Genistein Affects ER - But Not ER -Dependent ... - Semantic Scholar

0013-7227/02/$15.00/0 Printed in U.S.A.

Endocrinology 143(6):2189 –2197 Copyright © 2002 by The Endocrine Society

Genistein Affects ER␤- But Not ER␣-Dependent Gene Expression in the Hypothalamus HEATHER B. PATISAUL, MELISSA MELBY, PATRICIA L. WHITTEN,

AND

LARRY J. YOUNG

Center for Behavioral Neuroscience (H.B.P., P.L.W., L.J.Y.), Department of Anthropology (M.M., P.L.W.), and Department of Psychiatry and Behavioral Sciences (L.J.Y.), Emory University, Atlanta, Georgia 30329 Isoflavone phytoestrogens are growing increasingly popular because of their reported cardiovascular and anticarcinogenic properties, but the effects of these compounds in the brain are largely unknown. In a previous study, we found that an isoflavone supplement, containing a mixture of soy phytoestrogens, inhibited estrogen-dependent female sexual behavior and was antiestrogenic for both ER␣- and ER␤-dependent gene expression in the hypothalamus. Here we examined the impact of the soy isoflavone genistein, a major component of the supplement, on estrogen-dependent female sexual behavior and ER␣- and ER␤-dependent gene expression in the rat brain. Genistein, at a dietary concentration of 100 or 500 ppm had no effect on lordosis behavior in rats. However, at 500

ppm genistein had differential activity through ER␣ and ER␤ in the hypothalamus. Genistein had no effect, in either the presence or absence of 17␤-E2, on oxytocin receptor density in the ventromedial nucleus of the hypothalamus, an estrogendependent action thought to be regulated via ER␣. However, genistein increased ER␤ mRNA expression in the paraventricular nucleus of the hypothalamus by 24%, whereas 17␤-E2 decreased ER␤ mRNA expression by 26%, a process likely mediated by ER␤ itself. These results suggest that at this dose, genistein has antiestrogenic action through ER␤ in the paraventriculr nucleus but negligible activity through ER␣ in the brain. (Endocrinology 143: 2189 –2197, 2002)

P

genistein may have antiestrogenic actions through both ER␣ and ER␤ as well; thus, this paper investigates the action of genistein alone on female sexual behavior and estrogendependent gene expression in the brain. ER␣ and ER␤ are differentially expressed in the brain (13, 14), and the ligand binding region of ER␤ is only 55% homologous to the ligand binding region of ER␣ (15), suggesting that the two receptors may have differential effects on gene expression and behavior. In general, phytoestrogens have a higher relative binding affinity for ER␤ than ER␣ (16), suggesting that they may be more bioactive through ER␤ than ER␣. However, a variety of phytoestrogens have now been shown to bind to both ER isoforms in vitro and activate ER-dependent gene transcription (16), although at least one of these assays has demonstrated that the phytoestrogens can have mixed agonist/antagonist effects (17). In vitro competitive binding assays have shown that ER␤ has a higher relative binding affinity for both the coumestan phytoestrogen, coumestrol, and the isoflavone phytoestrogen, genistein, than ER␣ (16, 18). The affinity of genistein to ER␤ is 20-fold higher than the affinity of genistein to ER␣, and genistein has a substantially lower affinity to both ER␣ and ER␤ than 17␤-E2 (16, 18). Despite its extremely low relative binding affinity to ER␣, and preferential affinity to ER␤, genistein has been repeatedly shown to be a full agonist for ER␣ as well as ER␤ in a variety of cell culture receptor reporter-gene assays (16, 19) and proliferation (E-Screen) assays (20). Our in vivo work using both coumestrol and a genistein-rich isoflavone supplement has repeatedly demonstrated that, although the majority of phytoestrogens have been shown to be full agonists in vitro, they can act as antiestrogens in the brain in vivo (10 –12). The following experiments were undertaken to determine if genistein alone is sufficient to produce the antiestrogenic neuroendocrine and

HYTOESTROGENS ARE NONSTEROIDAL, estrogenlike compounds produced by numerous plant species, particularly the legumes. Traditional Asian and vegetarian diets, both of which have a high soy content, contain significant quantities of the phytoestrogen isoflavonoid genistein, which has been suggested to have a myriad of health benefits including anticarcinogenic properties (1, 2). Although the potentially beneficial effects of a high phytoestrogen diet including a reduction of hot flushes (3) and hepatic cholesterol production (4, 5) have been well publicized, very little is known about the potential impact of these compounds on the brain. Traditional estrogen replacement therapy (ERT) is known to confer a variety of cognitive benefits including a decreased risk of depression (6) and improved memory (7–9). As commercially prepared phytoestrogen supplements and soy-enhanced foods are becoming more widely available, and frequently advertised as a natural alternative to ERT, it is imperative to examine the effects of phytoestrogens on the brain. We have recently demonstrated that an isoflavone supplement containing significant quantities of genistein, as well as other phytoestrogens, was antiestrogenic for both ER␣and ER␤-dependent gene transcription in the brain, and interfered with estrogen-dependent sexual behavior (10). We have also shown that the coumestan phytoestrogen, coumestrol, is antiestrogenic through both ER subtypes in the brain (11, 12). The results of these experiments suggest that Abbreviations: EB, E2 benzoate; ER␣KO, ER␣ knockout; ERT, estrogen replacement therapy; GC-MS, gas chromotography-mass spectrometry; GEN 500, 500-ppm genistein diet; LQ, lordosis quotient; OTR, oxytocin receptor; OVX, ovariectomized; PND, postnatal day; PVN, paraventricular nucleus; RT, room temperature; TR-FIA, time-resolved competitive-binding fluoroimmunoassay; VMN, ventromedial nucleus of the hypothalamus.

