Fecal Lignan and Isoflavonoid Excretion in Premenopausal Women Consuming Flaxseed Powder1

Vol. 4. .153-358. June 1995 Cancer Epidemiology, Biomarkers & Prevention 353

Mindy S. Kurzer,2 Johanna W. Lampe, have shown important biological effects that contribute to their Margaret C. Martini, and Herman Adlercreutz potential as chemopreventive agents.

Department of Food Science and Nutrition, University of Minnesota, St. Paul, Lignans have been shown to inhibit skin, breast, colon, Minnesota 55108 [M. S. K., J. W. L.. M. C. M.], and Department of Clinical and lung cancer cell growth (1, 2). Consumption of flaxsecd, a Chemistry, University of Helsinki, Meilahti Hospital, Helsinki, Finland [H. A.] concentrated source of lignans (3), has been shown to inhibit mammary tumor promotion (4) and development of early mark- ens of risk of mammary cancinogenesis (5). Proposed mecha- Abstract nisms by which lignans may inhibit carcinogenesis include Lignans and isoflavonoids are diphenolic compounds antivinal (6) and antioxidant (7) activities. In addition, the found in plant foods, particularly whole grains and similarities in structure among lignans, estradiol, and the syn- legumes. They have been shown to have anticarcinogenic thetic antiestrogen tamoxifen (Fig. 1) suggest that lignans may properties in animal and cell studies, and have been also exert their anticarcinogenic effects in part as a result of associated with reduced cancer risk in epidemiobogical antiestrogenic effects. In fact, lignans have been shown to studies. In order to perform further epidemiobogical and inhibit placental (8) and adipocyte (9) estrogen synthesis; to metabolic studies on these compounds, it is necessary to inhibit cstnadiol-induced proliferation of MCF-7 human breast be able to monitor concentrations in biological samples. carcinoma cells (10); and to stimulate sex hormone-binding In this study, we examined the effects of consumption of globulin synthesis, with subsequent decrease in free estradiol flaxseed, a concentrated source of lignans, on fecal lignan (11). Flaxseed consumption has been shown to alter the men- excretion and evaluated the effect of high lignan strual cycle in premenopausal women in a potentially cancer- consumption on fecal excretion of isoflavonoids. Thirteen preventive direction (12). women were studied for two diet periods of three Isoflavonoids have been reported to inhibit cancer cell menstrual cycles each in a cross-over design. During the growth and development in lung (13), stomach (14), lcukocyte control period, they consumed their usual diets; during (15, 16), and breast (17, 18) carcinoma cells. Proposed mech- the treatment period they consumed their usual diets anisms include in vitro inhibition of bacterial growth (19, 20), supplemented with 10 g/day ground flaxseed. Feces were as well as in vitro inhibition of placental (8) and adipocyte (9, collected on days 7-1 1 of the last menstrual cycle in each 21) estrogen synthesis. Isoflavonoids have been shown in viva diet period. Five-day fecal composites were analyzed for to have antispasmodic effects in guinea pig ileum (22), antihy- lignans and isoflavonoids by isotope dilution gas pertensive effects in rats (23), and hypolipemic effects in chromatography-mass spectrometry. Fecal excretion of rabbits (24). the lignans enterodiol, enterolactone, and matairesinob Despite great interest in the physiological effects of big- increased significantly with flax consumption, from nans and isoflavonoids, little is known about human metabo- 80.0 ± 80.0 (SD) to 2560 ± 3100; 640 ± 480 to 10,300 lism, absorption, and actual exposure. Mammalian lignans are ± 7580; and 7.33 ± 10.0 to 11.9 ± 8.06 nmol/day, synthesized by cobonic microflona from plant precursors found respectively. There were no differences in fecal excretion naturally in many plant foods, particularly oilseeds and grains of the isoflavonoids, daidzein, equol, genistein, and (3, 25). Flaxsecd is an oilseed known to be a concentrated O-demethylangolensin. source of lignan precursors (3). Consumption of flaxsced ne- suits in large increases in urinary excretion of entenodiol and Introduction enterolactone, the primary mammalian lignans derived from bacterial metabolism of their plant precursors, secoisolanicires- Numerous components of plant foods have been associated mob and matainesinol (26). Entenolactone also originates from with decreased cancer risk. Among these are lignans and isofla- bacterial oxidation of enterodiol. vones, two groups of diphenolic compounds most concentrated Genistein and daidzein arc the two primary dietary isofla- in whole grains and legumes (Fig. 1). Epidemiobogical studies vones found in high concentrations in legumes. Genistein and have shown decreased cancer risk with increased excretion of daidzein can also originate from bacterial metabolism of their these compounds, while animal, cellular, and metabolic studies respective glycosides, as well as metabolism of two other isoflavones, biochanin A and formononetin. Daidzein can be metabolized further by intestinal microfbora to the isofla-

Received 9/2/94: revised 12/21/94: accepted 12/21/94. vonoids 0-Dma3 and equol (25, 27).

