QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content PubMed PubMed Citation

---

Supplement

Poster Presentation Abstracts

	Isoflavone quantification

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Sensitive HPLC Method for Profiling Plasma Phytoestrogens. Tarja Nurmi and Herman Adlercreutz. Institute for Preventive Medicine Nutrition and Cancer, Folkhälsan Research Center, Clinical Chemistry Division, University of Helsinki, Helsinki, Finland.

Phytoestrogens have several beneficial effects on human health, i.e., they lower the risk of hormone-dependent cancers, reduce menopausal symptoms and are known to prevent osteoporosis and coronary heart disease. A Western diet contains low amounts of phytoestrogens; therefore, sensitive analytical methods are required to determine the phytoestrogen profiles in nonsupplemented samples, especially when metabolites are also of interest. Traditional analyses are performed by using gas chromatography–mass spectrometry, which is sensitive and selective but also expensive because of multistage sample pretreatment and a high initial investment. HPLC methods using UV-detection are simpler to use and the devices are usually less expensive; however, the sensitivity is not high enough for reliable analysis of nonsupplemented human plasma samples. Therefore, we developed an HPLC method using coulometric electrode array detection for analyzing plasma phytoestrogens (daidzein, genistein, O-desmethylangolensin, equol, dihydrodaidzein, dihydrogenistein, secoislariciresinol, matairesinol, enterolactone and enterodiol). The method is especially useful for analyzing nonsupplemented plasma samples from Western populations. The method was evaluated by determining intra- and interassay precision, resolution, detection limits, linearities, retention time and detector response repeatability. The simple pretreatment required combined with high sensitivity makes the method a valuable tool in clinical and epidemiologic studies.

Development of Rapid Methods for Measuring Isoflavones in Human Urine. Chunyang Wang and Regina Wixon. South Dakota State University, Department of Nutrition and Food Science, Brookings, SD.

UV-visible spectrophotometry was a very successful technique in measuring isoflavones in soybeans and processed soy protein ingredients. The same approach was used in the development of a rapid method for determining isoflavone concentrations in urine samples. Urine samples (n = 30) with different levels of isoflavones were collected from a dietary intervention study. The original samples were scanned, and mathematical models were developed to predicate isoflavone concentrations on the basis of spectrophotometric properties. The method was found to be ineffective. Various simple sample treatments were conducted before the scanning. They included the following: sample dilution, extraction using tetra-butyl methyl ether and solid-phase extraction. The effectiveness of the UV-visible method was improved, as shown by higher correlation coefficients (R² = 0.3–0.6). Solid-phase extraction was found to be the most effective sample treatment before scanning. Predicating models were developed for genistein, daidzein, glycitein and equol. Other developmental efforts will also be reported. This development effort has great potential to be used in scanning a large number of samples. [Supported by the South Dakota Soybean Research and Promotion Council.]

Liquid Chromatography–Mass Spectrometry Analysis of Isoflavones in Naturally Brewed and Chemically Hydrolyzed Soy Sauce. Amanda Murray, Michelle Smith,* Marion Kirk* and Stephen Barnes.* Department of Pharmacology and Toxicology, Altamont School, Birmingham, AL and *University of Alabama at Birmingham, Birmingham, AL.

Previous reports showed that soy sauce contains very low amounts of isoflavones. However, the method of preparation of soy sauce was not taken into consideration. In this study, we compared the isoflavone composition and concentration of naturally fermented and chemically hydrolyzed soy sauces. Commercially available soy sauces (fermented, Kikkoman, Kikkoman Lite, Kimlan Ponlai; and chemically prepared, Angostura, Chun King) were analyzed by reversed-phase HPLC at 262 nm with a diode array detector and by HPLC–electrospray ionization–mass spectrometry (LC-ES-MS) using a triple quadrupole mass spectrometer. Isoflavones were recovered from the soy sauces either after precipitation with 4 vol of methanol or by C8 Sep-Pak cartridge extraction with and without preliminary treatment with ß-glucosidase. The extracts were taken to dryness, redissolved in 80% methanol, centrifuged and analyzed by HPLC and LC-ES-MS. Initial experiments using HPLC-UV analysis revealed that fermented soy sauce extracts, unlike the chemically prepared soy sauces, had a large number of UV-absorbing compounds. However, the complexity of the profiles prevented unequivocal identification of individual peaks as possible isoflavones and their ß-glucoside conjugates. Analysis of extracts by LC-ES-MS scanning over the m/z range 240–650 allowed the development of reconstructed ion chromatograms for each of the isoflavones in soy (m/z 253 for daidzein, m/z 269 for genistein and m/z 283 for glycitein). Two suspected isoflavone peaks (m/z 285 and m/z 301) corresponding to the addition of two hydroxyl groups to daidzein and genistein were also detected. No evidence was obtained for the presence of isoflavone ß-glucoside conjugates. Confirmation of the identity of each primary isoflavone was made by carrying out MS–MS experiments to obtain daughter ion mass spectra of [M–H]- parent molecular ions. This LC-ES-MS approach clearly established that fermented soy sauces contain each of the primary isoflavones (in the unconjugated form) and in 5–10 times greater concentrations (42–79 µg/mL for total isoflavone concentration) than in the chemically prepared soy sauces. Only minor differences in composition or concentrations of isoflavones were observed between regular and ‘lite’ forms of one of the soy sauces. The results suggest that natural fermentation, unlike chemical hydrolysis, produces soy sauces that contain unconjugated isoflavones, some of which have undergone additional metabolism.

Deuteration of Isoflavone Metabolites, Dihydrodaidzein and Dihydrogenistein. Kristiina Wähälä and Sirpa Rasku.University of Helsinki, Department of Chemistry, Laboratory of Organic Chemistry, University of Helsinki, Helsinki, Finland.

To study the metabolism and biological and physiologic effects of phytoestrogens, a sensitive and specific quantitative method is required. The isotope dilution–gas chromatography–mass spectrometry–selective ion monitoring (ID-GC-MS-SIM) technique using synthesized deuterated internal standards for the correction of losses during the procedure is used to quantitate phytoestrogens from food and biological fluids. An internal standard for quantitative MS must have no unlabeled species, and the isotope labels must remain stable under the analytical conditions used. For polyhydroxy aromatics, it is desirable that the reference compound contain three to five stable deuterium atoms because the unlabeled compound, derivatized with trimethylsilyl for GC, will show fairly intense m+1 and m+2 ions in its mass spectrum. Finally, the standard must be isomerically and isotopically pure. The synthesis of new stable [6,8,3',5'-D₄]-dihydrodaidzein, [3,6,8,3',5'-D₅]-dihydrodaidzein and [3,2',3',5',6'-D₅]-dihydrogenistein involves hydrogen-deuterium exchange of aromatic protons using D₃PO₄·BF₃/D₂O as a deuteration reagent and deprotonation of labile deuteriums from fully deuterated dihydrogenistein. The sites of deuteration were determined from the ¹H and ¹³C nuclear magnetic resonance (NMR) spectra by comparison with those of undeuterated compound. Deuterium-carrying carbon atoms appear as low intensity triplets in the proton noise-decoupled spectra compared with the intensive singlets of undeuterated compounds. The exchange order of hydrogens was determined from different deuteration and dedeuteration experiments by following the progress of the reaction by NMR. The isotopic purity of the product is determined from the mass spectra of trimethylsilylated product to avoid M-1 losses from phenolic hydroxyls. The new deuterium-labeled dihydrodaidzein and dihydrogenistein can be reliably used as reference compounds and introduced at the beginning of the analytical procedure because the deuterium labels are securely bound and will survive the various isolation, purification and derivatization steps. The ID/GC/MS/SIM method has now been used for the quantitation of these isoflavone metabolites in human urine.

Phytoestrogen Content of Various Natural Products. Tarja Nurmi and Herman Adlercreutz. Institute for Preventive Medicine Nutrition and Cancer, Folkhälsan Research Center, Clinical Chemistry Division, University of Helsinki, Helsinki, Finland.

Phytoestrogens, shown to possess several beneficial effects on human health, are of particular interest for the pharmaceutical industry. Phytoestrogens seem to lower the risk of hormone-dependent cancers and reduce menopausal symptoms; they are known to prevent osteoporosis and coronary heart disease in experimental studies. A Western diet contains amounts of phytoestrogens that are too low to allow their levels in plasma to become high enough for biological activity. Instead of changing the traditional diet, it is now possible to supplement it with isoflavone-containing concentrated natural products. Positive health effects may occur with moderate amounts of phytoestrogens and high amounts may not have any effects. There is a difficult dilemma, i.e., what is a moderate amount and what is a sufficient amount? In cancer treatment, on the other hand, very high amounts may be needed. The phytoestrogen content of natural products varies significantly, and the amounts recommended to be consumed are not always specified. Ten commercial products were analyzed by using HPLC with a coulometric electrode array detector. The analytical method was carefully standardized and optimized to separate the following isoflavones: daidzein, genistein, glycitein, formononetin, biochanin A, and daidzein- and genistein-7-O-glucosides and their malonyl-glucosides. The samples were analyzed before and after the hydrolysis to quantify the conjugate forms of the phytoestrogens occurring in the products. Phytoestrogen content of the concentrated natural products was slightly or moderately lower than the amounts claimed by manufacturers. However, in one preparation, the content of phytoestrogens was minimal.

HPLC: Electrochemical Detection of Isoflavones and Lignans in Human Plasma. Paul H. Gamache and Ian N. Acworth. ESA Inc., Chelmsford, MA.

Interest in potential health benefits of phytoestrogens has created the need for simple and reliable techniques for their measurement. A highly sensitive method that uses HPLC with coulometric array detection was developed for the determination of plasma phytoestrogens. Analytes were separated in 25 min by reversed-phase (C18) isocratic elution using a water/methanol/acetonitrile, 68:25:7 (v/v/v) mobile phase containing sodium acetate buffer (0.2 mol/L, pH 4.8) at a flow rate of 0.6 mL/min and column temperature of 42°C. Eight serial electrochemical sensors were used at potentials of 340, 470, 500, 530, 560, 620, 680 and 760 (mV vs. Pd) to generate voltametric response relationships for each analyte. After enzymatic hydrolysis, plasma samples were acidified, washed with hexane and extracted with diethyl ether. Combined extracts were evaporated and the residue was dissolved in 50% (by vol) aqueous methanol before HPLC–electrochemical detection (ECD) analysis. Lower limits of quantification (15% relative standard deviation) estimated from plasma extracts were 0.8, 1.2, 0.8 and 1.4 ng/mL of daidzein, enterolactone, equol and genistein, respectively. By least-squares regression, the response was linear with concentration for all analytes (r 0.99, six levels, 10.0–2500 ng/mL, two replicate extractions at each level on 2 d). Intra- and interassay variability for plasma augmented with 50 ng/mL of standards was <5.5% RSD (n = 10) and <7.7% RSD (4 d), respectively. Plasma levels (mean ± SD) obtained from 15 adult volunteers were 2.7 ± 3.9, 4.2 ± 3.4, <1.2 and 2.7 ± 2.1 ng/mL of daidzein, enterolactone, equol and genistein, respectively. The described method is suitable for measuring plasma phytoestrogen in human subjects at basal levels and after supplementation.

The Synthesis of ¹³ C-Labeled Phytoestrogens. Tara Fryatt, Nigel P. Botting and Mark F. Oldfield. University of St. Andrews, St. Andrews, Scotland.

Accurate analysis of the biological effects of phytoestrogens on human health has been hindered by the lack of stable internal standards for gas chromatography–mass spectrometry analysis and for metabolic studies. The primary objective of this work was to develop efficient syntheses of labeled phytoestrogens for use in studies such as these. The initial targets were ¹³C-labeled daidzein and genistein, compounds that are recognized as protective factors against the development of hormone-dependent cancers. Synthetic procedures for the preparation of the isoflavonoid phytoestrogens formononetin, biochanin A, daidzein and genistein, incorporating a single ¹³C label at the 4-position were developed. These procedures use adaptations of existing methodologies and suitable labeled precursors. Current work is focusing on multiply labeled ¹³C-isoflavonoid derivatives. Work is also being carried out on developing methods for the synthesis of ¹³C-labeled lignans, for example, secoisolariciresinol and matairesinol.

	Isoflavone absorption and metabolism

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Transport and Metabolism of Soy Isoflavones (Glycosides vs. Aglycones) in Caco-2 Cells and in Perfused Gut Segments. Aukje Steensma,* Hub P.J.M. Noteborn,* Marcel J. B. Mengelers,* Harry A. Kuiper*. *State Institute for Quality Control of Agricultural Products (RIKILT), Wageningen, The Netherlands and Wageningen Agriculture University, Department of Food Technology and Nutritional Sciences, Sub Department of Toxicology, Wageningen, The Netherlands.

Genistein and daidzein are present in minor amounts in soybeans and soy-derived foods, whereas their sugar-conjugated derivatives are present in relatively high amounts (1–3 mg/g product). Data on the bioavailability of isoflavones are scarce. In this study, Caco-2 cells derived from a human colon adenoma carcinoma were used as a model for intestinal absorption. Information was obtained on the transport, metabolism and mechanism of action of genistein, daidzein and their glycosides by growing the cells on semipermeable filters. These in vitro transport and metabolism data were compared with those from intestinal perfused segments of the rat. In Caco-2 cells, a significant difference in the transport and metabolism between the aglycones and their glycosides was observed. Genistein and daidzein added at the apical side were transported to the basolateral side, whereas their glycosides were hardly transported in this direction. In perfused gut segments, the transport of genistein was also higher than its glycoside. Furthermore, the transport of genistein was highest in the ileal segment, whereas there were no differences in transport of the glycoside in various other segments tested. In both model systems, the glycosides were metabolized to their respective aglycones. Genistein was metabolized mainly to sulfates and glucuronides in the Caco-2 cells and to glucuronides in the perfused gut segments. Our data also indicated that Caco-2 cells and rat segments contained exogenous or endogenous glucosidase activity (or both) because only aglycones could be detected at the basolateral side of the Caco-2 cells and in the resorbate of perfused gut segments.

Absorption of Isoflavone Aglycones and Glycosides in Postmenopausal Women. M. Richelle, S. Pridmore-Merten, S. Bodenstab,* I. Tavazzi, S. Jecklin* and E.A. Offord. Nestlé Research Center, Lausanne, Switzerland and *Nestlé Product Technology Centre, Konolfingen, Konolfingen, Switzerland.

Isoflavones naturally present in soybeans and soy-based foods are mainly in the glucoside form. It is believed that removal of the sugar moiety of the glucoside (by ß-glucosidase activity) is required for its absorption through the intestinal wall. We compared the bioavailability of isoflavone aglycones and glucosides from an isoflavone-rich extract consumed as a drink. The isoflavone-rich extract was hydrolyzed enzymatically by commercial ß-glucosidases to produce aglycones. Drinks were constituted by mixing the isoflavone-rich extracts with water, sugar and orange flavor. The pharmacokinetics of isoflavones in plasma were determined over 34 h in six postmenopausal women. After the ingestion of both soy drinks (aglycone or glucoside), plasma total isoflavone increased rapidly, reaching a maximal concentration of 4 µmol/L between 5 and 7 h. Thereafter, plasma isoflavones decreased slowly, leading to plasma concentrations that were still elevated at 34 h postabsorption. The ratio of plasma daidzein to genistein mimicked the ratio found in the soy drink. A low level of dihydroxydaidzein (0.2 µmol/L), a metabolite of daidzein, appeared in plasma after 4 h. The pharmacokinetics of plasma isoflavones were similar with both products. In conclusion, hydrolysis of glucoside before ingestion does not improve the bioavailability of isoflavones from isoflavone-enriched extracts.

Kinetic Models Describing In Vitro Transport and Metabolism of Isoflavones and Their Glycosides in Human Caco-2 Cells. Aukje Steensma,* Marcel J. B. Mengelers,* Hub P.J.M. Noteborn* and Harry A. Kuiper.* *State Institute for Quality Control of Agriculture Products (RIKILT-DLO), Wageningen, The Netherlands and Wageningen Agriculture University, Department of Food Technology and Nutritional Sciences, Sub Department of Toxicology, Wageningen, The Netherlands.

Kinetic models can be applied to describe the in vitro transport and metabolism of isoflavones by using intestinal epithelial cells (Caco-2). These models may enable a quantitative comparison of in vitro and in vivo bioavailability parameters. An extended kinetic model was developed in Caco-2 cells grown on semipermeable filters that described the transport of isoflavones from the apical to basolateral side and vice versa. The model also included the metabolic activity of the cells. Transport of the isoflavones across the intestinal cells was controlled by diffusion. However, the transport rate of the glycosides across Caco-2 cells was too low to enable a kinetic modeling. Additional metabolic studies were carried out to incorporate metabolic rates into the kinetic model. The metabolic rates obtained from the metabolism studies could be incorporated into the model used for describing the transport experiments without alterations.

Absorption and Deglycosylation of Isoflavone Glycosides in the Small Intestine. A. P. Wilkinson, J. M. Gee, A. J. Day, M. S. DuPont, P. W. Needs, G. W. Plumb, I. T. Johnson, M. R. A. Morgan* and G. Williamson. Institute of Food Research, Norwich Research Park, Norwich, UK and *Proctor Department of Food Science, University of Leeds, Leeds, UK.

The predominant isoflavones in soy are the glycosides, although low levels of the aglycones are also present. Isoflavones appear in plasma within 30 min of ingesting soy, indicating some absorption from the small intestine. The aim of this study was to investigate whether isoflavone aglycones are absorbed preferentially in the small intestine compared with their glycosides. Daidzin and daidzein absorption was studied with the use of an in vitro rat everted-gut model. Everted sacs of rat proximal jejunum were filled with physiologic saline and suspended for 15 min at 37°C in oxygenated (95% CO₂/5% O₂) Krebs buffer (pH, 7.2–7.4) containing either daidzein ( 0, 10, 100 or 1000 µmol/L) or daidzin ( 0, 1, 10 or 100 µmol/L). After the incubation, the serosal fluids were collected and stored. An ELISA for daidzein and HPLC were used to both quantitate amounts of daidzin or daidzein transported across the rat small intestine and investigate the chemical forms present in the serosal fluid. A similar degree of absorption of daidzin and daidzein was observed. Absorbed daidzin and daidzein were rapidly metabolized within the small intestinal enterocytes. The principal metabolite of both daidzin and daidzein found in the serosal fluid was daidzein-O⁷-glucuronide. The mechanisms involved in the absorption and metabolism of daidzein and daidzin may be as follows. Daidzein diffuses passively into the enterocyte where it is conjugated with glucuronic acid, a reaction mediated by UDP-glucuronyl transferase, forming daidzein-O⁷-glucuronide, which is subsequently transferred across the basolateral membrane into the serosal fluid. Two mechanisms may account for daidzin uptake and metabolism. It may enter the enterocyte by interaction with an active sugar transport mechanism. Once inside the cell, daidzin can be hydrolyzed by an intracellular cytosolic ß-glucosidase recently shown to be present in the human small intestine (Day et al. 1998 ). This enzyme can effectively hydrolyze daidzin and genistin. After daidzin hydrolysis, the liberated daidzein is conjugated with glucuronic acid as described above. Alternatively, daidzin may be hydrolyzed at the mucosal brush border membrane by another enzyme (lactase phlorizin hydrolase). Our recent studies have shown that this enzyme can hydrolyze isoflavone glycosides. Thus, at the mucosal brush border, daidzin may be hydrolyzed by lactase phlorizin hydrolase and the daidzein formed then diffuses into the enterocyte to be metabolized as described above. It is frequently inferred that absorption of isoflavone glycosides occurs only after their hydrolysis by ß-glucosidases associated with the large intestinal microflora. The results of this study show that both daidzin and daidzein can be absorbed from the small intestine and that there is little difference between the two in their rate of uptake.

Review of Glycitein Bioavailability and Biological Effects. Suzanne Hendrich, Tong T. Song, Yan Zhang, Gui-Juan Wang and Patricia A. Murphy. Food Science and Human Nutrition, Iowa State University, Ames, IA.

Glycitein (4',7-dihydroxy-6-methoxyisoflavone) constitutes 5–10% of total isoflavones in most soybean foods and ~40% of total isoflavones in soy germ (SoyLife). Glycitein disposition and bioavailability were assessed in humans and its estrogenicity was assessed in mice. Total isoflavones of 4.5 µmol/kg body weight were fed to seven men and seven women from soymilk or SoyLife in a single meal in a randomized crossover design. Urinary excretion and plasma content of isoflavones were analyzed by HPLC-UV. Interindividual variation in isoflavone disposition was controlled by selecting subjects of the moderate fecal isoflavone degradation phenotype (average in vitro fecal genistein half-life of 8.9 ± 4.3 h). Urinary excretion as a percentage of ingested dose differed significantly among isoflavones as follows: glycitein, 55%; daidzein, 46%; and genistein, 29% (P < 0.05). Plasma isoflavone contents after soymilk feeding paralleled soymilk isoflavone contents (genistein daidzein glycitein; P < 0.05) in both sexes. Plasma isoflavones paralleled SoyLife isoflavone contents in men (daidzein glycitein genistein), but at 6 h after dosing, plasma glycitein and genistein did not differ in women. The sex difference may be due to isoflavone biotransformation differences and may require further study. Weanling female B6D2F1 mice were dosed with glycitein or genistein (3.0 mg/d for 4 d) or diethylstilbestrol (0.03 µg/kg for 4 d). Uterine weight was increased by 150% by glycitein, 50% by genistein and 60% by diethylstilbestrol, all significantly greater than the control. Glycitein was threefold more estrogenic than genistein (P < 0.05), suggesting that although glycitein is present in lesser amounts in most soy ingredients and foods, it may be responsible in part for certain biological effects of soy-containing foods.

Metabolites of Dietary Phytoestrogens Daidzein, Genistein and Glycitein. S. Heinonen, K. Wähälä* and H. Adlercreutz. Folkhälsan Institute for Preventive Medicine, Nutrition and Cancer, and Department of Clinical Chemistry, University of Helsinki, Finland and *Laboratory of Organic Chemistry, Department of Chemistry, University of Helsinki, Finland.

Soy and soy-based foods are rich sources of the isoflavones daidzein, genistein and glycitein. The beneficial properties of these dietary phytoestrogens include prevention of hormone-dependent diseases, such as breast and prostate cancers, osteoporosis and cardiovascular disease. In this investigation, the metabolites of soy isoflavones were studied in humans after soy was consumed. The isolation and the characterization of the urinary metabolites were carried out with absorption chromatography on Sephadex LH-20 and gas chromatography–mass spectrometry (GC–MS). The structures of the isolated isoflavones were confirmed by using authentic reference compounds. We identified dihydrogenistein, 6'-OH-O-desmethylangolensin, and cis-4-OH-equol in addition to the known isoflavonoids, daidzein, genistein and glycitein, and the known metabolites, equol, O-desmethylangolensin and dihydrodaidzein, by comparing the retention times and the spectra of the urinary compounds with those of reference standards. For the first time, the metabolites of glycitein were investigated, and new compounds such as 5'-OH-O-desmethylangolensin and 5'-methoxy-O-desmethylangolensin were isolated and identified tentatively by GC–MS. The metabolic pathways for daidzein, genistein and glycitein are presented on the basis of the isolation and identification of these isoflavonoids from human urine.