2189

2190

Patisaul et al. • Genistein Is Antiestrogenic via ER␤

Endocrinology, June 2002, 143(6):2189 –2197

behavioral effects described previously using a commercially prepared isoflavone supplement containing a variety of different phytoestrogens, including genistein (10). Lordosis is a reflexive sexual posture made by female rodents in response to male mounting. Females will display this posture only when sexually receptive. Ovariectomized (OVX) females can be reliably induced into behavioral estrus by administering either E2 benzoate (EB) and progesterone sequentially, or small daily doses of EB alone (for a detailed review, see Ref. 21). Female sexual behavior is absent in ER␣ knockout (ER␣KO) but not ER␤ knockout (ER␤KO) mice, demonstrating that ER␣ is critically important for the regulation of female sexual behavior, including lordosis (22–24). Thus, if genistein is antiestrogenic, particularly through ER␣, it should attenuate the lordosis response in hormonally primed OVX females. To distinguish through which ER subtype genistein is acting, we have chosen to look at two regions of the hypothalamus, each of which contains predominantly one ER subtype. Oxytocin receptor (OTR) density in the ventromedial nucleus of the hypothalamus (VMN) was used as a marker for ER␣ activity because this region contains little or no ER␤ (25), and estrogen is known to regulate OTR via an ER␣-dependent mechanism (26) through an increase in OTR mRNA expression (27). If genistein is antiestrogenic through ER␣, it should either decrease OTR density in the absence of estrogen, or attenuate the estrogen-dependent up-regulation of this receptor in the presence of estrogen. ER␤ mRNA expression in the paraventricular nucleus (PVN) was chosen as a marker of ER␤ activity because this region is devoid of ER␣, and estrogen down-regulates ER␤ mRNA in this region (11). Again, if genistein is antiestrogenic through ER␤, it should either have an opposite effect of estrogen on ER␤ mRNA expression, or interfere with the estrogen-dependent down-regulation of this receptor. Because phytoestrogenic effects are dose specific, and higher doses are often needed in rats to produce effects comparable to those seen in humans (28), we were careful to choose a dietary dose of genistein that would produce plasma genistein levels equivalent to those seen in humans consuming an isoflavone supplemented, or traditional Asian diet. We also examined whether or not genistein is estrogenic in the uterus at the low doses selected for the neuroendocrine experiments by assessing uterine weight in juvenile rats. The juvenile rat uterotrophic assay has frequently been employed to assess the estrogenicity of various compounds, including a number of phytoestrogens (29). Because newly weaned pups have not yet reached puberty, their endogenous estrogen levels are low and their uteri immature. Administration of estrogen, or an estrogenic compound, stimulates uterine growth and thus greatly increases uterine weight. This simple test is therefore a good way to assess estrogenicity. Materials and Methods Juvenile rat uterotropic assay Animal care, maintenance, and surgery were conducted in accordance with the applicable portions of the Animal Welfare Act and the U.S. Department of Health and Human Services “Guide for the Care and Use of Laboratory Animals.” All juveniles were selected from an inhouse colony of Long Evans rats housed in a 12-h light, 12-h dark cycle

at 23 C and 50% humidity. Female pups (n ⫽ 50) were weaned on postnatal day (PND) 21 and randomly assorted into groups of 10. The dams of each pup were maintained on a semipurified, phytoestrogenfree diet (AIN-76A Test Diet, PMI Nutrition International, Inc., Richmond, IN) beginning on the day of parturition. On the day of weaning (PND 21), the pups (n ⫽ 10 per group) were placed on either the phytoestrogen-free diet, or a genistein (Indofine Chemical Company Inc., Somerville, NJ) diet at a concentration of either 500, 1000, 2000, or 4000 ppm. The diets were prepared by mixing the genistein powder into the semipurified diet at the desired concentration. On PND 26, all pups were killed by CO2 asphyxiation, the uteri removed and weighed, and blood collected and chilled over ice. Due to incorrect sexing at the time of weaning, each group ended up with only 7 or 8 females.

Estrogen RIA Immediately after collection, all blood samples were spun for 10 min at a relative centrifugal force of 9.8 to separate the plasma. The plasma was then extracted and stored at ⫺80 C until analysis. Each sample was ether extracted, and the estrogen content of each plasma sample was quantified by RIA with a modification of the Pantex Direct 125I Estradiol kit (Pantex, Santa Monica, CA) as described previously (30, 31). The assay has a sensitivity of 9 pg/ml.

Plasma genistein quantification and analysis We examined blood levels of genistein from adult Long Evans females (Charles River Laboratories, Inc., Raleigh, NC) (n ⫽ 6). Many of these animals were ultimately used for round 1 of the behavioral experiments described below. Treatment consisted of the sequential administration of our standard lab chow (Lab Diet 5001, PMI Nutrition International, Inc.), the phytoestrogen-free diet (AIN-76A), a 100 ppm genistein diet (GEN 100), a 500-ppm genistein diet (GEN 500), and a 0.35% isoflavone supplemented diet (SUPPL) used in our previous experiments (10). The supplement is called “Super Concentrated Isoflavones with Genistein and Daidzein” (Solgar, Leonia, NJ) and is widely available to consumers via health food and grocery stores. The genistein diets were prepared by mixing the genistein powder into the semipurified diet at the specified concentration. Treatment order was randomly distributed, and each treatment was preceded by 1 wk on the semipurified diet. Blood was collected after 6 d on each of the five diets, for a total of five samples per animal. All animals were lightly anesthetized with isoflurane (Abbott Laboratories, Abbott Park, IL) and blood was collected by nicking the tail of each animal with a sterile razor and dripping one drop of blood onto each of three to six wells on sample collection filter paper (Schleicher and Schuell, Keene, NH; no. 903). Papers were allowed to dry for approximately 4 h at room temperature, placed in a zip-lock bag, then stored in a sealed container at ⫺27 C until the time of the assay. Concentrations of genistein in dried blood spots were determined using a time-resolved competitive-binding fluoroimmunoassay (TRFIA) in which europium-labeled genistein competes with unlabeled genistein for (rabbit antigenistein) antibody binding sites. This assay was originally developed for measurement of genistein in plasma and serum by Wang et al. (32) and has been modified by us for use with dried blood spots (30). This method differs from the one developed by Wang et al. in that it does not involve hydrolysis or extraction of samples, and thus only measures the free (unconjugated) phytoestrogens and the 4⬘monosulfates and monoglucuronides (most conjugated phytoestrogens are in this form). The biggest discrepancy between the direct TR-FIA and gas chromotography-mass spectrometry (GC-MS) methods is due to 7⬘-compounds. Blood spot standards (range 0 – 640 ng/ml) were prepared by mixing serially diluted solutions of genistein (Sigma, St. Louis, MO) in 5% BSA Tris buffer (pH 7.75) 1:2 with (human) red blood cells that have been washed three times with normal saline (0.86 g NaCl/100 ml DI H2O). After mixing for 1 h, aliquots of blood spot standards and controls were pipetted onto sample collection papers (Schleicher and Schuell, no. 903), allowed to dry for approximately 4 h at room temperature (RT), and stored in a sealed container at ⫺27 C (i.e. treated identically to samples). Using a 1/8 inch hole punch, two spots (equivalent to 5.0 ␮l of plasma/serum or 10 ␮l of whole blood) from each of the blood spot standards, controls and samples were eluted in 100 ␮l of 0.5% BSA Tris