I This research was supported by NCI Grant N01-CN-05288-O1 and Minnesota Human studies of the physiological effects of dietary big- Agricultural Experiment Station Project 18-34. J. W. L. was supported by nans and isoflavones require accurate measurement of these National (‘ancer Institute Grant 5T32-CA-09607. Development of the method was supported by NIH Grant I ROI CA 56289-01. 2 To whom requests for reprints should be addressed, at Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Avenue, St. Paul,

MN 55108. 3 The abbreviation used is: 0-Dma, 0-demethylangolensin.

HO lignan consumption on isoflavonoid excretion. Entenobactone, enterodiol, matairesinob, genistein, daidzcin, equol, and 0-Dma were measured in 5-day fecab composites collected from 13 cIciOL healthy young women after 3 menstrual cycles on their habitual diets and after 3 menstrual cycles in which their usual diets Daidzein Genistein were supplemented with 10 g/day ground flaxseed. In addition to determining the fecal concentrations of individual com- OcHCHaNtcH3)2 pounds, correlations were performed to evaluate the associations among individual isoflavonoids and bignans.

Materials and Methods Subjects. This study was a substudy of a larger study of the effects of flaxseed consumption on lignan excretion and the menstrual cycle (12, 26). For the large study, 18 healthy pre- H3 “‘ H menopausal women were studied. Health status of the subjects Tamoxifen was verified by health history, routine blood and urine screen- ing, and physical exam. Selected subjects were sedentary, omnivorous nonsmokers, who typically consumed diets containing HO, <10 g dietary fiber/day. All subjects reported regular menstrual cycles and stable body weights for at least I year, had not used oral contraceptives for at beast 3 months, had not been on antibiotics for the previous month, and were not on any regular medication. For this study, 13 women selected from the original 18

Enterodiol Enterolactone subjects were willing to collect feces. The prestudy averages for age, height, weight, and body mass index of these 13 subjects

Fig. 1. Structures of selected isoflavones (daidzein and genistein), estrogens were 27.8 ± 4.3 year, 1.64 ± 0.05 m, 60.6 ± 8.5 kg, and (tamoxifen and estradiol). and lignans (enterodiol and enterolactone). 22.5 ± 2.2 kg/m2, respectively (Table 1). Study Design and Diet. The protocol for this study was ap- proved by the University of Minnesota Institutional Review compounds in biological fluids, yet few studies have reported Board Human Subjects Committee. The study design and diet such data. Adlercreutz et a!. (1 1, 28-30) have reported urinary have been described in detail previously (26). The study was and serum (31, 32) concentrations of entenolactone, enterodiol, performed as a balanced, randomized, cross-oven design fob- and matairesinol, as well as genistein, daidzein, equol, and bowing the individual menstrual cycles of each subject. After an 0-Dma, in cross-sections of omnivorous and vegetarian sub- acclimation period of I menstrual cycle, each subject was jects in Boston, Helsinki, and Japan. Xu ci a!. (33) have studied for 6 menstrual cycles. For the first 3 menstrual cycles, published the only human study of isoflavone bioavaibability, one-half of the subjects were on their habitual diets (control reporting plasma, urinary, and fecal daidzein and genistein diet) and one-half were on their habitual diets supplemented concentrations in adult women. Although they report fecal with 10 g/day ground flaxseed (Enneco Corp., Manitowoc, WI; isoflavone concentrations after consumption of isoflavone-nich treatment). For the last 3 menstrual cycles, the groups switched soymilk, Xu et a!. (33) include no baseline data. Recently, diets. Adlercreutz et a!. (34) described a method for simultaneous Subjects were free living during the entire study. Although measurement of fecal lignans and isoflavonoids, reporting pre- they consumed their habitual diets, they were given instructions liminary data in omnivorous and vegetarian women. There have to minimize isoflavone and lignan consumption by avoiding all been no further reports of fecal lignan concentrations. legumes (due to the high isoflavone content) and flaxsced- Fecal concentrations of isoflavonoids and lignans are of containing products (certain prepared bread and cereal items). interest for several reasons: (a) fecal concentrations may reflect Food intake was monitored by 3-day diet records kept twice per the direct exposure of colon cells to anticarcinogenic lignans menstrual cycle. Diet records were analyzed for nutrient con- and isoflavonoids; (b) the known antiviral and antifungal ad- tent with the use of a computerized nutrient analysis program tions of lignans and isoflavonoids may exert beneficial intralu- (Nutritionist III; N-Squared Computing, Silvcnton, OR). minal effects. Their antioxidant actions also may be important For the treatment period, each daily dose of 10 g/day via prevention of the activation of cocancinogens to carcino- ground flaxseed was packaged individually and stored frozen. gens; (c) since both lignans and isoflavones are metabolized by Subjects received weekly supplies of flaxseed, with instructions intestinal microflora, increased lignan consumption may alter to keep frozen until consumption. They were also instructed to lignan or isoflavone metabolism and excretion, perhaps by consume the entire contents of each individual package every altering microfloral numbers or enzymatic activities; and (d) day, splitting it into 2 or 3 servings and adding it raw to knowledge of fecal concentrations and the metabolic intenac- prepared foods. Daily flaxseed intake was recorded by each tions among these compounds will complete the metabolic subject, and uneaten portions were returned and weighed. picture necessary to perform further epidemiological studies. Sample Collection and Analysis. The 13 subjects collected This study is the first metabolic experiment to report feces on days 7-1 1 of the last menstrual cycle of each diet lignan concentrations in human feces before and after consump- period. The methods used for fecal collection and preparation tion of a concentrated source of plant lignans. We also report have been described previously (35). Each fecab sample was typical fecal isoflavonoid concentrations in female subjects collected in a plastic bag. Immediately after defecation, 10 ml consuming a low isoflavone diet, as well as the effect of high of 10% (w/v) ascorbic acid were poured over the feces, air was