P lasma Isoflavone Concentrations in American Men and Women Consuming Different Levels of Isolated Soy Protein for up to 6 Months. S. Teixeira, S. M. Potter* and J. W. Erdman, Jr. Division of Nutritional Sciences, University of Illinois at Urbana-Champaign, Urbana, IL and *Protein Technologies International, St. Louis, MO.

Isoflavones are plant compounds with mild estrogenic activity. These compounds are found in high concentrations in soybeans and in many soy foods. It has been known for some time that Asian populations with high intakes of isoflavone-rich products have high concentrations of these compounds in blood and concomitantly a lower incidence of cardiovascular disease, several types of cancer, osteoporosis and menopausal symptoms. As part of two larger studies (a men’s and a women’s study) on the effects of soy protein consumption on blood lipid concentrations, we measured plasma concentrations of several isoflavones by HPLC coulometric array detection (eight-channel detector) and HPLC mass spectrometry. The plasma levels of daidzein, dihydrodaidzein, O-desmethylangolensin, equol, genistein, p-ethyl phenol, glycitein and total isoflavones were analyzed in 81 men and 66 postmenopausal women. In both studies before the soy feeding, the subjects consumed a National Cholesterol Education Program Step I diet for 3 wk (men) or 2 wk (women). In the men’s study, 92 subjects, divided into five groups, were fed 50 g/d of a soy protein–casein mixture in different proportions (0:50, 20:30, 30:20, 40:10 and 50:0 g/d) for 6 wk, with corresponding levels of isoflavones (0, 29.1, 44.8, 62 and 95.1 mg aglycones/d). Blood isoflavones were measured at 0, 3 and 6 wk. In the women’s study, three groups (n = 81 subjects) were fed 40 g/d of soy protein with no isoflavones (casein only) or a moderate (55.6 mg aglycones/d) or high level (90 mg aglycones/d) of isoflavones. The subjects received the study protein for 6 mo. Blood isoflavones were measured at 0 and 24 wk. The effects of chronic consumption of different amounts of isoflavones on blood isoflavone concentrations were analyzed by multiple linear regression. The outcome measured was the isoflavone concentration for each subject, with treatment effects represented as each group contrasted with the control group (casein). In the men’s study, soy feeding resulted in higher plasma concentrations of all isoflavones except for O-desmethylangolensin and glycitein. In the women’s study, soy feeding resulted in higher plasma concentrations of all isoflavones analyzed. Plasma isoflavone concentrations after soy feeding, except for equol, were higher in women than in men. At baseline, plasma isoflavone concentrations were higher in men. [Supported in part by the Illinois Soybean Program Operating Board, the Illinois Council for Food and Agriculture Research, and Protein Technologies International.]

Novel Chlorinated and Nitrated Derivatives of Soy Isoflavones Formed by Proinflammatory Oxidants. Brenda J. Boersma, Rakesh P. Patel, Marion Kirk, Victor M. Darley-Usmar and Stephen Barnes. University of Alabama at Birmingham, Department of Pharmacology and Toxicology, Birmingham, AL.

Several chronic inflammatory diseases including atherosclerosis and many types of cancer are associated with the production of the proinflammatory oxidants hypochlorous acid (HOCl) and peroxynitrite (ONOO^-). Because the soy isoflavones have structural similarities with tyrosine, a biological target of HOCl and ONOO^-, we hypothesized that isoflavones react with HOCl and ONOO^-, thereby modulating the biological responses to these proinflammatory oxidants. In initial studies carried out in vitro using HPLC and liquid chromatography–mass spectrometry (LC–MS), we demonstrated that the isoflavones genistein and biochanin A are converted to their mono- and dichlorinated derivatives by HOCl, whereas daidzein forms a monochlorinated derivative. In the case of ONOO^-, mononitrated products of genistein and daidzein, but not biochanin A, were formed. Treatment of genistein with HOCl and sodium nitrite led to the formation of a doubly substituted chloro-nitro derivative. In biological experiments, we used polymorphonuclear neutrophils (PMN) that were activated with phorbol 12-myristate 13-acetate (PMA) to simulate a respiratory burst that generates several reactive oxygen species, including HOCl. When genistein, daidzein or biochanin A was added to this cellular system, monochlorinated derivatives identical to those found in our in vitro studies were detected by LC–MS. In addition, human leukemia (HL-60) cells, which can be differentiated with dimethylsulfoxide to form PMN-like cells, are activated with PMA to produce reactive oxygen species. Addition of each of the isoflavones to these cells also led to the formation of similar amounts of the monochlorinated products. These results indicate that the chlorinated derivatives of the isoflavones could be formed under pathologic conditions when reactive oxygen species are generated. By using the HL-60 cells, we have a model system in which we can investigate the formation and properties of these novel isoflavone metabolites in a renewable biological system.

Flavones and Isoflavones: Synthesis and Isotopic Labeling. Matthew R. Benton and Nigel P. Botting. School of Chemistry, University of St. Andrews, Andrews, Fife, UK.

At present, we are interested in the synthesis of isoflavones and flavones. Both of these classes of compounds are eliciting considerable interest because of their biological effects. In particular, they show anticancer activity and are being investigated as potential chemopreventive agents. The poster describes our studies on the synthesis of novel isoflavones and the development of improved routes for the synthesis of isoflavone metabolites (e.g., glucosides or glucuronides). Recently, we have also been developing methods for the synthesis of isoflavones containing multiple ¹³C atoms. These derivatives can be used as internal standards for analysis and also in metabolic studies. This work is now being extended to examine the synthesis of ¹³C-labeled flavones. Two of the initial targets are apigenin and naringenin labeled with three ¹³C atoms.

Maternal and Cord Blood Phytoestrogen Levels in Indonesian Women. Fabien S. Dalais, Andreanyta Meliala and Mark L Wahlqvist. International Health and Development Unit, Monash University, Alfred Hospital, Prahran, VIC, Australia.

Epidemiologic, cell, animal and human studies have suggested a potential beneficial link between phytoestrogens from soy and cancer, cardiovascular disease, bone metabolism and menopausal symptoms. Japanese individuals have among the lowest rates of hormone-dependent cancer and low rates of cardiovascular disease and menopausal symptoms. It has been shown that Japanese mothers and their infants have high levels of isoflavones in their plasma and cord blood. Indonesian individuals also consume high levels of soybean in the form of tempeh and tofu. The aim of this study was to determine whether Indonesian mothers and their infants were exposed to levels of isoflavones similar to those of the Japanese individuals and possibly received similar health benefits. Blood and cord blood samples were collected from 30 women giving birth at Bethesda Hospital in Yogyakarta, Indonesia. The maternal levels of the isoflavones genistein and daidzein (mean ± SEM) were 83.1 ± 11.7 and 28.9 ± 6.49 nmol/L, respectively, and the cord blood levels were 91.7 ± 12.5 and 33.9 ± 5.62 nmol/L, respectively. On average, these levels are marginally lower than those published for Japanese individuals. Because these individuals are exposed to isoflavones at an early stage of life, these compounds may modify hormonal metabolism and in turn alter disease profiles later in life.

Absorption of Soy Isoflavone Aglycone in Humans and Its Antiatherosclerotic Effect in Rabbits. Toru Izumi, Jun Yamakoshi, Akio Obata, Koichiro Tobe, Makoto Saito, Shigehiro Kataoka and Mamoru Kikuchi. Kikkoman Corporation, Research and Development Division, Noda City, Japan.

Isoflavone is one of the biologically active compounds in soybeans. There are two types of soy isoflavones, i.e., glucoside forms (IFG) and aglycone forms (IFA). The absorption of IFG has been reported but that of IFA has not yet been reported. We tested the absorption of IFA in humans and their antiatherosclerotic effect in rabbits. We measured the isoflavone concentration in plasma by HPLC after the intake of IFA and IFG by humans. IFA were absorbed faster and in larger amounts than were IFG from both low (0.11 mmol) and high (1.7 mmol) single intakes. The plasma concentration of genistein was higher than that of daidzein after intakes of IFA and IFG. We also tested the antiatherosclerotic effect of IFA and IFG in cholesterol-fed rabbits. IFA significantly inhibited the progression of atherosclerosis in a dose-dependent manner (IFA 1.0%, P < 0.01; IFA 0.33%, P < 0.05). In rabbits fed a diet containing 0.55% IFG (equimolar to 0.33% IFA), the progression was not significantly inhibited. Our data show that IFG are not directly absorbed from the gut but are converted to IFA by intestinal bacteria and then are absorbed. Thus, IFA are more effective than IFG for the prevention of atherosclerosis.

	Osteoporosis

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Soy Consumption and Bone Mineral Density in Older Japanese American Women in King County, Washington. Madeline M. Rice,* Andrea Z. LaCroix,* Johanna W. Lampe, Gerald van Belle,* Mark Kestin,*** Barbara L. Drinkwater, Amy B. Graves and Eric B. Larson*. *University of Washington, Seattle, WA; Fred Hutchinson Cancer Research Center, Seattle, WA; **Bastyr University, Seattle, WA; Pacific Medical Center, Seattle, WA; and University of South Florida, Tampa, FL.

The purpose of this study was to examine the relation between soy consumption and bone mineral density (BMD) in 267 older Japanese-American women aged 65–93 y. Soy consumption was measured using a 14-item soy food-frequency questionnaire. Soy isoflavone intake was estimated by using published isoflavone (genistein plus daidzein) content of soy foods. BMD of the hip and spine was measured by using dual-energy X-ray absorptiometry. Least-squares means general linear models were used to estimate mean BMD according to categories of current and lifetime soy consumption. Current soy isoflavone intake was grouped by tertiles (low, moderate, high). Lifetime soy intake was categorized as low (lowest current isoflavone tertile with same or less intake during adolescence, late 20s and late 40s), high (highest two tertiles with same or more intake during adolescence, late 20s and late 40s) and varied (all other women). All analyses were adjusted for age, weight, language spoken at the interview, age at menarche and postmenopausal years without estrogen. Femoral neck BMD in women who consumed high amounts of soy throughout life was 0.680 g/cm² compared with 0.628 g/cm² in women who consumed very little soy throughout their lifetime (P = 0.03). Among women currently using fiber supplements, lumbar spine BMD was 0.968 g/cm² in women in the highest tertile of current isoflavone intake compared with 0.843 g/cm² in women in the lowest tertile (P = 0.01). No association was observed between current isoflavone intake and lumbar spine BMD in women who were not using fiber supplements. Women using postmenopausal estrogen appeared to benefit most from high soy consumption, although this effect modification was not significant. Postmenopausal estrogen users who were high soy consumers had the highest BMD at all sites.

Urinary Isoflavone Levels and Several Factors That Influence Bone Metabolism in Postmenopausal Women. Chung Ja Sung, Sun Hae Choi and Byoung-Seob Ko.* Department of Food and Nutrition, Sookmyung Women’s University, Seoul, Korea and *Korea Institute of Oriental Medicine, Seoul, Korea.

Several studies showed that isoflavones in soy protein are responsible for its bone-sparing effects. Soy products contain large amounts of isoflavones, which have estrogenic and antiestrogenic properties, and may protect against postmenopausal osteoporosis. The purposes of this study were to investigate the association between urinary isoflavones and a bone resorption biochemical maker, deoxypyridinoline (DPD), and to investigate the correlation of DPD with several factors, including bone mineral density (BMD) of the spine and femoral neck. We interviewed 160 postmenopausal women aged 47–85 y and selected 60 women who had a higher frequency of soy product consumption to answer a questionnaire about soy food consumption. We administered a 24-h dietary recall, made anthropometric measurements and collected spot urine samples. BMD was assessed by dual-energy X-ray absorptionmetry. Urinary DPD and pH values were measured and major isoflavones (genistein and daidzein) were analyzed by using HPLC. The concentration of urinary isoflavones was negatively correlated with urinary DPD (P < 0.01). DPD level was negatively correlated with weight and height. Urinary isoflavone levels were positively correlated with age. With multiple regression analysis, DPD was negatively associated with urinary genistein and femoral BMD; spinal BMD was positively associated with femoral BMD, body mass index and urinary pH; and femoral BMD was negatively associated with DPD. We conclude that urinary isoflavone levels, urinary pH and body mass index affect urinary DPD and BMD. Therefore, consuming soy food can be one way to protect against bone resorption.

Effects of Soy Isoflavones, Daidzein, Genistein and Glycitein, on Bone Loss and Lipid Metabolic Pathway in Ovariectomized Rats. Hitoshi Ishida, Takehiko Uesugi,* Toshiya Toda* and Kuniro Tsuji. School of Pharmaceutical Science, University of Shizioka, Shizuoka Japan and *Fujicco Co. Ltd., Kobe, Japan.

It would be helpful to discover a natural dietary substance that would minimize the risk of bone loss and normalize lipid metabolism in postmenopausal women. Recently, some soy products containing isoflavones (e.g., daidzein, genistein or glycitein) were shown to have positive effects on bone density and to reduce abdominal fat in ovariectomized (ovx) rats. It was found that ipriflavone, which is structurally related to soy isoflavones, shows a similar effect. Therefore, we hypothesized that soy isoflavones might be effective in ameliorating the bone loss and hypercholesterolemia due to ovarian hormone deficiency. To test our hypothesis, we studied the effects of daidzein, genistein and glycitein on bone loss and lipid metabolism in ovx Sprague-Dawley rats (age 11 wk). Each compound was administered orally to ovx rats for 4 wk. The femurs of these rats showed significantly lower density and breaking strength than did those of sham-operated rats. These changes were largely prevented in rats that received daidzein, genistein or glycitein at a dose of 50 mg/(kg · d) and in rats that received subcutaneous estrone [7.5 mg/(kg · d)] as a positive control. Ovariectomy caused atrophy of the uterus and increased the ratio of the urinary excretion of pyridinoline and deoxypyridinoline to endogenous creatinine. This was prevented by administration of daidzein, glycitein or estrone but, interestingly, not genistein. With respect to food intake and body weight, rats in the ovx group had significantly higher final body weights and food intake than did rats in the sham-operated group. Daidzein and glycitein, like estrone, prevented this ovariectomy-induced increase in body weight gain and food intake, whereas genistein did not. Daidzein and glycitein reduced abdominal fat and the level of serum total cholesterol in comparison with the control group as did estrone. The results indicate that daidzein and glycitein seem to be proestrogenic compounds; genistein appears to have a mechanism and site of action different from those of these two compounds.

Relationship Between Urinary Isoflavones and Bone Metabolism in Postmenopausal Japanese Women. Y. Fukui, A. Miura,* Y. Nara,** T. Uesugi, K. Honda and Y. Yamori.* Fujicco Co. Ltd., Kobe, Japan; *WHO Collaborating Center for Research on Primary Prevention of Cardiovascular Diseases, University of Kyoto, Kyoto, Japan; Department of Environmental Preservation and Development, Graduate School of Human and Environmental Studies, University of Kyoto, Kyoto, Japan; and **Graduate School of Integrated Science and Art, University of East Asia, Shimonoseki, Japan.

Soy isoflavones are recognized as having beneficial effects on bone health. This study examines the effect of daily isoflavone intake on bone metabolism. We investigated the association of urinary isoflavones excretion with bone density or urinary bone resorption markers in postmenopausal Japanese women. Subjects included two populations as follows: middle-aged Japanese women (n = 39, 54.8 ± 3.3 y) living in Japan (JJ) and elderly Japanese women (n = 48, 74.4 ± 3.7 y) living in Hawaii (JH). Urine samples were collected for 24 h and were analyzed for the urinary isoflavones daidzein and genistein and bone resorption markers pyridinoline and deoxypyridinoline by HPLC. The bone density, given as the stiffness value, was estimated by an ultrasonic bone densitometer. All subjects were divided by stiffness values into three groups as follows: the higher stiffness, middle stiffness and lower stiffness groups. In both populations, the higher stiffness group had significantly higher urinary isoflavone (daidzein and genistein) excretions than did the lower stiffness group (JJ: 20.0 ± 12.8 and 8.6 ± 7.7 µmol/d; JH: 26.6 ± 24.5 and 9.7 ± 8.9 µmol/d, respectively, for the higher stiffness and lower stiffness groups; P < 0.05). Significant inverse correlations were found between urinary isoflavones excretion and bone resorption markers pyridinoline and deoxypyridinoline in JJ (r = -0.338, P < 0.05 and r = -0.387, P <0.05, respectively, for pyridinoline and deoxypyridinoline). Isoflavones taken from daily Japanese diets were related inversely to bone resorption markers and positively to bone density in menopausal women. Thus, a sufficient daily isoflavone intake is expected to contribute to the prevention of postmenopausal osteoporosis.

Effects of Soy Isoflavones on Blood Lipids, Blood Pressure and Biochemical Markers of Bone Metabolism in Postmenopausal Women. William W. Wong. USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, TX.

Cessation of estrogen production with menopause results in hypercholesterolemia, hypertension and elevated bone loss in postmenopausal women. Estrogen replacement therapy lowers blood lipids, increases arterial compliance and minimizes bone loss in postmenopausal women but increases the risk of cancer. Isoflavones, found in abundance in soybeans, are weak estrogens and were shown to inhibit mammary cancer cell formation and growth. To determine the effects of soy isoflavones on blood lipids, blood pressure and biochemical markers of bone metabolism in postmenopausal women, the fasting blood lipid concentrations, blood pressure and biochemical markers of bone metabolism of six postmenopausal women [ages 55.4 ± 3.5 y, (mean ± SD); body weight, 75.0 ± 10.7 kg; height, 164.7 ± 5.2 cm] ingesting 160 mg/d of soy isoflavones were measured before and after 6 wk of treatment. After isoflavone treatment, no significant changes were observed in body weight (-0.8 ± 2.1%, P = 0.40) or in blood concentrations of total cholesterol (0.5 ± 7.5%, P = 0.88) or LDL cholesterol (-0.5 ± 9.3%, P = 0.90). However, an increase of 10.2 ± 10.1% in apolipoprotein B100, the major lipoprotein in LDL cholesterol, was detected (P = 0.06). Blood concentrations of HDL cholesterol and its major lipoprotein, apolipoprotein A1, were elevated by 5 ± 14% and 3 ± 11%, respectively, although these increases were not significant (P 0.45). No significant change in systolic blood pressure was observed; however, there was a significant reduction in diastolic blood pressure of 12 ± 9% (P <0.03). Isoflavone treatment affected biochemical markers of bone resorption; urinary concentrations of deoxypyridinoline and calcium, serum concentration of parathyroid hormone and urinary calcium excretion were modified by -7 ± 41%, -12 ± 46%, +33 ± 60% and -5 ± 29%, respectively. Biochemical markers of bone formation also were altered by isoflavone treatment; serum concentrations of osteocalcin, bone alkaline phosphatase and insulin-like growth factor I were reduced by 9 ± 15, 16 ± 23 and 8 ± 27%, respectively. However, the changes in biochemical markers of bone metabolism were not significant. Although many of the changes in blood lipids and biochemical markers of bone metabolism induced by the soy isoflavone treatment were not significant because of the small sample size, the size of many of these changes was similar to that reported in postmenopausal women receiving estrogen replacement therapy. Therefore, further studies with a larger number of subjects, a longer treatment period, measurement of lumbar spine bone mineral content and bone density and measurement of calcium kinetics using the dual-tracer technique are warranted. [Funded by the USDA/ARS; the soy isoflavone tablets were provided by Schouten USA Inc., and the Pyrilinks-D kits, for the measurement of urinary Dpd, were kindly provided by Metra Biosystems, Inc.]

	Cancer

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Molecular Effects of Soy-Isoflavone Genistein in Prostate Cancer Cells. Joanne N. Davis,* Omer Kucuk* and Fazlul H. Sarkar.** Departments of *Cancer Biology, Internal Medicine, and **Pathology, Karmanos Cancer Institute, Wayne State University School of Medicine, Detroit, MI.

Prostate cancer is the second leading cause of cancer-related deaths in men in the United States, accounting for 36% of all male cancers and 13% of cancer-related deaths in men. Epidemiologic data provide convincing evidence that dietary factors play an important role in the etiology of cancer. We previously demonstrated that the dietary isoflavone genistein inhibits proliferation, induces apoptosis and modulates important cell cycle regulatory molecules, particularly p21WAF1 and cyclin B, in prostate cancer cells, and therefore may be a potential chemopreventive or therapeutic agent. To further elucidate the molecular mechanism by which genistein elicits its effects, we first investigated the role of a transcription factor NF-B; second, we measured prostate-specific antigen (PSA) levels in prostate cancer cells. NF-B was shown to protect cells against apoptosis by initiating prosurvival mechanisms. We investigated whether genistein modulates NF-B, particularly the inactivation, which may lead to the apoptosis observed in genistein-treated cells. Here we show that genistein decreases NF-B activity in prosate cancer cells in a dose-dependent manner. Prostate cancer cells treated with genistein at 30 and 50 µmol/L for 24 h resulted in reduced NF-B DNA binding. Using confocal microscopy, we showed that genistein blocks the translocation of NF-B p50 and p65 subunits from the cytoplasm to the nucleus, preventing NF-B activation and prohibiting DNA binding. Additionally, we demonstrated that genistein abrogates NF-B activation by two known inducers, H₂O₂ and tumor necrosis factor- (TNF-). Prostate cancer cells pretreated with 50 µmol genistein/L for 48 h inhibited NF-B DNA binding and blocked translocation of NF-B subunits to the nucleus when stimulated with either H₂O₂ or TNF-. These results suggest that the inactivation of NF-B by genistein may lead to the cell growth inhibition and apoptosis observed in genistein-treated cells. The most valuable tumor marker used for the detection and monitoring of prostate cancer is PSA. PSA, a member of the kallidrein family, is a serine protease secreted by prostate epithelial cells. PSA, which is able to cleave the predominant seminal vesicle protein, has been proposed as a candidate growth factor, cytokine or growth factor regulator and has been linked to tumor progression. Therefore, we investigated whether genistein has any effect on PSA expression and secretion in the androgen-sensitive prostate cancer cell line, LNCaP. LNCaP cells were treated with genistein at 0, 30 and 50 µmol/L for 3 d. The medium was collected and assayed for the presence of PSA. We observed that treatment with genistein at 30 µmol/L reduced PSA secretion by 50% and 50 µmol/L reduced PSA by 80% compared with control. Using immunohistochemistry and Western blot analysis, we determined that genistein inhibits PSA protein expression levels but did not affect the protein expression levels of another tumor-associated antigen, prostate-specific membrane antigen. These results indicate that genistein lowers the PSA levels in prostate cancer cells in vitro. In conclusion, the inactivation of NF-B and downregulation of PSA by genistein provide encouraging evidence to support genistein’s role as a chemopreventive and therapeutic agent for prostate cancer. These results also indicate that NF-B may play a pivotal role in genistein-induced apoptosis, providing a mechanism by which genistein promotes cell death.

The Specific Role of Genistein in Estrogen Metabolism. N. B. Kumar, K. Allen, A. Cantor, G. Shaw and C. E. Cox. H. Lee Moffitt Cancer Center and Research Institute at the University of South Florida, Tampa, FL.

