Reproductive safety studies with genistein in rats

Food and Chemical Toxicology 45 (2007) 1319–1332 www.elsevier.com/locate/foodchemtox

Reproductive safety studies with genistein in rats R. Michael McClain a,*, Erich Wolz b, Alberto Davidovich c, James Edwards b, Jochen Bausch b a

McClain Associates, Toxicology Department, 10 Powder Horn Terr, Randolph, NJ 07869, USA b DSM Nutritional Products Ltd., CH-Wurmisweg 576, CH-4303 Kaiseraugst, Switzerland c DSM Nutritional Products, 45 Waterview Blvd., Parsippany, NJ 07054, USA Received 2 January 2007; accepted 14 January 2007

Abstract Genistein is a phytoestrogen that occurs naturally in the diet and is found in a wide variety of plant-derived foods especially in soybeans and soy-based foods. There is wide spread interest in genistein and related phytoestrogens as chemopreventive agents for a variety of human diseases and cancers based on epidemiologic evidence of reduced cancer rates in populations with a high intake of soy. Soy, and hence its constituents, such as genistein, have been consumed at high levels in several Asian populations for many centuries without any apparent adverse effects and to the contrary, many health benefits have been associated with the ingestion of soy based foods. Concern has been raised, however, of potential adverse effects due to the estrogenic and other activities of the isoflavones and thus a comprehensive series of safety studies was performed with genistein. To assess the teratogenic and fetal toxic potential of genistein, several studies were conducted. Genistein was tested in an in vitro rat whole embryo culture assay (WEC), which is a preliminary screen, for fetotoxic and teratogenic potential, over a concentration range of from 1 to 100 lg/mL. Treatment related anomalies were observed at concentrations of P10 lg and at 100 lg/mL, all embryos were malformed. Two in vivo embryo fetal developmental safety studies were conducted with genistein by oral administration (gavage and dietary admix) in which there was no evidence for a teratogenic effect. In an oral (gavage) embryonic and fetal development pilot study, genistein was administered to rats at dose levels of 0, 20, 150 and 1000 mg/kg/day from days 6–20 of gestation to females that were allowed to litter and rear their offspring up to day 7 of lactation. A slight maternal toxicity at 1000 mg/kg/day was observed as indicated by decreased body weight and food consumption and at this dose, adverse effects in the pups were observed including increased pup mortality, poor general condition, reduced pup body weights, and reduced pup milk uptake. At the high dose of 1000 mg/kg, no external malformations were noted, however some minor visceral and skeletal variations were observed. At the low dose of 20 mg/kg/day, an increased mortality, reduced milk uptake, a decreased % male sex ratio, and decreased body weights during lactation were observed. Due to lack of effects at the mid dose and the small number of animals, a relationship to treatment was considered unlikely. In an oral (dietary admix) Prenatal developmental safety study, genistein was administered to rats at dose levels of 0, 5, 50, 100 and 500 mg/kg/day from day 5–21 of gestation. At 500 mg/kg, maternal body weight and food consumption were markedly reduced. The incidence of resorptions was markedly increased with a corresponding decrease in the number of live fetuses per dam. Fetal body weights were also reduced. No treatment-related teratogenic effects were noted during external, visceral and skeletal examination of fetuses or in body weight normalized anogenital distance. On the basis of these studies, it is concluded that genistein has no teratogenic potential in vivo at very high doses of up to 1000 mg/kg/ day by oral gavage in the embryo-fetal toxicity pilot study or up to 500 mg/kg/day by dietary admix in the Prenatal developmental study even though these doses were maternally toxic and fetal-toxic. In vitro, genistein had teratogenic potential at high concentrations in the WEC screening assay, however this was not predictive of the in vivo findings. On the basis of the definitive Prenatal development study, the NOAEL for maternal toxicity and adverse effects on embryonic development was considered to be 100 mg/kg/day when administered orally by dietary admix.  2007 Elsevier Ltd. All rights reserved. *

Corresponding author. Tel.: +1 973 895 1363; fax: +1 973 895 1393. E-mail address: [email protected] (R.M. McClain).

0278-6915/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2007.01.009

1320

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

Keywords: Genistein; Daidzein; Soy; Estrogen; Phytoestrogen; Isoflavone; Safety; Toxicity; Rats; Reproductive tract; Teratogenicity; Reproductive toxicity blood levels

1. Introduction Genistein is a phytochemical that occurs naturally in the diet and is found in a wide variety of plant-derived foods. The structure of genistein resembles that of endogenous estrogens (Fig. 1) and it has the capability of binding to the estrogen b receptor, but with a much lower affinity than estradiol. Once bound to the receptor genistein can increase the expression of estrogen responsive genes (Miksicek, 1995). Because of its ability to exert estrogenic activity, genistein and related compounds are referred to as phytoestrogens. The phytoestrogens from plant sources can be classified as isoflavones, coumestans, and lignans. These substances are found in a wide variety of foods derived from plants and the isoflavone group of phytoestrogens is found predominately in soybeans and foods containing soy products (Knight and Eden, 1994). The major isoflavones are genistein, daidzein and glycitein, which in their natural state exist primarily as glycosides (Sirtori, 2001). Processing of soybeans for food increases the hydrolysis of the glycoside and increases the concentration of the aglycone (Zhou and Erdman, 1997; Hutchins et al., 1995), which are more read-

Structure of Genistein:

OH OH

HO

O

O

OH Me S S

H S H

S R

H

ily absorbed from the GI tract than the glycoside conjugates (Setchell et al., 1984; Setchell, 2000; Izumi et al., 2000). After oral ingestion, the glycosides are hydrolyzed by bacteria in the large intestine and absorbed. The aglycones are then glucuronidated and sulfated in the liver and undergo enterohepatic recirculation and are excreted primarily in the urine (Barnes et al., 1996). The estrogenic effect of the isoflavones was first recognized due to impairment of fertility in grazing animals (Hsieh et al., 1998; Bennetts et al., 1946; Setchell et al., 1987a, 1987b). Genistein may act as an estrogen receptor agonist or mixed agonist/antagonist. Genistein binds to the estrogen receptor with an affinity of from 100 to 1000-fold less than that of estradiol and can compete with estradiol and displace it from its binding sites. Genistein binding to the estrogen receptor increases the expression of estrogen responsive genes (Hsieh et al., 1998; Wang et al., 1996) and this activity can be blocked by the estrogen receptor antagonist tamoxiphen (Wang et al., 1996). Thus, it has been demonstrated that genistein and related isoflavones have estrogenic activity mediated though the estrogen receptor and one would expect to see the physiologic effects of an estrogenic compound, although at a reduced potency as compared to estradiol. Soy has been a major component of the diet for several Asian populations for centuries, thus soy and its components, such as genistein, have been extensively consumed without any apparent adverse effects. Indeed, there has been considerable interest in soy and its components, such as genistein, for potential chemopreventive effects based on several epidemiologic studies, suggesting reduced cancer rates associated with the consumption of soy-based foods. Several cancers such as breast, prostate and colon and several diseases including osteoporosis, hypercholesteremia, menopausal symptoms, atherosclerosis and heart disease have been of particular interest (Goldwyn et al., 2000; Suthar et al., 2001a, 2001b). This manuscript reports the results of in vitro and in vivo studies with genistein to assess the developmental safety including an in vitro study in embryo culture, a pilot embryonic and fetal development study in rats and a full Prenatal developmental study in rats. Genistein was found not to have teratogenic potential in vivo at very high oral doses of up to 1000 mg/kg/day. In vitro, genistein had teratogenic potential at high concentrations in the WEC; however, this was not predictive of the in vivo findings.

HO Fig. 1. Structure of Genistein: Genistein (Bonestein ) is 4 0 ,5,7-Trihydroxyisoflavone (C15H10O5) with a molecular weight of: 270.24 CAS No. [446-72-0]. Estradiol is Estra-1,3,5(10)-triene-3,17-diol (17b) (C18H24O2) with a molecular weight of: 272.38. Cas No. [50-28-2]. TM

2. Materials and methods Synthetic genistein (DSM Nutritional Products formerly Roche Vitamins) was tested for embryo and fetal toxicity including an in vitro study in

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332 embryo culture, a pilot embryonic and fetal development study in rats and a full Prenatal developmental safety study in rats as follows: 1. An in vitro whole embryo culture (WEC) screening assay for embryotoxic and teratogenic potential tested over a concentration range of from 1 lg/mL (3.7 lmol/L) to 100 lg/mL (370 lmol/L) (Schmitt, 1997). 2. A pilot oral (gavage) embryonic and fetal development study in female Wistar (RORO) rats (10/group) at dose levels of 0 (control), 20, 150 and 1000 mg/kg/day administered from day of gestation 6–20 (Plassmann, 1997). This study, as a pilot study was not a Good Laboratories Practices (GLP) compliant study. 3. An oral (dietary admix) Prenatal developmental study in female Wistar rats (30/group) at dose levels of 0 (control), 5, 50, 100 and 500 mg/kg/ day administered from day 5–21 of gestation (Wolz et al., 2001 according to OECD draft Guidelines 414, June 2000). This study was GLP compliant.