Endocrinology, June 2002, 143(6):2189 –2197 2191

buffer in glass tubes. Tubes were covered with Parafilm, rotated gently (50 rpm) on an automatic shaker for 1 h at room temperature, and incubated overnight at 4 C. On the following day, after mixing at 50 rpm for 1 h at RT, standards, controls and samples (40 ␮l) were pipetted into prewashed coated wells (goat antirabbit IgG Microtitre Plates (Wallac, Inc. Turku, Finland). Next, 100 ␮l of genistein antibody (an antibody obtained from rabbits immunized with genistein-BSA conjugates, and a generous gift from H. Adlercreutz) solution (1:80,000 dilution in 0.5% BSA Tris buffer) was pipetted into each well, followed by 100 ␮l of europium-labeled genistein derivative [4⬘-O-carboxy-methyl-genistein (Wallac, Inc.)] solution (1:500,000 dilution in 0.5% BSA Tris buffer). Assay wells were covered and incubated on an automatic plate shaker for 90 min at 50 rpm and RT. After washing wells to remove unbound antigen using an automatic washer, 200 ␮l of DELFIA enhancement solution (Wallac, Inc.) were added to each well. Assay wells were covered and placed on an automatic plate shaker for 45 min at 50 rpm and RT. Fluorescence was measured using a 1232 DELFIA fluorometer (Wallac, Inc.). The log of the standard curve, with spline data reduction, was used for interpolation of results. Assay performance characteristics are as follows: sensitivity (8 ng/ml); average recovery (100.8 –107.5%); linearity (96.3–105.2%); intraassay coefficient of variation (6.8 –11.1, n ⫽ 10) and interassay coefficient of variation (11.1–17.7, n ⫽ 13). All values below the sensitivity level were reduced to zero. The majority of these values were for animals fed the phytoestrogen-free diet. One data point was identified as an outlier and dropped from the analysis.

been successfully employed by other experimenters to determine how plasma thyroid hormone levels interact with estrogen-induced lordosis behavior (34, 35). A new set of animals were purchased (Charles River Laboratories, Inc.) and housed in identical treatment conditions as round 1. Male (n ⫽ 12) and female (n ⫽ 26) Long Evans rats (Charles River Laboratories, Inc.) were acclimated to the vivarium for 1 wk. The females were then OVX under ketamine anesthesia. Beginning 3 wk post surgery, the animals were primed with EB and progesterone and paired with males as described in round 1 in order ensure that each animal had sexual experience. Testing began 1 wk after the completion of this initial priming, and all females were fed the phytoestrogen-free diet for the entire week before testing. The females were divided into four treatment groups: phytoestrogenfree diet and no estrogen (n ⫽ 8; control), phytoestrogen-free diet and estrogen (n ⫽ 6), GEN 500 diet and no estrogen (n ⫽ 8), and GEN 500 diet and estrogen (n ⫽ 6). EB was administered daily (0.2 ␮g/0.1 ml sesame oil) for 8 d and control animals were injected with sesame oil only (0.1 ml). The treatment diet was given daily beginning on the day of the first injection. Testing took place for 30 min, every other day, beginning on d 2, under red light in the first 4 h of the dark cycle as described in round 1. Again, all behavior was recorded and scored from the videotape as described in round 1.

Female sexual behavior

The adult female (n ⫽ 26) Long Evans rats used in round 2 of the behavioral experiments were left undisturbed in the vivarium on standard chow for 2 wk following the completion of the behavioral testing. After 3 d on the phytoestrogen-free diet, all animals were divided into the same four groups used for the behavioral experiments: phytoestrogen-free diet and no estrogen (n ⫽ 6), phytoestrogen-free diet and estrogen (n ⫽ 6), GEN 500 diet and no estrogen (n ⫽ 6), and GEN 500 diet and estrogen (n ⫽ 7). To ensure that estrogen administration was continuous across the entire treatment period, 17␤-E2 was administered by SILASTIC brand (Dow Corning, Midland, MI) capsule (6 mm in length, 1.98 mm inner diameter, 3.2 mm outer diameter). Capsules of a similar size have previously been shown to produce physiological levels of plasma E2 (10). All animals were implanted with either an estrogenfilled or empty capsule under isoflurane (Abbott Laboratories) anesthesia 2 d after the initial administration of the treatment diet. Four days after implantation, animals were killed by CO2 asphyxiation for a total of 6 d on the treatment diet and 4 d on the hormone treatment. Uteri were collected and weighed at the time the rats were killed. Blood was collected and spun down within 20 min of collection to isolate the plasma. All brains were immediately frozen on dry ice at the time the rats were killed. The plasma and brains were kept at ⫺80 C until use. The estrogen content of each plasma sample was quantified by RIA as was done for the serum from the juveniles used for the uterotropic assay. The brains were cut on a cyrostat into sections 20 ␮m thick at 80-␮m intervals and thaw mounted on Superfrost plus microscope slides (Fisher Scientific, Pittsburgh, PA). Serial sections were taken from the lateral septum to the caudal end of the VMN (corresponding to Bregma ⫺0.3 mm to ⫺3.0 mm).

Round 1 (100 ppm). Adult male (n ⫽ 8) and female (n ⫽ 32). Long Evans rats (Charles River Laboratories, Inc.) were housed in a 12-h light, 12-h dark cycle at 23 C and 50% humidity. Females were housed in groups of two, and males were individually housed. All animals were maintained on the phytoestrogen-free diet for at least 4 d before beginning the experiment. Following a 1-wk acclimation to the vivarium, all females were OVX under ketamine anesthesia. Within a few weeks after surgery, all females were induced into estrus with EB (10 ␮g in 0.1 ml sesame oil) and progesterone (500 ␮g in 0.1 ml sesame oil) and tested with males to confirm each male’s vigor and each female’s sexual responsiveness to ovarian hormones. Each female was tested at least twice with multiple males, and each male was tested at least four times. Males that did not reliably engage in sexual behavior after repeated exposures (reliability ⫽ mating in at least 80% of pairings) were not used in the study (n ⫽ 5 vigorous males). Testing began 2 wk after the completion of this initial priming, and all females were placed on the phytoestrogenfree diet 3 d before receiving treatment diet. The females were divided into four treatment groups (n ⫽ 8 per group): phytoestrogen-free diet and no estrogen (control), phytostrogenfree diet and estrogen, GEN 100 diet and no estrogen, and GEN 100 diet and estrogen. All animals were given progesterone. EB (10 ␮g in 0.1 ml sesame oil) was sc injected 48 h before testing, and progesterone (500 ␮g in 0.1 ml sesame oil) was sc injected 4 –5 h before testing. Animals not receiving EB were injected with sesame oil (0.1 ml) only. Testing took place in an arena identical in size and shape to the home cage of each animal, which was thoroughly cleaned between trials. The females were given 5 min to adapt to the arena before the male was introduced. Testing sessions were 30 min in length, and all interactions were videotaped under red light and scored from the tape. Receptivity was assessed using the lordosis quotient (LQ) as calculated by the number of lordosis responses in 10 min divided by the number of mount attempts made in the same 10 min multiplied by 100. To eliminate variability between subjects due to mount latency, the beginning of the 10 min scoring period was defined as 5 sec before the first mount attempt by the male. Round 2 (500 ppm). In the second round of testing, we used a higher dose of genistein and a behavioral screen that is more sensitive to antiestrogenic effects. In this test, rats are injected daily with low doses of EB and tested every other day. Typically, after a few days, the rats begin to show lordosis behavior even in the absence of progesterone. LQ, and thus presumably sexual receptivity, is directly proportional to the amount of estrogen administered (33). By giving small doses of EB gradually over several days together with a genistein diet, the effects of genistein on LQ can be determined across a range of estrogen levels. This technique has