Our goal was to evaluate the individual effectiveness of supplementing a group of premenopausal, breast cancer–free women with a dietary supplement of the isoflavone genistein (40 mg/d) in producing a change in sex hormones that are implicated in the initiation and promotion of breast cancer. Consecutively recruited premenopausal omnivorous women (n = 68), of all races and ethnicities, aged 25–55 y, were admitted to the study and randomly assigned to an experimental group supplemented with soy (40 mg genistein/d) or to a control group consuming a placebo for a 12-wk period. Changes in their anthropometric, nutritional and hormonal biomarkers from the early follicular phase were analyzed at baseline and after intervention. Preliminary analysis indicated that hormonal concentrations of free estradiol decreased by 78% in the group consuming genistein compared with 44% in the placebo group. Serum estrone and sex hormone–binding globulin levels were elevated in 47 and 46% of the subjects, respectively, in the experimental group compared with 26 and 40%, respectively, in the placebo group. In the experimental group, the menstrual cycle of 58% of the subjects increased by more than 2 d compared with 36% in the placebo group. These data suggest that increased genistein intake affects estrogen metabolism by altering the sex hormone concentrations that are implicated in breast cancer promotion or inhibition. [Supported by an NIH NCI grant RO3 CA72588-01A1.]

Effect of Soybean Saponins on the Growth and Antioxidant Defense of Human Hepatocarcinoma Cells. M.-K. Sung and M.-Y. Park. Department of Food and Nutrition, Sookmyung Women’s University, Seoul, Korea.

Carcinogenesis is a multistep process including initiation, promotion and progression. Recent studies indicated that oxygen free radicals, by-products of normal cellular respiratory processes as well as lipid peroxidation processes, induce cellular DNA damage, which is the most plausible mechanism for the initiation of carcinogenesis. Lipid peroxides also were shown to promote tumor cell growth. Saponins are amphiphilic compounds present in a variety of edible and nonedible plants. Recent studies indicated that saponins extracted from soybeans inhibit the formation of lipid peroxides in corn oil samples; this may be due to their ability to scavenge radicals. In this study, effects of soybean saponins on the growth, cellular lipid peroxidation and antioxidative enzyme activities of HepG2 cells were investigated. Effects of saponins were compared with -tocopherol and ascorbic acid. Cells (1–2 x 10⁷) were incubated for 24 h and then treated with tert-butylhydroperoxide (0.5 nmol/L for 45 min) to initiate lipid peroxidation followed by saponin treatment (300 µg/plate for 48 h). Cellular superoxide dismutase (SOD), glutathione peroxidase (GPX) and glutathione S-transferase (GST) activities were measured. Results showed that tert-butylhydroperoxide treatment significantly increased cellular malondialdehyde content. Cell growth was significantly decreased with saponin, -tocopherol and ascorbic acid treatment. Malondialdehyde content was significantly reduced by saponin (72%) and -tocopherol (40%). Soybean saponins significantly increased cellular SOD, GPX and GST activities. Ascorbic acid significantly decreased GPX activity. However, the activity of GST was not affected by either -tocopherol or ascorbic acid. These results indicate that soybean saponins possess considerable antioxidative capacity, exerting antiproliferative effects on tumor cells.

Soybean Saponins Inhibit the Formation of DNA Adducts in Human Colon and Liver Cells. H.-S. Jeon and M.-K. Sung. Department of Food and Nutrition, Sookmyung Women’s University, Seoul, Korea.

Numerous chemical carcinogens, activated to form electrophilic agents, react with DNA, which explains the induction of a heritable change in a cell leading to malignant transformation. This may be a main event in the initiation of carcinogenesis. Soybeans contain up to 2% saponins. Soybean saponins were shown to inhibit the growth of human colon carcinoma cells with low toxicity. Also, they were shown to decrease the ornithine decarboxylase activity that is directly related to cancer cell proliferation. These results indicate that soybean saponins are important modulators in the promotion stage of carcinogenesis. This study was performed to examine the effects of soybean saponins on DNA adduct formation, which is the most important reaction of carcinogens with cellular macromolecules initiating carcinogenesis. CCD-18Co and HepG2 cells were used as models for colon and liver cells, respectively. Cells (4–5 x 10⁵) were seeded and allowed to attach. After 18 d (CCD-18Co) and 2 d (HepG2) in culture, soybean saponins at a concentration of 0–50 µg/mL were added and incubated for 1 h. Preincubated tritiated aflatoxin B1 (12 nmol/L; specific activity 25 Ci/mmol) was added to each plate and incubated for a further 48 h. DNA was purified and aliquots of DNA samples were used to measure radioactivity by liquid scintillation counting. Results showed that soybean saponins significantly inhibited the formation of DNA-aflatoxin B1 adducts in CCD-18Co cells by 23.7, 50.7 and 49.4% when 10, 30 and 50 µg saponins/mL, respectively, were used. Amounts of DNA adducts in HepG2 cells were also significantly decreased by 37.3, 50.1 and 49.8%, respectively. These results indicate that soybean saponins may effectively reduce cellular DNA damage by carcinogens and can be regarded as potential chemopreventive agents.

Effects of Genistein on the Activities of Antioxidant Enzymes in Strenuously Exercised Rats. C. Chen and R. M. Bakhit. Department of Human Nutrition, Foods and Exercise, Virginia Polytechnic Institute and State University, Blacksburg, VA.

Male Sprague-Dawley rats (n = 48), aged 1 y, were assigned to four groups as follows: normal diet/sedentary, normal diet/exercise, genistein diet/sedentary and genistein diet/exercise. The AIN complete basal diet was supplemented with 500 mg genistein/kg. After 4 wk of consuming the experimental diets, the rats in the exercise groups underwent an acute exercise protocol of 22 m/min at 12° inclination for 1 h. Immediately after the exercise, blood was drawn from each rat and tested for antioxidant enzymes [superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GPX)]. Genistein concentration was measured in the plasma, liver and muscle. The average feed intakes were not significantly different among the four groups (P 0.05). Plasma, liver and muscle genistein concentrations were significantly higher in the rats undergoing acute exercise than in the nonexercised rats (P < 0.05). In addition, in the exercised rat groups, the genistein-fed rats had significant increases in the activities of two antioxidant enzymes, GPX and CAT, compared with exercised rats fed the normal diet (P < 0.05). GPX and CAT levels in the exercised genistein-fed rats were comparable to those of the sedentary rats. No significant changes were observed for SOD. The results suggest that acute strenuous exercise might lead to the release of relatively more genistein from storage tissue into circulation than would occur with no exercise. There is also an indication that 500 ppm genistein from dietary supplementation might diminish the disadvantages of increased production of free radicals resulting from acute exercise by maintaining the antioxidant defense systems in the acutely exercised rats. [Funded by the Virginia Soybean Association.]

Familial Breast Cancer Dietary Prevention Demonstration Trial. Cristina Bellati and Franco Berrino. Istituto Nazionale Tumori, Milan, Italy.

High serum levels of sex hormones and insulin-like growth factor I (IGF-I) usually precede breast cancer. We recently conducted a randomized dietary trial (the DIANA study) in which a comprehensive modification of Western dietary habits significantly reduced the bioavailability of such hormones (testosterone and estradiol decreased by 18%; sex hormone–binding globulin and IGF binding protein increased). The dietary intervention aimed at reducing insulin resistance and increasing the intake of phytoestrogens, both soy isoflavonoids and lignans from various sources. Insulin inhibits the liver synthesis of sex hormone–binding globulin and IGF-BP and stimulates the ovarian production of sex hormones, whereas phytoestrogens have the opposite effect. The penetrance of genes responsible for hereditary breast cancer is likely to be affected by several environmental factors, including ionizing radiation, tobacco smoking, oral contraceptive use and nutritional habits. The latter refer mainly to fruit and vegetable consumption and the dietary regulation of the availability of sex steroid hormones and IGF and related peptides. Evidence is increasing that these nutritional factors are more strongly related to genetics than to sporadic breast cancer. We are presently carrying out a case-only study (the COS Study) to test this interaction in several thousand European women who were diagnosed with breast cancer before age 40. Seven countries are involved in the study. In Italy, patients who most likely carry high penetrance mutations together with their healthy sisters are being invited to participate in a dietary prevention trial. The dietary modification strategy for the intervention group includes kitchen courses and behavioral and psychological support. Families randomly assigned to the control group receive the usual cancer prevention dietary recommendations, but active intervention will be postponed for several years. The endpoint will be breast cancer in healthy women and contralateral breast cancer in patients. A study including 400 mutation carriers per group has 80% statistical power to detect a 50% reduction in breast cancer incidence.

Urinary Phytoestrogens and Breast Cancer Risk in a Prospective Study in Postmenopausal Dutch Women. P.H.M. Peeters,* I. den Tonkelaar,* P. Van’t Veer,** L. Keinan-Boker,* C.M.J. Arts, H. Adlercreutz and J.H.H. Thijssen. *Julius Center for Patient Oriented Research, University Medical Center, Utrecht, The Netherlands; International Health Foundation, Utrecht, The Netherlands; **Division of Human Nutrition and Epidemiology, Wageningen University, Wageningen, The Netherlands; TNO Nutrition and Food Research Center Institute, Zeist, The Netherlands; Department of Clinical Chemistry, University of Helsinki, Helsinki, Finland; and Department of Endocrinology, UMCU, Utrecht, The Netherlands.

Phytoestrogens, naturally occurring compounds in many foods and especially in soy products, are defined as plant substances that are structurally or functionally similar to estradiol. Two former retrospective studies assessed the relationship between urinary phytoestrogen excretion and breast cancer risk; their results indicated a protective role for phytoestrogens. Urine samples in those studies were collected after breast cancer diagnosis, and urinary phytoestrogen levels (a short-term indicator) might have been affected by the disease state. We chose, therefore, to study the associations of certain phytoestrogens with breast cancer risk by using urinary specimens collected several years before breast cancer was diagnosed. Subjects were 88 postmenopausal women with breast cancer (cases) and 286 postmenopausal women without breast cancer (controls) selected from a cohort of women (n = 14,697) who participated in a breast cancer screening program, the DOM Project, in Utrecht, The Netherlands. Levels of genistein and enterolactone were determined by time-resolved fluorescent immunoassay and expressed in micromoles per mole creatinine. For each subject, the mean value for genistein and enterolactone was computed from two urinary samples collected 1 y apart. Odds ratios of the highest to the lowest tertiles of urinary phytoestrogen per creatinine concentrations were computed. Higher urinary genistein excretion was weakly associated with a reduced breast cancer risk, i.e., the odds ratio for the highest compared with the lowest tertile was 0.83; the 95% confidence interval (CI) was 0.46–1.51. In contrast, higher urinary enterolactone excretion was weakly and nonsignificantly associated with an increased breast cancer risk, i.e., the odds ratio for highest compared with lowest tertile was 1.43, 95% CI, 0.79–2.59. Tests for trends for both phytoestrogens were nonsignificant. We were not able to detect the previously reported firm protective effect of phytoestrogens. Such an effect may be smaller than expected or limited to premenopausal women.

Phytoestrogen and Androgen Levels in Australian Postmenopausal Women Diagnosed with Breast Cancer. A. L. Murkies, F. S. Dalais,* E. M. Briganti, D. L. Heal, H. G. Burge, M. L. Wahlqvist* and S. R. Davis. The Jean Hailes Foundation, Clayton, VIC, Australia; *International Health and Development Unit, Monash University, Clayton, VIC, Australia; Department of Epidemiology and Preventive Medicine, Monash University, The Alfred Hospital, Prahran, VIC, Australia.

Isoflavones are phytoestrogens, plant-derived compounds, that have been implicated as potential anticarcinogenic compounds by in vitro, animal, epidemiologic and human case-control studies. The role of endogenous androgens is still being investigated. The aim of this study was to assess the association between urinary excretion of isoflavones (daidzein and genistein), dietary composition and circulating androgens and their 24-h urinary metabolites in postmenopausal women and the risk of breast cancer. Cases (n = 18) and controls (n = 20) completed a detailed dietary questionnaire; a blood sample and a 24-h urine sample were collected. Analysis and detection of phytoestrogens were carried out by HPLC and UV. Nonparametric analysis was undertaken (Mann-Whitney U test) to compare groups. There were no significant differences in lifestyle and reproductive variables between groups. Urinary daidzein excretion was significantly lower in breast cancer patients [0 nmol/24 h; 0, 1375.5 (median and interquartile ranges)] than in control subjects (1558.2 nmol/24 h; 498, 2773.2; P = 0.02), and there was a trend for decreased genistein (P = 0.08) for patients compared with control subjects. Serum testosterone was elevated in the patients (1.2 nmol/L; 1, 1.6) more than in the control subjects (1 nmol/L; 0.7, 1.25; P = 0.05), whereas no significant differences were observed between the two groups for other hormonal measures or total fat, fiber and vegetable consumption. Women with breast cancer were found to have lower urinary isoflavone levels. The lower urinary phytoestrogens could be attributed to a reduced dietary intake due to future surgery for breast cancer. There was no difference in the usual diet as reported; however, the food intake questionnaire may not have been sufficiently specific for phytoestrogens and may have had limitations in evaluating actual phytoestrogens intake from known sources, such as processed foods. Larger studies are required to examine the association between phytoestrogens and breast cancer more thoroughly while addressing the issues of bias.

Serum Insulin-Like Growth Factor I Is Unaffected by Genistein in Wild-Type or Estrogen Receptor- Knockout Mice. Ruth S. MacDonald, William H. Thornton, Jr., J. Kevin Day and Dennis B. Lubahn. Nutritional Sciences Program, University of Missouri, Columbia, MO.

Consumption of soy foods is ecologically associated with a reduced risk of breast cancer. On the basis of animal experiments, the phytoestrogen genistein is suggested to be one component of soy that provides protection. Recent evidence suggests that the estrogen receptor interacts with growth factor receptors in several cell types to mediate cellular function. In cultured cells, genistein inhibits tyrosine kinase activity, which is an essential component of the insulin-like growth factor-I (IGF-I) signaling pathway. Hence, we examined changes in serum IGF-I in wild-type (WT) and estrogen receptor- knock-out (ERKO) mice fed genistein. WT and ERKO mice were fed a semipurified diet containing 0 (control) or 1 g genistein/kg diet beginning at age 3 wk. The objective of the study was to examine the protective effect of genistein on breast cancer development; hence two medroxyprogesterone pellets (40 mg each) were implanted subcutaneously into each mouse at 6 wk, and DMBA (1 mg/dose) was given orally at 9, 10, 12 and 13 wk. The mice continued to receive their respective dietary treatments for 14 mo. Body weight gain was slightly less in mice fed genistein than in controls throughout the study because of reduced food intake. Tumors developed in all of the WT mice between 29 and 45 wk, but none of the ERKO mice developed tumors. In the WT mice, no protective effect of genistein on breast tumor formation was found. Serum IGF-I concentrations were 199 ± 69, 512 ± 98, 262 ± 60 and 252 ± 85 ng/mL in the WT control, ERKO control, WT genistein and ERKO genistein groups, respectively. When analyzed as a 2 x 2 factorial by ANOVA, there were no significant differences by diet or genotype; however, there was a diet-genotype interaction. The range of serum IGF-I concentration in the WT mice was 13–281 ng/mL and in the ERKO mice was 127–935 ng/mL, including two mice with very high levels. The trend for higher IGF-I in the ERKO mice fed the control diet, which was suppressed by feeding genistein, suggests that genistein influenced serum IGF-I concentrations in these mice. We are continuing to examine the relationship between genistein consumption and the IGF-I axis in both WT and ERKO mice. This animal model provides a useful tool for examining the relationships among estrogen receptors, the IGF-I axis and dietary phytoestrogens. [Funded by Missouri Agricultural Experiment Station Interdisciplinary Regional Research Grant.]

Inositol Hexaphosphate Has an Antioxidant Function That Reduces GST-P⁺ Foci on Hepatocarcinogenesis in Rats. Hae-Jeoung Lee, Hyeon-duck Kim, Sang-A Lee and Haymie Choi. Seoul National University, Department of Food and Nutrition, Seoul, Korea.

Inositol and inositol hexaphosphate (phytate), which are found in plant foodstuffs such as seeds, grains, fruits and vegetables, were demonstrated to have anticancer, anti-cell-proliferation, and antioxidant functions. This study was designed to determine the effects of phytate on rat hepatocarcinogenesis and whether supplementing with inositol would enhance anticarcinogenic and antioxidant effects. Rats received a single intraperitoneal injection of diethylnitrosamine, were subjected to two-thirds hepatectomy 3 wk later and were killed 8 wk after the injection. One week before the partial hepatectomy, the rats were divided into three groups, and inositol or phytate was added to the drinking water (adjusted pH. 7.4); one group received 2% phytate, one received 2% inositol and one received 1% phytate + 1% inositol. In all three groups, the numbers and the areas of glutathione S-transferase–placental positive (GST-P⁺) foci were significantly decreased. Thiobarbituric acid–reactive substances (TBARS) content and catalase and GST activities were significantly reduced in rats fed inositol phosphates. TBARS content and catalase activities were the lowest in the phytate group because phytate potently inhibited the formation of the iron-derived hydroxyl radical and suppressed lipid peroxidation. Catalase and GST activities were not induced likely because of the antioxidant function of inositol phosphates. TBARS content was positively correlated with GST-P⁺ foci. It seemed that inositol phosphates helped the endogenous defense system during carcinogenesis by decreasing TBARS and H₂O₂. Therefore, the preventive effect against hepatocarcinogenesis can be explained in part by the antioxidant function of inositol phosphates, and there were no differences in GST-P⁺ foci, activities of catalase and GST, and TBARS with or without the inositol supplement.

Soy Intake and Colorectal Cancer in Chinese in North America and China. Anna H. Wu, Alice S. Whittemore, Marion Lee, Richard P. Gallagher and Zheng Shu. University of Southern California, Department of Preventive Medicine, Los Angeles CA.

Epidemiologic studies suggest that a high intake of soy foods may reduce the risk of certain types of cancer, including those of the breast, prostate and endometrium. However, the role of soy foods in colorectal cancer is less clear. Our purpose was to examine the relationship of soy foods to risk of colorectal cancer, with a focus on consistency in findings in North America and China, in men and women, or colon and rectal cancer. Cases included persons with histologically confirmed colorectal cancer who were identified through the population-based tumor registries of Los Angeles, San Francisco, Vancouver and Ontario (n = 432 in North America) and in Hangzhou, China (n = 473). Control subjects matched to cases by age, sex and study area were interviewed (n = 1296 in North America, n =1192 in China). Dietary and nondietary (including physical activity and body size characteristics) information was collected during home interviews. Odds ratios were estimated by using conditional logistic regression stratified jointly by age, sex and study area. The intake pattern for soy will be presented by continent, sex, and disease status. Risk of colorectal cancer was 40% lower in association with a high soy intake (at least two times per week compared with fewer than two times per month) for men in North America, but a similar reduction in risk was not observed for women in North America. In China, the data showed no clear patterns of altered risk for colorectal cancer with intake of individual soy foods or all soy foods combined. Further adjustment for other dietary and nondietary factors did not change these findings. Thus, no consistent or strong associations between colorectal cancer risk and consumption of soy foods were found in this study.

Radical-Scavenging Activity in Brown-Colored Soybean Seed Coat. Yashurio Takahata, Shu Furuta, Masakazu Takahashi and Ikuo Suda. Kyushu National Agricultural Experiment Station, Ministry of Agriculture, Forestry and Fisheries, Nishigoshi, Kumamoto, Japan.

Soybeans having a brown- or black-colored seed coat have been used in the daily diet traditionally and occasionally in Japan. However, when the effect of soybeans on human health is discussed, most of the background data are derived from a normal soybean that has a yellowish-white–colored seed coat. Here we investigated the 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging activities of a brown-colored soybean seed coat by using indigenous varieties and breeding materials. Among four indigenous varieties, Akita-Zairai showed the highest radical-scavenging activity. We crossed Akita-Zairai and a normal soybean cultivar to obtain progeny seeds as breeding materials. In the F3 seed generation, seeds were classified roughly into the following four phenotypes on the basis of their seed coat phenotype: yellowish-white, lusterless pale brown, lustrous pale brown and dark brown. DPPH radical-scavenging activity in the lustrous pale brown F3 seed coat was higher than that in dark-brown F3 and the original parental Akita-Zairai. The yellowish-white seed coat showed negligible activity as did the lusterless pale brown. The difference in luster seemed to be a key trait in the intensity of radical-scavenging activity. DPPH radical-scavenging activities paralleled the content of polyphenolics. Using an LH-20 column, we fractionated compounds demonstrating radical-scavenging activities and showed that DPPH radical-scavenging activities and proanthocyanidin content were concomitant in each fraction. We could demonstrate that proanthocyanidins are possible candidates for radical-scavenging activities in the brown soybean seed coat. Soymilk made from cotyledons of normal cultivar and seed coats of Akita-Zairai had 2.5-fold higher radical-scavenging activity than did soymilk made from normal soybean alone. This result indicates that the radical-scavenging activity in brown seed coats is retained after processing.

Effect of Estrogen and Estrogen Mimics for the Growth of MCF-7 Cells in Nude Mice. Keiji Sugi, Takahiko Gotoh, Yin Hong and Akihiro Ito. Department of Cancer Research, Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima, Japan.

The estrogen-responsive human breast cancer cell MCF-7 was examined for its growth in nude mice under various doses of 17-ß-estradiol (E₂), bisphenol A and a miso diet. Seven-week-old nude mice were inoculated subcutaneously in six sites each with 5 x10⁶ MCF-7 cells/site. Mice were treated with intramuscular injection of E₂ at 0.001, 0.01 and 0.1 µg/mouse; oral administration of bisphenol A at 10, 100 or 1000 ppm; or a diet containing 10% miso. Twelve weeks later, inoculated tumor sites and weight were scored. Tumor take (incidence) was highest in control mice (47%, 223 mg). For E₂ administration, they were 22%, 194 mg with 0.001 µg; 25%, 393 mg with 0.01 µg; and 39%, 257 mg with 0.1 µg. Similarly, bisphenol A showed a dose-dependent increase by the incidence of 29%, 565 mg with 10 ppm; 40%, 616 mg with 100 ppm; and 42%, 869 mg with 1000 ppm. For the miso group, the values were 17% and 185 mg. Immunohistochemical staining for antibodies to pS2 estrogen-regulated protein and BrdU showed high frequencies in tumor tissues.

Meta-Analysis of Soy Intake and Breast Cancer Risk. Bruce Trock, Leslie White Butler, Robert Clarke* and Leena Hilakivi-Clarke. Departments of Human Oncology, *Physiology and Biophysics, and Psychiatry, Lombardi Cancer Center, Georgetown University Medical Center, Washington, DC.