2.1. Test material The genistein used in these studies (Batch 60401B) was stored in a refrigerator under argon or nitrogen and protected from light. The material had a purity of 98.3%, 99.4%, and 98.3% as assayed for the in vitro study, the pilot study, and the Prenatal developmental studies, respectively. For the pilot oral gavage study, genistein was prepared fresh each day and suspended in a vehicle of 5 g carboxymethylcellulose, 5 mL benzyl alcohol, 9 g NaCl, 4 mL Tween 80 in 1000 mL distilled water. For the Prenatal developmental dietary admix study, genistein was mixed with genistein free microgranulated feed at concentrations calculated to achieve target dose levels based on the most recent body weight and food consumption data. Stability in the diet had been determined. The dietary admixtures were stored refrigerated in a sealed container, in a dry place under nitrogen and protected from light.

2.2. Animals The rats used for the pilot study were Wistar rats (RORO) supplied by BRL (Biological Research Laboratories) Ltd., CH-4414 Fullinsdorf, Switzerland. The range of body weights at the start of study was 183– 208 g. For the full Prenatal developmental study, the rats were Wistar: WIST Hanlbm: (SPF) from RCC Ltd. Biotechnology and Animal Breeding Division Wolferstrasse 4, CH-4414 Fullinsdorf, Switzerland. Rats were acclimatized for 5 days in the pilot study and for 16 days for the full Prenatal developmental study.

2.3. Housing Rats were individually housed in Macrolon Type III cages with wood shavings for bedding, under standard laboratory conditions: air-conditioned with approximately 10–15 air changes per hour; monitored for temperature (22 ± 3 C) and relative humidity (40–70%), 12 h artificial fluorescent light/12 h dark with background music played for at least 8 h during the light period. Tap water was supplied ad libitum in drinking bottles. Diet was provided ad-libitum (Kliba 25-343-4) for the pilot study. In the full Prenatal developmental study, genistein-free diet Kliba NIH31 -2050 (Provimi Kliba AG, CH-4303 Kaiseraugst, Switzerland, Batch No. 66/00) was available ad libitum and fed to all females during the acclimatization period.

2.4. Whole embryo culture (WEC) procedures Two in vitro whole embryo culture (WEC) assays for embryotoxic and teratogenic potential were conducted with genistein at concentrations of 0, 1, 10 and 100 lg/mL (0, 3.7, 37 and 370 lmol/L) in the first assay and 0, 3, 10 and 30 lg/mL (0, 11, 37 and 111 lmol/L) in a second assay. The WEC was performed according to a modified method of New (1978). Male and

1321

female Ibm:WSA rats were mated and the females were killed and hysterectomized on about day 9.5 of gestation (the afternoon following mating being counted as day 0.5). Thirty implantations were removed and cultivated in flasks containing a mixture (1:1) of heat inactivated male rat serum and William’s medium E. Three flasks each with 2 embryos were cultured in the absence of genistein (control) and for the remaining bottles (8 embryos per dose) in the presence of genistein dissolved/suspended in gelatin at various concentrations. At the end of a 48 h culture period, the embryos were assessed for anomalies, growth and degree of differentiation using a modified morphological scoring system of Brown and Fabro (1981). In addition, the embryo and the yolk sac were examined and scored for general signs of toxicity (e.g. heartbeat, rotation of the embryo into a concave position, and vascularization/blood circulation of the yolk sac). General toxic effects were not considered to indicate specific embryotoxicity, since general toxicity on the embryo will be induced by virtually any compound if added at sufficiently high concentrations (Piersma et al., 1995).

2.5. Pilot embryonic and fetal development study procedures 2.5.1. Treatment A pilot oral (gavage) embryonic and fetal development study was conducted in female Wistar (RORO) rats (10/group) with genistein suspensions at dose levels of 0 (control), 20, 150 and 1000 mg/kg/day. Genistein was administered orally by gavage, based on the most recent daily body weight determinations from day 6–20 of gestation, inclusive. The day of mating was designated day 0 of gestation. Dosing was performed in the morning at approximately the same time each day. 2.5.2. Mating A stock of untreated male rats was used for mating and one female and one male were placed together overnight. On the following morning, a copulation plug on the refuse tray or a vaginal plug indicated that mating had occurred (designated day 0 of gestation). This mating procedure was repeated until enough mated females were available, which were randomly assigned to groups and dosed in a random order. 2.5.3. In life observations of dams and pups during the study Rats were observed at least once daily for changes of behavior, general condition and signs of pharmacological effects. Body weight determinations were made for dams on days 0 and 6–21 of gestation and on days 1, 4 and 6 of lactation (day of parturition is designated day 1 of lactation). Pups were weighed on day 1, 4 and 6 of life. Food consumption measurements for dams were made on days 0, 7, 14 and 21 of gestation (21 if possible, depending on the parturition day) and on days 1 and 6 of lactation. Pup milk uptake was qualitatively estimated on days 2–6 by evaluation of the milk patch seen in the stomach through the abdominal wall. Litter size determinations were made on days of lactation 1–7. 2.5.4. Pup necropsy and sex determination All females were allowed to litter and rear their young up to day 7 of lactation after which they were euthanized by carbon dioxide inhalation and necropsied. The organs were examined macroscopically. The pups were euthanized by carbon dioxide inhalation on lactation day 7 and subjected to external examination and sex determination. A sex determination was also performed on day 1 of lactation. All pups were necropsied and examined for visceral alterations and a final sex determination using a fresh dissection method (Sterz, 1977). The skull was not opened and the brains were not examined. Following dissection, all skeletons were processed and stained with Alizarin Red S. Pups that were found dead during the study were also processed and stained with Alizarin Red. Skeletal examinations were only performed for pups from the control and high dose groups. Approximately one week after the expected date of delivery, non-littering females were euthanized (carbon dioxide inhalation), necropsied and a macroscopic examination was performed. Structural alterations in pups were defined according to four categories, namely,

1322

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

abnormality (synonymous with malformation), variation, retardation and incidental (changes considered to be artifactual). For each group the following indices were calculated: Female fertility index, Gestation index, Live birth index, Viability index, and Lactation index.

categories, namely, abnormality, variation, and skeletal variant. The skeletal variant category included certain types of skeletal variation such as number and shape of ribs, and the degree of ossification of phalangeal nuclei and sternebrae.

2.7. Statistical methods 2.6. Prenatal developmental safety study procedures 2.6.1. Treatment An oral (dietary admix) Prenatal developmental study was conducted in female Wistar rats (30/group) with genistein at dose levels of 0 (control), 5, 50, 100 and 500 mg/kg/day administered from day 5–21 of gestation. The study was designed to conform with OECD draft guideline 414, June 2000. 2.6.2. Mating After acclimatization, females were housed with the proven male breeders (one male:one female) in special mating cages until evidence of copulation was observed. The day that evidence of mating was observed was designated day 0 of gestation. Mated rats were assigned to groups using a computer-generated random algorithm. 2.6.3. Blood levels Plasma samples were taken from five females per group early on day 21 of gestation and were assayed by LC/MS for free genistein and genistein conjugates after enzymatic cleavage to provide data for both the levels of free and total genistein. 2.6.4. Observations Animals were checked for mortality, morbidity, signs of abortion or premature delivery, and adverse clinical signs at least twice daily. Body weights were recorded daily from day 0 until day 21 of gestation and food consumption was recorded for the following intervals: days 0–3, 3–5, 5–9, 9–12, 12–15, 15–18 and 18–21, days of gestation. The concentration of genistein in the diet to achieve the target dose was changed based on the body weight and food consumption from the previous interval. The intake of genistein ranged from 96% to 103% of the target dose.

2.7.1. Rat WEC assay statistical methods The results are statistically analyzed by the Chi-square and Fisher’s exact-test (for anomalies), and the ANOVA and Dunnett-test (for growth and differentiation parameters). 2.7.2. Pilot developmental study statistical methods In the tables for each group, numbers of individuals tested (N), means, standard deviations, or medians are indicated. For statistical analysis, the median was used. For the statistical examinations, the following tests are applied: Chi-square and Fisher’s Exact Test for categorical variables. ANOVA, and Dunnett’s Test for parameters approximately normally distributed. Kruskal–Wallis-Test and Mann–Whitney–Wilcoxon-Test for all other parameters. The data are generally expressed as the median. 2.7.3. Prenatal developmental safety study statistical methods The following statistical methods were used to analyze body weights, food consumption, uterus weights, anogenital distances, reproduction and skeletal examination data: Means and standard deviations of various data were calculated. If the variables can be assumed to follow a normal distribution, the Dunnett many-one t-test, based on a pooled variance estimate, were used for intergroup comparisons (i.e. single treatment groups against the control group). The Steel test (many-one rank test) was applied when the data cannot be assumed to follow a normal distribution. Fisher’s Exact test for 2 · 2 tables were applied if the variables could be dichotomized without loss of information. Ratios of anogenital distances to the cube root of body weight were calculated to normalize for different fetal body weights (Gallavan et al., 1999).