Brain collection and analysis

In situ hybridization ER␤ in situ hybridization was performed using a set of two 48-bp, S-labled oligonucleotide probes (GTG AGG GAC ATC ATC ATG GAG GCC TCG GTG AAG GGC ATG CTG GGA CGG, and GAG CTC CAC AAA GCC AGG GAT TTT CTT AGC CCA GCC AAT CAT GTG CAC). These oligos are complementary to nucleotides 714 –762 and 784 – 832 of rat ER␤ mRNA (GenBank accession no. 2801690) and are 60% and 77% homologous to ER␣, respectively. In situ hybridization was performed using a well-established protocol in our laboratory (36). Following in situ, the rinsed and dried sections were exposed to Kodak BioMax MR film (Eastman Kodak Co., Rochester, NY) with 14C-labled autoradiographic standards (Amersham Pharmacia Biotech, Arlington Heights, IL) for 32 d to produce autoradiograms for quantitative analysis. 35

2192



OTR autoradiography 125

OTR autoradiography was performed using I-d(CH2)5[Tyr(Me)2,Tyr9-NH2]OVT (NEN Life Science Products, Boston, MA) as described previously (26). After air drying, the slides were exposed to BioMax MR film (Kodak) for 48 h. 125I-Labled autoradiographic standards (Amersham Pharmacia Biotech) were included in the cassette for quantification.

Data analysis Utertropic assay. Uterine wet weights and endogenous estrogen levels were analyzed by one-way ANOVA (SYSTAT, SPSS, Inc., Chicago, IL) and group differences were identified using Fisher’s least significant difference post hoc test. Genistein blood assay. The data were log transformed to correct for nonhomogeneous variability then analyzed by a one-way-repeatedmeasures ANOVA. Group differences were identified using Fisher’s least significant difference post hoc test. Female sexual behavior. All behavior was scored from the videotapes by a single investigator, then validated by a second investigator. Both investigators were blind to the treatment groups. Number of mounts by the male, and the number of lordotic responses by the female, were scored for each 10-min test session and averaged across each of the four treatment groups. Lordosis was defined as complete dorsoflexion of the spine in response to a mount by the male as previously described in detail (37). A mount was defined as placement of both front limbs on the hindquarters of the female with or without intromission or ejaculation. All data for round 1, including male mounting, was analyzed by oneway ANOVA using SYSTAT, and group differences were identified using Fisher’s least significant difference post hoc test. For round 2, the data were analyzed by a two-way ANOVA, followed by a one-way ANOVAs examining the effects of treatment and a one-way repeated measures ANOVA examining the effect of test day. Group differences were identified by Fisher’s least significant difference post hoc test. Estrogen-dependent gene expression. ER␤ in situ hybridization and OTR film autoradiograms were analyzed using the public domain NIH Image for Windows program (a free downloadable program at http://rsb. info.nih.gov/nih-image/). Brain regions from three adjacent sections per subject were measured bilaterally from anatomically matched sections, and care was taken to ensure that the area of the regions selected for measurement did not differ by more than 5% between sections and subjects. For the ER␤ autoradiograms, optical densities were converted to nCi/g tissue equivalents, and for the OTR autoradiograms, optical densities were converted to dpm/mg tissue equivalents using 14C and 35 S standards (Amersham Pharmacia Biotech) respectively. All data were analyzed by one-way ANOVA using SYSTAT and group differences were identified using Fisher’s least significant difference post hoc test.

Results Uterotrophic assay

TABLE 1. Effect of dietary genistein on uterine weight and endogenous estrogen levels in intact, immature rats Genistein (ppm)

0 500 1000 2000 4000

n

Uterine weight (mg)

Estrogen (pg/ml)

8 7 7 7 7

47 ⫾ 3 88 ⫾ 6b 94 ⫾ 8b 132 ⫾ 11c 110 ⫾ 6b,c

31.5 ⫾ 7.8a 19.5 ⫾ 1.7b 21.0 ⫾ 4.5a,b 21.8 ⫾ 2.8a,b 13.2 ⫾ 0.8b,c

a

Genistein increased uterine weight in all treatment groups and significantly decreased endogenous estrogen levels in the 500 and 4000 ppm groups. (Values are means ⫾ SEM; P ⬍ 0.05; treatment groups with the same letter (a, b, c) are not statistically different from each other, but are statistically different from groups with a different letter.) TABLE 2. Mean plasma genistein concentrations of humans and female rats after phytoestrogen consumption Diet

Humans (␮M)

Western Japanese Supplementedd Infant formula 100 ppm GEN 500 ppm GEN

0.001– 0.02a 0.41a; 0.1–1.2b 0.14 – 0.38c 2–7b

Rats (␮M)

0.036a

Ref.

54 54, 55 41 42

0.057a 1.159a

Female, ovariectomized rats consuming a 100 ppm genistein (GEN) diet had plasma levels of free genistein equivalent to those seen in humans consuming a Western diet. Rats consuming a 500 ppm GEN diet had plasma levels of free genistein equivalent to those seen in human adults consuming a phytoestrogen supplemented diet, but lower than infants consuming soy infant formula. (a Results obtained using TR-FIA measuring free genistein; b results obtained using GC-MS measuring total genistein; c results obtained using LC-MS measuring total genistein; d rats consuming approx. 0.1 mg GEN per day, humans consuming approximately 5.6 mg GEN per day.)

levels. Females on the GEN 500 diet had the highest plasma genistein levels (1.159 ␮m) and plasma genistein levels produced by the GEN 100 diet (0.057 ␮m) were roughly equivalent to the levels produced by the supplemented diet (0.036 ␮m). The standard vivarium diet also produced plasma genistein levels in a range similar to those seen with the isoflavone supplemented or GEN 100 diets (0.063 ␮m). The GEN 100 diet was used for round 1 of the sexual behavior tests, and the GEN 500 diet was ultimately used for round 2 of the sexual behavior tests and the neuroendocrine assays.

Dietary exposure to genistein at 500, 1000, 2000, and 4000 ppm (100 mg/kg BW/d; 200 mg/kg BW/d; 400 mg/kg BW/d; 800 mg/kg BW/d) resulted in a significant increase (P ⬍ 0.01) in uterine wet weight over the controls, and an overall trend for a decrease in endogenous estrogen levels (P ⬍ 0.08; Table 1). Uterine weights in the 1000, 2000, and 4000 ppm group were at least or greater than 2-fold heavier than the controls.

Female sexual behavior

Plasma assay

Round 2. A steady increase in LQ was seen in animals given EB, while no increase in LQ was seen in animals injected with oil only. The GEN 500 diet had no effect on LQ when given in either the presence or absence of estrogen (Fig. 1B). Again, males showed an equal number of mount attempts across all treatment groups on each day.

