Dietary Soy Intake and Urinary Isoflavone Excretion Among Women from a Multiethnic Population'

Vol. 7, 613-6/9, July 1998 Cancer Epidemiology, Biomarkers & Prevention 6/3

Dietary Soy Intake and Urinary Isoflavone Excretion among Women from a Multiethnic Population’

Gertraud Maskarinec,2 Seema Singh, Lixin Meng, and incidence for ethnic groups in the United States (per 100,000 Adrian A. Franke women, age-adjusted to the 1970 United States population) was

Cancer Research Center of Hawaii, Honolulu, Hawaii 968 I 3 as follows: Caucasians, 1 1 1 .8; Native Hawaiians, 105.6; Japa- nese, 82.3; Chinese, 55; and Filipino, 73. 1 (6). These variations have been attributed to differences in dietary fat and fiber intake Abstract and possibly to soy consumption among Asian women (7-9). Isoflavones are present in soybeans and its products in The interest in dietary soy has arisen from findings related to the potential cancer-protective properties of isoflavones ( I 0, concentrations up to 300 mg/100 g, have estrogenic and antiestrogenic properties, and may be protective against 1 1 ). Isoflavones are plant products with estrogenic activity and are therefore phytoestrogens, which include a wide variety of hormone-related cancers. The purpose of this cross- phytochemicals such as coumestans and lignans (12, 1 3). The sectional study was to investigate the association between term phytoestrogen was coined after the observation that Aus- urinary isoflavone excretion and self-reported soy intake. A total of 102 women of Caucasian, Native Hawaiian, tralian sheep grazing on a certain type of subterranean clover had a high rate of infertility (14), an observation later attributed Chinese, Japanese, and Filipino ancestry completed a to the high isoflavone content of this plant. The isoflavones dietary questionnaire for soy products consumed during have a heterocyclic phenol structure that closely resembles the last year and during the 24-h period before urine collection. Overnight urine samples were analyzed for estrogens and enables them to bind to estrogen receptors and to exert weak estrogenic effects ( I 3, 15-1 7). They may act as coumestrol and the soy isoflavones genistein, daidzein, antiestrogens by competing with endogenous estrogen for re- and glycitein and their main human metabolites by ceptor binding, thereby possibly reducing the risk for breast reverse-phase high-pressure liquid chromatography. Soy cancer by decreasing the promotional effects of high levels of protein and isoflavone intake (predominantly from tofu) endogenous estrogens ( 10, 18, 19) or by altering estrogen were estimated using published nutritional databases. biosynthesis (20, 21). Alternative mechanisms suggested for Wilcoxon’s rank-sum test scores and Spearman rank isoflavones to prevent cancer and other chronic conditions are correlation coefficients were computed. modification of enzyme activity, such as topoisomerase or Japanese women excreted more daidzein, genistein, tyrosine protein kinases (22) that play a role in cell proliferation and glycitein than did Caucasian women, whereas and transformation, as well as breast cancer oncogene expres- Caucasian women excreted slightly more coumestrol. Soy sion (23), inhibition of angiogenesis (24), and apoptosis (25). intake differed significantly among ethnic groups. Dietary Soybeans contain high amounts (concentrations of up to soy protein and isoflavone intakes during the previous 100-300 mg/lOO g) of the glycosides of the isoflavones daid- 24 h were positively related to urinary isoflavone zein, genistein, and glycitein (26-29). Daidzein present in soy excretion [r5 = 0.61 (P < 0.0001) and 0.62 (P < 0.0001), foods is partially converted by the gut flora into equol and respectively]. Urinary excretion of isoflavones was also DMA3 as end products (30, 31). The other common dietary related to annual dietary soy protein and isoflavone isoflavone, genistein, is converted to p-ethylphenol (3 1). The intake [r = 0.32 (P < 0.0012) and 0.31 (P < 0.0016), uncertain metabolism of soy isoflavones led to the identifica- respectively]. The strong correlation between urinary tion of several minor intermediate metabolites (32). The isofla- isoflavone excretion and self-reported soy intake validates vones and their metabolites can be measured in food, plasma, the dietary history questionnaire that is now used in a urine, and feces using gas chromatography and HPLC (27, study exploring dietary risk factors for breast cancer. 33-41). Studies among various populations and dietary groups Introduction have shown large amounts of isoflavonoids in the plasma, The low risk for breast and prostate cancer in Asian populations urine, and feces of vegetarians; macrobiotics; and in Japanese combined with evidence from migration studies suggests that individuals consuming a traditional Japanese diet (37, 38). environmental factors such as diet may influence the occur- Whereas substantial variations among individuals in the excre- rence of these diseases (1-5). From 1988 to 1992, breast cancer tion of isoflavones and their metabolites have been described (39), information on ethnic differences in soy intake and the excretion of urinary isoflavones is limited (42). Hawaii is an