3. Results 3.1. In-vitro whole embryo culture study results

2.6.5. Caesarean section and post mortem examination On day 21 of gestation, the females were killed by CO2 asphyxiation and examined for macroscopic abnormalities with emphasis on the uterus, uterine contents, position of fetuses in the uterus and number of corpora lutea in each ovary. The uteri (and contents) of females with live fetuses were weighed for the calculation of the corrected (for uterine contents) body weight gain; the uteri were also weighed without contents. If no implantation sites were evident, the uterus was placed in an aqueous solution of ammonium sulfide to accentuate the visualization of implantation sites. The number and distribution of implantations in uterine horns, classified as empty implantation sites, embryonic resorptions, fetal resorptions, dead fetuses or live fetuses were recorded. The fetuses were removed from the uterus, sexed, weighed individually, and examined for external abnormalities. In addition, the anogenital distances were measured after the external examination of fetuses to assess any potential hormonal effects of genistein on the fetuses using a stereomicroscope. After sacrifice using sodium pentobarbital, fetuses were allocated to one of the following procedures: (1) Wilson’s slicing technique for examination of the viscera and brain (Wilson, 1965). One half of the live fetuses from each litter was fixed in Bouin’s fixative and was examined by serial sections of the head, thorax and abdomen. This included detailed examination of the major blood vessels and sectioning of the heart and kidneys. After examination, the sections were preserved in a solution of ethyl alcohol and glycerin. (2) The remaining fetuses were eviscerated and the carcasses processed through solutions of ethanol, glacial acetic acid with Alcian blue (for cartilage staining), potassium hydroxide with Alizarin red S (for clearing and staining ossified bone) and aqueous glycerin for preservation and storage. The skeletons were examined and preserved individually. Structural alterations in fetuses were defined according to three

In the Whole Embryo Culture screening assay (WEC) at genistein concentrations of 0, 1, 3, 10, 30, and 100 lg/mL (conducted in 2 assays), anomalies considered to be treatment-related were observed at concentrations of P10 lg/ mL (37 lmol/L) (Tables 1 and 2). At 100 lg/mL, all embryos were malformed. There were no embryonic changes at 1 lg/mL (3.7 lmol/L) or 3 lg/mL (11 lmol/L) considered to be anomalies. At 1 lg/mL, a single embryo had its head bent back, but in the absence of head shortening this observation was not classified as an anomaly. This finding was not seen at higher concentrations and was considered not compound related. There were no clear signs of general toxicity to the embryos at 10 lg/mL and 30 lg/ mL, however, the embryonic growth and differentiation were slightly impaired at these two concentrations. In contrast, distinct general toxic effects on the embryo and the yolk sac as well as adverse effects on growth and differentiation were seen at the highest tested concentration (100 lg/ mL). Since anomalies were observed (10 or 30 lg/mL) in the absence of general embryo or yolk sac toxicity, genistein is considered to exhibit a teratogenic potential in vitro in the WEC assay at concentrations of P10 lg/mL (37 lmol/L). There were no anomalies at 1 or 3 lg/mL.

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332 Table 1 Rat whole embryo culture summary of evaluations (first assay)

Table 3 Pilot embryo fetal development study maternal survival, pregnancy status and body weight gain

Dose group (lg/mL)

Embryos assessed Toxicity score Yolk sac toxicity Embryo toxicity Abnormalities Head/brain Sensory organs Heart Rump/tail/limbs Embryos totally malformed Embryos with abnormalities

0

1

10

100

6

8

8

8

0 0

0 0

1 1

16 8

0 0 0 0

0 0 0 0

1 3 1 4

– – – –

0 0

0 0

0 4

1323

8 8***

Genistein was tested at concentrations of 0, 1, 10 and 100 lg/mL. Six control and eight treated embryos per treated group were evaluated. Abnormalities and toxicity scores can exceed the number of embryos. *** p 6 0.001.

Table 2 Rat whole embryo culture summary of evaluations (second assay)

Rearing groups

Dose (mg/kg/day) 0

20

150

1000

Females Non pregnant Pregnant

10 2 8

10 1 9

10 1 9

10 1 9

Dams delivering With live pups %

8 8 100

9 9 100

9 9 100

9 9 100

0

1

0

2

Body weight (gestation) Day 6 Day 21 Weight change (6–21)

212 292 80

216 294 78

219 300 81

215 287 72

Body weight (lactation)a Day 1 Day 6 Weight change (1–6)

230 253 23

225 240 15

220 253 33

205 231 27

Complete litter loss a

Female Wistar rats (10/group) were treated by oral gavage at dose levels of 0, 20, 150, and 1000 mg/kg/day from GD 6–GD 20. a Data represent the median values.

Dose group (lg/mL) 0

3

10

30

Embryos assessed

6

8

8

8

Toxicity score Yolk sac toxicity Embryo toxicity

0 0

0 0

0 0

0 0

Abnormalities Head/brain Sensory organs Heart Rump/tail/limbs

0 0 0 0

0 0 0 0

5 13 4 1

12 21 9 6

Embryos totally malformed Embryos with abnormalities

0 0

0 0

0 5*

0 7**

Genistein was tested at concentrations of 0, 3, 10 and 30 lg/mL. Six control and eight treated embryos per treated group were evaluated. Abnormalities and toxicity scores can exceed the number of embryos. * p 6 0.05. ** p 6 0.01.

3.2. Pilot embryonic and fetal developmental safety study results 3.2.1. General observations, body weight and food consumption In the Pilot oral (gavage) embryonic and fetal development study conducted with genistein suspensions at dose levels of 0 (control), 20, 150 and 1000 mg/kg/day administered from days 6–20 of gestation, there were no maternal deaths and no adverse clinical signs. There was no significant effect on body weight up to 150 mg/kg/day. At 1000 mg/kg/day, body weights were slightly decreased as compared to controls during the first week of treatment and remained reduced throughout the lactation period (Table 3). Maternal food consumption was slightly reduced at 1000 mg/kg/day during treatment. During the lactation

period, at 20 and 1000 mg/kg/day, the median food consumption was reduced, but mainly in females with complete litter loss or high pup mortality and thus considered not a direct effect of treatment. 3.2.2. Reproduction and delivery parameters At 1000 mg/kg/day, 1 female had no milk and lost the entire litter. There was no effect of genistein on the fertility index, gestation index, number of females surviving delivery, duration of gestation, number of delivered pups, the number of implantation sites or the number of resorbed implants (Table 4). 3.2.3. Pup survival At 20 and 1000 mg/kg/day, pup mortality during lactation was significantly increased as compared to controls; however, pup mortality was significantly decreased at 150 mg/kg/day as compared to the controls. Pup mortality was 11, 25, 2 and 26% in the control, low, mid and high dose groups, respectively. One female at the low dose and two females at the high dose showed complete litter loss over days 1–7 of lactation. Poor general condition was observed in 1 pup, 2 pups or all pups in the litters of three females at 1000 mg/kg/day. There was an increased incidence of decreased milk uptake during the lactation period in pups at 20 and 1000 mg/kg/day, but not at 150 mg/kg/ day (Table 4). 3.2.4. Pup body weights and sex ratio Median pup body weights tended to be reduced at 20 and 1000 mg/kg/day, although not significantly. No decrease was observed at 150 mg/kg/day. The sex ratio (% males) was 50, 40, 55, and 53% in the control, low,

1324

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

Table 4 Pilot embryo fetal development study pregnancy and litter data incidence (%) Rearing groups

Table 5 Pilot embryo fetal development study incidence of pup visceral and skeletal abnormalities, variations and retardations incidence (percent)

Dose group mg/kg/day

Dose group (mg/kg/day)

0

20

150

1000

Females on study Females pregnant Fertility index (%) Females with live born Gestation index (%) Females surviving delivery Duration of gestation With stillborn pups (%) Females totally resorbed Pups delivered (total) Live born Live birth index (%) Stillborn

10 8 (80) 8 (100) 8 22 0 (0) 0 80 80 (100) 0

10 9 (90) 9 (100) 9 22 0 (0) 0 91 91 (100) 0

10 9 (90) 9 (100) 9 22 0 (0) 0 93 93 (100) 0

10 9 (90) 9 (100) 9 22 1 (11.1) 0 99 98 (99) 1

Females entire litter lost L days 1–7 (%)

0 (0)

1 (11.1)

0 (0)

2 (22.2)

Pups dead or missing L day 1 (%) L day 2–4 (%) L days 1–7 (%) Pups surviving 4 days Viability index (%) Pups surviving 7 days Lactation index (%) Implantation sites Per litter Resorbed implants (%)

0 (0) 8 (10) 9 (11.3) 72 (90) 71 (88.8) 87 11 7 (8)

1 (1.1) 17 (18.7) 23*(25.3) 73 (80.2) 68* (74.7) 103 12 12 (11.7)

1 (1.1) 1 (1.1) 2*(2.2) 91* (97.8) 91* (97.8) 101 12 8 (7.9)

2 (2) 15 (15.3) 26*(26.5) 81 (82.7) 72* (73.5) 109 13 10 (9.2)

Sex ratio Male/female Male (%)

35/35 (50.0)

27/41 (39.7)

50/41 (54.9)

38/34 (52.8)

Pups without milk Total L days 1–6

4

23

4

28

Female Wistar rats (10/group) were treated by gavage at dose levels of 0, 20, 150, and 1000 mg/kg/day from GD 6–GD 20. * p 6 0.05.

mid and high dose, respectively. The low dose had a lower percentage of males that was considered incidental since there was no effect at the higher doses (Tables 4 and 5). 3.2.5. Pup examinations No external malformations were observed in any of the groups. Visceral evaluation (Table 5) revealed an increase in the retardation finding of ‘‘thymus remnant’’ with no pups affected at 0 and 20 mg/kg/day and 2 and 5 pups from one litter at 150 and 1000 mg/kg/day, respectively. Thymus hypoplasia was observed in one pup at 20 mg/kg/day. The blood vessel variation ‘‘artery origin variant’’ increased in a dose- dependent manner, with no pups in the control group affected and 1.5, 5.5 and 5.6% of the pups affected at 20, 150 and 1000 mg/kg/day, respectively. This increase was significant at 1000 mg/kg/day with respect to the litter incidence (more than half of the litters affected at 1000 mg/kg), but not at 150 mg/kg/day or below. Other variations such as ‘‘innominate artery missing, innominate artery shortened’’ showed no clear dose-dependency and were consid-