Table I Subject characteristics and dietary intake”

Control diet Flax supplemented

Age (yr) 27.80 ± 4.30 (23.00-34.00) 27.80 ± 4.30 (23.0O-34.0()) Height (m) 1.64 ± 0.05 (1.55-1.75) 1.64 ± 0.00 (1.55-1.75) Weight (kg) 61.00 ± 9.00 (53.00-85.00) 61.()0 ± 9.00(51.00-85.00) BMI” (kg/m2) 22.50 ± 2.20 (20.10-27.70) 22.50 ± 2.20(20.10-27.70) Energy intake (kcal/day) 1834.00 ± 1 10.00 (1 182.00-2564.00) 2015.00 ± 191.00 (1085.00-3208.00) Protein intake (g/day) 79.00 ± 5.00(51.00-111.00) 82.00 ± 9.00(39.00-141.00) Fat intake (g/day) 75.00 ± 5.00 (42.00-103.00) 84.00 ± 8.00 (41(10-137(X)) Carbohydrate intake (g/day) 213.00 ± 17.00 (124.00-312.00) 249.00 ± 28.00(140.00-405(X)) Dietary fiber (g/day) 6.00 ± 1.00 (1.00-11.00) 11(X) ± 1.00(7.00-17.00)

“ Arithmetic X ± SD (range): n 13. I’ BMI, body mass index.