Received 2/17/98; revised 4/13/98; accepted 4/22/98. excellent location to explore the possible effects of soy product The costs of publication of this article were defrayed in part by the payment of consumption, due to its multiethnic population with a high page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‘ Supported by Department of the Army Grant I 7-96- 1 -6284 and NIH Grant 5-K12 CA 01708. 2 To whom requests for reprints should be addressed, at Cancer Research Center , The abbreviations used are: DMA. O-desmethylangolensin; 1-IPLC, high- of Hawaii. 1236 Lauhala Street, Honolulu, HI 96813. pressure liquid chromatography; id., inside diameter.

proportion of women of Asian ancestry who have maintained to three 25-ml disposable plastic tubes and stored at -70#{176}C some of their traditional dietary habits. The objectives of this until analyzed. cross-sectional study were to explore the variation in dietary Urinary Analysis. Urine samples were analyzed by diode a- soy intake and urinary isoflavonoid excretion among women of ray reverse-phase HPLC for soy isoflavones, their most corn- Japanese, Filipino, Native Hawaiian, Chinese, and Caucasian mon mammalian metabolites, and coumestrol (29). In brief, ancestry and to investigate the association between dietary soy frozen urine samples were thawed, vortex-mixed, and centri- intake and urinary isoflavone excretion. fuged at 850 x g for 5 mm. A 2.0-ml clear supernatant was mixed with 0.4 ml of 0.5 M triethyl-amine acetate buffer (pH Materials and Methods 7.0) and 20 p1 of flavone (120 ppm in 96% ethanol) as an internal standard. The mixture was incubated with 10 pA of Study Population. Hawaii has a multiethnic population with j3-glucuronidase (Boehringer # 127680; 200 units/ml, 0.1 approximately 24% Caucasian, 20% Japanese, 1 1% Filipino, tmoW5 pA) and 10 sl of arylsulfatase (Boehringer # 102890; 5 19% Native Hawaiian, 5% Chinese, and 21% other ethnic groups (43). Participating women were recruited from two large units/ml) for 1 h at 37#{176}Cin the dark followed by repeated extraction with diethylether. The combined organic phases mammography clinics in Honolulu. The institutional review were dried under nitrogen and redissolved in 100 .d of meth- boards at all participating organizations approved the study anol by sonication for 10-20 s and in 100 l of 0.2 M sodium protocols. Women interested in participating received dietary acetate buffer (pH 4.0). After centrifugation, 20 p1 of the clear questionnaires and consent forms with addressed return enve- solution were injected into the HPLC system. The internal lopes by mail. In appreciation for their cooperation, women standard isoflavone was analyzed in the same way in the same were sent a summary of their dietary intake of various macro- batch as urine extracts for internal standard recovery calculation and micronutrients at the end of the study. Mammograms for all participating women were obtained from the clinics after the purposes. HPLC analyses were carried out on a NovaPak C 18 (150 X 3.9 mm, inside diameter; 4 pm) reverse-phase column radiological evaluation was completed. Women with a history (Waters, Milford, MA) coupled to an Adsorbosphere Cl8 of breast cancer, a previous breast biopsy or surgical procedure, (10 X 4.6 mm, inside diameter; 5 m) direct connect guard or a mammogram with a suspicious lesion requiring biopsy column (Alltech, Deerfield, IL). Elution was performed at a were excluded from the study. Antibiotic use or other metabolic flow rate of 0.8 ml/min with the following step gradient: (a) conditions that might affect isoflavone uptake or metabolism were not assessed. 20% acetonitrile in acetic acid-water (10:90, v/v) for 16 mm; (b) 70% acetonitrile for 14 mm; and (c) 20% acetonitrile for 10 Study participants with mixed ancestry were classified mm. Analytes were monitored with a dual-channel diode array using both parents’ ancestry and the following rules. If a per- son’s ethnicity is recorded as a combination of Native Hawaiian detector at 260 nm during the entire HPLC run, at 280 nm during equal elution, and at 342 nm during coumestrol elution. and any other ethnicity, the summary code is Native Hawaiian. Observed peaks were scanned between 190 and 400 nm. These Otherwise, the summary code is the first stated non-Caucasian ethnicity. Although all Native Hawaiian women were of mixed phytoestrogens were identified by comparing retention times and UV absorption patterns with authentic standards analyzed ancestry, the majority (84%) of Asian women reported one in the same batch and with reported UV data (46). Concentra- ancestry only. tions of analytes in the urine were calculated with area units Dietary Assessment. A self-administered questionnaire was obtained from HPLC analyses and the slope of the calibration used to obtain demographic and reproductive information and a curve. The excretion rates are expressed in nanomoles/hour family history for each participant. The dietary history ques- after consideration of urine volume and adjustment for the time tionnaire collected the usual frequency of consumption and between last urine collection and previous void and internal portion size for more than 50 food items, alcoholic and nonal- standard recovery. Detection limits were 0.2 nmollh for genis- coholic beverages, and vitamin supplements during the previ- tein, 0.4 nmollh for daidzein and glycitein, 0.9 nmol/h for ous year. A separate questionnaire listing 12 soy-based foods equol, 1 .4 nmol/h for DMA, and 0.5 nmol/h for cournestrol. On with local names was used to collect information on soy con- 10 randomly selected urine samples, a repeat analysis was sumption. We used United States Department of Agriculture performed to test reliability of measurements. Food Composition tables (44) to compute soy protein intake and used isoflavone data from food analyses to estimate isofla- Statistical Analysis. After examining each variable separately vone intake (45). for the normality of the distribution, we computed Spearman rank correlation coefficients (47) to evaluate the relationship between Urine Collection. An overnight urine sample was collected for urinary isoflavone excretion and dietary intake of isoflavones and 102 study participants, and a 24-h dietary questionnaire for soy protein consumption during the previous 24 h and during the soy-based foods was obtained at the same time. Of 129 women previous yea. Because of the non-normality of these distributions, eligible for urine collection, 5 women refused to provide a we calculated K-statistics (47) to verify the correlations. A mean sample, 2 women were menstruating, 8 women were on vaca- coefficient of variation was computed for the 10 urine samples tion, 3 women lived too far away, and 9 women could not be with repeat measurements by dividing the ratio of the SD and the reached after repeated calling, resulting in a participation rate of mean of each repeat sample by 10 (47). Nonpaametric analyses 78%. A container with 0.2 g of ascorbic acid and 0.3 g of boric (47) were performed to evaluate differences among all ethnic acid (to prevent bacterial contamination and degradation of groups (Kruskal-Wallis test) and between two ethnic groups (Wil- analytes) was delivered to each woman. Women were in- coxon scores). Statistical analysis was performed using the SAS structed to collect the first urine sample in the morning and all statistical software package version 6.12 (SAS Institute, Inc., Cay, of the samples if they urinated during the night after going to NC). bed. They were asked to record the times of last urination before going to bed and last urine collection in the morning. The samples were stored in the participants’ refrigerators until Results pick-up in the morning and transported to the laboratory on ice. The study population included 42 Caucasian women, 25 Japa- After mixing and weighing, each urine sample was transferred nese women, 13 Chinese women, 1 1 Native Hawaiian women,