Pup weights (g) Day 1 Day 4 Day 6 Visceral examinations (pups) Total abnormalities Pup incidence (%) Litter incidence (%) Total variations Pup incidence (%) Litter incidence (%) Total retardations Pup incidence (%) Litter incidence (%) Skeletal examinations (pups) Total abnormalities Pup incidence (%) Litter incidence (%) Total variations Pup incidence (%) Litter incidence (%) Total retardations Pup incidence (%) Litter incidence (%)

0

20

150

1000

5.6 7.9 10.1

5.1 5.7 7.4

5.9 7.8 10.1

5.2 7.4 9.5

71

68

91

72

1 (1.4) 1 (12.5)

0 (0) 0 (0)

1 (1.1) 1 (11.1)

2 (2.8) 1 (14.3)

9 (12.7) 5 (62.5)

5 (7.4) 4 (50)

12 (13.2) 6 (66.7)

12 (16.7) 5 (71.4)

0 (0) 0 (0)

0 (0) 0 (0)

2 (2.2) 1 (11.1)

5 (6.9) 1 (14.3)

70

–

–

73

4 (5.7) 3 (37.5)

– –

– –

8 (11) 3 (42.8)

18 (25.7) 6 (75)

– –

– –

27 (37) 7 (100)

4 (5.7) 3 (37.5)

– –

– –

9 (12.3) 4 (57.1)

Female Wistar rats (10/group) were treated by gavage at dose levels of 0, 20, 150, and 1000 mg/kg/day from GD 6–GD 20. (–) Skeletal examinations were not performed for the 20 and 150 mg/day groups. (%) Median values. * p 6 0.05.

ered not related to treatment. Other isolated findings showing no dose dependency and considered not related to treatment included 1 pup at 150 mg/kg/day with missing testicle and epididymis, one pup with displaced kidney, 2 pups from one litter at 1000 mg/kg/day with testicle, small. Other findings also considered not treatment-related were convoluted ureter and persistent ductus botalli (1 control pup). Red dots on liver or kidney were an incidental finding in all or most pups from 2 litters at 150 mg/kg/day.

3.2.6. Pup skeletal examinations Skeletal evaluation revealed a slightly higher incidence of abnormalities of sternal elements (fused and misshaped) at 1000 mg/kg/day as compared to controls. An increase in the variation ‘‘bilateral thoracic extra ribs’’ was also noted. All extra ribs were rudimentary. The incidence of incompletely ossified thoracic centers was slightly higher at the high dose level. Pups that died earlier during the study also showed sternal findings such as additional or fused sternal element, different variations and some retardations. The

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

latter are expected due to the fact that these pups were younger when examined. 3.3. Prenatal developmental safety study results 3.3.1. Maternal and litter data In an oral (dietary admix) Prenatal developmental study, genistein was administered at dose levels of 0 (control), 5, 50, 100 and 500 mg/kg/day from day 5–21 of gestation. At the end of the treatment period, mean female body weights were markedly reduced in the high dose group of 500 mg/kg/day (21.3% as compared to controls), but were similar to controls at 50 and 150 mg/kg/ day, respectively. The slight reduction in body weights of females in the low dose group of 5 mg/kg was considered to be incidental and the consequence of a slightly smaller litter size (due to a small increase in preimplantation loss). Body weight gain (corrected for uterus weight) was also reduced on day 21 in high dose females (1.8% vs +10.8% in controls). In the mid doses and low dose, body weight gain, corrected for uterine weight, were similar as compared to controls (+7.8, +8.0 and +7.8%, respectively) (Table 6). Mean food consumption during the treatment period was markedly reduced in females in the high dose group of 500 mg/kg/day (17.8% vs controls) while food consumption of females in the 50 and 150 mg/kg/day groups was similar to that of controls. Low dose females (5 mg/ kg/day) had a slightly reduced food consumption that was considered incidental due the lack of a dose response. Plasma data for the analyses of free genistein and total genistein (free plus conjugated) after enzymatic cleavage showed that the level of free genistein in the plasma was less than 5% of total genistein. The plasma levels increased with dose in a dose proportional manner and reached a

1325

high level of 22873 nmol/L at the high dose of 500 mg/ kg/day. Thus, the animals were highly exposed (Table 7). Among females at 500 mg/kg/day with live fetuses at Caesarean section, postimplantation loss was markedly increased (36% vs 5.4% in controls) due to a marked increase in embryonic resorptions; 13 high dose dams had total resorptions. The incidence of embryonic resorptions in the mid doses and low dose was similar to controls. As a consequence of the increased incidence of resorptions at the high dose, there was a corresponding decrease in the mean number of fetuses per dam as compared to controls (7.3 vs 11.8% in controls). The mean numbers of live fetuses at the mid doses and the low dose were similar to controls (Table 8). In females at 500 mg/kg/day with live fetuses, absolute uterus weights were significantly decreased. (3.0 g vs 4.4 g in controls) and there was a small decrease in relative uterus weight. Uterus weights in the mid doses and low dose were similar to that in controls. In one non-pregnant female at 5 mg/kg/day, both uterus horns contained clear fluid, a common finding that occurs spontaneously in rats of this strain. In two females at 500 mg/kg/day, (both with total post-implantation loss)

Table 7 Prenatal developmental safety study plasma levels of free and total genistein on gestation day 21 (nmol/L) Genistein dose (mg/kg/day)

Free Total

0

5

50

100

500

0 0

0 436 ± 193

87 ± 86 2600 ± 1240

190 ± 134 4912 ± 4289

474 ± 195 22873 ± 10196

Plasma samples were taken from five females per group early on gestation day 21 and were assayed by LC/MS for genistein and genistein conjugates after enzymatic cleavage. All values below the limit of quantification (18 nmol/L) are set to zero.

Table 6 Prenatal developmental safety study maternal body weights and food consumption Dose (mg/kg/day) 0

5

50

100

500

Females Pregnant Non pregnant Dams with live fetuses

30 30 0 30

30 27 3 25

30 29 1 29

30 27 3 26

30 30 0 17

Body weight (g) ± SD Day 5 Day 21 Change 5–21 (%) Adjusted change (%)

223 ± 15 328 ± 22 (+47) (+10.8)

218 ± 14 309 ± 30** (+42) (+7.8)

222 ± 15 319 ± 21 (+44) (+8.0)

223 ± 12 317 ± 19 (+42) (+7.8)

219 ± 15 258 ± 19** (+18) (1.8)

Food consumption Day 5–21 (g/day) % of control (5–21)

27.5 –

25.3 (8)

26.3 (4.4)

27.1 (1.5)

22.6 (17.8)

Female Wistar rats (30/group) were treated orally (admix) at dose levels of 0, 5, 50, 100, and 500 mg/kg/day from GD 5–GD 20. Only dams with at least 1 live fetus were used for body weight and food consumption determinations. Adjusted change body weight refers to change (5–21) adjusted for uterus weight. Data represent the mean and standard deviation. ** P 6 0.01.

1326

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

Table 8 Prenatal developmental safety study pregnancy and embryo-fetal data Dose (mg/kg/day)

Females on study Females pregnant Females with live fetuses Corpora lutea Mean ± SD Preimplantation loss Percent of corpora lutea Mean ± SD Implantation sites Percent of corpora lutea Mean ± SD Postimplantation loss Percent of implantation sites Mean ± SD Number of dams affected Implantation site scars Embryonic/fetal deaths Embryonic resorptions Percent of implantation sites Mean ± SD Number of dams affected Fetal resorptions Percent of implantation sites Mean ± SD Number of dams affected

0

5

50

100

500

30 30 30 410 13.7 ± 1.8 37 (9) 1.2 ± 1.2 373 (91) 12.4 ± 1.6 20 (5.4) 0.7 ± 0.7 16 0 20 16 (4.3) 0.5 ± 0.8 14 4 (1.1) 0.1 ± 0.3 4

30 27 25 319 12.8 ± 1.7 35 (11) 1.4 ± 1.3 284 (89) 11.4 ± 1.8 16 (5.6) 0.6 ± 1.0 11 0 16 14 (4.9) 0.6 ± 0.7 11 2 (0.7) 0.1 ± 0.4 1

30 29 29 385 13.3 ± 1.5 32 (8.3) 1.1 ± 1.0 353 (92) 12.2 ± 1.7 11 (3.1) 0.4 ± 0.7 8 0 11 11 (3.1) 0.4 ± 0.7 8 0 (0) 0±0 0

30 27 26 347 13.3 ± 1.9 31 (8.9) 1.2 ± 1.3 316 (91) 12.2 ± 2.1 10 (3.2) 0.4 ± 0.7 7 0 10 9 (2.8) 0.3 ± 0.6 7 1 (0.3) 0±0 1

30 30 17 220 12.9 ± 2.1 26 (11.1) 1.5 ± 2.3 194 (88) 11.4 ± 1.8 70 (36.1)** 4.1 ± 2.8* 15 0 70 66 (34)** 3.9 ± 3.0* 15 4 (2.1) 0.2 ± 0.6 3

Female Wistar rats (30/group) were treated orally (admix) at dose levels of 0, 5, 50, 100, and 500 mg/kg/day from GD 5–GD 20. * P 6 <0.05. ** P 6 0.01.

both uterine horns contained fluid which in one case was clear and in the other hemorrhagic as a consequence of total embryonic/fetal resorptions. 3.3.2. Fetal data Fetal body weights calculated on a litter basis were reduced in fetuses from high dose females (4.2 g vs 4.9 g in controls) and similarly on an individual basis. Among fetuses from the 5, 50 and 100 mg/kg/day groups, body weights were comparable to controls. There was a higher percentage of male fetuses in litters at the high dose (53.2% males vs 44.2% in the controls) (Table 9). 3.3.3. Anogenital distance The anogenital distance is listed in Table 9 as both the absolute and body weight normalized values for which there are statistically significant differences. Since there were differences in body weight among the groups, the body weight normalized values were used. Although normalized (for fetal body weight), the anogenital distances (mm) for male fetuses were slightly, but statistically significantly decreased at the low dose of 5 mg/kg/day (1.27 vs 1.33 in vehicle controls). However, they were similar to control values in the 50, 100 and 500 mg/kg groups (1.29, 1.29 and 1.30, respectively). In the absence of a dose relationship, the reduced anogenital distances at the low dose were considered incidental and not related to treatment.