forced from the bag, and the fecal sample was then sealed and data. Results were considered significant at levels of P < 0.05. stored at -20#{176}C. A 5-day fecal composite was created by Data are presented as both nontransformed and geometric weighing all individual fecal samples for each subject to de- means. For correlation analyses of log-transformed data, 0.1 (ermine total wet weight, adding 0.1% (w/v) sodium azide, and was entered in place of zero values. homogenizing the feces. Aliquots of the composite were frozen at -20#{176}C for later bignan and isoflavonoid determination. Duplicate abiquots were freeze dried for determination of dry Results weight. Diet and Body Weight. Diet and body weight results for the Fecal lignans and isoflavonoids were analyzed with the 18 subjects of the large study have been published previously use of a modification of the urinary method developed by (26). Mean consumption of energy and macronutnients for the Adlencneutz et a!. (29). Fecal lignans and isoflavonoids were 13 subjects in this substudy are shown in Table 1. These data extracted and separated by ion exchange chromatography and include the contribution from the 10 g/day flaxseed, which quantified by capillary gas chromatography-mass spectrometny included 184 Id (44 kcal), 2 g protein, 4 g fat, and 4 g dietary in the selective ion-monitoring mode. The complete fecab fiber/day. Flax consumption resulted in no significant differ- method has been published separately (34). Briefly, deuterated ences in consumption of energy, carbohydrate, protein, or fat bignan and isoflavonoid standards were added to 0.3-0.6-g expressed as g/day on percentage energy. Fiber intake increased fecab samples. After extraction and filtration in a 1 :9 acetone: significantly (P < 0.0001), from 6 to 1 1 g/day, with flax ethanol solution, samples were left overnight in 70% ethanol to supplementation. precipitate proteins. Samples were then run through a series of There were no significant differences in body weight or Sep-Pak C1,, (Waters Associates, Inc., Milford, MA), DEAE- body mass index between the control and flaxsecd treatment Sephadex-acetate (Pharmacia Fine Chemicals, Uppsaba, Swe- periods. den), and QAE-Sephadex-canbonate (Pharmacia Fine Chemi- Fecal Lignans and Isoflavonoids. Fecal enterolactone, en- cabs) columns. Following denivatization, analyses were carried terodiob, and matairesinol excretion increased significantly with out with the use of a Hewlett-Packard 5995 quadrapole gas flaxseed consumption, when expressed pen g wet weight, per g chromatography-mass spectrometry instrument. Samples were dry weight, or per day (Table 2). Although the entenodiob: analyzed in triplicate, and all samples from a given subject were enterolactone ratio appeared to increase with flaxseed, the in- analyzed in the same run. Triplicate quality control samples crease was not significant. There were no significant effects of were also analyzed in each run. flaxseed consumption on fecal isoflavonoid excretion (Table 3). Intra-assay coefficients of variation for the fecab method Within the control diet, when results were expressed pen g were found to be 6.4, 10.1, 5.1, 6.2, 6.2, 9.8, and 7.4% for dry weight, per g wet weight, on per day, all three isoflavonoids 0-Dma, equol, enterodiol, enterolactone, daidzein, genistein, (0-Dma, daidzein, and genistein) were correlated positively and matainesinol, respectively. Interassay coefficients of van- with each other (Tables 4 and 5). The lignan entenodiol was ation were 8.0, 12.1, 5.5, 8.6, 9.3, 24.7, and 6.8% for 0-Dma, correlated positively with each of the isoflavonoids, as well as equob, enterodiob, entenobactonc, daidzein, genistein, and ma- with the lignan precursor matairesinol. tairesinol, respectively. The recoveries of added phytoestrogens When correlations were performed within the flax treat- were between 80.0 and 103.6%. ment period, 0-Dma was no longer associated significantly Statistics. Although bignans and isoflavonoids have been hy- with daidzein and genistein, although daidzein and genistemn pothesized to have many physiologically significant effects, it were still correlated positively with each other. The correlations is unknown whether it is the concentration or total quantity between enterodiol and the isoflavonoids 0-Dma, daidzein, and excreted that is the important parameter to be followed. As a genistein were lost, although entenodiol was still associated result, all data were analyzed on a pen g dry weight, per g wet positively with matairesinol. When results were expressed as weight, and pen day basis. Statistical analyses were performed per g wet weight, enterolactone was associated negatively with with the use of the Statistical Analysis System (36). The effect its precursor, matairesinol. When expressed as pen day, equol of diet on fecal bignan and isoflavonoid excretion was evaluated was associated positively with matairesinol and enterobactone. with the use of pained t tests. Pearson correlations were performed within each diet period to evaluate associations among the individual lignans and isoflavonoids. Since some of the Discussion variables were not normally distributed, t tests and correlations This is the first report of a metabolic study in which lignan were performed on log-transformed as well as nontransformed concentrations were measured in human feces before and after

Table 2 Fecal lignan excretion

Control diet Flax supplemented

Enterodiol (nmol/g dry wt) 3.77 ± 2.65 (2.65)” 1 15.00 ± 123.00” (46.70)’

(nmol/g wet wt) 1 .03 ± 0.63 (0.76) 30.50 ± 33.00” (13.30)’ (nmollday) 80.00 ± 80.00(60.00) 2,560.00 ± 3,100.00’ (980.00)’

Enterolactone (nmol/g dry wt) 30.90 ± 24.30(21.80) 416.00 ± 248.00’ (326.00)’ (nmol/g wet wt) 8.76 ± 6.41 (6.31) 127.00 ± 95.30” (27.70)’

(nmol/day) 640.00 ± 480.00(450.00) 10,300.00 ± 7,580.00’ (6,820.00)’

Matairesinol (nmol/g dry wt) 0.28 ± 0.31 (0.08) 0.52 ± 0.2Y (0.17) (nmol/g wet wt) 0.07 ± 0.08 (0.03) 0.14 ± 0.04” (0.13) (nmol/day) 7.33 ± 10.()0 (2.01) 11.90 ± 8.06 (9.71)

Enterolactone 0.24 ± 0.36 (0.12) 1.05 ± 2.35 (0.14)

Enterodiol Total lignans (nmol/g dry wt) 34.90 ± 24.80 (26.60) 531.00 ± 223.00’ (493.00)’ (nmol/gjwet wt) 9.86 ± 6.47 (7.71) 158.00 ± 87.70’ (141.00)’ (nmol/day) 727.00 ± 510.00(550.00) I 2,87 1.00 ± 8,430.00” (10,337.00)’

“ Arithmetic X ± SD (geometric mean); n = 13. 5 Significantly different from control diet using paired t tests; P < 0.0001.

‘ Significantly different from control diet using paired t tests; P < 0.01.

‘I Significantly different from control diet using paired t tests; P < 0.001. Significantly different from control diet using paired t tests; P < 0.05.