Table 1 Mean soy protein intak e and urinary i soflavonoid excretio n among 102 wo men in Hawaii b y ethnicity”

...... Kruskal-Wallts test Chinese Filiptno Native Hawaiian Japanese Caucasian Others

No. 13 7 II 25 42 4 NA” Age(yr) 55(13) 47(9) 50(4) 50(10) 53(11) 45(6) 4.7(0.45) Soy protein intake during the previous 13.3 ( 18.6) 1.9 (3.4) 8.5 ( 15.5) 9. 1 ( I I .9) 2.8 (6.6) 7.9 (7.7) IS. I (0.01) 24 hours (g/day) Soy protein intake during the previous 4.8(4.4) 1.5 (1.8) 4.6 (3.7) 6.1 (6.8) 1.8 (2.7) 6.6 (4.4) 26.7 (0.00()I) yea (g/day) Dietary isoflavone intake during the 38.2 (56.9) 5.0 (8.8) 22.2 (40.5) 31.3 (43.7) 6.9 ( l7.6 17.7 ( 19.8 16.0 (OJX)7) previous 24 hours (mg/day) Dietary isoflavone intake during the I I .9 ( I I .0) 5.2 (7.5) I 2. 1 ( I 2.4) 18.9 (27.0) 5.2 (8.6) 1 6.8 ( I I .5) 24.3 (0.0()02) previous year (mg/day) Urinary isoflavone excretion (nmol/h) 307.6 (458.7) 77.6 (I 13.7) 293.7 (689.9) 724.7 (956.7) 138.9 (298.6) 81.1 ( 134.5) 18.1 (0.0()3) Daidzein (nmollh) 187 (305) 44 (66) 179 (428) 442 (661 ) 79 ( 157) 42 (65) 16.2 (0.006) Genistein(nmol/h) 65(105) 12(17) 72(162) 134(186) 28(62) 22(35) 18.2(0.003) Glycitein(nmol/h) 24(34) 4(8) 20(49) 88(186) 14(36) 8(17) 18.7(0.002) DMA (nmollh) II (29) 14(34) 16(32) 58(146) 16(65) 0(0) 4.0(0.55) Equol (nmol/h) 20(55) 0(0) 7 (23) 1.8 (9.2) 2 (10) 9 (17) 4.7 (0.45) Coumestrol(nmol/h) 0(0) 3(9) 0(0) 0.1(0.5) 13(70) 0(0) 3.2(0.66)

“ Unless otherwise indicated, mean values are shown. followed by SD in parentheses. 5 NA, not applicable.