High soy intake in Asian countries was proposed as a factor contributing to the low breast cancer risk for Asian women. However, soy is being marketed and recommended to the public as if a clear protective effect was established when, in fact, the epidemiologic data are rather limited. Because in vivo and in vitro data show estrogenic effects for genistein, the major component of soy, it is important that associations between soy and breast cancer risk be evaluated before recommendations can be made with safety, especially for women who already have breast cancer. Therefore, we performed a meta-analysis of epidemiologic studies examining soy and breast cancer risk. A literature search was based on key words associated with soy or specific isoflavones and phytoestrogens, and breast cancer. An Internet search was also conducted to identify unpublished data. We analyzed data by using the measures of soy intake provided in the studies and also normalized intake to daily grams of soy protein. Odds ratios (OR) were pooled by using Mantel-Haenszel methods, and random effects models were used in the presence of significant heterogeneity of OR across studies. A total of nine studies (eight case-control, one cohort) were included in the analysis. Five studies were done in Asian women living in Asian countries; one was conducted in women with Asian ancestry living in the West; and three were conducted in non-Asian populations. A great deal of variability was observed in the size of the OR and in the level of soy intake that was defined as high intake. A modest significant reduction in risk was associated with high soy intake over all studies [OR = 0.87; 95% confidence interval (CI): 0.80, 0.96]. However, this effect was confined to premenopausal women (OR = 0.80; 95% CI: 0.71, 0.90). There was no protective effect at all in postmenopausal women (OR = 1.01; 95% CI: 0.86, 1.19). There was also no significant effect of soy in women in Asia (OR = 0.95; 95% CI: 0.83, 1.09). Although there is some evidence of a small reduction in premenopausal breast cancer risk associated with soy intake, the number of studies is small, measurement of soy intake is crude and control of confounding factors is inconsistent. Interpretation of these results is further complicated by the similar reductions in risk associated with widely varying soy intakes and the low percentage of subjects consuming soy in studies of non-Asians. There was also a lack of consistency in the effects on hormonal measures observed in soy feeding studies. Coupled with the fact that some studies have suggested potentially adverse effects of soy, these data suggest that recommendations for women to increase their soy intake to prevent breast cancer or prevent its recurrence are premature and that larger, more rigorously controlled studies are required.

Premenopausal Equol Excretors Show Plasma Hormone Profiles Associated with Lowered Risk of Breast Cancer. A. M. Duncan, B. E. Merz-Demlow, X. Xu, W. R. Phipps* and M. S. Kurzer. Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN and *Department of Obstetrics-Gynecology, University of Rochester, Rochester, NY.

Increased urinary excretion of equol, a metabolite of the isoflavone daidzein, has been associated with lowered risk of breast cancer (Ingram et al. 1997 ). This risk reduction has generally been presumed to be a consequence of increased isoflavone consumption. However, only 30–40% of the population excretes more than trace amounts of equol regardless of isoflavone intake. Accordingly, we hypothesized that the observed protective effect of equol may be due to hormonal differences between equol excretors and nonexcretors. To evaluate the effects of equol status per se, we compared plasma hormone and binding globulin concentrations between premenopausal equol excretors (n = 5) and nonexcretors (n = 9) consuming identical isoflavone doses (10, 64 and 128 mg/d) as components of soy protein isolates for 3.5 menstrual cycles each. P < 0.05 was considered significant. Urinary equol for excretors far exceeded that of nonexcretors, even at the lowest dose. At all doses, equol excretors generally had lower concentrations of estrone, estrone sulfate, testosterone, androstenedione, dehydroepiandrosterone, dehydroepiandrosterone sulfate and cortisol and higher concentrations of sex hormone–binding globulin and midluteal progesterone, a hormonal pattern generally consistent with lowered breast cancer risk. Thus, the association of lowered breast cancer risk with equol excretion may largely reflect the tendency of equol excretors to have more favorable hormonal profiles rather than merely reflecting increased isoflavone intake. Equol may be a marker for the presence of colonic bacterial enzyme activity that increases fecal steroid excretion. Alternatively, equol itself, even with very modest isoflavone intake, may exert beneficial effects on the regulation of endogenous hormones. [Supported by NIH grants CA-66016 and MO1-RR00400, and a gift from Protein Tech. International.]

Phytoestrogen Effect on Human Breast Epithelium. N. J. Bundred, D. F. Hargreaves,* W. R. Miller, M. Morton, I. McFadyen, A. Howell,** S. A. Roberts* and C. S. Potten.* Academic Department of Surgery, South Manchester University Hospital, Manchester, UK; *Paterson Institute of Cancer Research, Manchester, UK; Department of Surgery, Edinburgh Royal Infirmary, Edinburgh, Scotland; **Academic Department of Medical Oncology, Christie Hospital, Manchester, UK; and Tenovus Institute, Cardiff, Wales.

Epidemiologic studies suggest that phytoestrogen consumption prevents breast cancer, but the mechanism is unclear. To determine the effect of dietary soy protein supplementation on estrogenic- and androgenic-induced proteins in breast secretions, we conducted two randomized controlled trials of 14 d and 3 mo of treatment with either placebo or soy protein in premenopausal women. Levels of phytoestrogens rose in both serum (P 0.001 for genistein and daidzein) and nipple aspirate (genistein: 387 ± 433 ng/mL before treatment; 430 ± 430 ng/mL after 14 d of treatment) after supplementation. Nipple aspirate pS₂ (an estrogen-induced protein) rose and apolipoprotein D (apo D) fell after 14 d and 3 mo of supplementation (Table 1 ). Phytoestrogens exert an estrogenic stimulus on human breast as determined by increased secretion of estrogenic proteins.

Epidemiologic evidence suggesting a correlation between diets high in soybean and overall low cancer mortality rates, especially those of colon, breast and prostate, has given impetus to identifying components in soybeans responsible for their anticancer properties. We isolated a soybean cDNA encoding the small subunit peptide of a cotyledon-specific 2S albumin (Gm2S-1). The peptide (named lunasin) has a unique, highly acidic carboxyl end. A chimeric gene encoding the lunasin peptide tagged with green fluorescent protein arrested cell division and caused abnormal spindle fiber elongation, chromosomal fragmentation and cell lysis when transiently transfected into murine embryo fibroblast, murine hepatoma and human breast cancer cells. Deletion of the acidic carboxyl end abolished the antimitotic effect. Immunolocalization of lunasin and an immunobinding assay using synthetic peptides revealed the preferential adherence of lunasin to chromatin. Lunasin is the first antimitotic peptide whose cDNA was cloned and the first from a common food source. We also showed that the exogenous application of lunasin peptide to as low as 125 nmol/L inhibits the in vitro transformation of mouse embryo fibroblast cells (C3H 10T 1/2) into tumorous foci by the carcinogens 3-methyl cholanthrene and 7,12-dimethylbez[a]anthracene. The lunasin gene has potential application as an antimitotic cancer therapeutic agent, and the lunasin peptide may be an important cancer-preventive compound in soybeans. Furthermore, its antimitotic property suggests that lunasin could play a native role in arresting mitosis that initiates the cell expansion phase of seed development in which DNA endoreduplication and synthesis of storage proteins, lipids and carbohydrates occur.

Potential Effects of Combined Genistein Isoflavone and the Glucocorticoid Dexamethasone on Testicular Cells. James Kumi-Diaka, Andre Butler and Vu Nguyen. Florida Atlantic University, Davie, FL.

A series of experiments was performed to assess the response of testicular cells TM3, TM4 and GC-1 spg to the exposure of genistein, dexamethasone and a combination of genistein and dexamethasone. The trypan blue exclusion assay was used to determine the percentage of viability, and the lactate dehydrogenase cytotoxicity test was used to assess the degree of treatment-induced cytotoxicity on each cell type. A second series of experiments was performed to study and determine the percentage of treatment-induced apoptotic cell death on each cell line with the DNA-Tunnel assay ApopTagR kit (TdT-enzyme and digoxigenin-dUTP reaction). Genistein induced a concentration-dependent dual effect of growth promotion (<10 >µg/mL) and inhibition (10 µg/mL) on testis cells, with a significant degree of sensitivity between the cells. Genistein, dexamethasone and the genistein-dexamethasone combination induced significant apoptosis on testicular cells; there was synergism in induction of apoptosis in testis cells with the genistein-dexamethasone combination treatment. The synergistic actions of genistein and dexamethasone in the induction of apoptosis may be of clinical and pathophysiologic research interest because of the chemopreventive-therapeutic potential of genistein and the clinical and pharmacologic application of dexamethasone.

Effects of Chronic vs. Acute Phytoestrogen Supplementation on Antioxidant Status, DNA Damage and Insulin-Like Growth Factor. Jayne V. Woodside, Erica E. Denholm, Maeli J. Campbell, Ian S. Young* and Anthony J. C. Leathem. Department of Surgery, University College London, London, UK and *Department of Clinical Biochemistry, The Queen’s University of Belfast, Institute of Clinical Science, Belfast, Northern Ireland.

Phytoestrogens, a group of plant compounds, may play a preventative role in endocrine-responsive cancers and coronary heart disease. However, the range of physiologic effects that these compounds may possess remains to be fully elucidated. Phytoestrogens are proposed to act as antioxidants and protect against DNA damage, but this work has been carried out only in vitro. In addition, like tamoxifen, phytoestrogens may be able to lower insulin-like growth factor (IGF) levels. We conducted a feeding study examining the effects of acute (a single 80-mg load) vs. chronic (80 mg/d for 7 d) administration of phytoestrogens in healthy female volunteers (n = 10) on various indices of antioxidant and lipid status, background DNA damage assessed by using the COMET assay, and IGF-1 and binding protein-1 levels. There were no significant changes in antioxidant status, as measured by FOX1 assay, and lag time to oxidation or total cholesterol, HDL cholesterol or triglyceride concentrations in either phase of the study. Levels of background DNA damage in buccal cells were also unchanged after supplementation. However, the mean DNA damage in lymphocytes was nonsignificantly reduced after 1 wk of supplementation (mean ± SD baseline and postweek, respectively: 4.07 ± 2.81 and 2.10 ± 0.86, P = 0.066). Levels of IGF-1 were unchanged after the single 80-mg load. However, after 1 wk of supplementation, IGF-1 concentrations were raised (mean ± SD baseline and postweek, respectively: 190.9 ± 85.0 and 280.1 ± 95.4 ng/mL, P <0.01). Levels of IGF binding protein-1 were not altered by phytoestrogens in either the acute phase or the chronic phase of the study. This study provides a rationale for carrying out further placebo-controlled studies looking at phytoestrogen consumption and levels of IGF and DNA damage.

Soy Protein Has an Anticarcinogenic Activity by Reducing GST-P⁺ Foci but Not by Apoptosis in Rat Hepatocarcinogenesis. Haymie Choi, Hyeon-duck Kim, Jo-hye Hwang and Hae-jeoung Lee. Seoul National University, Department of Food and Nutrition, Seoul, Korea.

This study was designed to examine the effect of dietary proteins in rat hepatocarcinogenesis. Six-week-old Sprague-Dawley rats were fed diets containing soy protein or casein at 10 and 20%. Two weeks after the diet feeding, hepatocarcinogenesis was initiated by diethylnitrosamine and promoted by 2-acetylaminofluorene and two-thirds partial hepatectomy. The rats were killed at 4 d or 8 wk after the diethylnitrosamine injection. The areas and numbers of glutathione S-transferase–placental positive (GST-P⁺) foci and apoptotic hepatocytes, the contents of thiobarbituric acid–reactive substances (TBARS) and total glutathione (GSH), and the activities of GST and glucose 6-phosphatase (G6Pase) were measured to investigate the anticarcinogenic effect of soy protein. The areas and numbers of GST-P⁺ foci were significantly lower in the soy protein groups than in the casein groups. The areas and numbers of apoptotic hepatocytes were significantly increased in the casein groups and decreased in the soy protein groups. From these results, apoptosis cannot be the reason for the decrease in GST-P⁺ foci in soy protein groups. TBARS and GSH contents were not significantly different between groups. G6Pase activity, a biomarker of membrane stability, was increased at 4 d and 8 wk in the soy protein groups. These increased G6Pase activities were inversely related to the decrease of GST-P⁺ foci. Soy proteins seem to be more anticarcinogenic than casein by decreasing the preneoplastic lesions, which may be explained in part by an increase in membrane stability rather than by apoptosis during the initiation and promotion periods.

Postprandial Changes in Male Sex Hormones after Meals Containing Soy (as Tofu), Lean Meat or Fatty Meat. R. Habito and M. J. Ball. School of Biological Sciences, Deakin University, Burwood, VIC, Australia.

Prostate disease is considered to be sex hormone dependent, and dietary and lifestyle factors could affect sex hormone concentrations in a manner that may alter the risk for prostate disease. Because individuals spend a considerable portion of their time in a postprandial state, we investigated the effect of different meals on sex hormone concentrations, including the effect of one meal with soy protein containing isoflavones. The effects of three isocaloric meals were evaluated sequentially in 15 healthy men at 7-d intervals. The meals were a lean meat meal, a tofu meal (both containing 28% energy from protein and 20% energy from fat) and a fatty meat meal (54% energy from fat). Blood samples were obtained at baseline (fasting, premeal) and at 2, 3 and 6 h after each meal for analysis of serum testosterone, sex hormone–binding globulin (SHBG) and free androgen index. There was a significant fall in testosterone within 2 h after both the tofu and lean meat meals. The changes in SHBG concentrations were significantly more positive after the tofu meal than the lean meat meal at 2 and 3 h (P = 0.04 and 0.01, respectively). The 2-h serum testosterone and the decremental area under the curve were significantly less negative after the fatty meat meal in which there was little change from baseline. Lower biologically active sex hormone concentrations after the lean meat and particularly after soy protein, as tofu, compared with fatty meats may provide long-term benefits in reducing the risk of a disease such as prostate cancer that appears to be sex hormone dependent.

Regulation of Male Sex Hormone Levels by Isoflavone Intake. Jong-Sang Kim, Hyun-Mi Son and Chong-suk Kwon.* School of Food Science, Inje University, Kimhae, Korea and *Department of Food Science and Nutrition, Andong National University, Andong, Korea.

Soy isoflavones were reported to inhibit 5--reductase, which is involved in the conversion of testosterone into dihydrotestosterone, an active form of male sex hormone. We hypothesized that feeding soy or isoflavones should cause a change in the plasma dihydrotestosterone level in male rats. Rats fed a diet containing either soy flour or semipurified isoflavones for 1 wk showed a significantly lower level of plasma dihydrotestosterone, but not testosterone, than did the rats fed the control diet. This hormonal change caused by soy flour or isoflavones may be partially responsible for the prevention of prostate cancer resulting from the consumption of soy.

Effects of Genistein and Radiation on HeLa Cells. Shung-jun Yang, Mingliang Jiang and Sameer Rafla. Department of Radiation Oncology, New York Methodist Hospital, Weill Medical College, Cornell University, Brooklyn, NY.

Much effort has been made toward enhancing cure rates of advanced tumors by combined use of radiotherapy with antineoplastic drugs, including natural compounds of plant origin. Genistein, one of the major flavonoids isolated from soybeans, has been linked to the low incidence of breast, prostate and colon cancers observed in some Asian countries. Moreover, increased public awareness of the importance of a healthful diet has led to the widespread use of soybean products and other natural nutritional supplements. The ability of genistein to inhibit the activities of protein-tyrosine kinase and DNA topoisomerase II, enzymes involved in DNA synthesis and cell proliferation, is well documented. In addition, topoisomerase II plays an important role in repair of DNA damage. Genistein is thus a promising candidate for clinical application as an anticancer agent. Thus, understanding possible interactions of genistein with conventional medical treatment such as radiation would be beneficial in treating cancer or other disorders such as arteriovenous malformation or arterial restenosis. In this work, we treated human HeLa cells, which are cervical cancer derived, with genistein and X-rays, individually or in combination. Modifications in cell cycle distribution, certain cell cycle checkpoint genes and single-cell clonogenic survival were studied. Genistein cytotoxicity was found to be dependent on drug concentration in the culture medium and the duration of cell contact. For a 24-h treatment, concentrations of genistein <5 >µmol/L did not consistently reduce cell survival, whereas those 100 µmol/L killed most cells. Cells preincubated with genistein at 20 µmol/L for the same period and then irradiated with graded doses of X-rays showed lower survival levels than did counterpart cells preincubated in normal medium containing no drug and similarly irradiated. For example, a 3- and 5-Gy dose of radiation reduced survival to 30 and 14%, respectively, in control cells and 18 and 7%, respectively, in cells preincubated with genistein. Genistein appeared to enhance radiation lethality, suggesting beneficial application of combined use to achieve increased cell kill. Studies on other cellular changes are ongoing.

A Double-Blind Randomized Study on the Effects of the Isoflavone P-07 on the Endometrium. Georgina E. Hale, Sanjay Agarwal, Claude Hughes, Stanley J. Robboy and Marcia Bievre. Cedars-Sinai Medical Center, Los Angeles, CA.

Endometrial cancer is the most common malignancy of the female genital tract and the fourth most common malignancy after breast, lung and colon cancer. It is well established that Asian women have a much lower incidence of endometrial cancer than do Caucasian women and that Asian women consume much higher amounts of pulses (tofu and other soy products) in their diet. Many of the known risk factors associated with endometrial cancer are related to prolonged exposure to estrogen such as estrogen hormone therapy, early age at menarche, late age at menopause and obesity. How dietary factors can influence this estrogen exposure has been of interest to many investigators, and in one case-control study in Hawaii, a high consumption of tofu and other soybean products was found to be associated with a decreased risk of endometrial cancer independently of other known risk factors (Goodman et al.1997 ). Pulses are rich in isoflavones, in particular, genistein. As early as 1966, it was shown that subcutaneously administered genistein partially reversed the uterotropic effect of estrogen in rats (Folson et al. 1989 ), and more recently, dietary soy protein isolate was shown to partially reverse the uterotropic effects of estrogen in monkeys (Foth and Cline 1998 ). This reversal was reflected in a decrease in the Ki-67 antigen (a nuclear antigen marker of proliferation) but not a decrease in endometrial thickness. The Ki-67 antigen is an excellent marker of endometrial proliferation. Investigators have shown a decrease in Ki-67 antigen staining with progesterone and increased staining in association with endometrial hyperplasia and histological atypia compared with normal endometrium. To investigate the direct effect of isoflavones on human endometrium, we designed a randomized placebo-controlled study in women of late reproductive age to look at the Ki-67 antigen staining of endometrial samples before and after isoflavone administration. We are including women between the ages of 45 and 50 y. This age group is associated with having decreasing levels of progesterone relative to estrogen, making them theoretically at increased risk of the proliferative influence of endogenous estrogen. Exclusion criteria include the use of the oral contraceptive pill or hormone replacement therapy within the past 12 mo, a current (and within the past 3 mo) intrauterine device, a regular isoflavone-rich diet over the past 6 wk, cigarette smoking during the past 12 mo and chronic disease. The isoflavone supplement we chose is a red clover extract (P-07; Novogen Limited); each tablet contains a predominance of the isoflavones biochanin A (4'-methoxy-5,7-dihydroxyisoflavone) and genistein (4',5,7-trihydroxyisoflavone) and smaller amounts of daidzein (4',7-dihydroxyisoflavone) and formononetin (4'-methoxy-7-hydroxyisoflavone), totaling 43 mg unglycosylated isoflavones. The following tests will be performed at baseline and 3 mo: serum estrogen, progesterone, hormone-binding globulin and inhibin B; fasting lipids, glucose and insulin; full blood count, urea and electrolytes; fasting urinary bone marker; and transvaginal ultrasound estimation of endometrial thickness, uterine artery doppler and endometrial biopsy. Participants will be required to take a tablet of placebo or the red clover extract daily and adhere to a specified low isoflavone diet. Isoflavone excretion will be measured in two 24-h urine collections at baseline and 3 mo.

Urinary Isoflavonoids Are Reliable Biomarkers for Soy Intake and Indicate Reduced Breast Cancer Risk. Adrian A. Franke, M.C. Yu,* W. Zheng, L. Le Marchand and L. J. Custer. Cancer Research Center of Hawaii, University of Hawaii, Honolulu, HI; *University of Southern California/Norris Comprehensive Cancer Center, Los Angeles, CA; and School of Public Health and Cancer Center, University of South Carolina, Columbia, SC.

Increasing evidence suggests that certain dietary phytoestrogens, particularly isoflavonoids and lignans, may protect against chronic disorders including heart disease, bone loss. and breast, prostate and colon cancer. Epidemiologic studies concerned with the assessment of the role of phytoestrogens in health and disease require fast, reliable and affordable techniques to measure these phytochemicals in humans favorably through noninvasive protocols. We developed fast and selective HPLC methods to analyze isoflavonoids, including their conjugates and metabolites, from various human fluids and from nonhuman primate tissues suitable for metabolic, clinical, epidemiologic and pharmacologic studies. Analytes were identified and quantitated by photodiode array, fluorometric and electrochemical detection and, most recently, by mass spectrometric analysis after negative electrospray ionization. Applications of this technique showed that isoflavone levels in human milk, plasma and urine increase linearly with increasing soy doses and disappear (depending on the dose) 1–2 d after soy exposure. Urine was preferred over other body fluids as a biological matrix of isoflavone analysis for epidemiologic studies because of the noninvasive nature of its collection, its low biologic hazard, good subject compliance, the fact that urinary isoflavones reflect a greater time period of soy exposure and its high concentration of isoflavones. Total urinary isoflavone excretion was found to correlate very well with intake of soy products as assessed by a 1-y food-frequency questionnaire in 50 multiethnic Hawaiians (daidzein: r = 0.47, genistein: r = 0.34, glycitein: r = 0.37, equol: r = 0.50). A very similar result was observed by a food-frequency questionnaire in populations known to consume larger amounts of soy products, including 60 Chinese women from Shanghai (total isoflavones: P trend = 0.018) and 147 Chinese from Singapore (total isoflavones: P trend = 0.04). In a breast cancer case control study with 120 Chinese women from Shanghai, we observed a protective effect of isoflavones and, to a lesser extent, of lignans. Women in the tertile of highest urinary isoflavone excretion experienced a 46, 30, 59 and 50% decreased risk to develop breast cancer vs. those in the tertile with lowest daidzein, genistein, glycitein and total isoflavone excretion, respectively. These findings support a potential breast cancer preventive effect achieved by soy consumption in populations that eat soy foods habitually.

	Heart disease and lipids

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Dietary Soy Isoflavones Reduce Plasma LDL Cholesterol and Atherosclerosis in a Human Apolipoprotein B Transgenic Mouse Model. F. M. Steinberg and M. M. Braun. Department of Nutrition, University of California, Davis, CA.

This study used a novel mouse model of atherosclerosis to test the hypothesis that a soy diet containing isoflavones would be cardioprotective, resulting in lower plasma LDL cholesterol (LDL-C) and smaller atherosclerotic lesion area compared with a soy diet lacking isoflavones. The +IF-Soy diet contained 0.93 mg total isoflavones/g diet, whereas the comparison diet (-IF-Soy) used ethanol-washed soy protein containing 0.04 mg total isoflavones/g diet. The human apolipoprotein B transgenic (h ApoB tg) mouse overexpresses the h ApoB gene, resulting in a very human-like lipoprotein profile with high levels of circulating LDL-C. Female weanling h ApoB tg mice (+IF-Soy; n = 6, -IF-Soy; n = 6) were housed individually and fed for 20 wk. Blood samples were taken at baseline and at 20 wk for lipid and lipoprotein separation by HPLC and colorimetric analysis of cholesterol. The aortic sinus region (150 mm) was evaluated for fatty streak lesions after serial sections were stained with oil red O and counterstained with Harris’ hematoxylin. Baseline values for total and LDL-C were not different between groups (102.1 vs. 83.6 mg/dL, and 56.2 vs. 47.2 mg/dL for -IF-Soy and +IF-Soy diets, respectively). Final values for total-C were not significantly different between groups (291.3 vs. 249.1 mg/dL for -IF-Soy and +IF-Soy diets, respectively ). However, final LDL-C values were significantly higher in the -IF-Soy group (204.4 vs. 135.2 mg/dL, P < 0.05). Atherosclerotic lesion analysis of the aortic sinus area showed greater mean lesion area per section in the -IF-Soy group than in the +IF-Soy (83,454 vs. 34,766 µm²). These results demonstrate that isolated soy protein that contains isoflavones is more beneficial than isolated soy protein without isoflavones (and other possible components) with regard to indices of cardiovascular disease in this mouse model of human atherosclerosis.