Ta ble 3 Fecal isoflavonoid excretion” Table 4 Correlations among isoflavonoids and lignans expressed as per g wet weight or per g dry weight” b Control diet Flax supplemented Control diet Flax supplemented Daidzein (nmol/g dry wt) 5.32 ± 1 1.14 (1.85)” 4.53 ± 7.32 (1.85) Expressed per g dry wt (nmol/g wet wt) 1.54 ± 3.50 (0.55) 1.30 ± 2.05 (0.51) 0-Dma and daidzein r = 0.72; P = 0.006 r = 0.49; P = 0.09 (nmollday) 120(X) ± 260(X) (40.00) 80.00 ± 1 10.00 (4().00) 0-Dma and genistein r = 0.78; P = 0.002 r = 0.27; P = 0.39

Daidzein and genistein r = 0.81; P = 0.0008 r = 0.83; P = 0.0004 Equol Enterodiol and 0-Dma r 0.60; P 0.03 r -0.06; P 0.85 (nmollg dry wt) 0.29 t 0.21 (0.22) 0.52 0.45 (0.34) Enterodiol and daidzein r = 0.67; P = 0.01 r = 0.28; P = 0.36 (nmol/g wet wt) 0.09 ± 0.07 (0.06) 0.15 ± 0.14 (0.10) Enterodiol and genistein r = 0.74; P = 0.004 r 0.49; P 0.09 (nmol/day) 6.66 ± 5.43 (4.63) 14.60 ± 16.70 (7.23) Enterodiol and r = 0.59; P = 0.03 r = 0.60; P = 0.03 Genistein matairesinol

(nmol/g dry wt) 2.58 6.32 (0.35) 2.02 ± 2.66 (0.73) Expressed per g wet wt (nmol/g wet wt) 0.77 ± 1.98(0.11) 0.52 ± 0.66 (0.21) 0-Dma and daidzein r = 0.73; P = 0.005 r = 0.53; P = 0.06 (nmollday) 58.50 ± 147(10 (8.13) 34.40 ± 37.80 (15.30) 0-Dma and genistein r = 0.72; P = 0.005 r = 0.28; P = 0.36

0-Dma Daidzein and genistein r = 0.80; P = 0.0009 r = 0.81; P 0.0008

(nmollg dry wt) 2.44 ± 3.63 (0.52) 1.92 ± 3.23 (0.37) Enterolactone and r = -0.04; P = 0.89 r = -0.70; P = 0.007 (nmol/g wet wt) 0.75 ± 1.16 (0.16) 0.62 ± 1.02 (0.12) matairesinol (nmollday) 55.40 ± 86.60 (12.30) 37.20 ± 57.30 (8.40) Enterodiol and 0-Dma r = 0.61; P = 0.03 r = -0.05; P 0.87 Enterodiol and daidzein r = 0.64; P = 0.02 r = 0.22; P = 0.47 Daidzeinlequol 24.20 ± 37.60 (8.6()) 14.00 ± 16.70 (5.57) Enterodiol and genistein r = 0.73; P = 0.005 r = 0.42; P = 0.15

Total isoflavonoids Enterodiol and r = 0.53; P = 0.06 r 0.56; P 0.04 (nmol/g dry wt) 10.60 ± 20.40 (4.59) 8.99 ± 12.20 (4.71) matairesinol (nmol/g wet wt) 3.15 ± 6.41 (1.31) 2.59 ± 3.47 (1.35) “ All correlations were done on log-transformed data. (nmol/day) 241.00 ± 483.00 (94.60) 166.00 ± 188.00 (98.80) I, Correlations shown are those that were significant within at least one diet period.

“ There were no significant differences due to treatment.

I’ Arithmetic X ± SD (geometric mean); n = 13. matairesinol) after consumption of flaxseed. Daily fecal excretion of enterobactone increased 16-fold, entenodiob in- consumption of a concentrated source of plant lignan precur- creased 32-fold, and matairesinol increased I .6-fold. Urinary sons. The plant precursors of enterodiol and cntcnolactone are bignan excretion in the same subjects showed a 7-fold in- considered to be secoisolanicinesinol and matainesinol, respec- crease in enterolactone, from 4.06 to 29.8 mol/day, and a tively. Consumption of flaxseed, thought to contain primarily 26-fold increase in enterodiol, from 0.87 to 22.6 tmol/day, secoisobaniciresinol, leads to enterodiob formation in the intes- with no increase in urinary matairesinol excretion (26). tine. The enterodiol can be oxidized to entenolactone by intes- There was a greater increase in enterodiol excretion than in tinal microflora. entenolactone excretion with flax consumption, which likely As expected, we found significantly increased daily reflects the increased production of enterodiol from its pre- fecal excretion of the bignans (entenodiol, enterolactone, and cursor secoisolaniciresinol.