7 Filipino women, and 4 women of other ethnicities (2 African- women (x2 12.6; P = 0.0004) and by Native Hawaiian American women, 1 Indonesian woman, and I Vietnamese women (x2 6.5; P = 0.01 1 ). Ethnic differences in the woman). The age of the study population ranged from 36-80 excretion of DMA, equol, and coumestrol were not statistically yeas, with a mean age of 53 years. The coefficient of variation significant. The ratio of urinary isoflavone excretion to soy for the repeat measurements of urinary isoflavone excretion intake during the previous 24 h was approximately twice as was 9.01 %, indicating high reliability of the laboratory proce- high for Japanese women as it was for all other women (Fig. I). dure. Urinary isoflavone excretion showed a strong correlation of The mean urinary excretion for isoflavones was more than 0.61 with dietary soy protein intake during the previous 24 h four times higher in Japanese women than it was in Caucasian (Table 2). A weaker but still significant correlation of 0.32 was women (Table I; Fig. I). Urinary isoflavone excretion in Chi- obtained between urinary isoflavone excretion and self-re- nese and Native Hawaiian women was more than twice as high ported dietary soy protein intake during the previous year. as that in Caucasian women. Filipino women and the women of Because dietary soy protein was predominantly obtained from other ethnicities had the lowest urinary isoflavone excretion tofu and isoflavone intake was estimated using concentrations rate. Whereas only 1 Japanese woman did not excrete any in food, the correlation between soy protein and isoflavone isoflavones, 16 (38%) Caucasian women, 3 (27%) Native Ha- intake was greater than 0.9 (Table 2). Excluding observations waiian women, I (25%) woman of other ethnicity, 2 (15%) with nondetectable isoflavone levels did not affect the size of Chinese women, and 1 (14%) Filipino woman did not excrete the correlation coefficients. The nonparametric sc-statistic be- any isoflavones. For all ethnic groups, daidzein was excreted at tween urinary isoflavone excretion and soy intake during the the highest rate, followed by genistein and glycitein (Table 1; previous 24 h and during the previous year was 0.46 (95% Fig. 2). The excretion of urinary isoflavone metabolites and confidence interval, 0.32-0.59) and 0.23 (95% confidence in- coumestrol showed great variation among ethnic groups (Table terval, 0.06-0.42), respectively, confirming the results from I ; Fig. 2). Equol was excreted at the highest rate by Chinese the correlation analysis. women, whereas all other women excreted equol at a very low rate. Coumestrol was excreted at higher rates by Caucasian women than it was by any other group. Discussion The mean dietary intake of soy protein during the previous In this cross-sectional study, we examined urinary isoflavonoid year was three times higher for Japanese women than it was for excretion and soy intake in 102 healthy women from different Caucasian women (Table 1). Soy consumption for Chinese and ethnic groups. Japanese, Chinese, and Native Hawaiian women Native Hawaiian women was more than twice as high as that in were found to consume higher amounts of soy protein than Caucasian women. Filipino women had the lowest consumption were Caucasian and Filipino women. Likewise, women with of soy protein during the previous year. A similar pattern of soy Japanese, Chinese, or Native Hawaiian ancestry excreted more protein consumption was obtained for intake during the previ- isoflavones in urine than did Caucasian and Filipino women. ous 24 h, except that Chinese women reported the highest soy All of the ethnic groups investigated excreted daidzein at the intake (Table 1). highest rate, followed by genistein. This is in good agreement Assessment by the Kruskal-Wallis test revealed significant with previous reports on urinary isoflavone excretion (39, 40, differences among ethnic groups for dietary soy and isoflavone 48-50). Glycitein was excreted at very low levels, and great intake during the previous yea and the previous 24 h (Table 1) variation was observed in the excretion of isoflavone metabo- as well as significant differences for urinary excretion of daid- lites in all groups. Urinary isoflavone excretion was signifi- zein, genistein, and glycitein. The pairwise comparison be- cantly related to soy protein intake during the previous 24 h and tween various ethnic groups revealed a significantly higher during the previous year as measured by a food frequency excretion of daidzein by Japanese women than by Caucasian questionnaire.

14 . 800

700 12

i600 .;; 10 .5001

Fig. I. Mean soy protein intake and un- 400 nary isoflavonoid excretion among 102 multiethnic women. 0, soy protein intake during the previous 24 h (g/day); . soy E .3O0 protein intake during the previous yea (g/ day); U, urinary isoflavones (nanomoles/ 0 04 hour).