Analysis with the TR-FIA assay revealed detectable plasma genistein (free) levels in animals fed either the standard vivarium diet, the 0.35% isoflavone supplemented diet, the GEN 100 or GEN 500 diets (Table 2). Animals fed the phytoestrogen-free diet did not have detectable genistein

Round 1. Estrogen-treated animals showed a robust induction of lordosis behavior (LQ ⫽ 76.3%) compared with the controls (LQ ⫽ 7.5%; P ⬍ 0.001; Fig. 1A). Genistein-treated animals (LQ ⫽ 9.1%) were similar to the controls and genistein intake had no effect on estrogen-treated animals (LQ ⫽ 74%). Males showed an equal mean number of mount attempts across all treatment groups (data not shown).



FIG. 1. Effect of dietary genistein on lordosis behavior in ovariectomized (OVX) female rats in (A) round 1 and (B) round 2. Animals in round 1 (A) were given a single injection of 10 ␮g EB or an equivalent injection of sesame oil followed 48 h later by a single injection of 500 ␮g progesterone then tested 4 h following the progesterone injection. Genistein treated animals received the GEN 100 diet. Animals in round 2 (B) were given daily injections of 2 ␮g EB or an equivalent injection of sesame oil for 8 d. Testing took place every other day beginning on d 2 of the injections. Genistein treated animals received the GEN 500 diet. A, The GEN 100 diet did not affect lordosis behavior in either the presence or absence of estrogen after the sequential administration of estrogen and progesterone. B, OVX females exposed daily to 2 ␮g EB and fed the GEN 500 diet had the same lordosis quotient as females exposed daily to the same dose of EB but fed a genistein-free diet. OVX females treated with genistein alone did not differ from controls. (Values are means ⫾ SEM; *, P ⬍ 0.001.)

OTR binding

Plamsa estrogen concentrations and uterine responses 125

2

9

Binding with the I-d(CH2)5[Tyr(Me) ,Tyr -NH2]OVT ligand resulted in a strong signal in the VMN (Fig. 2A). Quantitative analysis of the signal on the film autoradiograms revealed that treatment with 17␤-E2 resulted in a 165% increase in OTR binding in the VMN compared with the controls (P ⬍ 0.001), and treatment with the GEN 500 diet did not alter this effect (P ⬍ 0.721; Fig. 2B). Treatment with the GEN 500 diet in the absence of estrogen produced no significant change in OTR expression compared with the controls. ER␤ mRNA expression

Hybridization with the antisense oligonucleotide probe for ER␤ mRNA resulted in strong signals for ER␤ mRNA in the PVN (Fig. 3A) but no detectable levels in the VMN (not shown). This is consistent with previous reports of ER␤ distribution in the brain (13, 14, 38, 39). Quantitative analysis of the signal on the film autoradiograms showed that treatment with 17␤-E2 resulted in a 26% decrease in ER␤ mRNA expression in the PVN (Fig. 3B; P ⬍ 0.02) compared with the control group. In contrast, ingestion of the GEN 500 diet resulted in a 24% increase (P ⬍ 0.02) compared with the control group. Treatment with both 17␤-E2 and the GEN 500 diet resulted in a 34% decrease in ER␤ mRNA signal compared with the control group (P ⬍ 0.002), which was not significantly different from the group treated with 17␤-E2 alone (P ⬍ 0.398).

E2 levels in animals given blank capsules averaged 3.67 pg/ml compared with an average of 244.1 pg/ml for those given the implants containing 17␤-E2. Gonadally intact rat females typically have serum 17␤-E2 levels around 86.5 pg/ml during proestrus and 26.4 pg/ml during diestrus (40). Thus, the E2 levels achieved using the implants were about 2.5 times higher than physiological levels during proestrus, and the E2 levels in the OVX animals given blank capsules were even lower than physiological levels seen during diestrus. Treatment with the GEN 500 diet did not significantly increase uterine weight (67 ⫾ 5 mg) compared with the controls (69 ⫾2 mg), or diminish the uterotropic effects of estrogen (352 ⫾27 mg) when given in combination (354 ⫾30 mg). This is markedly different from the results of an immature rat uterotropic bioassay, which showed that the GEN 500 diet significantly increased uterine weight in juvenile females compared with the controls. Discussion

This is the first study to demonstrate that the isoflavone phytoestrogen, genistein, has differential activity on ER␣and ER␤-dependent gene expression in the hypothalamus, and the first to show that genistein is antiestrogenic for ER␤-dependent gene expression in vivo. A GEN 500 diet had the opposite effect of 17␤-E2 on ER␤ mRNA expression in the PVN but no effect on the estrogen-dependent up-regulation

2194


FIG. 2. Effects of estrogen and GEN 500 diet on OTR binding in the VMN of the adult, female, ovariectomized rat brain. A, Film autoradiograms showing OTR binding in the VMN of rats treated with no 17␤-E2 and no genistein (E⫺G⫺), no 17␤-E2 and GEN 500 diet (E⫺G⫹), 17␤-E2 and no genistein (E⫹G⫺) and both 17␤-E2 and 500 ppm dietary genistein (E⫹G⫹). B, Estrogen increased OTR binding in the VMN, and the GEN 500 diet did not alter this effect. The GEN 500 diet had no effect on OTR binding in the VMN in either the presence or absence of estrogen. (Results are expressed as the mean ⫾ SEM; *, P ⬍ 0.001 compared with the controls.)

of OTR binding in the VMN in either the presence or absence of estrogen. The absence of activity through ER␣ in the VMN may partially explain why neither a 100 ppm genistein diet (GEN 100) nor the GEN 500 diet altered lordosis behavior. These results are in contrast to the antiestrogenic activity through both ER␣ and ER␤ seen with a soy isoflavone dietary supplement containing a mixture of phytoestrogens, including genistein (10) and suggest that the antiestrogenic actions of the supplement on female sexual behavior and OTR density are mediated by some other component of the supplement than genistein. These genistein doses used in these experiments were too low to cause uterine growth in adult OVX rats but substan-


FIG. 3. A GEN 500 diet had the opposite effect of estrogen on ER␤ mRNA expression in the PVN of the adult, female, ovariectomized rat brain. A, Film autoradiograms showing ER␤ mRNA expression in the PVN of rats treated with no 17␤-E2 and no genistein (E⫺G⫺), no 17␤-E2 and the GEN 500 diet (E⫺G⫹), 17␤-E2 and no genistein (E⫹G⫺) and both 17␤-E2 and the GEN 500 diet (E⫹G⫹). B, Estrogen decreased, but the GEN 500 diet increased, ER␤ mRNA expression in the PVN. The GEN 500 diet had no effect on ER␤ mRNA expression in the presence of estrogen. (Results are expressed as the mean ⫾ SEM; *, P ⬍ 0.02 compared with the controls.)

tially increased uterine weight in juvenile animals. This increase was coupled with a trend for a decline in endogenous estrogen levels, suggesting that genistein both interferes with endogenous estrogen production, and has estrogenic activity in the uterus. Although the increase in uterine weight with genistein treatment in juveniles was substantial, there was no dose response relationship. This may be because the majority of doses used for this study were high enough to produce maximal increases in uterine weight (equal to or greater than 2-fold).