Table 5 Correlations among isoflavonoids and lignans expressed expected if the entenodiol precursors were increased signifi- as per day” cantly with the addition of flaxseed, while the isoflavonoids, found mainly in the habitual diets, remained unchanged. The Control diet Flax supplemented associations between enterodiol and the isoflavonoids observed 0-Dma and daidzein r 0.75; P = 0.003 r 0.47; P 0.11 during the control period likely reflect common sources in the 0-Dma and genistein r = 0.71; P = 0.006 r 0.21; P 0.49 habitual diets. Daidzein and genistein r = 0.83; P = 0.0(104 r 0.81; P 0.0009 In conclusion, this is the first report of a metabolic study Equol and matairesinol r = 0.39; P = 0.19 r 0.62; P 0.01 in which lignan concentrations were measured in human feces Equal and enterolactone r 0.45; P = 0.13 r 0.60; P 0.03 before and after consumption of a concentrated source of plant Enterodiol and 0-Dma r = 0.62; P = 0.02 r 0.08; P 0.79 lignan precursors. It is also the first report of fecal isoflavonoid Enterodiol and daidzein r - 0.72; P = 0.006 r 0.26; P 0.39

Enterodioi and genistein r = 0.74: P = 0.004 r = 0.46; P = 0.11 concentrations in subjects consuming a typical Western diet,

Enterodiol and matairesinol r = 0.63; P = 0.02 r 0.56; P 0.04 which is low in isoflavones. Consumption of 10 g/day ground flaxseed resulted in significantly increased fecal concentrations “ All correlations were done on log-transformed data. I, Correlations shown are those that were significant within at least one diet period. of the lignans entenobactone, enterodiol, and matairesinol, while no changes were observed in fecal concentrations of the isoflavonoids daidzein, equol, genistein, on 0-Dma. To more fully understand the exposure of differing populations to these com- The fecal cnterodiol:entenobactonc ratio in our 13 subjects pounds, as well as the interactions among them, it will be increased from 0.24 to 1.05 with flax consumption. Using an in necessary to perform further dietary studies in which bignan and vitro fermentation system to evaluate lignan production by isoflavone intakes are controlled and levels of the compounds numerous foods incubated with human fecab bacteria, Thomp- and their precursors are measured in urine, blood, and feces. son et a!. (3) reported the enterodiol:entcrobactone ratio pro- Improved understanding of the concentrations of lignans and duced by flaxseed flour to be 3.4. Although it is difficult to isoflavones in biological fluids will enable us to perform further compare these very different methodologies, both studies show epidemiological studies in which bignans and isoflavonoids greaten entenodiol than entenobactone production, likely reflect- may be used as biomarkens of cancer risk. ing the higher concentration of secoisobaniciresinol than matairesinol in flaxseed. In our study, low matairesinol excretion after flaxseed Acknowledgments consumption suggests that there is very little matairesinol We thank Susan Fredstrom for assisting with study coordination; Anja Koskela, present in the flaxseed. At the same time, correlation analyses Sirkka Adlercreutz, and Inga Wiik for technical assistance; and the 13 subjects for their cooperation. showed fecab matairesinol excretion to be associated positively with fecal enterodiol excretion. This suggests that although matairesinob levels were low, flax may have been the primary References source of matainesinol just as it is known to be the primary 1. Kardono, L. B., Tsauri, S., Padmawinata, K., Pezzuto, J. M., and Kinghorn, source of the enterodiob precursor secoisolariciresinob. On the A. D. Cytotoxic constituents of the bark of Plumeria rubra collected in Indonesia. other hand, bow levels of matainesinob may compete with 5cc- J. Nat. Prod., 53: 1447-1455, 1990. oisolanicinesinol for the same demethylating and dehydroxylat- 