For female fetuses, normalized anogenital distances were slightly reduced at the 5 mg/kg/day group with borderline significance in the 50 mg/kg/day group (0.55 and 0.55, respectively vs 0.57 in vehicle controls). The normalized anogenital distance was the same in the controls as in the 100 and 500 mg/kg/day groups (0.57 and 0.57, respectively vs 0.57 in the vehicle controls). In the absence of a dose relationship, reduced anogenital distances in the 5 and 50 mg/kg/day groups for female fetuses were also considered to be incidental and not related to treatment (Table 9). 3.3.4. External examinations There were no external abnormalities considered related to treatment with genistein. In the controls, 1/353 fetuses had a dome-shaped skull (possible hydrocephalus). In the low dose, 1/268 fetuses had an encephalocele and an additional fetus had agnathia. In the mid dose, 1/342 fetuses had agnathia and an additional fetus had a thread-like tail. No abnormal findings were noted in the 306 fetuses of the upper mid dose or in the 124 fetuses of the high dose (Table 9). 3.3.5. Visceral examination During the visceral examination of fetuses, no abnormal findings considered related to treatment with genistein were noted. In the control group, 1/169 fetuses had dilated lateral cerebral ventricles and 14/169 fetuses (12 litters) had

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

1327

Table 9 Prenatal developmental safety study fetal data Dose (mg/kg/day) 0

5

50

100

500

Total fetuses Percent of implantation sites Mean ± SD Live fetuses Dead fetuses Sex ratio of fetuses M/F Male (%)

353 (95) 11.8 ± 1.6 353 0 156/197 (44.2)

268 (94) 10.7 ± 1.6 268 0 125/143 (46.6)

342 (97) 11.8 ± 1.9 342 0 161/181 (47.1)

306 (97) 11.8 ± 2.2 306 0 143/163 (46.7)

124 (64)** 7.3 ± 3.0* 124 0 66/58 (53.2)

Fetus weights (Individual) (g) Males Mean ± SD Females Mean ± SD

156 5 ± 0.4 197 4.7 ± 0.4

125 5 ± 0.4 143 4.8 ± 0.4

161 5 ± 0.5 181 4.7 ± 0.4

143 4.9 ± 0.3 163 4.7 ± 0.4

66 4.3 ± 0.6** 58 4.2 ± 0.4**

External examination External abnormalities Pup incidence (%) No. litters

1/353 (0.3) 1

2/268 (0.8) 2

2/342 (0.6) 2

0/306 (0) 0

0/124 (0) 0

Visceral examination Visceral abnormalities Pup incidence (%) No. litters

15/169 (8.9) 15

7/127 (5.5) 7

14/164 (8.5) 14

11/144 (7.6) 11

6/60 (10) 6

Skeletal examination (Bone) Skeletal abnormalities Pup incidence (%) No. litters

4/184 (2.2) 4

6/141 (4.3) 6

3/178 (1.7) 3

3/162 (1.9) 3

5/64 (7.8) 4

Anogenital distance (mm) ± SD Male fetuses Normalized (body weight) Female fetuses Normalized (body weight)

2.28 ± 0.3 1.33 ± 0.2 0.96 ± 0.1 0.57 ± 0.1

2.17 ± 0.3* 1.27 ± 0.2* 0.92 ± 0.2 0.55 ± 0.1

2.20 ± 0.3 1.29 ± 0.2 0.92 ± 0.2* 0.55 ± 0.1*

2.18 ± 0.3* 1.29 ± 0.2 0.95 ± 0.1 0.57 ± 0.1

2.11 ± 0.4** 1.30 ± 0.2 0.91 ± 0.2 0.57 ± 0.1

Female Wistar rats (30/group) were treated (admix) at dose levels of 0, 5, 50, 100, and 500 mg/kg/day from GD 5–GD 20. * P 6 <0.05. ** P 6 0.01.

a left umbilical artery. In the low dose group, 1/127 fetuses had dilated lateral cerebral ventricles and shortened lower jaw and tongue and 6/127 fetuses (6 litters) had a left umbilical artery. In the lower mid-dose group, 1/164 fetuses had dilated lateral cerebral ventricles, shortened lower jaw and tongue; 1/164 fetuses had absence of the tail and 12/164 fetuses (10 litters) had a left umbilical artery. In the upper mid dose group, 11/144 fetuses (9 litters) had a left umbilical artery. In the high dose group, 6/60 fetuses (6 litters) had a left umbilical artery. Left umbilical artery is a common finding in rats of this strain and was observed in all the groups at a comparable incidence. This is considered to be a variation without functional consequences (Table 9). 3.3.6. Skeletal examinations (abnormal findings) (Bone) During skeletal examination of the fetuses, abnormal skeletal findings for bone were noted in 4/184 fetuses in the control group, 6/141 fetuses in the low dose group, 3/ 178 fetuses in the lower mid dose group, 3/162 fetuses in the upper mid dose group, and 5/64 fetuses in the high dose group. These were commonly observed abnormal findings

that included: asymmetrically ossified or abnormally shaped sternebral centra; wavy or broadened ribs, cervical ribs, abnormally shaped or fused vertebral bodies and others. The frequency of these findings was similar in all groups and therefore considered not to be treatment related (Table 9). 3.3.7. Cartilage examination of fetuses (abnormal findings) During cartilage examination of the fetuses, abnormal findings were noted in 3/184 fetuses of the control group, 5/141 fetuses at 5 mg/kg/day, 3/178 fetuses at 50 mg/kg/ day, 2/162 fetuses at 100 mg/kg/day, and 2/64 examined fetuses (in 2 l) at 500 mg/kg/day. The frequencies of these abnormal findings (asymmetrically chondrified or abnormally shaped intersternebral cartilages, broadened costal cartilages and others) were comparable in all groups, showed no dose-dependency and were therefore considered incidental (Not shown). 3.3.8. Skeletal examinations (stage of development) The stage of fetal skeletal development was generally similar in the control and genistein treated groups. The

1328

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

skeletal variations were those commonly observed and on a litter basis, there were no significant differences between the control and high-dose treated animals. On a fetus basis, there were a few minor increased incidences at the high dose that included: digit 2 proximal phalanx, non-ossified, right and left side, talus non-ossified, right and left side, and toes 2–5 proximal phalanx, non ossified, right and left side. These differences were attributed to the reduction in fetal body weight at 500 mg/kg/day and were considered not a test item specific effect (Not shown). 3.3.9. Cartilage examination of fetuses (stage of development) The stage of cartilage development was generally similar in the control and genistein treated groups. The skeletal variations were those commonly observed and on a litter basis, there were no significant differences between the control and high-dose treated animals with the exception of decreased incidence of incompletely chondrified intersternebral cartilage. On a fetus basis, at 100 mg/kg/day, there were increased and decreased incidences, depending on the site, of incompletely chondrified intersternebral cartilage (numbers 2–6). There was an increased incidence of incompletely chondrified xiphoid cartilage at 100 mg/kg/ day. The observed differences were considered to be incidental because of lack of dose response relationship (Not shown). 4. Discussion and conclusions Isoflavones (genistein, daidzein, glycitein, and their conjugated forms) structurally resemble estradiol and have estrogenic activity, but with a binding affinity to the estrogen receptor 100–1000 times less than estradiol. The reproductive effects of the isoflavones were first noted as a result of infertility in grazing animals, captive cheetahs and California quail (Kaldas and Hughes, 1989; Kurzer and Xu, 1997). Ewes grazing on subterranean or red clover in Australia were rendered permanently infertile. The clover was determined to contain various isoflavones including genistein and daidzein, along with their precursors, biochanin A and formononetin. Isoflavones were also implicated in infertility and liver failure in a group of captive cheetahs in which the commercial diets contained high concentrations of daidzein and genistein suggesting that the presence of dietary isoflavones were factors (Setchell et al., 1987a, 1987b). A comprehensive review of the reproductive and developmental toxicity of genistein has been performed (NTP CERHR, 2006). In the studies reported here, genistein was tested primarily for teratogenic potential in vitro and in vivo. In the in vitro rat whole embryo culture (WEC) assay, used as a preliminary screen for embryotoxic and teratogenic potential, genistein was tested over a concentration range of from 1 to 100 lg/ml. No adverse effects on embryonic development were observed at 1 or 3 lg/ ml. Treatment related anomalies were observed at concen-