The juvenile uterotropic assay revealed that a significant increase in uterine weight occurred at a dose of at least 500 ppm but gave no indication as to whether this was a relatively high or low dose for a rat. To choose a physiologically relevant dose for these neuroendocrine studies, we had to determine what dietary dose of genistein produces plasma levels equivalent to those found in humans consuming an isoflavone supplemented or traditional Asian diet. We used a modified version of the TR-FIA genistein assay developed by Wang et al. (32). Plasma levels have traditionally been measured using GC-MS or HPLC. TR-FIA is more sensitive than either of those two methods, although less specific. Circulating genistein is present in both conjugated and unconjugated forms. The unconjugated or “free” genistein (aglycone form) is the hormonally active form, and the TRFIA method primarily detects the “free” form. Wang et al. have validated the TR-FIA method against GC-MS using human plasma and serum samples and have given us the opportunity to modify it for use with dried blood spot samples from humans and rats. Thus, this is the first study to use TR-FIA for rat dried blood spots. The serology study was conducted primarily to determine what dietary concentration of pure genistein would produce plasma levels of genistein (free) equivalent to those produced by the 0.35% isoflavone supplemented diet used in earlier experiments (10). The GEN 100 diet produced plasma levels equivalent to those produced on the 0.35% isoflavone supplemented diet, and thus was chosen for the first round of behavioral experiments. It is interesting to note that animals fed the standard vivarium chow also had plasma levels in this range. This is not totally unexpected, however, given that the primary ingredient in this diet is soy meal. The TR-FIA used for the analysis is highly sensitive but can cross-react with daidzein (2.5%) and other phytoestrogens (32). Thus, the values for the vivarium diet and the isoflavone supplemented diet may be artificially high. The plasma values obtained from this study are all in the range of human plasma levels (Table 2). The GEN 100 diet produced plasma genistein levels slightly higher than plasma levels in humans consuming a traditional Western diet (41). The 500 ppm genistein diet produced plasma genistein levels higher than plasma levels in humans consuming an isoflavone supplemented diet, but similar to plasma levels in humans consuming a traditional Asian diet and lower than plasma levels in infants consuming soy infant formula (42). OTR binding density in the VMN is thought to be mediated by ER␣ (26) through an increase in OTR mRNA expression (27). Our previous work suggests that ER␤ mRNA expression in the PVN is likely mediated by ER␤ (11). Although the GEN 500 diet produced no change in lordosis behavior or OTR regulation in the VMN, there was a significant increase in ER␤-mRNA expression in the PVN. Collectively, these results suggest that the regulation of ER␤mRNA expression in the PVN does not significantly impact lordosis behavior, but that the estrogen-dependent gene expression regulated via ER␣ probably does. This is consistent with multiple studies that have conclusively demonstrated that ER␣ is required for the normal expression of both male and female sexual behavior using ER␣KO mice (22, 43, 44).


Although the significance of ER␣ on estrogen-dependent behavior has been well documented, the role of ER␤ is virtually unknown. Reproductive behavior is relatively normal in both male and female ER␤KO mice (23), indicating that the role of ER␤ in the elicitation and mediation of reproductive behaviors is minimal at best. Although the expression and regulation of ER␣ mRNA in the brain have been well characterized (45– 47), little is known about the regulation of ER␤ mRNA expression in the brain. This is now the third time we have demonstrated that a phytoestrogen, or combination of phytoestrogens, alter ER␤ mRNA expression in the PVN in the opposite direction of estrogen. Given that both in situ hybridization (38, 39, 48) and immunocytochemical studies (14) have failed to detect ER␣ in the rat PVN, this effect is likely to be mediated exclusively via ER␤. Collectively, this suggests that, as a group, the phytoestrogens may be generally antiestrogenic through ER␤ in vivo. Given how little is currently known about the role of ER␤ in the brain, it is difficult to predict what behavioral or neuroendocrine endpoints disruption of ER␤ by phytoestrogens is likely to affect. However, our results clearly demonstrate that the phytoestrogens are not merely mimicking the effects of endogenous estrogen in the brain. At first glance, the effects of genistein on estrogen-dependent gene expression and behavior described in this study may appear to be in contrast to our earlier work demonstrating that a commercially available, phytoestrogen supplement containing a mixture of phytoestrogens, including genistein, antagonized both ER␣- and ER␤-mediated gene expression in the female rat brain, and attenuated estrogendependent female sexual behavior. However, in vitro assays have demonstrated that the relative estrogenic potency of genistein for ER␤ is 30 times higher than for ER␣, and the binding affinity of genistein for ER␣ is 20-fold lower than for ER␤ (16). The results of the present study suggest that genistein alone could be responsible for the down-regulation of ER␤ mRNA expression in the PVN by the isoflavone supplement used in previous experiments (10), but that other elements in the supplement are likely responsible for the effects on ER␣-dependent gene expression and behavior. Given the relatively low binding affinity of genistein for ER␣, it is not surprising that genistein had no effect on the estrogen-dependent regulation of OTR in the VMN, an effect known to be mediated by ER␣ (26). It could be that higher doses are needed to elicit an effect. The dose chosen for this experiment was selected because it produced plasma genistein levels equivalent to those seen in animals consuming the phytoestrogen supplement used in our earlier experiments. High micromolar concentrations of genistein are needed in vivo to inhibit aromatase (3.5 ␮m) and a variety of kinases (2.6 –111 ␮m) (for a review see Ref. 28) and blood levels for this experiment are much lower (1.159 ␮m; Table 2). Additionally, although genistein deposition is abundant in digestive and reproductive tissues, particularly the liver and the prostate, genistein may not be detectable at even picogram levels in the brain at dietary doses below 500 ppm (49, 50), further suggesting that genistein activity in the brain at low physiological doses may be limited. Although genistein has been shown to have transcriptional activity through both ER␣ and ER␤ in cell transfection