2. Hirano, T., Fukuoka, K., Oka, K., Naito, T., Hosaka, K., Mitsuhashi, H., and ing enzymes, leading to decreased conversion of matairesinol to Matsumoto, Y. Antiproliferative activity of mammalian lignan derivatives against the human breast carcinoma cell line, ZR-75-1. Cancer Invest., 8: 595-602, 1990. enterolactone in the presence of high secoisolaniciresinob levels. This is the first report of basal levels of fecal isoflavonoids 3. Thompson, L. U., Robb, P., Serraino, S., and Cheung, F. Mammalian lignan production from various foods. Nutr. Cancer, 16: 43-52, 1991. in subjects consuming a typical Western diet, which is expect- edly low in isoflavones. Although flaxseed is a concentrated 4. Serraino, M., and Thompson, L. U. The effect of flaxseed consumption of the initiation and promotional stages of mammary carcinogenesis. Nutr. Cancer. I 7: source of plant lignans, it is a minor source of isoflavones. Our 153-159, 1992. finding that fecal isoflavonoids did not change with flaxseed 5. Serraino, M., and Thompson, L. U. The effect of flaxseed supplementation on consumption was therefore unsurprising. Although consump- early risk markers for mammary carcinogenesis. Cancer Lett., 60: 135-142, 1991. tion of high bevels of bignans did not significantly influence 6. Schroder, H. C., Merz, H., Steffen, R., Muller, W. E., Sam, P. 5., Trumm, S., isoflavonoid excretion, it did appear to influence isoflavonoid Schulz, J., and Eich, E. Differential in s’itro anti-HIV activity of natural lignans. metabolism by intestinal microflora. The significant association Z. Naturforsch., 45: 1215-1221, 1990. between fecal 0-Dma and its isoflavone precursor daidzein, 7. Yamashita, K., Nohara, Y., Katayama, K., and Namiki, M. Sesame seed seen during the control diet, was not observed during the flax lignans and ‘y-tocopherol act synergistically to produce vitamin E activity in rats. J. Nutr., 122: 2440-2446, 1992. treatment period. This suggests that high bignan consumption may redirect daidzein metabolic pathways away from 0-Dma 8. Adlercreutz, H., Bannwart, C., W#{228}h#{228}lS,K.,M#{228}kelS,T., Brunow, G., Hase, T., Arosema, P. J., Kellis, J. T., Jr., and Vickery, L. E. Inhibition of human aromata.se formation and toward equob formation. In fact, with flax con- by mammalian lignans and isoflavonoid phytoestrogens. J. Steroid Biochem. Mol. sumption, 0-Dma excretion decreased and equob excretion in- Biol., 44: 147-153, 1993. creased, although neither change was statistically significant. 9. Wang, C., MSkelS, T., Hase, T., Adlercreutz, H., and Kurzer, M. S. Lignans On the other hand, we did not find a significant correlation and flavonoids inhibit aromatase enzyme in human preadipocytes. J. Steroid between fecab daidzein and equol excretion. This is not surpris- Biochem. Mol. Biol., 50: 205-212, 1994. ing since it is well known that some people do not synthesize 10. Mousavi, Y., and Adlercreutz, H. Enterolactone and estradiol inhibit each equol significantly after consumption of a concentrated daid- other’s proliferative effect on MCF-7 breast cancer cells in culture. J. Steroid zein food source such as soy (37). Since we did not separate out Biochem. Mol. Biol., 4!: 615-619, 1992. equol producers from nonproducers, it is expected that any 1 1. Adlercreutz, H., Mousavi, Y., Clark, J., HOckerstedt, K., H#{228}mSlainen, E., WShSlS, K., Makel#{228},T., and Hase, T. Dietary phytoestrogens and cancer: in euro correlations involving equol would be difficult to show. and in vivo studies. J. Steroid Biochem. Mol. Biol., 41: 331-337, 1992. The associations between entenodiol and the isoflavonoids 12. Phipps, W. R., Martini, M. C., Lampe, J. W., Slavin, J. L., and Kurzer, M. S. daidzein, genistein, and 0-Dma observed during the control Effect of flax seed ingestion on the menstrual cycle. J. Clin. Endocrinol. Metab., diet were lost during the flax treatment period. This would be 77: 1215-1219, 1993.