trations of P10 lg/mL (37 lmol/L) and 100 lg/ml (370 lmol/L). At 100 lg/ml, distinct general toxic effects on the embryo and the yolk sac were observed. In the Pilot embryo-fetal developmental study, conducted as a preliminary screen for teratogenic potential, maternal body weight and food consumption were decreased at the high dose of 1000 mg/kg/day indicating a mild maternal toxicity. At the 1000 mg/kg/day dose, adverse effects in the pups included increased pup mortality, poor general condition, reduced body weights, and reduced milk uptake. At the high dose of 1000 mg/kg/ day, no external malformations were observed. In the visceral examination, there was no increase in the overall incidence of visceral findings. Skeletal examination revealed a higher incidence of minor abnormalities (sternum) or variations (mainly rudimentary extra ribs) as compared to controls. Some effects were also noted at the low dose of 20 mg/kg/day including, increased pup mortality, decreased pup body weights, decreased pup milk uptake, however at the mid-dose of 150 mg/kg/day, no adverse effects on the pups were noted. A decrease in the sex ratio of percent males was observed at the low dose, but not the higher doses. The relatively minor changes observed at the low dose, but not the mid-dose were considered incidental and not related to treatment due to a lack of dose response at the mid-dose and the small number of animals used for this pilot study. On the basis of this pilot study in a relatively small number of animals, genistein administered by oral gavage at a dose of up to 1000 mg/kg/day provided no evidence for a teratogenic effect. The effects at the high dose may be treatment related or represent a non-specific effect due to maternal toxicity at the high dose. In the main GLP Prenatal developmental safety study with larger group sizes (30 mated females/group), treatment with genistein caused a marked reduction of food consumption and maternal body weight gain at the high dose of 500 mg/kg/day and a markedly increased incidence of post implantation loss (resorptions), which resulted in a significant reduction in mean litter size. Fetal weights were markedly reduced at 500 mg/kg/day. The body weight normalized anogenital distance (AGD), measured as an indicator of estrogenic activity was not considered affected in this study. No increase in the incidence of external, visceral or skeletal malformations were observed. An increased incidence of non-ossification of certain bones of the paws was observed in fetuses at 500 mg/kg/day was attributed to the reduction in fetal body weight. Thus, in the main Prenatal developmental study, genistein administration showed fetal toxicity at 500 mg/kg/day, but there was no evidence for any teratogenic potential up to and including the highest dose level tested of 500 mg/kg/day. Overall, genistein was found not to have teratogenic potential in vivo in the pilot study or the definitive in vivo Prenatal developmental study, even at very high doses in the rat (up to 1000 and 500 mg/kg/day, respectively). In vitro, genistein had teratogenic potential in the WEC screening assay at concentrations P10 lg/mL (37 lmol/

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

L) there can be many reasons for differences in in-vivo and in vitro findings, which could include metabolism. In vivo, genistein is rapidly conjugated and genistein in the plasma is present predominantly in a conjugated form (Holder et al., 1999; Doerge et al., 2000). In the WEC screening assay, no metabolizing system is present other than that in the embryo itself. Furthermore, concentrations can be achieved in vitro that cannot be achieved in vivo. Doerge et al. (2001) studied the placental transfer of genistein in pregnant rats after dietary administration or gavage treatment. After a 75 mg/kg/ gavage dose, the fetal serum for total (conjugated plus free) and free (aglycone) genistein were approximately 20-fold and 5-fold respectively, lower than the maternal levels. After the 75 mg/kg dose of genistein, the maternal levels for total and free genistein were 4410 and 780 nM, respectively. Fetal serum concentrations were 220 and 60 nM, respectively indicating significant placental transfer. In the WEC, however, the concentration of genistein that began to show anomalies was 10 lg/mL (37 lmol/L or 37000 nM) of genistein (free) which is 168-fold and 600fold higher than the fetal serum levels of total and free genistein respectively, as measured by Doerge et al. (2001). Although in vitro assays such as the WEC are useful screening or investigative tools, the important point is that genistein has no evidence of teratogenic potential in vivo in a well-conducted study over a very high dose range. Although there is a large number of reproductive studies conducted with genistein administered at various times during gestation, lactation, and/or postweaning (NTP, 2006; NTP CERHR, 2006), there are few conventional (regulatory) reproductive toxicity studies with genistein in the published literature and there are no conventional teratology studies. Since the focus of this manuscript is on the teratogenic potential of genistein, studies with exposure in early gestation during the sensitive period of organogenesis would be the most relevant. There are several studies in which genistein was administered only during lactation (e.g., Lewis et al., 2003; Cotroneo et al., 2001; Nagao et al., 2001; Brown and Lamartiniere, 2000; Lamartiniere et al., 1998; Strauss et al., 1998.) or only post-weaning (e.g. Fritz et al., 2003; Fritz et al., 2002), which will not be discussed further. Most of the studies with genistein involved treatment during some part of gestation and lactation. In addition, there are a few studies in which genistein was administered only during late gestation and several multigeneration studies, which would involve treatment with genistein during gestation and lactation through two or more generations. The studies that were conducted with treatment over some part of gestation are indicated below along with the dose levels and the days over which the females were treated. Although these studies are somewhat limited to the extent that they involve natural delivery and that there was no detailed visceral or skeletal examination, there was no indication of a teratogenic effect in any of these studies.

1329

The studies that were conducted with exposure beginning in early gestation include: Wisniewski et al. (2003, 2005), treated pregnant rats and mice with genistein at a concentration of 5 and 300 mg/kg feed throughout gestation and lactation. You et al. (2002) treated pregnant SD rats with genistein at a concentration of 300 and 800 ppm in the diet from the beginning of gestation through weaning. Kang et al. (2002) tested exposures to pregnant SD rats to genistein from GD 6 to PND 20 at doses of 0.4 and 4 mg/kg/day. Flynn et al. (2000b) studied the effect of genistein administered to rats in the diet at concentrations of 0, 250, 500, and 1250 ppm (equivalent to 0, 2, 20, and 100 mg/ kg/day) from GD 7 through PND 77. Casanova et al. (1999), studied the effects of genistein on SD rats fed diets containing 0.02 or 0.1% of genistein during gestation and through PND 21. Fritz et al. (1998) exposed pregnant SD rats to genistein from 2 weeks prior to mating to PND 21 at concentration of 25 and 250 mg/kg diet. There was no evidence for a teratogenic potential in any of these studies. Although not as relevant, there was also no evidence of a teratogenic effect in several other studies in which genistein was administered later in gestation and lactation including: Masutomi et al. (2003), fed SD rats genistein in the diet at levels of 0, 20, 200 and 1000 ppm from GD 15 to PND 10. Fielden et al. (2002) examined pregnant B6D2F1 mice exposed to genistein administered by oral gavage at dose levels of 0, 0.1, 0.5, 2.5, and 10 mg/kg/day from GD 12 to PND 21. Roberts et al. (2000), treated rats with genistein at a concentration of 5 mg/kg diet from GD 17 to PND 21. Awoniyi et al. (1998), exposed pregnant rats to genistein in the diet (5 mg/kg diet) from GD 17 through PND 21. The most comprehensive published study with genistein is a multigeneration study conducted at NCTR (NTP, 2006). Delclos et al. (2001) fed pregnant SD rats genistein at dietary concentrations of 0, 5, 25, 100, 250, 625, and 1250 ppm from gestation day (GD) 7 through lactation. The concentration of 1250 ppm provided a range of intake over this period of from 70 to 97 mg/kg/day of genistein. Genistein treatment resulted in effects on multiple estrogen sensitive organs in both male and female rats, most of which were observed at the highest dose level and were generally consistent with estrogenic activity. Body weight and food consumption of the treated dams prior to parturition showed a decreasing trend with a significant reduction in the highest dose. At 1250 ppm, there was a decrease in the live birth weight and in males, there was a decrease in the weight of the ventral prostate. Histopathologic findings in males included ductal alveolar hyperplasia and hypertrophy in the mammary glands at concentrations of 25 ppm or more; aberrant or delayed spermatozoa at 1250 ppm and decreased sperm in the epididymis at 625 and 1250 ppm. However, in contrast to the histopathologic findings, there were no differences in testicular spermatid or epididymal sperm counts among the groups. Dalu et al. (2002), in a multigeneration study,