.- 200 0

2 100

0 Chinese Filipino Native Japanese Caucasian Others

Hawaiian

Ethnicity

Thus far, only a study from California (42) has reported ethnic groups was obtained in our study (Fig. 1 ). This varia- urinary isoflavone excretion by ethnic group. In that study with bility may be due to the nature of gastrointestinal absorption of 50 subjects, no statistically significant difference among van- soy isoflavones and metabolism by intestinal bacteria (34, 52) ous ethnic groups for isoflavone excretion was found, but the such as Lactobacilli, Bacteroides, Bifidobacterium, and Cbs- small group of Japanese women (n 5) excreted slightly more tridia (53). Variations in dietary habits can lead to differences isoflavones than did other ethnic groups. Similar to our results, in the gut flora (54). The intestinal bacteria vary widely be- Japanese women excreted very low levels of coumestrol corn- tween individuals, and only selected strains are capable of pared with Caucasian women. The comparison between the hydrolyzing plant 3-glucosides, such as isoflavones occurring Californian study (42) and our study is limited by the differ- in unfermented soy foods, and performing further flavonoid ences in the ethnic groups (Caucasian, Latina, African Amer- metabolism by ring fission (55, 56). Evidence for that hypoth- ican, and Japanese versus Caucasian, Chinese, Filipino, Japa- esis was provided in a well-controlled liquid diet study (52) in nese, and Native Hawaiian) and the fact that 24-h urine samples which seven women consumed three soy milk meals/day. Re- were collected in California (42). The advantage of our inves- covery of daidzein was significantly greater than that of genis- tigation was a larger sample size (102 women) from a multi- tein (P < 0.01 ) in women excreting small amounts of fecal ethnic population in which women of Asian ancestry have isoflavones. Anaerobic incubation of isoflavones with human maintained some of their traditional dietary habits. feces showed that the intestinal half-life of daidzein and genis- Other studies with small convenience samples have exam- tein is 7.5 and 3.3 h, respectively. Lower absorption levels of med urinary phytoestrogen excretion for various populations, isoflavones may explain our finding that Japanese, Chinese, related them to different diets (37, 38), and described a great and Native Hawaiian women had relatively high intakes of soy variation in urinary isoflavones and their metabolites. Urinary protein, but when compared to Japanese women, excretion of isoflavone excretion increased as much as 1000-fold (39) when urinary isoflavones was low for both Chinese and Native Ha- soy was added to a typical Western diet. Similarly, urinary waiian women (Fig. 1). Variations in the time when soy foods excretion of isoflavonoids was directly related to dose when I 1 were consumed could also be responsible for this observation. men and 9 women consumed soy protein in a controlled cx- For this study, only one overnight urine sample was col- perirnental diet trial (5 1) or when randomly added to the usual lected for easier implementation and higher compliance, result- diet (45). Soy doses (grams of beans) correlated excellently ing in a participation rate of 78%. Franke and Custer (39) with the observed isoflavonoid amounts excreted in urine dur- recommended a combination of three urine samples collected ing the first 24 h when healthy subjects were challenged with every other night for each subject. This would allow integration different amounts of roasted soybeans (39). of an individual’s phytoestrogen exposure over approximately Inter- and intraindividual differences were reported in a I week, a period that generally reflects usual dietary habits. study of six women and five men who consumed 44-96 g of Multiple 24-h urine samples would be a more valid measure- roasted soybeans (39). All individuals excreted genistein and ment of isoflavonoid excreted in urine than one overnight urine daidzein, but some did not excrete any or excreted only low sample but would make participation more difficult and costly. levels of other isoflavonoid metabolites. A similar variation in Another limitation of our study was the poor representa- the excretion of metabolites and coumestrol among various tion of some ethnic groups. The representation of Native Ha-

3500 1000 3000 800 2500 2000 600 . N :#{176}1500 B 0 400 1000 ! 8 200 500 #{176} #{176} 8 : 0 0 l 0 8 8 o 8

C F H J 0 W C F H J 0 W

Ethnicity Ethnicity 700 200

600 150 500

400 . _ 100 Fig. 2. Urinary excretion of ,,, coumestrol, isoflavones, and their ‘ 300 V metabolites (nanomoles/hour). C, ( 200 8 ,, Chinese; F, Filipino; H, Native Ha-

waiian; J, Japanese; 0, others; W, 100 #{176} Caucasian. 0 0 : 0

C F H J 0 W C F H J 0 W

Ethnicity Ethnicity

500 700

600 0 400

500

- 300 400 0 n 200 ; 300 0 0 (_) 200 100 100 0 0 0 0 #{176} 0 0 0 0 0 0 0 6 0 8

‘ I I I I I t I I t C F H J 0 W C F H J 0 Ethnicity Ethnicity

Table 2 Correlations between urinary isoflavone excretion and dietary intake of soy protein and isoflavones

p First variable Second variable r”

Urinary isoflavone excretion (nmol/h) Soy protein intake during previous year (g/day) 0.32 0.0012 Urinary isoflavone excretion (nmol/h) Soy protein intake during previous 24 h (g/day) 0.61 0.0001 Urinary isoflavone excretion (nmollh) Isoflavone intake during previous yea (mg/day) 0.3 1 0.0016 Urinary isoflavone excretion (nmol/h) Isoflavone intake during previous 24 h (mg/day) 0.62 0.0001 Soy protein intake during previous year (g/day) Soy protein intake during previous 24 h (g/day) 0.53 0.0001 Isoflavone intake during previous year (mg/day) Isoflavone intake during previous 24 h (mg/day) 0.55 0.0001 Soy protein intake during previous 24 h (g/day) Isoflavone intake during previous 24 h (mg/day) 0.96 0.0001

“ Spearman rank correlation coefficient.