13. Lee, Y. S., Seo, J. S., Chung, H. T., and Jang, J. J. Inhibitory effects of 26. Lampe. J. W., Martini, M. C., Kurzer, M. S., Adiercreutz, H., and Slavin, J. biochanin A on mouse lung tumor induced by benzo(a)pyrene. J. Korean Med. L. Urinary lignan and isoflavonoid excretion in premenopausal women consum- Sci., 6: 325-328, 1991. ing flaxseed powder. Am. J. Clin. Nutr., 60: 122-128, 1994.

14. Matsukawa, Y., Marui, N., Sakai, T., Satomi, Y., Yoshida, M., Matsumoto, 27. Setchell, K. D. R., and Adlercreutz, H. Mammalian lignans and phyto- K.. Nishino, H., and Aoike, A. Genistein arrests cell cyle progression at G2-M. estrogens: recent studies on their formation, metabolism and biological role in Cancer Res., 53: 1328-1331, 1993. health and disease. in: I. R. Rowland (ed), Role of the Gut Flora in Toxicity and 15. Traganos, F., Ardelt. B., Halko, N., Bruno, S., and Darzynkiewicz, Z. Effects Cancer, pp. 315-345, London: Academic Press, 1988. of genistein on the growth and cell cycle progression of normal human lympho- 28. Adlercreutz, H., Fotsis, T., Bannwart, C., WSh#{228}lS,K., M#{228}kel#{228},T.,Brunow, cytes and human leukemic MOLT-4 and HL-60 cell. Cancer Res., 52: G., and Hase, T. Determination of urinary lignans and phytoestrogen metabolites, 62(X)-6208, 1992. potential antiestrogens and anticarcinogens, in urine of women on various habit- 16. Hirano. T., Gotoh, M., and Oka, K. Natural flavonoids and lignans are potent ual diets. J. Steroid Biochem., 25: 791-797, 1986. cytostatic agents against human leukemic HL-60 cells. Life Sci., 55: 1061-1069, 1994. 29. Adlercreutz, H., Fotsis, T., Bannwart, C., WShSlS, K., Brunow, G., and Hase, T. Isotope dilution gas chromatographic-mass spectrometric method for the 17. Peterson, G., and Barnes, S. Genistein inhibition of the growth of human breast cancer cells: independence from estrogen receptors and the multi-drug determination of lignans and isoflavonoids in human urine, including identifica- resistance gene. Biochem. Biophys. Res. Commun., 179: 661-667, 1991. tion of genistein. Clin. Chim. Acta., 199: 263-278, 1991. 18. Monti, E., and Sinha, B. K. Antiproliferative effect of genistein and adria- 30. Adlercreutz, H., Honjo, H., Higashi, A., Fotsis, T., HSm#{228}lainen, E., Hase- mycin against estrogen-dependent and -independent human breast carcinoma cell gawa, T., and Okada, H. Urinary excretion of lignans and isoflavonoid phy- lines. Anticancer Res., 14: 1221-1226, 1994. toestrogens in Japanese men and women consuming a traditional Japanese diet. 19. Osawa, K., Yasuda, H., Maruyama, T., Morita, H., Takeya, K., and Itokawa, Am. J. Clin. Nutr., 54: 1093-1100, 1991. H. Isoflavonones from the heartwood of Swartzia polvphvlla and their 31. Adlercreutz, H., Fotsis, T., Lampe, J., WShSlS, K., M#{228}kel#{228},T.,Brunow, G., antibacterial activity against cariogenic bacteria. Chem. Pharmacol. Bull. and Hase, T. Quantitative determination of lignans and isoflavonoids in plasma of (Tokyo). 40: 2970-2974, 1992. omnivorous and vegetarian women by isotope dilution gas chromatography-mass 20. linuma, M., Tanaka, T., Mizuno. M., Yamamoto, H., Kobayashi, Y., spectrometry. Scand. J. Clin. Lab. Invest. Suppl., 215: 5-18, 1993. and Yonemori, S. Phenolic constituents in Erv:hrina x bidwilli and their activity 32. Adlercreutz, H., Markkanen, H., and Watanabe, S. Plasma concentrations of against oral microbial organisms. Chem. Pharmacol. Bull. (Tokyo), 40: 2749- phyto-estrogens in Japanese men. Lancet, 342: 1209-1210, 1993. 2752, 1992. 21. Campbell. D. R., and Kurzer, M. S. Flavonoid inhibition of aromatase 33. Xu, X., Wang, H-J., Murphy, P. A., Cook, L., and Hendrich, S. Daidzein is enzyme activity in human preadipocytes. J. Steroid Biochem. Mol. Biol., 46: a more bioavailable soymilk isoflavone than is genistein in adult women. J. Nutr., 381-388. 1993. 124: 825-832, 1994. 22. Herrera, M. D., Marhuenterodiola, E., and Gibson, A. Effects of genistein, an 34. Adlercreutz, H., Fotsis, T., Kurzer, M. S., W#{228}h#{228}l#{228},K.,MSkelS, T., and Hase, isoflavone isolated from Genusta tridentaza, on isolated guinea-pig ileum and T. Isotope dilution gas chromatographic-mass spectrometric method for the guinea-pig ileai myenteric plexus. Planta. Med., 58: 314-316, 1992. determination of unconjugated lignans and isoflavonoids in human feces, with 23. Wu, E. S., Loch, J. T. 3rd, Toder, B. H., Borrelli, A. R., Gawlak, D., Radov, preliminary results in omnivorous and vegetarian women. Anal. Biochem. 225: L. A., and Gensmantel, N. P. Flavones. III. Synthesis, biological activities, and 101-108, 1995 conformational analysis of isoflavone derivatives and related compounds. J. Med. 35. Adlercreutz, H. Diet and sex hormone metabolism. in: I. R. Rowland (ed), Chem., 35: 3519-3525, 1992. Nutrition, Toxicity and Cancer, pp. 137-195. Boston: CRC Press, 1991. 24. Gopalan, R., Gracias, D., and Madhavan, M. Serum lipid and lipoprotein fractions in bengal gram and biochanin A induced alterations in atherosclerosis. 36. SAS Institute, Inc. Statistical Analysis System, Ver. 6.08 (CMS). Cary, NC: Indian Heart J., 43: 185-189, 1991. SAS Institute, Inc., 1993. 25. Borriello, S. P., Setcheil, K. D. R., Axelson, M., and Lawson, A. M. 37. Setchell, K. D. R., Borriello, S. P., Hulme, P., Kirk, D. N., and Mclean, M. Production and metabolism of lignans by the human faecal flora. J. AppI. Nonsteroidal estrogens of dietary origin: possible roles in hormone-dependent Bacteriol., 58: 37-43, 1985. disease. Am. J. Clin. Nutr., 40: 569-578, 1984.

M S Kurzer, J W Lampe, M C Martini, et al.

Cancer Epidemiol Biomarkers Prev 1995;4:353-358.

Updated version Access the most recent version of this article at: http://cebp.aacrjournals.org/content/4/4/353

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/4/4/353. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.