2196


assays, the data from this study suggest that at a dietary dose of 500 ppm, genistein is only active via ER␤ in the brain. The different effects found in the transfection-reporter assays and our own in vivo findings, may be due to metabolism of genistein after consumption, or to interactions of the ligandreceptor complex with accessory binding proteins not present in the transfection assay tissue, but found in the PVN. Given our findings, it is possible that genistein is only active through ER␤ and not ER␣. If this proves to be the case, it would make genistein an extremely useful compound for exploring the individual and specific roles of ER␣ and ER␤ in the brain. However, it is essential that the effects of genistein over a much wider range of doses be explored before this assessment can be confirmed. Genistein is one of the most common and abundant isoflavones, found in thousands of plant species, particularly the legumes. Soy and soy-based foods, including soy infant formula, contain the highest concentrations of genistein (51, 52). Soy and isoflavone supplements, like the one used in these experiments, also contain significant quantities of genistein (53). These and other phytoestrogen-rich foods are being heavily advertised as a healthy alternative to traditional ERT and are increasing in popularity as their beneficial effects on the heart and bones become more widely publicized. We have now demonstrated that genistein, coumestrol, and a commercially prepared isoflavone supplement are antiestrogenic in the brain (10, 11). The supplement produced these effects in adult, OVX, rats at levels lower than those needed to cause uterine proliferation in juvenile rats. All of the doses used for these studies were low enough to produce plasma levels in rats that equate to plasma levels of phytoestrogens commonly seen in humans (Table 2). Although effects on humans are difficult to predict from studies on rats, the results of these studies suggest that although phytoestrogens may confer a variety of benefits, including a reduced risk of heart disease and cancer due to their estrogenic activity in the periphery, their antiestrogenic actions in the brain may indicate that they do not confer the same mental health benefits as traditional ERT (6 –9, 53). Acknowledgments We gratefully acknowledge Herman Adlercreutz (University of Helsinki, Helsinki, Finland) and his research team for collaborating with us on the TR-FIA analysis of the rat serum samples, Catherine Sharer and Betsy Russell (Emory University, Atlanta, GA), for their technical assistance, Isadora Rapoport and Margarette Shegog (Emory University) for their assistance with behavioral testing and scoring, and especially Eleni Antzoulatos Aultman (Emory University) for her assistance with many aspects of this project. Received December 3, 2001. Accepted February 8, 2002. Address all correspondence and requests for reprints to: Heather B. Patisaul, Ph.D., Center for Behavioral Neuroscience, Emory University, 954 Gatewood Road Northeast, Atlanta, Georgia 30329. E-mail: [email protected]. This work was funded by the National Science Foundation and Technology Center for Behavioral Neuroscience, Emory University (IBN 9876754), and through the generous financial support of the ARCS Foundation.

References 1. Barnes S 1995 Effect of genistein on in vitro and in vivo models of cancer. J Nutr 125:S777–S783

Patisaul et al. • Genistein Is Antiestrogenic via ER␤ 2. Santell RC, Kieu N, Helferich WG 2000 Genistein inhibits growth of estrogenindependent human breast cancer cells in culture but not in athymic mice. J Nutr 130:1665–1669 3. Nagata C, Shimizu H, Takami R, Hayashi M, Takeda N, Yasuda K 1999 Hot flushes and other menopausal symptoms in relation to soy product intake in Japanese women. Climacteric 2:6 –12 4. Herrington D 2000 Role of estrogens, selective estrogen receptor modulators and phytoestrogens in cardiovascular protection. Can J Cardiol 16:5E–9E 5. Anderson J, Smith B, Washnock C 1999 Cardiovascular and renal benefits of dry soybean intake. Am J Clin Nutr 70(Suppl):464S– 474S 6. Whooley MA, Grady D, Cauley JA 2000 Postmenopausal estrogen therapy and depressive symptoms in older women. J Gen Intern Med 15:535–541 7. Duff SJ, Hampson E 2000 A beneficial effect of estrogen on working memory in postmenopausal women taking hormone replacement therapy. Horm Behav 38:262–276 8. Kimura D 1995 Estrogen replacement therapy may protect against intellectual decline in postmenopausal women. Horm Behav 29:312–321 9. Jacobs DM, Tang MX, Stern Y, Sano M, Marder K, Bell KL, Schofield P, Dooneief G, Gurland B, Mayeux R 1998 Cognitive function in nondemented older women who took estrogen after menopause. Neurology 50:368 –373 10. Patisaul HB, Dindo M, Whitten PL, Young LJ 2001 Soy isoflavone supplements antagonize reproductive behavior and ER␣- and ER␤-dependent gene expression in the brain. Endocrinology 142:2946 –2952 11. Patisaul H, Whitten P, Young L 1999 Regulation of estrogen receptor beta mRNA in the brain: opposite effects of 17␤-estradiol and the phytoestrogen, coumestrol. Brain Res Mol Brain Res 67:165–171 12. Jacob DA, Temple JL, Patisaul HB, Young LJ, Rissman EF 2001 Coumestrol antagonizes neuroendocrine actions of estrogen via the estrogen receptor ␣. Proc Soc Exp Biol Med 226:301–306 13. Shughrue PL, Lane MV, Merchenthaler I 1997 Comparative distribution of estrogen receptor ␣ and ␤ mRNA in the rat central nervous system. J Comp Neurol 388:507–525 14. Li X, Schwartz PE, Rissman EF 1997 Distribution of estrogen receptor-␤-like immunoreactivity in rat forebrain. Neuroendocrinology 66:63– 67 15. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-Å 1996 Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930 16. Kuiper GG, Lemmen JG, Carlsson B, Corton JC, Safe SH, Van Der Saag PT, Van Der Berg B, Gustafsson JA 1998 Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor ␤. Endocrinology 139:4252– 4263 17. Bowers JL, Tyulmenkov VV, Jernigan SC, Klinge CM 2000 Resveratrol acts as a mixed agonist/antagonist for estrogen receptors ␣ and ␤. Endocrinology 141:3657–3667 18. Kuiper GG, Carlsson B, Grandien K, Enmark E, Ha¨ggblad J, Hilsson S, Gustafsson J-Å 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ␣ and ␤. Endocrinology 138: 863– 870 19. Maggiolini M, Bonofiglio D, Marsico S, Panno ML, Cenni B, Picard D, Ando S 2001 Estrogen receptor ␣ mediates the proliferative but not the cytotoxic dose-dependent effects of two major phytoestrogens on human breast cancer cells. Mol Pharmacol 50:595– 602 20. Gutendorf B, Westedorf J 2001 Comparison of an array of in vitro assays for the assessment of the estrogenic potential of natural and synthetic estrogens, phytoestrogens and xenoestrogens. Toxicology 166:79 – 89 21. Pfaff D 1999 Drive. Cambridge, MA: MIT Press 22. Ogawa S, Eng V, Taylor J, Lubahn DB, Korach KS, Pfaff DW 1998 Roles of estrogen receptor-␣ gene expression in reproduction-related behaviors in female mice. Endocrinology 139:5070 –5081 23. Ogawa S, Chan J, Chester AE, Gustafsson JA, Korach KS, Pfaff DW 1999 Survival of reproductive behaviors in estrogen receptor ␤ gene-deficient (␤ERKO) male and female mice. Proc Natl Acad Sci USA 96:12887–12892 24. Rissman EF, Wersinger SR, Fugger HN, Foster TC 1999 Sex with knockout models: behavioral studies of estrogen receptor ␣. Brain Res 835:80 –90 25. Shughrue PJ, Komm B, Merchenthaler I 1996 The distribution of estrogen receptor-␤ mRNA in the rat hypothalamus. Steroids 61:678 – 681 26. Young LJ, Wang Z, Donaldson R, Rissman EF 1998 Estrogen receptor ␣ is essential for induction of oxytocin receptor by estrogen. Neuroreport 9:933–936 27. Bale TL, Dorsa DM 1995 Regulation of oxytocin receptor messenger ribonucleic acid in the ventromedial hypothalamus by testosterone and its metabolites. Endocrinology 136:5135–5138 28. Whitten PL, Patisaul HB 2001 Cross-species and interassay comparisons of phytoestrogen action. Environ Health Perspect 109:5–23 29. Whitten PL, Russell E, Naftolin F 1994 Influence of phytoestrogen diets on estradiol action in the rat uterus. Steroids 59:443– 449 30. Worthman C, Stallings J 1997 Hormone measures in finger-prick blood spot samples: new field methods for reproductive endocrinology. Am J Phys Anthropol 104:1–22 31. Worthman C, Stallings J, Hoffman L 1990 Sensitive salivary estradiol assay for monitoring ovarian function. Clin Chem 36:1769 –1773 32. Wang GJ, Lapcik O, Hampl R, Uehara M, Al-Maharik N, Stumpt K, Mikola