waiian women in our study was only 1 1 % as compared with 15. Martin. P. M., Horwitz, K. B., Ruyan, D. S., and McGuire, W. L. Phytoestro- 19% in the population, and the representation of Filipino gen interaction with estrogen receptors in human breast cancer cells. Endocri- nology, 103: 1860-1867, 1978. women in our study was 6% as compared with 1 1% in the 16. Shutt, D. A.. and Cox, R. I. Steroid and phytoestrogen binding to sheep population. Caucasian and Japanese women were overrepre- uterine receptors in vitro. J. Endocrinol., 52: 299-310, 1972. sented, probably because of differential insurance coverage and 17. Folman, Y., and Pope, G. S. The interaction in the immature of potent mammography utilization. Language was a barrier for a few oestrogens with coumestrol, genistein and other utero-vaginotrophic compounds women who recently immigrated. of low potency. J. Endocrinol.. 34: 215-225, 1966. This study is one of the first to investigate soy intake and I 8. Folman, Y., and Pope, G. S. Effects of norethisterone acetate, dimethylstil- isoflavone excretion with a larger sample size of healthy boestrol, genistein and coumestrol on uptake of [3H]oestradiol by uterus, vagina and skeletal muscle of immature mice. Endocrinology. 44: 213-218, 1969. women of various ethnic origins. Dietary soy protein and isofla- 19. Makela, S., Pylkkanen, L., Santti, R., and Adlercreutz, H. Role of plant vone intake based on self-reporting for the previous 24 h and estrogens in normal and estrogen-related altered growth of the mouse prostate. In: for the previous year were positively related to the urinary Euro Food Tox III, pp. 135-139. Schwerzenbach, Switzerland: Institute of Tox- excretion of isoflavones. The results of this study suggest icology of Swiss Federal Institute of Technology and University of Zanch. 1991. differences in urinary excretion of isoflavones by ethnic group 20. Makela, S., Poutanen, M., Lehtimaki, J., Kostian, M. L., SantO, R., and that are related to self-reported dietary soy intake and to dif- Vihko, R. Estrogen-specific l7(3-hydroxysteroid oxidoreductase type I (E.C. 1 . I . 1 .62) as a possible target for the action of phytoestrogens. Proc. Soc. Exp. ferential intestinal absorption patterns. The strong correlation Biol. Med., 208: 51-59, 1995. between urinary isoflavone excretion and self-reported soy 21. Adlercreutz, H., Bannwart, C., Wahala, K., Makela, T., Brunow, G., Hase, T., intake validates the questionnaire that is now used in a study Arosemena, P. J., Kellis, J.. and Vickery, L. E. Inhibition of human aromatase by exploring dietary risk factors for breast cancer. mammalian lignans and isoflavonoid phytoestrogens. J. Steroid Biochem. Mol. Biol., 44: 147-153, 1993.

22. Markovits, J., Linassier, C., Fosse, P., Couprie, J., Pierre, J., Saucier, J. M., Acknowledgments Le Pecq, J. B., and Larsen, A. K. Inhibitory effects of the tyrosine kinase inhibitor Many thanks go to the staff at the mammography clinics, in particular, to Ginnnie genistein on mammalian DNA topoisomerase II. Cancer Res., 49: 51 1 1-51 17, Coggins (Clinical Manager at Kapiolani Women’s Center, Honolulu, HI) and Dr. 1989. Daniel Henshaw (Kaiser Permanente. Honolulu, HI). We acknowledge Dr. Li- 23. Ogawara, H., Akiyama, T., Watanabe, S. I., Ito, N., and Kobori. M. Inhibition Ching Lyu’s expertise in designing the soy food questionnaire and Laurie J. of tyrosine protein kinase activity by synthetic isoflavones and flavones. J. Custer’s laboratory skills in performing the isoflavone analysis. Without the time Antibiot. (Tokyo). 42: 340-343, 1989.

and effort contributed by the participating women. this project would have been 24. Fotsis, T., Pepper, M., Adlercreutz, H., Hase, T., Montesano, R., and impossible. Their participation is greatly appreciated. Schweigerer, L. Genistein, a dietary ingested isoflavonoid, inhibits cell proliferation and in vitro angiogenesis. J. Nutr., 125 (Suppl.): 790S-797S, 1995. 25. Kyle. E., Neckers. L., Takimoto, C.. Curt, G., and Bergan, R. Genistein- References induced apoptosis of prostate cancer cells is preceded by a specific decrease in I. Pakin. D. M., Muir, C. S., Whelan, S. L., Gao, Y. T., Ferlay. J., and Powell, focal adhesion kinase activity. Mol. Pharmacol., 51: 193-200, 1997. J. Cancer Incidence in Five Continents. IARC Scientific PubI. No. 120. Lyon, 26. Tsukamoto, C., Shimada, S., Igita, K., Kudou, S., Kokubun, M., Okubo, K., 1992. France: IARC. and Kitamura, K. Factors affecting isoflavone content in soybean seeds: changes 2. Armstrong. B. K.. and Mann. J. I. Diet. In: M. P. Vessey and M. Gray (eds.), in isoflavones, saponins, and composition of fatty acids at different temperatures Cancer Risks and Prevention, pp. 168-198. Oxford, United Kingdom: Oxford during seed development. J. Agric. Food Chem., 43: 1 184-1 192, 1995.

University Press, 1985. 27. Coward, L., Barnes, N. C., Setchell, K. D. R., and Barnes, S. The antitumor 3. Hirayama. T. Epidemiology of breast cancer with special reference to the role isoflavones, genistein and daidzein, in soybean foods of American and Asian ofdiet. Prey. Med., 7: 173-195, 1978. diets. J. Agric. Food Chem., 41: 1961-1967, 1993.