Lipoprotein Effects of Dietary Soy Protein Isolate before and after Alcohol Extraction. Hans Meinertz and Karin Nilausen.* Department of Medicine B, National University Hospital, Copenhagen, Denmark and *Department of Medical Anatomy, University of Copenhagen, Copenhagen, Denmark.

Animal studies show beneficial effects of alcohol-extractable components of soy protein isolate on both atherogenic and antiatherogenic plasma lipoproteins. We tested the hypothesis that those components produce beneficial lipoprotein effects similarly in humans, in which case soy protein might be prepared to contain optimal amounts of alcohol-extractable components for use in preventive diets. Twelve healthy, normolipidemic subjects, six premenopausal women and six men, consumed three liquid-formula diets of identical composition except that they contained either unextracted or ethanol-extracted soy protein or casein. After 28–32 d, the unextracted soy protein diet did not lower total, LDL and VLDL cholesterol or apo B nor did it increase HDL cholesterol significantly compared with the extracted soy protein diet, although it did raise plasma apo AI 10% (P < 0.05). Furthermore, in women, VLDL cholesterol, total triglycerides and apo AII increased 40, 36 and 6% (P < 0.05), respectively. In men, however, the unextracted soy protein diet decreased the ratio of LDL to HDL cholesterol by 22% (P < 0.05). Most remarkably, unextracted soy protein increased the atherogenic lipoprotein(a) ~335% (P < 0.001) in both women and men. We conclude that alcohol-extractable components of soy protein increase several lipid risk factors and that in young women, in particular, the total lipid risk factor level appears to be increased. These data suggest that extracted rather than unextracted soy protein should be used in antiatherogenic diets.

Effect of Genistein Injections on Borderline Hypertension in Conscious Rats. Nikolai P. Breitkopf, Kathleen M. Eyster, John L. Williams and Douglas S. Martin. South Dakota Soy Cardiovascular Research Consortium, Division of Physiology and Pharmacology, School of Medicine, University of South Dakota, Vermillion, SD.

Evidence indicates that soy isoflavones decrease plasma lipids, reduce atherosclerosis and alter coronary vascular reactivity. Although the effect of soy isoflavones on blood pressure has received comparatively little attention, direct injection of genistein attenuated hypertension development in the spontaneously hypertensive rat. The goal of this study was to determine whether genistein has an antihypertensive effect in the borderline hypertensive rat and to determine the optimal dose for this effect. Female borderline hypertensive rats were obtained at age 4 wk and treated with intraperitoneal injections of genistein in three doses (1, 10 and 25 mg/kg) or vehicle (dimethyl sulfoxide 10%). Injections were given daily for 4 mo. The rats then had arterial and venous catheters implanted chronically for the recording of mean arterial pressure (MAP) and intravenous drug administration, respectively. MAP reactivity was assessed by graded intravenous infusion of norepinephrine [0.05–1 µg/(kg · min)]. In the borderline hypertensive rats treated with vehicle, MAP averaged 140 ± 4 mm Hg. There was a slight decrease in MAP to 135 ± 9 mm Hg in the rats treated with injections of genistein at 1 mg/kg. MAP was reduced further in the group treated with 10 mg/kg (123 ± 4 mm Hg). No further decline in MAP was observed in the rats treated with 25 mg/kg (122 ± 3 mm Hg). Thus, there is an optimal dose of genistein beyond which no further blood pressure–lowering effect is obtained. In addition, we investigated the hypothesis that the blood pressure–lowering effect of genistein injections was due to a reduction in the vascular sensitivity to norepinephrine. The dose-response curves for norepinephrine were displaced only slightly by genistein treatment. The maximal effect was obtained with the genistein dose of 25 mg/kg, which increased the median effective dose for norepinephrine from 0.2 to 0.3 µg/(kg·min). Nevertheless, the genistein-induced shifts in the norepinephrine dose-response curves were not large enough or consistent enough to account for the blood pressure–lowering effect of genistein. In summary, direct injection of genistein attenuated hypertension development in the borderline hypertensive rat. The antihypertensive effect of genistein reached a maximum at a dose of 10 mg/(kg·d) and was not mediated via a decrease in the pressor reactivity to norepinephrine. [Supported by grants from the South Dakota Soybean Research and Promotion Council and the Faculty Development Funds of the University of South Dakota School of Medicine.]

Cardiovascular Benefits of Foods Containing Soy Isoflavones and -Linolenic Acid. Lee Astheimer, Leisa Ridges, Barbara Meyer, Rachel Sunderland, Katherine Moerman and Peter Howe. Smart Foods Centre (ARC Key Centre for Teaching and Research), University of Wollongong, NSW, Australia.

We examined the effects of 8 wk of consumption of a specific set of food products containing soy, linseed and canola on plasma lipids, fatty acid profiles of plasma and red blood cells, and platelet thromboxane production in 20 postmenopausal women (body mass index = 29.6 ± 4.3 kg/m²). The daily supplement provided a total of 32 mg daidzein, 13 mg genistein and an estimated 6 g -linolenic acid (LNA). Compliance was evaluated through unscheduled diet recalls, food acceptability questionnaires and urinary isoflavone levels at 3 and 8 wk. Combined urinary isoflavones (daidzein and genistein) increased dramatically from 0.23 ± 0.094 to 6.36 ± 1.15 mg/24 at 3 wk but fell to 2.58 ± 0.507 mg/24 h at 8 wk, coincident with reporting of reduced acceptability of some high daidzein foods. Significant decreases in total (9.94 ± 1.69%), LDL (12.54 ± 2.31%), and non-HDL (9.81 ± 2.24%) cholesterol were noted after 3 wk (paired t tests; P < 0.01), but these were no longer significant after 8 wk, perhaps as a result of reduced compliance. Slight but significant increases in plasma LNA as well as the longer-chain n-3 fatty acids, docosahexaenoic and eicosapentaenoic acids, were noted in plasma after 8 wk. However, a concomitant increase in the incorporation of these long-chain n-3 fatty acids into red blood cells was not evident. Similarly, platelet thromboxane production, known to decrease with long-chain n-3 supplementation, was unaffected. Because the predominant effect of n-3 supplementation on plasma lipids is reduction of triglycerides with a possible transient increase in LDL, the beneficial changes of plasma lipids observed in this study are more likely to be attributable to the increased consumption of phytoestrogens, that is, soy isoflavones and linseed lignans. Lignan levels and their potential effects on plasma lipids are now being assessed.

Antihypertensive Effect of Dietary Soy: Functional Role of Nitric Oxide. Doug Martin, Kathleen M. Eyster, John L. Williams and Nikolai P. Breitkopf. South Dakota Soy Cardiovascular Research Consortium, Division of Physiology and Pharmacology, School of Medicine, University of South Dakota, Vermillion SD.

Phytoestrogens such as genistein may have beneficial effects in coronary artery disease, osteoporosis and breast cancer. Studies in other laboratories showed that genistein reduced arterial responses to angiotensin II, acute injection of genistein altered coronary artery reactivity in monkeys and chronic intraperitoneal injection of genistein attenuated the development of hypertension in the spontaneously hypertensive rat (SHR). We tested the hypothesis that dietary intake of genistein through a soy-based diet would attenuate the development of hypertension in SHR via a nitric oxide mechanism. Female SHR underwent ovariectomy at age 4 wk. The rats were fed a custom diet containing whole soy or casein (control) from age 5 wk. Mean arterial blood pressure (MAP; mm Hg) and heart rate (HR; bpm) were recorded from conscious unrestrained rats after 8–12 wk of consuming the diets. These variables were measured before and after blockade of nitric oxide synthase with L-NAME (25 µmol/kg). ANOVA indicated a significant effect of diet on MAP. Post-hoc analysis revealed a significant reduction in MAP in the SHR fed soy. In the soy-fed rats, MAP averaged 150 ± 4 mm Hg (n = 10), whereas in the casein-fed rats, MAP was 164 ± 3 mm Hg (n = 12). Injection of L-NAME into the soy-fed rats caused MAP to increase by 34 ± 3 mm Hg (n = 8). In the casein-fed rats, the L-NAME–induced pressor response averaged 32 ± 3 mm Hg (n = 11). These values were not significantly different. Accordingly, these data suggest that nitric oxide is not the primary factor mediating the soy-induced decrease in blood pressure. [Supported by a grant from the South Dakota Soybean Research and Promotion Council.]

Role of Amino Acid Composition of Dietary Proteins in Regulation of Hypercholesterolemia. Elzbieta M. Kurowska,* Isabelle Giroux,David J. Freeman and Kenneth K. Carroll.* Departments of *Biochemistry and Pharmacology and Toxicology, University of Western Ontario, London, Ontario, Canada.

Substitution of dietary soy protein for animal proteins is known to be associated with a reduction of plasma total and LDL cholesterol, especially in individuals with hypercholesterolemia. This could be related to high content of arginine (Arg) and the low content of lysine (Lys) and methionine (Met) in soy protein compared with animal proteins. Our previous experiments showed that high dietary levels of Lys + Met in rabbits induce hypercholesterolemia, and this is counteracted in part by the addition of Arg. In this study, we postulated that these effects could be mediated by the amino acids themselves or by their metabolic products and that the regulation could occur directly in the liver. The results demonstrated that Met deamination products and a Met transamination intermediate, homocyst(e)ine, are unlikely to be important in the hypercholesterolemia induced by Lys + Met. However, an involvement of other Met transamination products was not excluded because feeding a diet rich in Lys + Met led to the accumulation of liver phosphatidylcholine, a compound that requires methyl groups from Met for its synthesis. Studies in the human liver cell line HepG2 showed increases in net production of apolipoprotein B (apo B), the protein component of LDL, after 24 h incubation of these cells with serum from rabbits fed the diet enriched with Lys + Met, whereas decreases were shown after a similar incubation with high levels of Arg. Medium apo B levels were not altered by exposure of cells to an excess of Lys + Met. The results suggested that an intact Arg and metabolites of Lys, Met or both (rather than intact Lys, Met or both) could be involved in regulation of LDL metabolism in the liver induced directly by dietary amino acids. [Supported by the Heart and Stroke Foundation of Ontario and by the Ontario Soybean Growers’ Marketing Board.]

A Prospective Study on Urinary Phytoestrogen Excretion and the Risk of Fatal Myocardial Infarction in Postmenopausal Women. Yvonne T. Van der Schouw, Jos H. H. Thijssen,* Herman Adlercreutz, and Diederick E. Grobbee. Julius Center for Patient Oriented Research, University Medical Center, Utrecht, The Netherlands; *Endocrinology Laboratory, University Medical Center, Utrecht, The Netherlands; and Meilahti Hospital, Helsinki, Finland.

The purpose of this study is to investigate prospectively whether urinary excretion of genistein and enterolactone, as a marker for phytoestrogen intake, is associated with fatal myocardial infarction risk. Estrogen replacement therapy has been associated with a reduced cardiovascular disease risk. Phytoestrogens are a family of plant compounds that have been shown to have both estrogenic and antiestrogenic properties. Accumulating evidence suggests that phytoestrogens may confer health benefits related to cardiovascular diseases. We studied the relation between phytoestrogen excretion and fatal myocardial infarction in a prospective cohort study of 12,239 women living in the city of Utrecht, who were initially aged between 52 and 67 y. The vital status of these women was followed between 1976 and 1995 (168,513 y). Concentrations of genistein and enterolactone in one overnight urine sample were determined in urine samples of 250 women who died of myocardial infarction and in a reference group of 250 women who did not. Concentrations of genistein and enterolactone were measured with time-resolved fluorescence immunoassays and adjusted for creatinine excretion. Data were analyzed by using a nested case-control analysis. With Poisson regression, we quantified the association between quartiles of genistein and enterolactone excretion and death caused by myocardial infarction. Higher urinary genistein excretion was not associated with reduced myocardial infarction mortality risk; the risk ratio for the highest compared with the lowest quartile was 0.29, 95% confidence interval 0.76–1.04. Results for enterolactone excretion were comparable. In conclusion, we were not able to detect a protective effect of phytoestrogen excretion on myocardial infarction mortality risk.

Oral Genistein Does Not Retard the Development of Hypertension in Growing Spontaneously Hypertensive Rats. Roger A. King and Richard J. Head. CSIRO Health Sciences and Nutrition, Adelaide, SA, Australia.

Genistein has been shown to impair the constrictor response of isolated blood vessels to the neurotransmitter norepinephrine in vitro (Toma et al. 1995 ). Oral treatment with isoflavones from soy (Nestel et al. 1997 ) and clover (Nestel et al. 1999 ) for 5–10 wk has also been shown to improve vascular compliance in women, but blood pressure (BP) was not reduced in normotensive menopausal women (Nestel et al. 1999 ) or in men with high normal BP (Hodgson et al. 1999 ). The spontaneously hypertensive rat (SHR) is a common model of human essential hypertension in which BP rises sharply after ~4–5 wk of age accompanied by hypernordrenergic innervation of blood vessels and vascular hypertrophy, with BP ~50–70 mm Hg above their normotensive Wistar Kyoto (WKY) controls at 3 mo of age (King et al. 1992 ). The SHR model allows the study of the entire period of hypertension development, which is not possible in human studies. In this study, we sought to confirm that genistein impairs the constrictor response of isolated rat aortic rings to norepinephrine and to determine whether genistein, administered orally during the period of development of hypertension, reduced BP or aortic hypertrophy in SHR. In Experiment 1, the effect of genistein at 50 µmol/L on the constrictor response of aortic rings to graded doses of norepinephrine from 0.12 nmol/L to 0.8 µmol/L was studied in WKY rats. Genistein decreased the tension developed by ~50% at all but the lowest concentration of norepinephrine (P < 0.05). In Experiment 2, SHR were fed an isoflavone-free purified diet at age 5 wk, when BP is similar to that of WKY controls. Genistein treatment was commenced immediately at a rate of 3 or 6 mg/kg body weight as a once-daily oral gavage in 25 mmol Na₂CO₃/L. Control rats received vehicle. Blood pressures were measured by a tail-cuff method before the commencement of treatment and after 4 and 8 wk of treatment, at which time the rats were killed. Heart weight and abdominal aorta weight-length ratio were measured as indices of hypertrophy. There was no significant difference in any of the measured indices between the two treatment doses; thus the results were combined. There was no effect of genistein on BP, body weight, heart weight or aortic weight-length ratio (Table 1).We confirmed that genistein impaired the constrictor response of isolated rat aortae to norepinephrine; however, oral genistein did not influence the development of hypertension or aortic hypertrophy in SHR.

The Effect of Isolated Soy Protein on Total Homocysteine Levels in Adult Peritoneal Dialysis Patients. Carol J. Boushey, Sara Long Anderson* and M. René Malinow.Department of Food and Nutrition, Purdue University, West Lafayette, IN; *Food and Nutrition Program, Southern Illinois University, Carbondale, IL; and Oregon Regional Primate Research Center, Oregon Health Sciences University, Beaverton, OR.

The purpose of this investigation was to examine the effects of isolated soy protein on total homocysteine (tHcy) concentrations in individuals treated with peritoneal dialysis. Twenty-six patients (58% males) were originally recruited. Ages ranged from 36 to 73 y with an average of 55 y. Subjects were provided 35 g of isolated soy protein in the form of a beverage powder and baked products to consume every day in addition to their usual diets. Nonfasting serum samples were collected at baseline, 6 and 12 mo. Analyses of serum samples included tHcy, folate, pyridoxal 5' phosphate (pyridoxine) and cobalamin (vitamin B-12). Similar to results from other studies of dialysis-dependent, end-stage renal disease (ESRD) patients, the average tHcy level in this group at baseline, 28.7 µmol/L, was greater than expected in the general population. Values for individuals with a tHcy level in the upper tertile at baseline (i.e., 31.6 µmol/L) were compared with those for individuals below the upper tertile (i.e., 31.6 µmol/L). Individuals in the upper tertile for tHcy had lower levels of folate, vitamin B-12 and pyridoxine; however, none reached statistical significance. Only 14 individuals completed the protocol up to mo 6 of measurement and only eight individuals completed the protocol up to the last month (mo 12). Individuals with a tHcy level in the upper tertile at baseline were compared with individuals below the upper tertile to examine reduction in tHcy. The average reductions in tHcy in the highest tertile group (16.6 µmol/L at 6 mo and 33.5 µmol/L at 12 mo) were significantly greater than the other tertiles: 3.7 µmol/L at 6 mo (P = 0.085) and 2.1 µmol/L at 12 mo (P = 0.025 ). This may indicate that individuals with ESRD may not be able to lower their homocysteine levels to those of the general population by using the current therapy. For these individuals, that intervention may be feasible only for those exceeding tHcy levels of ~30 µmol/L. When nutrient metabolites were examined by tertiles of homocysteine, the nutrients that increased in concentration were folate and vitamin B-12. This suggests that these two nutrients may have contributed to the reductions in tHcy. This is of great interest because the isolated soy product fed to these subjects is considered to be only a fair source of folate. It is possible that other substances present in the soy product may protect individuals from the loss of folate and vitamin B-12 through the dialysate. In addition, exploration of additional therapies (e.g., very high supplementation of folic acid) may be appropriate.

Soy Intake Pattern and Relation with Serum Cholesterol in the Hong Kong Chinese Population. Suzanne C. Ho, Jean Woo, Sophie Leung and Aprille Sham. Department of Community and Family Medicine, Chinese University of Hong Kong, Shatin, Hong Kong.

The study aims to describe the pattern of soy intake in the Hong Kong Chinese population and the influence of socioeconomic status on soy intake. It is hypothesized that higher soy intake is associated with lower levels of cholesterol and LDL cholesterol. The importance of the hypothesis is that although geographical variation of some common diseases such as coronary heart diseases have been attributed to the differences in soy intake, few data on the soy consumption pattern and its association with the risk factors for heart disease in the Asian setting have been reported. The study will contribute to the understanding of the soy-disease relationship. A dietary survey was conducted as part of a territory-wide population-based cardiovascular risk factor study. A total of 500 men and 510 women completed the dietary intake study. The dietary assessment was based on a semiquantitative food-frequency questionnaire comprising 253 food items; 10 soy foods commonly consumed in the local population were among the items included. Of the study population, 88% had consumed some form of soy in the past week; 68% of the soy-derived isoflavones were obtained from tofu, 8% were from fried tofu and 8% were from soymilk (vitasoy). The mean amount of isoflavone intake per week in the whole population was 135.2 mg (± 155.7); and 154.8 mg (± 157.3) in those who had consumed any of the soy items. Men had a higher intake of isoflavones than women, but no difference was observed after adjustment was made for the total energy intake (17.7 mg/MJ for men and 16.9 mg/MJ for women). Older subjects and subjects with higher levels of education tended to have higher soy consumption. We observed a significant negative linear relation between soy intake and LDL cholesterol in men. The overall intake of soy foods was modest only in the Hong Kong Chinese population. There was apparent beneficial effect of soy intake in the improvement of lipid profile in men.

Soy Protein Reduces Arterial LDL Concentration and LDL Delivery in Both Diabetic and Nondiabetic Cynomolgus Monkeys. Janice D. Wagner, Li Zhang, Kathryn A. Greaves and Dawn C. Schwenke. Winston-Salem, NC.

We previously showed that soy protein consumption improves lipoprotein concentrations and reduces progression of atherosclerosis in cynomolgus monkeys. The mechanism for these beneficial effects is unclear. The purpose of this study was to determine potential mechanisms for the atheroprotective effects and to determine whether these effects extend to diabetic monkeys. As such, we designed an experiment with a 2 x 2 factorial design in which adult male monkeys (n = 23) were fed an atherogenic diet; the protein source was soy isolate (n = 11) or casein lactalbumin (n = 12), and monkeys were either control (n = 12) or streptozotocin-induced diabetic (n = 11). Diabetics had significantly increased fasting glucose and glycated hemoglobin levels; this relationship was not affected by protein consumption. Soy consumption significantly reduced total cholesterol and increased HDL cholesterol concentrations in both control and diabetic monkeys. To determine plasma and arterial LDL metabolism, radiolabeled LDL was injected before necropsy. ¹²⁵I-LDL was injected 48 h and ¹³¹I-tyramine cellobiose-LDL was injected 1 h before necropsy to determine arterial LDL concentration and permeability, respectively. An increase in whole-body plasma LDL fractional catabolic rate was found with soy consumption. Soy reduced arterial LDL concentration across all arterial sites and LDL permeability in the carotid bifurcation and internal carotid arteries. Arterial delivery of LDL was calculated as LDL permeability x plasma LDL concentration and was also reduced across all arterial sites. There was no additional effect of diabetes. In conclusion, these beneficial effects of soy on both plasma and arterial LDL metabolism would be expected to reduce atherosclerosis progression and were found in both control and diabetic monkeys.

A Soy Diet Improves Vascular Reactivity in Spontaneously Hypertensive Rats. Mark G. Lounsbery, Kathleen M. Eyster, Douglas S. Martin and John L. Williams. South Dakota Soy Cardiovascular Research Consortium, University of South Dakota School of Medicine, Vermillion, SD.

Recent studies demonstrated that a diet that contains soy foods can lower blood cholesterol and may improve vascular reactivity during severe atherosclerosis. The goal of these studies was to determine whether a soy-based diet improves vascular reactivity in spontaneously hypertensive rats (SHR). Four-week-old male, ovariectomized (ovx) female, or sham female SHR or normotensive ( Wistar Kyoto, WKY) rats were fed a diet containing either soymeal or casein for 10 wk. Posterior cerebral arteries (~250 µm) were removed from anesthetized rats and placed in an arteriograph. The system was perfused with physiologic saline; pressure, flow, gases, pH and temperature were controlled. Cumulative doses (10^-10–10^-5 mol/L) of 2-methylthio-ATP (2MSATP), an agent that causes release of nitric oxide from the endothelium, were administered luminally for 10 min at each dose. Arterial diameter was measured with a microscope and video system. In general, 2MSATP caused vasodilation at higher doses, but the magnitude of response varied considerably among groups. Dilator responses were similar in male WKY rats consuming casein vs. soy diets (14.4 ± 3.3% vs. 10.9 ± 1.8%, respectively; means ± SEM). In contrast, in male SHR, vasodilation was much greater (P < 0.05, repeated measures ANOVA) in soy-fed rats (17.8 ± 4.0%) than in casein-fed rats (9.5 ±1. 9%). Results were qualitatively similar in ovx rats, but the magnitude of response to 2MSATP was much smaller. Although increases in diameter were not significantly different between dietary groups of ovx WKY rats, vasodilation was greater in soy-fed ovx SHR rats than in casein-fed ovx SHR rats (7.0 ± 1.3% and 3.4 ± 1.1%, respectively; P < 0.05). In both sham WKY and SHR rats, 2MSATP caused a small dose-response effect that was similar in the soy and casein groups. Our findings demonstrate that a diet containing soy foods can improve or restore vasodilator function in hypertensive rats. Furthermore, our findings demonstrated sex differences in dilator responsiveness to 2MSATP and in the ability of soy foods to improve vascular reactivity. [Supported by a grant from the South Dakota Soybean Research and Promotion Council.]