33. 34. 35.

36.

37.

38.

39.

40.

41.

42. 43.

44.

H, Wahala K, Adlercreutz H 2000 Time-resolved fluoroimmunoassay of plasma daidzein and genistein. Steroids 65:339 –348 Hardy D, DeBold J 1971 The relationship between levels of exogenous hormones and the display of lordosis by the female rat. Horm Behav 2:287–297 Dellovade TL, Zhu Y, Krey L, Pfaff D 1996 Thyroid hormone and estrogen interact to regulate behavior. Proc Natl Acad Sci USA 93:12581–12586 Morgan MA, Dellovade TL, Pfaff D 2000 Effect of thyroid hormone administration on estrogen-induced sex behavior in female mice. Horm Behav 37: 15–22 Wang ZX, Liu Y, Young LJ, Insel TR 2000 Hypothalamic vasopressin gene expression increases in both males and females postpartum in a biparental rodent. J Neuroendocrinol 12:111–120 Hardy D, DeBold J 1971 Effects of mounts without intromission upon the behavior of female rats during the onset of estrogen induced heat. Physiol Behav 7:643– 645 Alves SE, Lopez V, McEwen BS, Weiland NG 1998 Differential colocalization of estrogen receptor ␤ (ER␤) with oxytocin and vasopressin in the paraventricular and supraoptic nuclei of the female rat brain: an immunocytochemical study. Neurobiology 95:3281–3286 Hrabovszky E, Kallo I, Hajszan T, Shughrue PJ, Merchenthaler I, Liposits Z 1998 Expression of estrogen receptor-␤ messenger ribonucleic acid in oxytocin and vasopressin neurons of the rat supraoptic and paraventricular nuclei. Endocrinology 139:2600 –2604 Bale TL, Dorsa DM, Johnston CA 1995 Oxytocin receptor mRNA expression in the ventromedial hypothalamus during the estrous cycle. J Neurosci 15: 5058 –5064 Doerge DR, Chang HC, Churchwell MI, Holder CL 2000 Analysis of soy isoflavone conjugation in vitro and in human blood using liquid chromatography-mass spectrometry. Drug Metab Dispos 28:298 –307 Setchell KDR, Zimmer-Nechemias L, Cai J, Heubi JE 1997 Exposure of infants to phyto-oestrogens from soy-based infant formula. Lancet 350:23–27 Ogawa S, Taylor JA, Lubahn DB, Korach KS, Pfaff DW 1996 Reversal of sex roles in genetic female mice by disruption of estrogen receptor gene. Neuroendocrinology 64:467– 470 Rissman EF, Wersinger SR, Taylor JA, Lubahn DB 1997 Estrogen receptor


45. 46. 47. 48. 49. 50. 51. 52. 53.

54. 55.

function as revealed by knockout studies: neuroendocrine and behavioral aspects. Horm Behav 31:232–243 Simerly RB, Young BJ 1991 Regulation of estrogen receptor messenger ribonucleic acid in rat hypothalamus by sex steroid hormones. Mol Endocrinol 5:424 – 432 Lauber AH, Mobbs CV, Muramatsu M, Pfaff DW 1991 Estrogen receptor messenger RNA expression in rat hypothalamus as a function of genetic sex and estrogen dose. Endocrinology 129:3180 –3186 Lauber AH, Romano GJ, Mobbs CV, Pfaff DW 1991 Estradiol regulation of estrogen receptor messenger ribonucleic acid in rat mediobasal hypothalamus: an in situ hybridization study. J Neuroendocrinol 2:605– 611 Shughrue P, Lane M, Scrimo P, Merchenthaler I 1998 Comparative distribution of estrogen receptor-␣ (ER␣) and ␤ (ER␤) mRNA in the rat pituitary, gonad, and reproductive tract. Steroids 63:498 –504 Chang HC, Churchwell MI, Delclos KB, Newbold RR, Doerge DR 2000 Mass spectrometric determination of genistein tissue distribution in diet-exposed Sprague-Dawley rats. J Nutr 130:1963–1970 Coldham NG, Sauer MJ 2000 Pharmacokinetics of [14C]genistein in the rat: gender-related differences, potential mechanisms of biological action, and implications for human health. Toxicol Appl Pharmacol 164:206 –215 Franke AA, Custer LJ, Cerna CM, Narala K 1995 Rapid HPLC analysis of dietary phytoestrogens from legumes and from human urine. Proc Soc Exp Biol Med 208:18 –26 Setchell KD, Zimmer-Nechemias L, Cai J, Heubi JE 1998 Isoflavone content of infant formulas and the metabolic fate of these phytoestrogens in early life. Am J Clin Nutr 68:1453S–1461S Setchell KD, Brown NM, Desai P, Zimmer-Nechemias L, Wolfe BE, Brashear WT, Kirschner AS, Cassidy A, Heubi JE 2001 Bioavailability of pure isoflavones in healthy humans and analysis of commercial soy isoflavone supplements. J Nutr 131:1362S–1375S Uehara M, Arai Y, Watanabe S, Adlercreutz H 2000 Comparison of plasma and urinary phytoestrogens in Japanese and Finnish women by time-resolved fluoroimmunoassay. Biofactors 12:217–25 Adlercreutz H, Fotsis T, Watanabe S, Lampe J, Wahala K, Makela T, Hase T 1994 Determination of lignans and isoflavonoids in plasma by isotope dilution gas chromatography-mass spectrometry. Cancer Detect Prev 18:259 –271