4. Nomura, A. M. Y., Lee, J., Kolonel, L. N., and Hirohata, T. Breast cancer in 28. Wang, H., and Murphy, P. A. Isoflavone content in commercial soybean two populations with different levels of risks for the disease. Am. J. Epidemiol., foods. J. Agric. Food Chem., 42: 1666-1673, 1994. 119: 496-502, 1984. 29. Franke. A. A., Custer, L. J., Wang. W., and Shi, S. J. HPLC analysis of 5. BuelI, P. Changing incidence of breast cancer in Japanese-American women. isoflavonoids and other phenolic agents from foods and from human fluids. Proc. J. NatI. Cancer Inst., 5/: 1479-1483, 1973. Soc. Exp. Biol. Med., 211: 163-173, 1998. 6. Miller. B. A.. Kolonel, L. N., Bemstein, L., Young, J. J. L., Swanson, G. M., 30. Price. K. R., and Fenwick, G. R. Naturally occurring oestrogens in foods: a West, D., Key. C. R., Liff, J. M., Glover, C. S., and Alexander, G. A. Racial/ review. Food Addit. Contam., 2: 73-106, 1985. Ethnic Pattems of Cancer in the United States 1988-1992. NIH Publ. No. 31. Setchell, K. D. R.. and Adlercreutz, H. Mammalian lignans and phyto- 96-4104. Bethesda, MD: National Cancer Institute. 1996. estrogens. Recent studies on their formation. metabolism and biological role in

7. Hankin, J. H. Role of nutrition in women’s health: diet and breast cancer. health and disease. In: I. Rowland (ed), Role of the Gut Flora in Toxicity and J. Am. Diet. Assoc., 93: 994-999, 1993. Cancer, pp. 315-345. London: Academic Press, 1988.

8. Willet, W. C.. Stampfer. M. J., Colditz, G. A., Rosner, B., Hennekens, C. H., 32. Joannou, 0. E., Kelly. G. E., Reeder, A. Y., Waring, M., and Nelson, C. A and Speizer. F. E. Dietary fat and risk of breast cancer. N. EngI. J. Med., 316: urinary profile study of dietary phytoestrogens. The identification and mode of 22-28, 1986. metabolism of new isoflavonoids. J. Steroid Biochem. Mol. Biol., 54: 3-4, 1995. 33. Adlercreutz, H., Fotsis, T., Bannwart, C., Wahala, K., Brunow, G., and Hase, 9. Zanidzc, D.. Lifanova. Y., Maximovitch, D., Day. N. E.. and Duffy, S. W. Diet, T. Isotope dilution gas chromatographic-mass spectrometric method for the alcohol consumption and reproductive factors in a case-control study of breast determination of Iignans and isoflavonoids in human urine, including identifica- cancer in Moscow. mt. i. Cancer, 48: 493-501, 1991. tion of genistein. CIin. Chim. Acta, 199: 263-278, 1991. 10. Adlercreutz, H. Westem diet and Westem diseases: some hormonal and 34. Setchell, K. D. R.. Welsh, M. B., and Lim, C. K. High performance liquid biochemical mechanisms and associations. Scand. J. Clin. Lab. Investig., 2014: chromatographic analysis of phytoestrogens in soy protein preparations with 3-23, 1990. ultraviolet light, electrochemical and thermospray mass spectrometric detection. II . Messina, M., and Barnes, S. The role of soy products in reducing risk of J. Chromatogr. B., 386: 315-323, 1987. cancer. J. NatI. Cancer Inst., 83: 541-546, 1991. 35. Adlercreutz, H., Fotsis, T., Lampe, J., Wahala, T., Makela, T., Brunow, G., I 2. Miksicek, R. J. Estrogenic flavonoids: structural requirements for biological and Hase, T. Quantitative determination of lignans and isoflavonoids in plasma of activity. Proc. Soc. Exp. Biol. Med., 208: 44-50, 1995. omnivorous and vegetarian women by isotope dilution gas chromatography-mass I 3. Kuiper, G., CaIsson, B., Grandien, K., Enmak, E., Haggblad, J., Nilsson, S., spectrometry. Scand. J. Clin. Lab. Investig., 53: 5-18, 1993. and Gustafsson, J. A. Comparison of the ligand binding specificity and transcript 36. Franke, A. A., Custer, L. J., Cema, C. M., and Narala, K. K. Quantitation of tissue distribution of estrogen receptors a and 3. Endocrinology, 138: 863-870, phytoestrogens in legumes by HPLC. J. Agric. Food Chem., 42: 1905-1913, I 997. 1994.

14. Bennetts, H. W., Underwood, E. J., and Sheir, F. L. A specific breeding 37. Adlercreutz, H., Honjo, H., and Higashi, A. Urinary excretion of Iignans and problem of sheep on subterranean clover pastures in westem Australia. Aust. Vet. isoflavonoid phytoestrogens in Japanese men and women consuming a traditional J.. 22: 2-12, 1946. Japanese diet. Am. J. Clin. Nutr., 54: 1093-1 100, 1991.

