J. Clin. Nutr., 66: 46-51. 1997. 69. Makela, 5, Poutanen, M., Kostian, M. L., Lehtimaki, J., Strauss, L., Saniti, R., 63. Setchell, K. D. R., and Adlercreutz, H. Mammalian lignans and phytoestro- and Vihko, R. Inhibition of l7-hydroxysteroid oxidoreductase by flavonoids in gens. Recent studies on their formation, metabolism and biological role in health breast and prostate cancer cells. Proc. Soc. Exp. Biol. Med., 217: 310-316, 1998. and disease. In: Rowland, I. (ed). Role of the Gut Flora in Toxicity and Cancer. 70. Liehr, J. G. Catecholestrogens in the induction of tumors in the kidney of the pp. 315-345. London: Academic Press, 1988. Syrian hamster. Adv. Pharmacol., 42: 824-828. 1998. 64. Kelly. G. E.. Joannou, G. E., Reeder. A. Y., Nelson, C., and Waring. M. A. 71. Shou, M., Korzekwa, K. R., Brooks, E. N., Krausz, K. W., Gonzalez, F. J., The variable metabolic response to dietary isoflavones in humans. Proc. Soc. Exp. and Gelboin, H. V. Role of human hepatic cytochrome P450 lA2 and 3A4 in the Biol. Med.. 208: 40-43, 1995. metabolic activation of estrone. Carcinogenesis (Lond.). 18: 207-214. 1997. 65. Kurzer, M. S., and Xu, X. Dietary phytoestrogens. In: R. E. Olson. D. M. 72. Aoyama, T., Korzekwa, K., Nagata, K., Gillette, J., Gelboin, H. V., and Bier. and D. B. MaCormick (eds.), Annual Review of Nutrition, Vol. 17, pp. Gonzalez, F. J. Estradiol metabolism by complementary deoxyribonucleic acid- 353-381. Palo Alto: Annual Review, Inc.. 1997. expressed human cytochrome P450s. Endocrinology. 126: 3101-3106, 1990.

Downloaded from cebp.aacrjournals.org on September 26, 2021. © 1998 American Association for Cancer Research. Effects of soy isoflavones on estrogen and phytoestrogen metabolism in premenopausal women.

X Xu, A M Duncan, B E Merz, et al.

Cancer Epidemiol Biomarkers Prev 1998;7:1101-1108.

Updated version Access the most recent version of this article at: http://cebp.aacrjournals.org/content/7/12/1101

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/7/12/1101. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.