Soy Isoflavone, Daidzein, Attenuates Postovariectomy Blood Pressure Elevation by Accelerating Nitric Oxide Production. Y. Yamori,* T. Teramoto,* T. Noguchi,* Y. Sekine,* Y. Nara** and K. Ikeda.* *Department of Environmental Preservation and Development, Graduate School of Human and Environmental Studies, University of Kyoto, Kyoto, Japan; Fujicco Co., Ltd., Kobe, Japan; and **Graduate School of Integrated Science, University of East Asia, Shimonoseki, Japan.

Recent studies suggested that estrogen-induced cardiovascular protection might be mediated by an increased synthesis of vascular nitric oxide (NO). In this study, we investigated the effects of isoflavones with weak estrogenic activity on blood pressure and vascular NO actions in ovariectomized stroke-prone spontaneously hypertensive rats. Female stroke-prone spontaneously hypertensive rats (n = 32; 15 wk old) were divided into four groups. Groups 1 and 2 were fed the standard diet and groups 3 and 4 were fed daidzein (D)- and genistein (G)-mixed standard diets, respectively (0.16 mmol/100 g standard diet). At age 19 wk, group 1 (Sham) was sham-operated and groups 2 (ovx), 3 (ovx + D) and 4 (ovx + G) were ovariectomized. All rats were killed at age 22 wk and their thoracic aortae were checked for vascular NO release indirectly by precontracting aortic rings with prostaglandin F₂ and by observing the effect on tone changes induced by N-monomethyl-L-arginine, an L-arginine analog that competes for NO synthase and inhibits NO formation. Systolic blood pressure and urinary NO metabolites were measured before and after surgery. Before surgery, systolic blood pressure and urinary NO metabolites did not differ among all groups. After surgery, however, systolic blood pressure was significantly lower (P < 0.01) and urinary NO metabolites were significantly higher (P < 0.05) in ovx + D than in ovx. Basal release of NO in ovx + D was significantly higher than that in ovx (P < 0.005). These results indicate that daidzein attenuates the development of hypertension by accelerating vascular NO production. This new result supports the several epidemiologic and animal study findings suggestive of a possible cardiovascular protective effect of soy intakes.

Effect of Active Peptides from Soy Proteins on Cholesterol Homeostasis in Cell Cultures. M. R. Lovati, C. Manzoni, E. Gianazza, E. M. Kurowska* and C. R. Sirtori. Institute of Pharmacological Sciences, Faculty of Pharmacy, Milano, Italy and *Centre for Human Nutrition, Department of Biochemistry, University of Western Ontario, London, ON, Canada.

An activation of LDL receptors was recently described in a human hepatoma cell line (Hep G2) exposed to purified (+’) subunits from 7S soy globulin; ß chains were ineffective (Lovati et al. 1998 ). By using a mutant soy cultivar devoid of the ’ subunits (Keburi), further experiments in HepG2 cells demonstrated a marked LDL-receptor activation by the ’ vs. subunit. A commercial isoflavone-free, heat-hydrolyzed soy concentrate (Croksoy), found effective in type II hypercholesterolemic patients, showed an LDL-receptor activation similar to that of the 7S globulin when incubated under the same conditions (Manzoni et al. 1998 ). To assess the final identity of the putative peptide (or peptides) responsible for the biochemical effect, experiments were performed in Hep G2 cells exposed to either synthetic peptides corresponding to specific sequences differing among the +’ and ß subunits or peptides coming from the in vitro digestion (trypsin + pepsin) of Croksoy. Moreover, the ability of the whole 7S globulin, its +’ and ß subunits, and whole Croksoy to interfere in the apoB secretion in the medium as well as in sterol biosynthesis was evaluated in the same model. Hep G2 cells were preincubated for 24 h at 37°C in minimal essential medium with 5% lipoprotein-deficient serum in the presence or absence of synthetic and in vitro–produced peptides at different concentrations (10^-4 or 10^-6 mol/L for synthetic peptides, 0.5 mg/mL for digested Croksoy). After addition of ¹²⁵I-LDL, cells were incubated at 37°C for a further 4 h to determine the uptake and degradation of LDL. Increased ¹²⁵I-LDL uptake (+41%) and degradation (+10%) vs. controls were shown after Hep G2 incubation with a synthetic peptide (10^-4 mol/L, MW 2271) corresponding to the 127–150 positions of the consensus sequence present in the ’ and not in the ß subunit of the 7S globulin. Cells exposed to Croksoy enzyme digestion products showed a marked up-regulation of LDL receptors vs. that in controls as well as vs. HepG2 cells incubated with undigested Croksoy (¹²⁵I-LDL uptake: +58% and +17%, respectively; ¹²⁵I-LDL degradation: +113% and +26%, respectively). Among soy-derived products, only the 7S soy globulin showed an inhibitory activity on apolipoprotein B secretion (-80% secretion) and ¹⁴C-acetate incorporation (-70, -60 and -48% into free and esterified cholesterol and triacylglycerol, respectively) when tested in Hep G2 cells at a concentration of 1.0 mg/mL. These findings support the hypothesis (Sirtori et al. 1998 ) that one or more peptides escaping detection in vivo may reach the liver after intestinal digestion, thus eliciting a cholesterol-lowering effect. Moreover, these data indicate that the protein moiety, devoid of isoflavone components, is responsible for the biochemical effect.

Effect of Soy Protein on LDL Cholesterol in Hypercholesterolemic Costa Rican Children. A. G. Arauz, G. Padilla, M. Roselló, S. Guzmán, S. Rodriguez, L. Cunningham and J. M. Ordovás. Costa Rican Institute for Research and Education on Nutrition and Health (Inciensa), Tres Rios, Costa Rica.

A diet low in cholesterol, total fat and saturated fat is the first step in the treatment of hypercholesterolemia. However, many children cannot achieve plasma cholesterol goals by using this dietary approach. It has been long recognized that there is a hypocholesterolemic effect of soy protein when it is partially substituted for animal protein. The objective of our study was to evaluate the effect of soy protein on LDL cholesterol reduction when soy products are used as part of a regular diet at home. The product used were Archer Daniels Midland isolated protein and textured protein (soy meat and soymilk). We carried out a clinical random study including 26 hypercholesterolemic children; 13 were in the control group and 13 were in the intervention group (mean age 10.3 ± 2.04 y, range 7–15 y). During the 8-wk experimental period, both groups consumed a Step 1 diet (<30% total fat, <10% saturated fat, <250 mg cholesterol). The intervention group also consumed the soy products as part of their regular meals. Food intake and adherence to the protocol were assessed during weekly home visits. Daily soybean protein intake in the intervention group was 20.88 ± 5.59 g/d, range 14.24–33 g/d. Preliminary data from 18 children (9 in each group) showed that the mean LDL-cholesterol reduction was significantly (P < 0.005) higher in the intervention group (15.5%) than in the control group (5.5%). There were no significant differences in total cholesterol reduction. The addition of soy protein to a Step 1 diet in hypercholesterolemic children was associated with a specific reduction in LDL cholesterol. We therefore propose that soy protein may offer a nonpharmacologic alternative to reducing elevated LDL cholesterol in children. [Supported by Inciensa-FODESAF, Costa Rica; American Soybean Association and United Soybean Board.]

Effects of an Isoflavone Supplement on Risk Factors for Breast Cancer and Coronary Heart Disease in Premenopausal Women. S. J. Smith, G.S.-M. Chan, S. Samman and P. M. Lyons-Wall. Human Nutrition Unit, Department of Biochemistry, University of Sydney, Sydney, NSW, Australia.

Traditional soy-based Asian diets are associated with a lower prevalences of breast cancer and coronary heart disease. One dietary component that could contribute to this effect is the isoflavones, found in especially high concentrations in soy. Isoflavones are diphenolic compounds with a structure similar to that of estradiol that may act as both estrogens and antiestrogens, depending on the target tissue and hormonal status of the individual. The aim of this study was to determine the effects of an isoflavone supplement on plasma concentrations of estradiol and androgen precursors, lipids and lipoproteins. We hypothesized that the isoflavones would produce favorable hormonal and lipid profiles. Participants were healthy women (n = 14; mean age 27.5 ± 8.2 y; body mass index: 21.3 ± 0.5, measured as kg/m²) with regular menstrual cycles, stable body weight, low soy intake and not taking an oral contraceptive pill. The women consumed 86 mg/d of isoflavones in tablet form for two menstrual cycles followed by placebo for an equivalent period, or vice versa. Blood samples were collected during the second cycle of each treatment period for measurement of plasma steroid hormones (estradiol, progesterone, testosterone, androstenedione, dehydroepiandrosterone sulfate and sex hormone–binding globulin). Plasma lipids and lipoproteins were also measured. The isoflavone supplement had no significant effects on concentrations of steroid hormones, sex hormone–binding globulin, total cholesterol or triacylglycerol, although there was a unexpected trend toward a 15% increase in mean estradiol concentration (P = 0.06). Consistent with the rise in estradiol, there was a trend toward an increase in HDL₃ cholesterol; in the group of women who received isoflavones followed by placebo, HDL₃ cholesterol continued to rise further during the placebo period and at the end of the study, it was 11% higher than at baseline (P = 0.024). The results do not support the hypothesis that isoflavones can reduce risk of breast cancer by lowering concentrations of estradiol or precursor androgens per se. However the potential beneficial effect of isoflavones on HDL metabolism in relation to cardiovascular risk and, in particular, the interaction with steroid hormones during the menstrual cycle require further investigation.

High Dietary Cholesterol Diets May Attenuate Hypocholesterolemic Effects of Soy Consumption. Kathryn A. Greaves, Martha D. Wilson, Lawrence L. Rudel and Janice D. Wagner. Wake Forest University School of Medicine, Department of Pathology, Section on Comparative Medicine, Winston-Salem, NC.

We previously reported that one mechanism for the hypocholesterolemic effects of soy may be a reduction in cholesterol absorption. In this abstract, we report that reductions in cholesterol absorption may be attenuated by high amounts of cholesterol in the diet. Results from two studies examining 20 wk of soy protein consumption in ovariectomized cynomolgus monkeys are reported. All monkeys were fed either casein-lactalbumin (CAS) or isolated soy protein (SOY) as a protein source. In the first study, monkeys were fed diets containing 0.28 mg cholesterol/kcal. In the second study, monkeys were fed diets containing 0.36 mg cholesterol/kcal. Cholesterol absorption was determined by using the fecal isotope ratio method. Data are presented as means ± SEM. A significant reduction in cholesterol absorption was found in monkeys consuming isolated soy protein with 0.28 mg cholesterol/kcal compared with monkeys consuming casein-lactalbumin protein (SOY 66 ± 3% vs. CAS 74 ± 2%). However, there was no significant difference between the SOY and CAS groups consuming 0.36 mg cholesterol/kcal (SOY 63 ± 2% vs. CAS 66 ± 3%). Additionally, non-HDL cholesterol concentrations were significantly lowered in the SOY group compared with CAS group consuming 0.28 mg cholesterol/kcal (SOY 118 ± 14% vs. CAS 326 ± 37%). This was not the case in the monkeys consuming 0.36 mg cholesterol/kcal (SOY 324 ± 51% vs. CAS 403 ± 51%). Although many studies in humans suggest that soy protein is more beneficial in hypercholesterolemic individuals, most of these studies have included a low cholesterol diet in their design. The results from this study suggest that a threshold of dietary cholesterol consumption may exist above which soy protein may not be effective in lowering plasma cholesterol.

Promoting Effect of Fermented Soy Flour on the Antioxidant Activity of Coenzyme Q-10. Joe Vinson and Jimmy Li.* Department of Chemistry, University of Scranton, Scranton, PA and *Department of Food Science, Rutgers University, Piscataway, NJ.

Soy food and fermented soy food have demonstrated many health benefits, including hypocholesterolemic effects. Coenzyme Q-10 (CoQ-10), the important antioxidant for energy production in human tissues, was also found to be beneficial to human health. Oral CoQ-10 supplementation was reported to decrease the oxidation of LDL. In our study, pure CoQ-10 was combined into a fermentation medium at a concentration of 7.5% (dry basis) in soy flour. Then the soy fermentation product and pure CoQ-10 were tested for their antioxidant activity against the oxidation of human lipoprotein, mainly LDL; the 50% inhibition concentration (IC₅₀) was determined to evaluate the antioxidant activity against lipoprotein oxidation. At the same CoQ-10 level, the fermentation product was found to be 9.6 times more active than pure CoQ-10. Accelerated storage stability in both solid and liquid form was also tested for those samples. CoQ-10 in the pure chemical form decomposed much more quickly than it did in the soy fermentation product.

Effect of Soy Protein Foods on LDL Oxidation and Ex Vivo Sex Hormone Receptor Activity—a Controlled Crossover Trial. David J. A. Jenkins,* Cyril W. C. Kendall,** Marcella Garsetti,*** Rachel S. Rosenberg-Zand, Chung-Ja Jackson,Sanjiv Agarwal, A. Venket Rao,Eleftherios P. Diamandis, Tina Parker,* Dorothea Faulkner, Vladimir Vuksan* and Edward Vidgen.* *Clinical Nutrition and Risk Factor Modification Center, St. Michael’s Hospital, Toronto, ON, Canada; Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; ** Department of Food Science Technology and Microbiology, Division of Human Nutrition, University of Milan, Milan, Italy; Laboratory Services Division, University of Guelph, Guelph, ON, Canada; and Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, ON, Canada.

Plant-derived estrogen analogs (phytoestrogens) may confer significant health advantages, including cholesterol lowering, antioxidant activity and possibly reduced cancer risk. However, concern has also been raised that phytoestrogens may be endocrine disrupters and major health hazards. We therefore assessed the effects of soy foods as rich sources of isoflavonoid phytoestrogens on LDL oxidation and sex hormone receptor activity. Hyperlipidemic subjects (n = 31) consumed two low fat metabolic diets for 1 mo in a randomized crossover study. Major differences between test and control diets were the increases in soy protein foods (33 g/d of soy protein) providing 86 mg/(2000 kcal · d) of isoflavones and the doubling of soluble fiber intake. Fasting blood samples were obtained at the start, wk 2 and wk 4; 24-h urine samples were collected at the end of each phase. Soy foods increased urinary isoflavone excretion in subjects consuming the test diet compared with the control diet (3.8 ± 0.7 vs. 0.0 ± 0.0 mg/d; P < 0.001). The test diet lowered oxidized LDL concentrations measured both as conjugated dienes in the LDL fraction (56 ± 3 vs. 63 ± 3 mmol/L, P < 0.001) and as the ratio of conjugated dienes to LDL cholesterol (15.0 ± 1.0 vs. 15.7 ± 0.9, P = 0.032), even in subjects already taking vitamin E (400–800 mg/d). No significant difference was detected in ex vivo sex hormone activity between urine samples from the test and control periods. In conclusion, consumption of high isoflavone foods was associated with reduced levels of circulating oxidized LDL, even in subjects taking vitamin E, with no evidence of increased urinary estrogenic activity. Soy consumption may reduce cardiovascular risk without increasing risk for hormone-dependent cancers.

Combined Effect of Vegetable (Soy) Protein and Soluble Fiber Added to a Standard Cholesterol-Lowering Diet. David J. A. Jenkins,* Cyril W. C. Kendall,** A. Venket Rao, Sanjiv Agarwal,Peter J. H. Jones,** Mahmoud Raeini** Jon A. Story, Edward Vidgen,* Larry C Griffin,and Stephen C. Cunnane. *Clinical Nutrition and Risk Factor Modification Center, St. Michael’s Hospital, Toronto, ON, Canada; Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; **School of Dietetics and Human Nutrition, McGill University, Ste Anne de Bellevue, QC, Canada; Department of Food and Nutrition, Purdue University, West Lafayette, IN; Loblaw Brands Ltd., Toronto, ON, Canada.

Our goal was to determine whether a combination of two plant components, vegetable protein and soluble fiber, further reduced serum lipids when incorporated into the currently advocated low saturated fat diet. Hyperlipidemic men and women (n = 31) consumed two low fat (7% of total energy from saturated fat), low cholesterol <80 mg cholesterol/d) metabolic diets for 1 mo in a randomized crossover study. Major differences between test and control diets were the increased amount of vegetable protein (93 vs. 23% of total protein), of which 23 g/d was soy, and the doubling of soluble fiber. Fasting blood samples were obtained at the start and end of each phase. Fecal collections were obtained on the last 3 d of each phase. Compared with the low fat control diet, the test diet lowered total cholesterol (6.2 ± 1.2%, P < 0.001), LDL cholesterol (6.7 ± 1.7%, P < 0.001), apolipoprotein B (8.2 ± 1.2%, P < 0.001), and the ratios of LDL to HDL cholesterol (6.3 ± 2.0%, P = 0.004) and apolipoprotein B to apolipoprotein A-I (5.4 ± 1.5%, P = 0.001). A combination of vegetable protein and soluble fiber significantly improved the lipid-lowering effect of a diet low in saturated fat. The results support expanding current dietary advice to include increased vegetable protein and soluble fiber intakes so that the gap in effectiveness between a good diet and drug therapy is reduced.

Effect on Serum Lipids and Oxidized LDL of Supplementing Self-Selected Low Fat Diets with Soluble Fiber, Soy, and Vegetable-Protein Foods. Cyril W. C. Kendall,* ** David J. A. Jenkins* ** Edward Vidgen,* ** Marcella Garsetti,* ** Larry C. Griffin, Sanjiv Agarwal,** A. Venket Rao,** Stephen C. Cunnane,** Philip W. Connelly,^# Lawrence A. Leiter,** Vladimir Vuksan* ** and Robert Josse. *Clinical Nutrition and Risk Factor Modification Center and Department of Medicine, Division of Endocrinology and Metabolism, St. Michael’s Hospital, Toronto, ON, Canada; **Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, ON, Canada; Department of Food Science Technology and Microbiology, Division of Human Nutrition, University of Milan, Milan, Italy; Loblaw Brands Limited, Toronto, ON, Canada; and Departments of Biochemistry and ^#Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada.

Increased intake of soluble fiber and soy protein may improve the blood lipid profile and reduce oxidized LDL. The goal of this study was to assess any benefits of providing soy protein and soluble fiber foods to hyperlipidemic subjects already consuming low fat, low cholesterol therapeutic diets. Twenty hyperlipidemic men and postmenopausal women completed 8 wk of tests and control dietary treatments in a randomized crossover design as part of an unlimited National Cholesterol Education Program Step-2 therapeutic diet (7% saturated fat and 200 mg/d of cholesterol). During the test phase, foods high in soy, other vegetable proteins and soluble fiber were provided. During the control phase, low fat dairy and low soluble-fiber foods were provided. Fasting blood lipids and apolipoproteins were measured at 4 and 8 wk of each phase. Subjects receiving the test diet consumed 12 ± 2 g/d of soy protein from the foods they chose. Direct comparison of test and control treatments indicated a rise in HDL cholesterol concentration after consuming the test diet (6.4 ± 2.4%, P = 0.013) and a significantly reduced ratio of total to HDL cholesterol (-5.9 ± 2.3%, P = 0.020). The proportion of conjugated dienes in the LDL cholesterol fraction was significantly reduced (8.5 ± 3.3%, P = 0.020) as a marker of oxidized LDL. A combination of acceptable amounts of soy, vegetable protein and soluble fiber foods as part of a conventional low fat, low cholesterol therapeutic diet is effective in further reducing serum lipid risk factors for cardiovascular disease.

Effects of Yeast Fermentation on the Antioxidant Activities of Soy Flour. Joe Vinson, Pratima Bose and Jimmy Li.* Department of Chemistry, University of Scranton, Scranton, PA and *Department of Food Science, Rutgers University, Piscataway, NJ.

Soy food and fermented soy food have shown many health benefits, including hypocholesterolemic effects. Much research is being done on the bioactivities in soy food and fermented soy food. In our study, organic and nonorganic soy flours and their yeast fermentation products (Soynatto, Bio-Foods, West Patterson, NJ) were tested for their total phenol levels and their antioxidant activities against the oxidation of human lipoprotein (80% in LDL). The 50% inhibition concentration (IC₅₀) was determined according to the inhibition of LDL oxidation vs. soy concentration curves. The fermentation products were found to be 5.5–10.0 times more active than nonfermented soy flours, and this depended on the difference between soy flours. On the other hand, the fermentation did not change the total phenol content significantly.

	Other topics

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Daidzein Exerts a Marked Effect on Body Composition in Male Hamsters. Kathy B. Hanson, Tong T. Song,* Charles T. Peterson, Sun-ok Lee, Suzanne Hendrich, Patricia A. Murphy and D. Lee Alekel. Center for Designing Foods to Improve Nutrition; Department of Food Science and Human Nutrition; Iowa State University, Ames, IA; *Protein Technologies International, St. Louis, MO; and Department of Statistics, Iowa State University, Ames, IA.

The purpose of this study was to determine whether dietary isoflavones have an estrogen-like effect on bone, body composition and food efficiency (weight gain/food intake) in growing male and female hamsters. We hypothesized that dietary isoflavones (either a mixture of genistein, daidzein, and glycitein or daidzein alone) would increase food efficiency and exert a positive effect on bone mineral density and body composition by decreasing the proportion of weight gain as fat, regardless of sex. Golden Syrian hamsters (n = 80; aged 6–8 wk) were randomly assigned to one of the following four diet regimens (10 males and 10 females per treatment): casein without added isoflavones (CAS-), casein plus daizein (CAS + DAID), soy protein isolate containing isoflavones (SPI+) and soy protein isolate deficient in isoflavones (SPI-). After 10 wk of unlimited feeding, the hamsters were killed and carcasses were analyzed by dual-energy X-ray absorptiometry. Statistical analysis using ANOVA indicated that sex (P <0.0001) exerted a predominant effect on all outcomes [bone mineral density, lean mass, percentage of body fat (%BF) and food efficiency], with females having higher values for each variable except %BF. Contrary to our hypothesis, no treatment effects were found in females. However, in males, CAS + DAID suppressed weight gain (P = 0.03), food efficiency (P = 0.03) and %BF (P = 0.003), whereas it increased lean mass (P = 0.008) compared with SPI+. In contrast, %BF was greater in SPI+ than CAS-, illustrating that the mixture of isoflavones had a different physiologic effect than did daidzein alone. In estrogen-sufficient females, isoflavone intake had no effect, whereas daidzein had a marked effect on body composition in males.

Adipose Differentiation of 3T3-L1 Cells by Soybean Isoflavones. Keizo Sekiya. Shikoku National Agricultural Experiment Station, Ministry of Agriculture, Forestry and Fisheries, Zentsuji, Kagawa, Japan.