38. Goldin, B. R., Adlercreutz, H., Gorbach, S. L., Woods, M. N., Dwyer, J. T., 48. Kelly, G. E., Nelson, C.. Wann, W. A., Joannou, G. E., and Reeder, A. Y. Conlon, T., Boh, E., and Gershoff, S. N. The relationship between estrogen levels Metabolites of dietary (soya) isoflavones in human urine. Clin. Chim. Acta, 223: and diets of Caucasian American and Oriental immigrant women. Am. J. Clin. 9-22. 1993. Nutr., 44: 945-953, 1986. 49. Ingram. D.. Sanders, K., Kolybaba, M., and Lopez, D. Case-control study of 39. Franke, A. A., and Custer, L. J. High-performance liquid chromatographic phyto-oestrogens and breast cancer. Lancet, 350: 990-997, 1997. assay of isoflavonoids and coumestrol from human urine. J. Chromatogr. B., 662: 47-60, 1994. 50. Xu, X., Wang, H-J., Murphy, P. A., Cook, L., and Hendrich, S. Daidzein is 40. Kar. S. C., Lampe, J. W., Hutchins, A. M., and Slavin, J. L. Urinary a more bioavallable soymilk isoflavone than is genistein in adult women. J. Nutr., isoflavonoid excretion in humans is dose dependent at low to moderate levels of 124: 825-832, 1994. soy-protein consumption. Am. J. Clin. Nutr., 66: 46-51, 1997. 51. Kirkman, L. M., Lampe, J. W., Campbell, D. R., Martini, M. C., and Slavin, 41. Lu, L. J. W., Lin, S. N., Grady, J. J., Nagamani, M., and Anderson, K. E. J. L. Urinary lignan and isoflavonoid excretion in men and women consuming Altered kinetics and extent of urinary daidzein and genistein excretion in women vegetable and soy diets. Nutr. Cancer, 24: 1-12, 1995. during chronic soy exposure. Nutr. Cancer, 26: 289-302, 1996. 52. Xu, X., Harris, K. S., Wang, H. J., Murphy, P. A., and Hendrich, S. 42. Horn-Ross, P. L., Barnes, S., Kirk, M., Coward, L., Parsonnet, J., and Hiatt, Bioavailability of soybean isoflavones depends upon gut microflora in women. J. R. A. Urinary phytoestrogen levels in young women from a multiethnic popula- Nutr., 125: 2307-2315, 1995. tion. Cancer Epidemiol. Biomark. Prey., 6: 339-345, 1997. 43. Hawaii Department of Business, Economic Development and Tourism. State 53. Gobrach, S. L., Plaut, A. G., Nahas, L., Weinstein, L., Spanknebel, G.. and of Hawaii Data Book 1995: A Statistical Abstract. Honolulu, Hi: State of Hawaii, Levitan, R. Studies of intestinal microflora. II. Microorganisms of the small 1995. intestine and their relations to oral and faecal flora. Gastroenterology, 53: 856- 44. United States Department of Agriculture. United States Department of Ag- 867, 1967. riculture Nutrient Database for Standard Reference. Riverdale, Maryland: United 54. Heneghan, J. B. Alimentary tract physiology interaction between the host and States Department of Agriculture, 1993. its microbial flora. In: I. R. Rowland, (ed), Role of the Gut Flora in Toxicity and 45. Franke, A. A., Custer, L. J., Tanaka, Y., and Maskarinec, G. Isoflavonoids in Cancer, pp. 39-78. San Diego, CA: Academic Press, 1988. soy drinks and availability in human body fluids. Proc. Am. Assoc. Cancer Res., 38: 112, 1997. 55. Winter, J., Moore, L. H.. Dowell, V. R., and Bokkenheuser, V. D. C-ring cleavage of flavonoids by human intestinal bacteria. AppI. Environ. Microbiol., 46. Franke, A. A., Custer, L. J., Cema, C. M., and Narala, K. Rapid HPLC 55: 1203-1208, 1989. analysis of dietary phytoestrogens from legumes and from human urine. Proc. Soc. Exp. Biol. Med., 208: 18-26, 1995. 56. Miyake, Y., Yamamoto, K., and Osawa, T. Metabolism of antioxidant in 47. Rosner, B. Fundamentals of Biostatistics. Boston, MA: PWS-Kent Publish- lemon fruit (Citrus limon BURM. f.) by human intestinal bacteria. J. Agric. Food ing Co., 1990. Chem., 45: 3738-3742, 1997.

Downloaded from cebp.aacrjournals.org on October 2, 2021. © 1998 American Association for Cancer Research. Dietary soy intake and urinary isoflavone excretion among women from a multiethnic population.

G Maskarinec, S Singh, L Meng, et al.

Cancer Epidemiol Biomarkers Prev 1998;7:613-619.

Updated version Access the most recent version of this article at: http://cebp.aacrjournals.org/content/7/7/613

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/7/7/613. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.