1330

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

treated both male and female rats of the F0 generation with genistein in the diet at concentrations in the diet of up to 500 ppm in which there was no effect on organ weights or histopathology of the male reproductive tract in either the F1 or F2 generation and testosterone and dihydrotestosterone were elevated in the F1 but not the F2 males as compared to controls. Flynn et al. (2000a) studied the effect of genistein on maternal behavior in rats fed at doses up to 40 mg/kg/day through 2 generations in which there were no severe effects on maternal nursing behavior. There was no indication of a teratogenic potential of genistein in any of the 3 multigeneration studies. 4.1. Treatment during late gestation In 2 studies conducted with genistein administered only later in gestation (GD 15 or 16 to GD 20) Hilakivi-Clarke et al. (1998) found in mice treated with 20 lg/day SC, that genistein increased the density of terminal buds in the mammary glands and reduced epithelial differentiation and accelerated puberty (vaginal opening). Levy et al. (1995), found that 5 or 25 mg/day/rat genistein administered to dams SC decreased birth weight but had no effect on parturition, dead pups, sexually dimorphic nucleus volume or estrus cycle in offspring. Genistein decreased anogenital distance in males and females and delayed puberty in females. Although there is no evidence for a teratogenic potential with genistein the overall effects of genistein on reproductive parameters are compatible with its weak estrogenic activity. However, the reported effects of genistein on reproduction are inconsistent among the various studies. For example, the anogenital distance was measured in the Prenatal developmental toxicity study as a marker of estrogenic activity, which was considered not altered (normalized for body weight). In studies conducted with exposure during gestation, several studies found changes in the AG distance (e.g. Levy et al., 1995; Wisniewski et al., 2003, 2005; Casanova et al., 1999) whereas several others did not (Delclos et al., 2001; Masutomi et al., 2003; Fielden et al., 2002; You et al., 2002; Kang et al., 2002; Fritz et al., 1998). In general, it appears that the effect of genistein on reproductive parameters is greater in female than male offspring. High doses in males reduce testosterone levels and/ or with some effect on testes weights and spermatogenesis (Delclos et al., 2001; Wisniewski et al., 2003; Roberts et al., 2000), whereas no effect was observed in some others (Wisniewski et al., 2005; Masutomi et al., 2003; Fielden et al., 2003; You et al., 2002; Kang et al., 2002). In females, some authors report there is an advancement of puberty (Hilakivi-Clarke et al., 1998; You et al., 2002; Casanova et al., 1999) with no effect observed in some of the other studies (Kang et al., 2002; Awoniyi et al., 1998; Fritz et al., 1998). In the mammary gland changes included ductal alveolar hyperplasia and /or hypertrophy and depending on the time of administration, an increase or decrease

in mammary gland differentiation (Delclos et al., 2001; Hilakivi-Clarke et al., 1998) with no effect on mammary gland morphology reported in some other studies (Fielden et al., 2002). Increased uterine weights with hypertrophy (Casanova et al. (1999) have also been reported. In general, with high doses of genistein there can be decreased survival and birth body weights of offspring. With respect to human pharmacokinetics, the overall target dose for chemoprevention is approximately 50 mg of isoflavones per day. Much of the justification regarding the selection of a dose of 50 mg/day (0.5–1.0 mg/kg/day) is based on the presumed average intake of Asian populations that may range from 0 to 125 mg per day (0–2 mg/ kg/day). More careful studies have determined the median isoflavones intake is 30–40 mg/day in Asian populations (Barnes, 2003). Morton et al. (2002) determined the serum concentrations of isoflavones in a large group of Japanese men and women over forty years of age as compared to serum samples from a United Kingdom population. The mean genistein concentration in Japanese males was 493 nmol/ L and in Japanese females 502 nmol/L as compared to 33 and 28 nmol/L for UK men and women, respectively. Similar values were reported by Yamamoto et al. (2001), with serum concentrations in a group of Japanese for genistein of 475 nmol/L with an estimated intake (FFQ) for genistein of 31 mg/day. These intakes were approximately 700 times higher than for US Caucasians. In Asian males and females, the total genistein plasma concentrations are thus approximately 500 nmol/L. In the Prenatal developmental safety study, the average maternal total genistein plasma levels on day 21 of gestation were approximately, 436, 2600, 4912, and 22,873 nmol/L at doses of 5, 50, 100 and 500 mg/kg/day, respectively (Table 7). This results in exposure multiples (rat/human based the plasma levels), of approximately 1, 5, 10 and 46 times the human Asian blood genistein levels at dose levels of 5, 50, 100, and 500 mg/kg/day, respectively. 5. Conclusions Overall, a large number of studies have been conducted with genistein involving treatment during gestation with natural delivery, which have provided no evidence for teratogenic effects. This is consistent with the data from both the Pilot embryo-fetal developmental study and the GLP Prenatal developmental study reported here. On the basis of the studies in this manuscript, it is concluded that genistein has no teratogenic potential, in vivo at very high doses of up to 1000 mg/kg/day by oral gavage in the Pilot feto-embryo-toxicity study or up to 500 mg/kg/ day by dietary admix in the definitive Prenatal developmental study even though these doses were maternally toxic and fetal-toxic. In vitro, genistein had teratogenic potential at high concentrations in the WEC assay used as a screening tool, however this was not predictive of the in vivo effects. On the basis of the Prenatal developmen-

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

tal study, the NOAEL for maternal toxicity and adverse effects on embryonic and fetal development was considered to be 100 mg/kg/day when administered orally by dietary admix. References Awoniyi, C.A., Roberts, D., Veeramachaneni, D.N., Bradley, S., Hurst, B.S., Tucker, K.E., Schlaff, W.D., 1998. Reproductive sequelae in female rats after in utero and neonatal exposure to the phytoestrogen genistein. Fertility and Sterility 70, 440–447. Barnes, S., Sfakianos, J., Coward, L., Kirk, M., 1996. Soy isoflavinoids in cancer prevention. Underlying biochemical and pharmacological issues. Advances in Experimental Medicine and Biology 401, 87–100. Barnes, S., 2003. Phytoestrogens and osteoporosis: what is a safe dose? British Journal of Nutrition 89 (Suppl. 1), S101–S108. Bennetts, H.W., Underwood, E.J., Shier, F.L., 1946. A specific breeding problem of sheep on subterranean clover pastures in Western Australia. Australian Veterinary Journal 22, 2–12. Brown, N.A., Fabro, S., 1981. Quantitation of rat embryonic development in vitro: a morphological scoring system. Teratology 24, 65–78. Brown, N.M., Lamartiniere, C.A., 2000. Genistein regulation of transforming growth factor-alpha, epidermal growth factor (EGF), and EGF receptor expression in the rat uterus and vagina. Cellular Growth and Differentiation 11, 255–260. Casanova, M., You, L., Gaido, K.W., Archibeque-Engle, S., Janszen, D.B., Heck, H.A., 1999. Developmental effects of dietary phytoestrogens in Sprague Dawley rats and interactions of genistein and daidzein with rat estrogen receptors alpha and beta in vitro. Toxicological Sciences 51, 236–244. Cotroneo, M.S., Wang, J., Eltoum, I.A., Lamartiniere, C.A., 2001. Sex steroid receptor regulation by genistein in the prepubertal rat uterus. Molecular and Cellular Endocrinology 173, 135–145. Dalu, A., Blaydes, B.S., Bryant, C.W., Latendresse, J.R., Weis, C.C., Delclos, B.K., 2002. Estrogen receptor expression in the prostate of rats treated with dietary geinstein. Journal of Chromatography B 777, 249–260. Delclos, K.B., Bucci, T.J., Lomax, L.G., Latndresse, J.R., Warbritton, A., Weis, C.C., Newbold, R.R., 2001. Effects of dietary genistein exposure during development on male and female CD (Sprague Dawley) rats. Reproductive Toxicology 15, 647–663. Doerge, D.R., Chang, H.C., Churchwell, M.I., Holder, C.L., 2000. Analysis of soy isoflavones conjugation in vitro and in human blood using liquid chromatography. Drug Metabolism and Disposition 28, 298–307. Doerge, D.R., Churchwell, M.I., Chang, H.C., Newbold, R.R., Delclos, K.B., 2001. Placental transfer of the soy isoflavone genistein following dietary and gavage administration to Sprague Dawley rats. Reproductive Toxicology 15, 105–110. Fielden, M.R., Fong, C.J., Haslam, S.Z., Zacharewski, T.R., 2002. Normal mammary gland morphology in pubertal female mice following in utero and lactational exposure to genistein at levels comparable to human dietary exposure. Toxicology Letters 133, 181–191. Fielden, M.R., Samy, S.M., Chou, K.C., Zacharewski, T.R., 2003. Effect of human dietary exposure levels of genistein during gestation and lactation on long-term reproductive development and sperm quality and mice. Food and Chemical Toxicology 41, 447–454. Flynn, K.M., Ferguson, S.A., Delclos, K.B., Newbold, R.R., 2000a. Effects of genistein exposure on sexually dimorphic behaviors in rats. Toxicological Sciences 55, 311–319. Flynn, K.M., Ferguson, S.A., Delclos, K.B., Newbold, R.R., 2000b. Multigenerational exposure to dietary genistein has no severe effects on nursing behavior in rats. Neurotoxicology 21, 997–1001. Fritz, W.A., Coward, L., Wang, J., Lamartiniere, C.A., 1998. Dietary genistein: perinatal mammary cancer prevention, bioavailability and toxicity testing in the rat. Carcinogenesis 19, 2151–2158.