Antidiabetic and hypoglycemic drugs were recently reported to enhance adipose differentiation of 3T3-L1 preadipocytes. In this experiment, the effect of soybean constituents on adipose differentiation was investigated by using mouse 3T3-L1 preadipocytes. Active constituents were purified and identified. 3T3-L1 cells were grown as monolayer cultures at 37°C in DME-medium supplemented by 10% fetal bovine serum under an atmosphere of 5% CO₂/95% air. Confluent cell cultures were then treated and converted to adipocytes by culture in the presence of insulin and soybean extract samples. After ~10 d, glycerol-3-phosphate dehydrogenase activity and triglyceride accumulation were measured as markers of adipose differentiation. Insulin-sensitive glucose uptake and lipoprotein lipase activity were also assayed in the treated cells. Crude soybean extract was found to enhance the differentiation. Active constituents were purified and identified as isoflavones (e.g., daidzein or genistein). Aglycones were more active than glycosides. In the daidzein-treated cells, insulin-sensitive glucose uptake and lipoprotein lipase activity were increased. In this experiment, it is proposed that the activation of adipose function by newly differentiated adipocytes seems to be an important factor in explaining the mechanism for the effect of soybean in the prevention and treatment of chronic diseases.

Vitamin E Content and Radical-Scavenging Activity of Soymilk Made from Lipoxygenase-Lacking (Triple-Null) Soybean. Ikuo Suda, Yoichi Nishiba and Shu Furuta. Kyushu National Agricultural Experiment Station, Ministry of Agriculture, Forestry and Fisheries, Nishigoshi, Kumamoto, Japan.

Soybeans constitute an important food ingredient because of their low cost and high nutritional quality with physiologic function. However, normal soybean seeds contain three lipoxygenase isozymes: L-1, L-2 and L-3. These enzymes are responsible for the generation of undesirable grassy-beany flavors that limit the wide use of soybeans in food products. Because of their low level of beany flavor, lipoxygenase-lacking (triple-null) soybeans (e.g., Ichihime or L-star) can produce excellent, tasty soymilk and soymilk-related products. In this study, we compared the vitamin E (tocopherol) content and 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging activity in soymilk made from normal soybeans and lipoxygenase-lacking (double-null and triple-null) soybeans. Vitamin E content and DPPH radical-scavenging activity were determined by using HPLC and electron spin resonance spectrometry, respectively. Soymilk made from triple-null soybeans showed a higher vitamin E content and radical-scavenging activity than did normal and double-null soybeans. Vitamin E potency of triple-null soybean soymilk was 1.4- to 2.8-fold higher than that of parental normal soybean soymilk. Triple-null soybeans could improve the nutritional quality and functionality of soybean products in addition to improving the flavor and could offer advantages for the prevention of lifestyle-related disease (chronic disease) induced by recent food habits in Japan.

Radical-Scavenging Activities of Boiled Extracts of Soybean. Shu Furuta, Ikuo Suda, Yasuhiro Takahata and Yoichi Nishiba. Kyushu National Agricultural Experiment Station, Ministry of Agriculture, Forestry and Fisheries, Kumamoto, Japan.

Various soybean cultivars have different surface colors, such as yellow, green, red-brown and black. We investigated the radical-scavenging activities of soybean cultivars in vitro. The boiled soybean extracts obtained from cultivars with a black surface (containing anthocyanins) showed the highest scavenging activities toward 1,1-diphenyl-2-picrylhydrazyl and superoxide radicals.

Beneficial Effects of Soy Protein on Renal Function in Type 1 Diabetic Patients at Risk for Nephropathy. T. J. Hanna, P. Fanti and J. W. Anderson. Department of Internal Medicine, University of Kentucky, Lexington, KY.

Diabetic nephropathy is a primary contributor to illness and death in individuals with type 1 diabetes. An increase in urinary albumin excretion and elevated glomerular filtration rate (GFR) are characteristics of early diabetic nephropathy. The objective of this research was to evaluate the potential role of soy protein as a therapeutic agent in the prevention of renal disease in type 1 diabetes. Seven young (29.6 ± 1.9 y) volunteers with type 1 diabetes under good glycemic control (hemoglobin A_1c 7.2 ± 0.5%) participated in this 20-wk, crossover, dietary intervention pilot study. GFR was measured by using technetium 99m pentetate (Tc^99m-DTPA) clearance and urinary albumin, and albumin-creatinine ratios were determined from 24-h urine collections. After a 4-wk run-in period to assess baseline characteristics, subjects consumed a diet containing 55 g/d of soy protein for 8 wk. Soy protein was in the form of a soy patty, soy beverage and soy pasta. After 8 wk of consuming the diet based on soy protein, subjects were asked to resume their normal diet based on animal protein for an additional 8 wk. Subjects received extensive dietary counseling throughout the duration of the study. Specifically, they were instructed to reduce animal protein intake during the soy period. However, dietary recalls indicated that soy protein supplemented rather than substituted animal protein during the soy period. Consequently, total protein intake was higher during the soy period (1.4 g/kg body weight) compared with baseline (1.1 g/kg body weight) and the control period (1.0 g/kg body weight). Despite higher total protein intake during the soy period, the GFR was lower at the end of this period [147 ±12 mL/(min · 1.73 m²)] compared with baseline [154 ± 14 mL/(min · 1.73 m²)] and with the end of the control period [162 ± 16 mL/(min · 1.73 m²)]. When only the five subjects with GFR values consistent with hyperfiltration [GFR 120 mL/(min · 1.73 m²⁾] were included in this computation, the decrease in GFR during the soy period was more pronounced. Reductions of 6.3 and 8.6% in GFR in these hyperfiltering subjects were noted when the soy period was compared with baseline and the soy-free control period, respectively. Only two subjects had abnormal urinary albumin excretion at baseline, and these were the only individuals to experience a reduction in the albumin-creatinine ratio during the soy period. Interestingly, both subjects had a reduction in GFR as well during this period. From our small sample population, it may be hypothesized that subjects with clinically relevant microalbuminuria (30–300 mg/d) benefit the most from a soy-based diet. Additionally, incorporating soy into the diet of these type 1 diabetes patients had a significant benefit on serum total and LDL cholesterol levels in all seven subjects. After 8 wk of consuming a diet based on soy protein, there was an average 7% reduction in total cholesterol and 13% reduction in LDL cholesterol. When subjects returned to their control diet based on animal protein, both total and LDL cholesterol returned to values greater than those at baseline. In summary, our small sample population appears to demonstrate that subjects with clinically relevant microalbuminuria (30–300 mg/d) may benefit the most from a soy-based diet and that soy protein exerts beneficial effects on lipid profiles in type 1 diabetes. In conclusion, this pilot study suggests that incorporating soy protein into the diet of individuals with type 1 diabetes may be of value in the prevention of nephropathy.

Development of a Well-Liked Muffin Containing High Levels of Soy Protein and Isoflavones. Sandra C. Shehadeh and Raga M. Bakhit. Virginia Polytechnic Institute and State University, Blacksburg, VA.

Soy protein isolates (SPI) and the isoflavones they contain have received much attention in recent literature for their effectiveness in decreasing the risk of cardiovascular disease and cancer. These SPI are desirable because of their light color, their bland flavor and the fact that they can be mixed into bakery products accepted by the general public. A soy blueberry muffin developed at Virginia Tech underwent physical evaluation along with sensory evaluation by a consumer hedonic survey. Each muffin contained 11.3 g protein (9.3 g SPI) and 15.82 mg total isoflavones (from HPLC analysis). Moisture content was 43.3% and water activity was 0.954 at 22.0°C. Compression displacement was 13.0 mm with a maximal load of 1.71 kg. Freezing and thawing did not affect any physical measurement. This amount of SPI per muffin is much higher than the 5–6 g that is common in other clinical recipes, thus allowing consumers to eat small quantities of baked products and still obtain desired levels of soy protein intake. Sensory evaluation by 170 subjects in southwest Virginia indicated that overall (appearance, flavor, consistency), the muffin was well-liked by 85% of the participants; 50% said that they would buy the soy muffins at least two times per week, and 95% said they would eat them again. The soy blueberry muffin recipe that we used is reproducible for clinical studies in which SPI supplementation is desired. Developing soy products that are well liked by the public is vital for long-term compliance for maximal health benefit.

Positive Effects of Soy Supplementation on Macronutrient Intake in Postmenopausal Women. Janine E. Albers, Randall A. McCoy and Eric E. Thomas. South Dakota Soy Cardiovascular Research Consortium, Department of Family Medicine, University of South Dakota School of Medicine, Sioux Falls, SD.

Protein needs increase and energy needs decrease as people age. Unfortunately, protein needs are often not met in the aging population. Soy supplementation may provide benefits of increasing protein intake with minimal increases in energy intake in the elderly. The purpose of this study was to investigate the effects of soy protein supplementation on dietary macro- and micronutrient intake in an independently living aging population. Thirteen postmenopausal women, 55 y of age, were randomly assigned to receive 40 g soy protein with and without isoflavones along with their usual diet in a double-blind, randomized, crossover design. A baseline period was conducted in which usual nutrient intake information was obtained. The baseline period was followed by a 4-wk diet intervention period, an intervening 1- to 2-wk washout period, and a second 4-wk diet intervention. Soy was provided in the form of muffins. Measurements were taken by using 4-d food records during each period of the study. Comparisons were made between the two soy supplementation periods and between the soy supplementation periods and usual diet intake. Supplementing the subjects’ diets with soy protein resulted in a significant increase (P < 0.001) in overall protein intake while producing a significant decrease (P = 0.019) in the intake of fat. Energy intake did not change significantly between study periods. Results from this study show that dietary protein can be increased without an increase in fat or energy intake in the diet of an aging population through the use of soy protein.

Efficacy of a Food Mix Based on Soy Protein Isolate in Improving the Nutritional Status of Malnourished Children (Age 1–2 y). Usha Chandrasekhar, Rajammal P. Devadas, W. Hilda Sahayarani and M. Vimalarani. Department of Food Science and Nutrition, Avinashilingam University, Coimbatore, India.

The study was conducted to compare growth and development of malnourished children fed a porridge fortified with soy protein isolate (SPI) with that of children fed the standard food mix provided by the Nutrition Intervention Programme of India’s Integrated Child Development Services (ICDS) to children 2 y of age. Malnourished children (n = 1200) aged 1–2 y, with grade II malnutrition, selected from Coimbatore, Tamil Nadu, India, participated in the study. The energy gap of the children’s existing diet was identified through a standardized food intake and a 3-d weighed food record, which formed the basis for determining the level of SPI-fortified porridge for supplementation. Children were divided into four groups as follows: group I (SPI 62; n = 200 ) fed porridge fortified with 62 g SPI to bridge the existing energy gap in the community; group II (Control; n = 400) unsupplemented control; group III (SPI 49; n = 400) fed porridge fortified with 49 g of SPI to provide the same amount of energy as in ICDS food mix; and group IV (ICDS; n = 200) fed the standard ICDS food mix. Monthly records of heights, weights, three monthly assessment of head circumference, skinfold and chest measurements, initial and final hemoglobin levels, and number of respiratory incidences (morbidity pattern) were evaluated. After 10 mo of the feeding program, the mean increments in height and weight were as follows: group I, 3.76 cm, 3.48 kg; group II, 0.82 cm, 0.65 kg; group III, 2.15 cm, 1.76 kg; and group IV, 1.32 cm, 0.96 kg, respectively. These increments in growth parameters were significantly (P < 0.01) superior in group I compared with all other groups. Group III children exhibited the next highest increment, which was also significant (P < 0.01) compared with groups I and IV. Group IV children also had a significant (P < 0.01) increase in these variables compared with group II (Control) children. These trends indicate the beneficial effect of intervention in general and specifically bring out the possible role of SPI-fortified porridge in improving the nutritional status of young growing children.

Good Compliance and Few Side Effects with Consumption of Natural Soymilk in a 2-y Study. Eva Lydeking-Olsen and Jens-Erik Beck Jensen.* Institute for Optimum Nutrition, Copenhagen, Denmark and *Copenhagen University Hospital, Hvidovre, Denmark.

Previous studies in animals and humans showed a beneficial effect of soy products on bone metabolism and risk factors for heart disease. It is therefore important to evaluate the consumption of soy products. The aim of the study was to evaluate the use of, compliance with and side effects of soymilk in a 2-y study from 6- and 12-mo data. Postmenopausal Caucasian women (n = 106), mean age 58 y (range 41–75 y) were recruited and randomly assigned to one of the following four treatment groups: placebo, soymilk with isoflavones (100 mg/d), natural transdermal progesterone or both soymilk with isoflavones and natural transdermal progesterone. All participants received a broad-spectrum food supplement (Osforte) containing 680 mg calcium (citrate and carbonate), 300 mg magnesium (amino chelate), 20 mg silicium (sodium metasilicate), 15 mg zinc (amino chelate), 6 mg manganese (amino chelate), 3 mg boron (proteinate), 2 mg copper (amino chelate), 200 mg vitamin C, 40 mg pyridoxine, 200 IU vitamin D-3 and 1 mg vitamin K-1. The soymilk was enriched with calcium to 120 mg/100 mL and the daily intake was 400 mL; total calcium intake from diet, soymilk and food supplement was 1200–1500 mg/d. The study included courses on healthful cooking, how to use soymilk, exercise, balance training and relaxation techniques; one weekend course was followed by single-day courses every 3 mo for the first year and with individual follow-up during the second year. Main endpoints were changes in bone measurements (dual energy X-ray absorptiometry), lipids and biochemical markers of bone metabolism. Six participants withdrew from the study within 3 mo; three withdrew for social reasons, one relapsed with nonestrogen-receptor–positive cancer, one disliked the diet and one became severely nauseated from the soymilk. Mean compliance (controlled by pill-intake count, diaries recording the use of soymilk and weighing of leftover progesterone cream) was 97% (range 52–102%), 94% (range 50–117%) and 97% (range 39–100%) for food supplement, soymilk and progesterone cream, respectively. Participants were requested to call if any undesirable events occurred. Intervention was usually stopped until the undesirable events were over and then restarted to clarify whether the complaints were caused by the intervention. If this turned out to be the case, the reaction was reproduced twice and a decision about cessation was made. Mild stomach upset was experienced by 11% of the participants if the supplement was taken without food; 29% of the participants experienced mild side effects not leading to cessation of intake, predominantly digestive troubles, of which half were temporary; included in this group, 9% had undesired weight gain (1–7 kg), mainly from fluid retention. In addition, 6% had side effects leading to cessation of the soymilk intake; these reactions occurred 2–11 mo into the study; two subjects experienced severe nausea, flatulence and diarrhea, and four subjects had both digestive and generalized symptoms (tiredness, fluid retention, feeling of malaise, sleep disturbances, shortness of breath and joint pains). Symptoms cleared within days after cessation of soymilk. Two of these six subjects had disturbed liver parameters that normalized within weeks after cessation of soymilk. It is unclear whether the soymilk caused these disturbances because one subject had inactive hepatitis C and the other developed a coagulation disturbance 2 wk after she stopped the soymilk. Of the participants who received progesterone cream, 22% experienced mild side effects of skin irritation and temporary breast tenderness. One subject had to stop using the cream because of severe local skin irritation. No changes were seen in hematological variables. Of 13 subjects with self-reported cow’s milk intolerance, 11 tolerated soymilk. Beneficial effects occurred in 18, 14 and 8% for food supplement, soymilk and progesterone cream groups, respectively. The food supplement improved nail and hair quality; some subjects felt more energetic, and sleep quality improved. Soymilk relieved 12 subjects of chronic constipation, irritable bowel symptoms or both. We conclude that soymilk is a suitable alternative or supplement to other calcium sources in the diet. It is well tolerated and side effects are mostly mild and temporary, with a prevalence similar to those for cow’s milk or calcium supplements. [Soymilk and placebo were provided by ALPRO, Belgium, Pro-derma and placebo were provided by Phillips Nutritionals, USA, and Osforte by Genese A/S, Denmark. The study is supported by grants from the Danish Ministry of Health, Fondation Idella, Glunz and Jensen Foundation, Civilingenieur Frode V. Nyegaard and Wife’s Foundation, and Director E. Danielsen and Wife Foundation.]

	Isoflavone intake

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Soy Consumption and Nutrient Intakes of American Adults Based on the CSFII Data Base 1994–1996. V. Juturu, G. Hsheih and P. M. Kris-Etherton. Department of Nutrition, Pennsylvania State University, University Park, PA.

The purpose of this study was to evaluate the effect of consuming soy products on the energy and nutrient intakes of free-living male and female adults and children relative to not consuming soy products by using data from the Continuing Survey of Food Intake by Individuals (CSFII) conducted from 1994 to 1996. More Caucasians (4.22%) reported soy consumption than did other groups (Asians, 1%; blacks, 0.3%; American Indians, 0.03%; and others, 0.25%). Of CSFII respondents, 5.8% consumed soy products. Compared with soy nonusers, soy users had a higher intake of energy, protein, polyunsaturated fatty acids, dietary fiber and dietary carbohydrate and a lower intake of saturated fatty acids (SFA) (P < 0.0001). Total fat, monounsaturated fatty acids, dietary cholesterol, calcium and vitamin B-12 intakes were similar for soy users and nonusers. Energy intake was higher in adult soy users than nonusers (P < 0.0001). Body mass index was 25.8 and 24.9 kg/m² for male and female soy users and 26.6 and 26.2 kg/m² for male and female nonusers, respectively (P < 0.0001). Mean adequacy ratio was higher in soy users (90) than nonusers (86; P < 0.001). The nutrient profile of the diet of soy users was superior to that of nonusers and met some dietary recommendations, mainly by reducing SFA and cholesterol, for the prevention of chronic diseases.

Intake of Dietary Phytoestrogens in Postmenopausal Women:The Framingham Heart Study. Miriam J. J. de Kleijn, Yvonne T. van der Schouw, Peter W. F. Wilson,* Diederick E. Grobbee and Paul F. Jacques.Julius Center for Patient Oriented Research, University Medical Center Utrecht, Utrecht,The Netherlands; *Framingham Heart Study, Boston University School of Medicine, Framingham, MA; and Jean Mayer, U.S. Department of Agriculture Human Nutrition Research Center on Aging at Tufts University, Boston, MA.

The purpose of the study was to estimate the intake of dietary isoflavones, coumestans and lignans in healthy Western postmenopausal women. Many plant foods contain isoflavones, coumestans and lignans. Only a few published studies examined the dietary intake of phytoestrogens in the general Western population. The potentially positive health effects of phytoestrogens are most relevant to postmenopausal women. We studied 939 postmenopausal Caucasian women who entered the fifth examination of the Framingham Offspring Study (1991–1995) and completed the Willett Food Frequency Questionnaire (FFQ). By searching the medical and agricultural literature and contacting experts, we identified food sources of phytoestrogens. We determined the concentrations of the different isoflavones (daidzein, genistein, formononetin and biochanin A), coumestans and lignans (matairesinol and secoisolariciresinol) for each food item on the FFQ. We scored the concentration of phytoestrogens in each food item into seven categories and multiplied the score by the serving size of the food. The amounts in each food item were multiplied by the frequency of the consumption of that food and then summed across foods to obtain the total intake of each phytoestrogen. Median total daily intake of the phytoestrogens is presented with interquartile ranges in parentheses. The estimated daily median intake of the isoflavones was as follows: daidzein, 39 µg (24–57 µg); genistein, 70 µg (28–120 µg); formononetin, 31 µg (13–44 µg); and biochanin A, 6 µg (2–11 µg). The median total intake of isoflavones was 154 µg (98–235 µg), and the main source was beans and peas. The estimated daily intake of coumestans was 1 µg (0–2 µg), and the main source was broccoli. The estimated daily median intake of matairesinol was 25 µg (16–36 µg ) and of secoisolariciresinol was 534 µg (383–763 µg). The median total intake of lignans was 565 µg (407–788 µg), and the main source was fruits other than citrus fruits or berries. Dietary intake of isoflavones, coumestans and lignans in healthy postmenopausal Caucasian women in the United States is low. Despite the low intakes, recommendations for changes in the diet of postmenopausal women to increase dietary phytoestrogens may be premature before the health benefits of phytoestrogens are clearly demonstrated.

Soy Consumption in Taiwanese Children in Taipei. A.K.-F. Hsiao and P. M. Lyons-Wall. Human Nutrition Unit, Department of Biochemistry, University of Sydney, Sydney, NSW, Australia.

Soybeans have been a staple food in the traditional Asian cuisine for many centuries. Epidemiologic evidence showed that lifetime soy consumption could contribute to the low rates of heart disease and cancer observed in Asian communities, yet controversy remains concerning whether eating soy is beneficial for children mainly because of the presence in soy of the weak estrogenic isoflavones. This study examined the range and frequency of intake of soy foods in a group of healthy Taiwanese children, aged 8–9 y living in Taipei (n = 66; weight: 32.4 ± 0.9 kg; height: 134 ± 0.8 cm). The city of Taipei was divided into three geographical areas, and within each, one public school was randomly selected. Approximately equal numbers of boys and girls participated from each school, and parents were requested to fill in the questionnaire on behalf of their children. The range and frequency of intake of soy foods were determined by using a specially designed frequency questionnaire that included soy products available in local shops and markets in Taipei. Seven frequency categories were offered ranging from ‘3 times/d‘ to ‘never,‘ and each item was allocated an appropriate serving size. The children consumed a diversity of soy products (23 different items), reflecting the use of soy as a traditional staple food in the Taiwanese cuisine. The most commonly consumed items were soy drink (mean intake 556 g/wk), soft tofu (303 g/wk), hard-pressed tofu (255 g/wk), soybean curd jelly (221 g/wk) and soy sauce (122 g/wk). Soy products were introduced at age 1.9 y (mean age) as soy drink (40%), tofu (40%), soybean curd jelly (9%) or other products (11%). Estimated total intake of isoflavones in the group was 255 mg/wk. The average daily intake of isoflavones in this study (36.6 mg/d) was comparable to intakes that have resulted in physiologic effects in young women. However, no effects have been reported in children, either anecdotally or in the scientific literature. It is possible that the potential estrogenic effects of isoflavones are lowered in children as a result of the first-pass effect in the liver, in which isoflavones are absorbed and converted to inactive metabolites in the liver before entering the peripheral circulation to act on target tissues. Alternatively, the tissues of children could be less sensitive than those of adults to the effects of isoflavones.

Urinary Isoflavone and Lignan Excretion by Premenopausal Women in Seattle, Washington. H. Skor, S. Schwartz, E. Fitzgibbons, D. Scholes, C. Chen, K. Wähälä* and J. Lampe. Fred Hutchinson Cancer Research Center, Public Health Sciences, Seattle, WA; *University of Helsinki, Department of Chemistry, Helsinki, Finland; and Group Health Cooperative of Puget Sound Center for Health Studies, Seattle, WA.

Soy foods are increasingly available to consumers, but few data exist on the measurement of and levels of exposure to isoflavones in the general U.S. population. Over a 24-mo period (4/97 to 3/99), we measured excretion of isoflavones and lignans by Seattle women participating in a case-control study of uterine leiomyomata. Premenopausal women (n = 217) provided two overnight urine collections 48 h apart. Isoflavone and lignan excretion expressed as nanomoles per milligram of creatinine was highly correlated with excretion expressed as nanomoles per hour (P = 0.0001). Correlations between the two measures of genistein, daidzein, O-desmethylangolensin, total isoflavones (sum of isoflavonoids), and the lignans enterolactone and enterodiol (nmol/h) were 0.34, 0.32, 0.76, 0.41, 0.65 and 0.41, respectively (P = 0.0001). Geometric mean excretions (nmol/h) were as follows: genistein, 9.1; daidzein, 19.8; total isoflavones, 45.0; and total lignans (enterolactone plus enterodiol), 100.0. Isoflavone and lignan excretion was detectable in at least one collection in most women as follows: genistein, 81%; daidzein, 93%; O-desmethylangolensin, 49%; enterolactone, 99%; and enterodiol 96%. However, when excretion was examined separately for each collection, the frequency of detectable levels of isoflavones in women decreased; for example, for genistein, 63% of women excreted detectable levels on a single night but only 46% excreted genistein on both nights. Asians (n = 39) had higher levels of genistein (P = 0.02), daidzein (P = 0.04) and total isoflavones (P = 0.03) than did non-Asians, but the non-Asians excreted higher concentrations of lignans (P = 0.002). Despite increased soy food availability and consumer awareness of the possible benefits of soy, we did not observe an increase in isoflavone excretion over the 24 mo in the whole group or in non-Asians alone. In conclusion, in the Seattle area, a large proportion of women are exposed to isoflavones, even though at low levels.