1331

Fritz, W.A., Eltoum, I.E., Cotroneo, M.S., Lamartiniere, C.A., 2002. Genistein alters growth but is not toxic to the rat prostate. Journal of Nutrition 132, 3007–3011. Fritz, W.A., Cotroneo, M.S., Wang, J., Eltoum, I.E., Lamartiniere, C.A., 2003. Dietary diethylstilbestrol but not genistein adversely affects rat testicular development. Journal of Nutrition 133, 2287–2293. Gallavan, R.H., Holson, J.F., Stump, D.G., Knapp, J.F., Reynolds, V.L., 1999. Interpreting the toxicologic significance of alterations in anogenital distance: potential for confounding effects of progeny body weights. Reproductive Toxicology 13, 383–390. Goldwyn, S., Lazinsky, A., Wei, H., 2000. Promotion of health by soy isoflavones: efficacy, benefit and safety concerns. Drug Metabolism and Drug Interaction 17, 261–289. Hilakivi-Clarke, L., Cho, E., Clarke, R., 1998. Maternal genistein exposure mimics the effect of estrogen on mammary gland development in female mouse offspring. Oncology Reports 5, 609–616. Hsieh, C.Y., Santell, R.C., Haslam, S.Z., Helferich, W.G., 1998. Estrogenic effects of genistein on the growth of estrogen receptor-positive human breast cancer (MCF-7) cells in vitro and in vivo. Cancer Research 58, 3833–3838. Holder, C.L., Churchwell, M.I., Doerge, D.R., 1999. Quantification of soy isoflavones, genistein, daidzein and conjugates in rat blood using LC/ ES-MS. Journal of Agriculture and Food Chemistry 47, 3764–3770. Hutchins, A.M., Slavin, J.L., Lampe, J.W., 1995. Urinary isoflavinoid, phytoestrogen and lignan excretion after consumption of fermented and unfermented soy products. Journal of the American Dietary Association 95, 545–551. Izumi, T., Piskula, M.K., Osawa, S., Obata, A., Tobe, K., Saito, M., Kataoka, S., Kubota, S., Kikuchi, M., 2000. Soy isoflavone aglycones are absorbed faster and in higher amounts than their glucosides in humans. Journal of Nutrition 130, 1695–1699. Kaldas, R.S., Hughes, C.L., 1989. Reproductive and general metabolic effects of phytoestrogens in mammals. Reproductive Toxicology 3, 81– 89. Kang, K.S., Che, J.H., Lee, Y.S., 2002. Lack of adverse effects in the F1 offspring maternally exposed to genistein at human intake dose level. Food and Chemical Toxicology 40, 43–51. Knight, D.C., Eden, J.A., 1994. A review of the clinical effects of phytoestrogens. Obstetrics and Gynecology 87, 897–904. Kurzer, M.S., Xu, X., 1997. Dietary Phytoestrogens. Annual Review of Nutrition 17, 353–381. Lamartiniere, C.A., Zhang, J.X., Cotroneo, M.S., 1998. Genistein studies in rats: potential for breast cancer prevention and reproductive and developmental toxicity. American Journal of Clinical Nutrition 68, 1400S–1405S. Levy, J.R., Faber, K.A., Ayyash, L., Hughes, C.L., 1995. The effect of prenatal exposure to the phytoestrogen genistein on sexual differentiation in rats. Proceedings of the Society of Experimental Biology and Medicine 208, 60–66. Lewis, R.W., Brooks, N., Milburn, G.M., Soames, A., Stones, S., Hall, M., Ashby, J., 2003. The effects of the phytoestrogen genistein on the postnatal development of the rat. Toxicological Sciences 71, 74–83. Masutomi, N., Shibutani, M., Takagi, H., Uneyama, C., Takahashi, N., Hirose, M., 2003. Impact of dietary exposure to methoxychlor, genistein or diisononyl phthalate during the perinatal period on the development of the rat endocrine/reproductive systems in later life. Toxicology 192, 149–170. Miksicek, R.J., 1995. Estrogenic flavonoids: structural requirements for biological activity. Proceedings of the Society of Experimental Biology and Medicine 208, 44–50. Morton, M.S., Arisaka, O., Miyake, N., Morgan, L.D., Evans, B.A.J., 2002. Phytoestrogen concentrations in serum from Japanese men and women over forty years of age. Journal of Nutrition 132, 3168– 3171. Nagao, T., Yoshimura, S., Saito, Y., Nakagomi, M., Usumi, K., Ono, H., 2001. Reproductive effects in male and female rats of neonatal exposure to genistein. Reproductive Toxicology 15, 399–411.

1332

R.M. McClain et al. / Food and Chemical Toxicology 45 (2007) 1319–1332

New, D.A., 1978. Whole embryo culture and the study of mammalian embryos during organogenesis. Biological Reviews Cambridge Philosophical Society 53, 81–122. NTP, 2006. Multigenerational Reproductive Toxicology Study of Genistein (Cas No. 446-72-0) in Sprague Dawley Rats (Feed Study), Document TR 539, National Toxicology Program. NTP CERHR, 2006. National Toxicology Program Center for the Evaluation of Risks to Human Reproduction, expert panel evaluation of genistein and soy formula, March 15–17, 2006, CERHR website (http://cerhr.niehs.nih.gov) and available in printed text from CERHR in May 2006. Piersma, A.H., Attenon, P., Bechter, A.H., Govers, M., Krafft, N., Schmid, B.P., Stadler, J., Verhoef, A., Verseil, C., 1995. Interlaboratory evaluation of embryotoxicity in the postimplantation rat embryo culture. Reproductive Toxicology 3, 275–280. Plassmann, S., 1997. Ro 24-2076/000 Segment II: Oral study for effects on embryo-fetal development in the rat. Hoffmann La Roche Ltd., Basle Research Report B-0164609. Roberts, D., Veeramachaneni, D.N., Schlaff, W.D., Awoniyi, C.A., 2000. Effects of chronic dietary exposure to genistein, a phytoestrogen, during various stages of development on reproductive hormones and spermatogenesis in rats. Endocrinology 13, 281–286. Schmitt, G., 1997. The teratogenic potential of genistein: whole embryo culture assay. Hoffmann La Roche Ltd., Basle Research Report B0167288. Setchell, K.D., 2000. Absorption and metabolism of soy isoflavones-from food to dietary supplements and adults to infants. Journal of Nutrition 130, 654S–655S. Setchell, K.D., Borriello, S.P., Hulme, P., Kirk, D.N., Axelson, M., 1984. Nonsteroidal estrogens of dietary origin: possible roles in hormonedependent disease. American Journal of Clinical Nutrition 40, 569– 578. Setchell, K.D., Gosselin, S.J., Welsh, M.B., Johnston, J.O., Balistreri, W.F., Kramer, L.W., Dresser, B.L., Tarr, M., 1987a. Dietary estrogens-a probable cause of infertility and liver disease in captive cheetahs. Gastroenterology 93, 225–233. Setchell, K.D., Gosselin, S.J., Welsh, M.B., Johnston, J.O., Balistreri, W.F., Kramer, L.W., Dresser, B.L., Tarr, M.J., 1987b. Dietary estrogens – a probable cause of infertility and liver disease in captive cheetahs. Gastroenterology 93, 225–233. Sirtori, C.R., 2001. Risks and benefits of soy phytoestrogens in cardiovascular diseases, cancer, climacteric symptoms and osteoporosis. Drug Safety 24, 665–682. Sterz, H., 1977. Routine examination of rat and rabbit fetuses for malformation of internal organs combination of Barrow’s and Wilson’s methods. In: Neubert, D., Merker, H.J., Kwasigroch, T.E.

(Eds.), Methods in Prenatal Toxicology. Stuttgart Georg Thieme, 113– 125. Strauss, L., Makela, S., Joshi, S., Huhtaniemi, I., Santti, R., 1998. Genistein exerts estrogen-like effects in male mouse reproductive tract. Molecular and Cellular Endocrinology 144, 83–93. Suthar, A.C., Banavalikar, M.M., Biyani, M.K., 2001a. Pharmacological activities of genistein, an isoflavone from soy (Glysine max): part Ianti-cancer activity. Indian Journal of Experimental Biology 39, 511– 519. Suthar, A.C., Banavalikar, M.M., Biyani, M.K., 2001b. Pharmacological activities of genistein, an isoflavone from soy (Glysine max): part IIanti-cholesterol activity, effects on osteoporosis and menopausal symptoms. Indian Journal of Experimental Biology 39, 520–525. Wang, T.T., Sathyamoorthy, N., Phang, J.M., 1996. Molecular effects of genistein on estrogen receptor mediated pathways. Carcinogenesis 17, 271–275. Wilson, J.G., 1965. In: Wilson, J.G., Warkany, J. (Eds.), Teratology, Principles and Techniques. The University of Chicago Press, Chicago and London, pp. 265–277. Wisniewski, A.B., Klein, S.L., Lakshmaman, Y., Gearhart, J.P., 2003. Exposure to genistein during gestation and lactation demasculinizes the reproductive system in rats. Journal of Urology 169, 1582–1586. Wisniewski, A.B., Cernetich, A., Gearhart, J.P., Klein, S.L., 2005. Perinatal exposure to genistein alters reproductive development and aggressive behavior in male mice. Physiology and Behavior 84, 327– 334. Wolz, E., Pfannkuch, F., Schmitt, G., Howald, M., Knuppe, C., Becker, H., 2001. Ro 24-2076/000 (Genistein): Prenatal developmental toxicity study in the rat. Hoffmann La Roche Ltd., Basle Regulatory Document RDR-1005025. Yamamoto, S., Sobue, T., Sasaki, S., Kobayashi, B., Arai, Y., Uehara, M., Adlercreutz, H., Watanabe, S., Takahashi, T., Litoi, Y., Iwase, Y., Akabane, M., Tsugane, S., 2001. Validity and reproducibility of a selfadministered food frequency questionnaire to assess isoflavone intake in a Japanese population in comparison with dietary records in blood and urine isoflavones. Journal of Nutrition 131, 2741–2747. You, L., Casanova, M., Bartolucci, E.J., Fryczynski, M.W., Dorman, D.C., Everitt, J.I., Gaido, K.W., Ross, S.M., Heck, H.A., 2002. Combined effects of dietary phytoestrogen and synthetic endocrineactive compound on reproductive development in Sprague Dawley rats: genistein and methoxychlor. Toxicological Sciences 66, 91– 104. Zhou, J.R., Erdman, J.W., 1997. Chemical effects on processing and food preparation on carotenoids and soy and garlic phytochemicals. In: Lachance, P.A. (Ed.), Nutraceuticals: Designer Foods III. Garlic soy and licorice. Food and Nutrition Press, Trumbull, C.T, pp. 23–37.