Phytoestrogen Concentrations in Japanese Women and Men. M. S. Morton, O. Arisaka,* A. Miyake,* L. D. Morgan and B.A.J. Evans.Bioclinical Services, International, Units 1–3 Willowbrook Laboratories, Cardiff, UK; *Juntendo University School of Medicine, Tokyo, Japan; and Department of Child Health, UWCM, Cardiff, UK.

The soy-derived isoflavones daidzein and genistein have been proposed to be cancer-protective agents in Japanese and other Asian populations. This study reports the concentrations of the isoflavonoids daidzein, genistein and equol in the serum of Japanese women (n = 125) and men (n = 102) as determined by a robust gas chromatography–mass spectrometric method. The women were aged 40–89 y (mean 66.9 y), and the men were aged 40–85 y (mean 64 y). For comparison, data are also presented for British women (n = 60) and men (n = 50). The mean serum levels of daidzein and genistein in Japanese women were 62.6 ng/mL (range 0–611 ng/mL) and 135.6 ng/mL (range 0–1133 ng/mL), respectively. The mean equol concentration was 13.8 ng/mL (range 0–241 ng/mL) with only 32.5% of the women producing equol concentrations 5 ng/mL. In British women, the mean concentrations of daidzein and genistein were 3.7 and 8.1 ng/mL, respectively. The mean concentrations of daidzein and genistein in the sera of Japanese men were 71.1 ng/mL (range 0–577 ng/mL) and 133.2 ng/mL (range 0–1160 ng/mL), respectively. The mean concentration of the isoflavone equol was 23.8 ng/mL (range 0–461 ng/mL), and 57 (56%) of these samples had equol concentrations 5 ng/mL. For British men, the mean concentrations of daidzein and genistein were 4.63 ng/mL (range 0–16.4 ng/mL) and 9.2 ng/mL (range 0–31.8 ng/mL), respectively. Clearly, from this study, Japanese women and men have much higher levels of daidzein and genistein in their serum than do British women and men. A life-long exposure to these soy-derived phytoestrogens may be responsible in part for the lower incidences of breast and prostate cancer observed in Japanese and other Asian populations.

	Menopause symptom relief

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
A Comparison of the Effect of Treatment with Oral Micronized Estradiol or Phytoestrogen-Containing Soy Protein Isolate on Tail Skin Temperatures of Ovariectomized Rats. Yuanlong Pan, Mary S. Anthony, Misti Kapps and Thomas B. Clarkson. Comparative Medicine Clinical Research Center, Wake Forest University School of Medicine, Winston-Salem, NC.

Whether phytoestrogen-containing soy supplements have beneficial effects on hot flushes of postmenopausal women and how those effects, if any, compare with estrogen replacement therapy has been uncertain. Some researchers found a 22% reduction in severity of hot flushes in postmenopausal women treated with soy supplements but no change in numbers of hot flushes. Small effects on hot flush frequency and severity were found by one researcher. No significant differences between soy and control treatments were found in two other studies. These observations are in contrast to the 90% reduction in hot flushes with estrogen replacement. Uncertainty exists about whether the dose of soy isoflavones (30–60 mg/d) was too low. We used ovariectomized retired breeder rats to approach these uncertainties experimentally. The treatment groups were as follows: the control group was fed a diet based on casein and lactalbumin; the Soy(-) group was fed alcohol-washed soy protein isolate with the phytoestrogens extracted; the Soy(+) group was fed phytoestrogen-containing soy protein (equivalent to a woman’s dose of 144 mg/d of isoflavones), which is two to three times higher than that in the studies with women; and the 17-ß-estradiol (E₂) group was fed oral micronized estradiol at a dose equivalent to a woman’s dose of 1 mg/d. A temperature transponder was taped to the surface of the tail to measure temperature. Tail skin temperature was significantly increased within 1 wk after ovariectomy. The animals were pair-fed during treatment (21 d). Soy(-) had no effect on skin temperature. E₂ had a large effect on skin temperature (~1.4°C reduction from Control). Soy(+) was intermediate between the E₂ and Control (~0.8°C reduction from Control). In conclusion, in the rat model system, soy phytoestrogens, even at high doses, have a modest effect on skin temperatures that is about half that of estradiol.

Effect of Isoflavone-Rich Soy Protein Isolate on Menopausal Symptoms in Perimenopausal Women. D. Lee Alekel, A. St. Germain and C. T. Peterson.* Department of Food Science and Human Nutrition, Human Metabolic Unit, Center for Designing Foods to Improve Nutrition and *Department of Statistics, Iowa State University, Ames, IA.

Soy isoflavones, estrogen-like substances structurally and functionally similar to 17-ß-estradiol, may play a role in relieving menopausal symptoms. In this double-blind, 24-wk study, we examined the effects of isoflavone-rich (80.4 mg/d) soy protein isolate in relieving menopausal symptoms and compared two methods for collecting menopausal symptoms, i.e., a menopausal diary and a menopausal index. Subjects (n = 69) were randomly assigned to treatment as follows: isoflavone-rich soy (n = 24), isoflavone-deficient soy (n = 24) or whey control (n = 21) protein. At baseline, midtreatment and post-treatment, subjects used a diary to record hot flushes and night sweats. Every 6 wk, a menopausal index was used to assess symptoms. Urinary isoflavones indicated excellent compliance. Repeated measures ANOVA indicated no treatment effect using the menopausal diary or menopausal index, respectively, on change in frequency of hot flashes (P = 0.065 or P = 0.18) and night sweats (P = 0.85 or P = 0.92). Chi-square analyses indicated no treatment effect on severity of hot flushes or night sweats at any time when either tool was used except for night sweats at midtreatment (using the diary), which were significantly lower (P = 0.036) in the control group. Chi-square analyses revealed no treatment effect on frequency or severity of insomnia, limb numbness, headaches or fatigue at any time point. Similarly, there was no treatment effect on mood swings, libido, vaginal dryness or urinary frequency and urgency. However, all treatment groups in general reported a decline in menopausal symptoms, indicating a strong placebo effect. On the basis of these results, isoflavones may not provide the once-anticipated relief of vasomotor or other menopausal symptoms.

A Randomized Trial of a Soy Beverage in the Treatment of Menopausal Symptoms in Postmenopausal Breast Cancer Survivors. Cheri Kutynec, Ivo Olivotto, Jerilynn Prior,* Greg Hislop, Keith Chambers, Karen Gelmon and Edith Templeton. BC Cancer Agency, Vancouver Cancer Centre, Vancouver, BC, Canada; *University of British Columbia, Canada; and Vancouver Hospital and Health Sciences Centre, Vancouver, BC, Canada.

Menopausal symptoms are often distressing and difficult to tolerate in women who have undergone treatment for breast cancer. Treatments such as chemotherapy and tamoxifen may induce menopause and increase its symptoms, and conventional treatment using hormone replacement therapy is contraindicated. In both postmenopausal and Asian women, the consumption of soy-rich diets containing phytoestrogens has been associated with a reduction in hot flushes. The purpose of this randomized, double-blind, placebo-controlled trial was to evaluate the acceptability and effectiveness of a soy beverage (containing 80 mg isoflavones) in the treatment of menopausal symptoms in women with a history of treated breast cancer. Eligible women were included in the study if they were symptomatic for hot flushes, were 4 mo or more post-treatment (may be taking tamoxifen) and postmenopausal. Women were excluded if they had had a recurrence of cancer, had liver or kidney disease, were using hormone replacement therapy or antibiotics, or smoked. After enrollment, women completed a validated instrument called the daily menopause diary for 4 wk at baseline and for 12 wk while consuming 500 mL/d of a soy beverage or placebo. At baseline and 4, 8 and 12 wk, women completed a questionnaire to determine their consumption of soy, alcohol and caffeine and use of alternative therapies. The measurement of serum follicle-stimulating hormone, alkaline phosphatase and gamma GT was obtained at baseline, and additional samples of serum and urine were stored at baseline, 6 and 12 wk for future analysis. At present, 135 out of a goal of 160 women are enrolled, and 60 women have completed the study. Results will be discussed. [Funded by The National Cancer Institute of Canada and the Canadian Breast Cancer Research Initiative.]

Effect on Biochemical Parameters of an Oral Soy Extract Used in the Treatment of Vasomotor Symptoms in Menopausal Women. D. H. Upmalis, F. C. Cone and C. Lamia. Personal Products Company, North Brunswick, NJ.

The objective was to determine biochemical changes resulting from the use of an oral soy isoflavone extract for relief of hot flushes in menopausal women with the use of a double-blind, randomized, balanced, parallel group, Phase II, outpatient, multicenter (16 sites) study. Menopausal women (n = 177), mean age 55 y, with five or more hot flushes per day, were randomly assigned to receive either two soy isoflavone extract tablets (containing a total of 25 mg genistin and daidzin per tablet) or two placebo tablets once daily for 90 d. Physical and gynecological exams and biochemical evaluations were performed at admission and completion. Soy isoflavone demonstrated a significant reduction in the severity of hot flushes over the 12 wk and there was a trend toward significance in reduction of frequency of hot flushes and night sweats. There was no effect on endometrial thickness. Mean biochemical changes are given in Table 1 . Soy isoflavone extract was effective in reducing frequency and severity of menopausal hot flushes but in this study showed little effect on biochemical variables that have been shown to be affected by dietary soy. The change in fasting glucose requires further investigation to confirm and characterize this observation. The findings confirm the utility of soy isoflavone extract in the relief of menopausal hot flushes and underline the importance of characterization of isoflavone extracts because their metabolic effect can be markedly different from that of the parent dietary source.

Improvement of Menopausal Symptoms and Quality of Life by Oral Soy Extract Tablets. David H. Upmalis, Brinda Wiita, Frederick L. Cone, Cathleen A. Lamia, Michelle Warren* and Lynn Bradley. Advanced Care Products, North Brunswick, NJ; *Columbia University, New York, NY; and Johns Hopkins University, Baltimore, MD.

The objective was to characterize the effects of soy isoflavone extract tablets on menopausal symptoms and quality of life. Estrogen-deficient menopausal women (n = 177) experiencing five or more hot flushes daily participated in a double-blind, randomized, placebo-controlled, 12-wk multicenter study of Optisoy soy extract tablets (a total 50 mg/d of genistin and daidzin). Symptom-based quality of life was measured with the Women’s Health Questionnaire, a validated menopausal quality-of-life scale. Hot flushes and night sweats, recorded on diary cards, were the primary efficacy criteria. Severity of hot flushes was reduced to a significantly greater extent by soy compared with placebo. The number of hot flushes also decreased significantly in the first 6 wk. Patient responses to the questionnaire revealed a significant (P = 0.04) reduction from baseline in sleep problems and a slight improvement in memory problems in the soy group (P = 0.06). Soy isoflavone extract tablets relieved menopausal symptoms effectively and improved quality of life.

	Safety issues

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 
Long-Term Adverse Effects After Developmental Exposure to Genistein. Retha R. Newbold, Wendy N. Jefferson, Elizabeth Padilla Banks and Bill C. Bullock. Developmental Endocrinology Section, Laboratory of Toxicology, National Institute of Environmental Health Sciences, Research Triangle Park, NC.

Exposure to synthetic and naturally occurring estrogenic compounds results in permanent alterations in the developing organism if exposure occurs during the critical stages of differentiation. Reports from our laboratory showed that perinatal exposure of mice to diethylstilbestrol (DES) results in impaired fertility, structural malformations and lesions of the reproductive tract. In fact, long-term consequences of exposure to DES at 2 µg/(pup · d) on d 1–5 include a high incidence (95%) of uterine carcinoma in animals 18 mo of age. Doses <2 >µg/(pup · d) also result in long-term adverse effects. Thus, the developing reproductive tract appears to be extremely sensitive to perturbation by compounds with estrogenic activity. Because the nutritional and pharmaceutical use of phytoestrogens has increased over the past few years, mainly because of its reported beneficial effects, we investigated the potential risks posed by genistein if exposure occurs early in development. We tested the possibility that developmental exposure to this compound would influence morphological, functional and biochemical markers known to be estrogen sensitive. Uterine epithelial cell proliferation and induction of uterine lactoferrin and complement C3 were increased in response to genistein. Although lactoferrin was previously reported to be constitutively produced in the uteri of neonatally DES-treated mice as early as 2 mo before the development of uterine carcinoma, the role of this protein in the induction of neoplasia, either as a marker or contributing factor, remains to be determined. However, lactoferrin and other marker proteins are induced by developmental exposure to genistein. The interaction of genistein with the estrogen receptors and ß was investigated. Further, mice were followed to evaluate a potential increased risk for histological abnormalities, including uterine tumors later in life. Many of the long-term effects observed after DES treatment, including uterine adenocarcinoma, were observed after developmental exposure to genistein. Similarities between the effects of DES and genistein point to the need for further mechanistic studies in phytoestrogens and the role in long-term effects that follow developmental exposure.

Estrogen Receptor Estrogen Receptor ß and Lactoferrin Expression in Reproductive Tract Tissues after Treatment with Genistein or Diethylstilbestrol during Development. Wendy N. Jefferson, Sabine E. Kulling, Elizabeth Padilla Banks, Jennifer Hagelbarger, John F. Couse* and Retha R. Newbold. Laboratory of Toxicology and *Laboratory of Reproductive Toxicology, NIEHS, Research Triangle Park, NC and Institute of Food Chemistry, University of Karlsruhe, Karlsruhe, Germany.

Permanent alterations occur in both male and female mice after exposure during development to synthetic and naturally occurring estrogenic compounds. Reports from our laboratory showed that perinatal exposure of mice to diethylstilbestrol (DES) results in impaired fertility and structural malformations and lesions of the reproductive tract. Long-term consequences of exposure to DES on d 1–5 included a high incidence (95%) of uterine carcinoma. Similar long-term effects were shown with genistein treatment. Thus, the developing reproductive tract appears to be uniquely sensitive to compounds with estrogenic activity. To study the developmental effects of genistein, which is found in many soy products and was documented to have estrogenic activity, we tested the possibility that neonatal exposure to this compound would alter estrogen-signaling pathways in estrogen target tissues and compared the changes with tissues similarly exposed to DES. First, the ontogeny of estrogen receptor (ER) , ER-ß and lactoferrin (LF) from fetal d 14 through postnatal d 26 was mapped in the developing reproductive tract of control female CD-1 mice by using a ribonuclease protection assay and immunohistochemical analysis. ER- was present in both the ovary and reproductive tract; ER-ß was present in the ovary but not in the uterus. LF could not be detected in uterine tissue until about postnatal d 19. After neonatal treatment on d 1–5 with genistein (500–50,000 µg/kg) or DES (0.01–1000 µg/kg), reproductive tract tissues were compared with those of nonexposed control mice to determine whether the treatment altered the expression of these proteins. Mice exposed to genistein or DES showed an increase in LF expression in the uterus on d 5, the last day of treatment. Preliminary data from neonatally exposed genistein or DES mice showed alterations in both ER- and ER-ß expression in the ovary and reproductive tract. The dose response and time course for ER- and ER-ß expression during and after estrogen treatment are being determined. The relationship of altered ER-, ER-ß and LF expression and long-term adverse consequences in the female reproductive tract is unknown; however it is currently being investigated.

Dietary Soy Phytoestrogens Decrease Brain Calcium-Binding Proteins but Do Not Alter Hypothalamic Androgen-Metabolizing Enzymes in Adult Male Rats. Edwin D. Lephart, Kenneth D. Setchell,* Herman Adlercreutz and K. Scott Weber. Neuroscience Center, Brigham Young University, Provo, UT; *Clinical Mass Spectrometry Center, Children’s Hospital Center, Cincinnati, OH; and Folkhälsan Research Center, Department of Clinical Chemistry, University of Helsinki, Finland.

Phytoestrogens, estrogen-like molecules found in many plants, recently have received a great deal of investigative attention because of their potential protective effects against age-related diseases (e.g., cardiovascular disease and osteoporosis) and hormone-dependent cancers (i.e., breast and prostate cancer). Calbindin (CALB) and calretinin (CALRET) are calcium-binding proteins that are potentially important in the development and function of the central nervous system, acting as potential neuroprotective factors against programmed cell death (apoptosis) associated with neurogenerative diseases, such as Parkinson’s and Huntington’s diseases. On the other hand, the aromatase and 5-reductase enzymes represent the major pathways of steroid hormonal action in the central nervous system, especially in hypothalamic regions during perinatal and postnatal development in which they play important regulatory roles in neuroendocrine functions, reproductive endocrine physiology and sexual behaviors. To determine whether a soy diet containing relatively high levels of phytoestrogens may influence brain androgen-metabolizing enzymes and calcium-binding protein levels in adult male rats, we examined aromatase and 5-reductase enzyme activities in the medial basal hypothalamic and preoptic area (MBH-POA) and characterized MBH-POA and amygdala (AMY) CALB and CALRET levels from rats fed a phytoestrogen-free vs. a phytoestrogen-containing diet. Male Sprague-Dawley rats were randomly assigned to two treatment groups as follows: phyto-free diet or phyto-600 diet. The phyto-600 diet contained phytoestrogens at 600 µg/g; the phytoestrogen content in the phyto-free diet was below the limits of HPLC detection. After 4 wk of consuming the diets (ages 70–100 d), brain tissue sites were collected, MBH-POA aromatase and 5-reductase levels were determined, plasma and brain phytoestrogen concentrations were measured, and MBH-POA and AMY CALB and CALRET levels were assayed by Western analysis. The total plasma phytoestrogen concentration in the phyto-600 group was significantly higher (2460 ng/mL) than in the phyto-free group (32 ng/mL). The MBH-POA content of phytoestrogens was also significantly higher in the phyto-600 (69 ng/g) than in the phyto-free group (9 ng/g). Surprisingly, there were no significant differences in the MBH-POA androgen metabolizing enzymes (i.e., aromatase or 5-reductase). However, there was a significant decrease in CALB levels in the MBH-POA brain site (and CALRET in the MBH-POA and AMY brain sites) of the phyto-600 compared with the phyto-free–fed rats. These data suggest that dietary phytoestrogens (or the lack thereof) for a relatively short interval (~4 wk) can significantly alter brain calcium-binding proteins in adult male rats. This alteration may have important implications for the neuroprotective effects of these Ca²⁺-regulating molecules at the time of neurogenesis, during postnatal development and especially in aged animals.

Normal Growth and Immune Function of Newborn Term Infants Fed Soy Formulas for One Full Year. Karin M. Ostrom and John B. Lasekan. Ross Products Division, Abbott Laboratories, Columbus, OH.

Healthy, newborn term infants fed soy-based infant formulas (SOY) in a randomized, masked parallel study were evaluated for growth, serum biochemistry values, antibody response to routine childhood vaccinations and morbidity. For their first year of life, infants (n = 186) were randomly fed one of two formulas (one commercially available and one experimental) containing soy protein isolate. A nonrandomized reference cohort of 81 infants was fed mixed feedings (MF) consisting exclusively of human milk to at least age 2 mo followed by human milk, a commercially available cow’s milk–based formula or both to age 1 y. Groups did not differ by sex, ethnicity or birth weight. Growth was normal and weight, length and head circumference were similar among the three groups. All serum biochemistry values for the three groups were within the normal infant reference ranges. The response to Hemophilus influenzae type b conjugate (Hib), diphtheria toxoid, oral polio and tetanus toxoid vaccines, differential white blood cell count and lymphocyte subsets indicated normal maturation of the immune system for all three groups of infants. For the infants fed SOY, the response to Hib vaccine was similar to or greater than that for infants fed MF, and the responses to the other vaccines were similar to those of infants fed MF. Of all infectious morbidity and antibiotic usage data collected, only physician- reported infectious diarrhea was less for infants fed MF than for infants fed SOY. This 1-y study demonstrated that growth and immune response of infants fed SOY was normal and was similar to growth and immune response of infants fed HM.

Low Comparative Reactivity of Soy Proteins and Soy-Based Infant Formulas in an Animal Model of Food Allergy. Christopher T. Cordle, Geralyn Duska-McEwen, Karen L. Courtad, Melissa R. Nameth and William T. Malone. Ross Products Division, Abbott Laboratories, Columbus, OH.

The most common infant food allergen is cow’s milk, probably because cow’s milk–based infant formula is the most widely used infant feeding substitute for human milk. Infants who develop allergies to cow’s milk react to its casein, whey protein or both. The use of soy-based formula in infants allergic to cow milk is widely practiced but not officially endorsed by pediatric societies. Clinical results for prophylactic use of soy in patients at high risk to develop atopic disease conflict. When animal models of food allergy were used to compare cow’s milk–based with soy-based formulas and casein or whey with soy protein isolate, soy formulas and soy proteins were consistently less reactive. We used a hyperimmunization animal models to rank the immunologic reactivity potential of several protein systems that might be used for infants allergic to cow’s milk or infants at high risk to develop food allergy. Laboratory animals were hyperimmunized with casein, whey, rice protein, rice bran protein, pea protein, oat protein or soy protein. Immune sera were collected and antibodies to the immunizing proteins were quantitated by using sensitive, antigen-specific ELISA methods. Preimmunization antibody titers were also determined as controls for the immunization process and ELISA specificity. Data were expressed as log immune response [IR; (log d 35 titer) - (log d 0 titer)]. Geometric mean IR values for the immunogens were compared by ANOVA; this analysis indicated that the proteins fall into three statistically distinct reactivity categories. Cow’s milk proteins are highly reactive, i.e., casein IR was 4.49 and whey IR was 4.46. Rice, rice bran and pea proteins showed the same intermediate reactivities, i.e., the IR were 3.99, 3.96 and 3.63, respectively. Oat and soy proteins were least reactive and not different from each other; the IR were 2.71 and 2.22, respectively. The IR of a hypoallergenic casein hydrolysate was 1.28. It is clear that soy and oat proteins are allergenic for some patients. However, well-controlled and standardized animal model data support the current practice of using soy-based infant formula and oatmeal for infants with cow’s milk intolerance. The reduced immunologic reactivity of soy and oat may represent intrinsic characteristics of these proteins that might be further exploited in the management of food allergies.

Immunological Study of a Soybean-Based Liquid Food. Leonardo M. Vanella, Oscar A. Brarda, Raúl V. Boudet and Marcela Permigiani. Fundación Médica Científica Centro República, Córdoba, Argentina.

The purpose of this study was to evaluate the immunogenicity of a soybean-based liquid food (SBLF; AdeS Natural), and to establish its antigenicity and allergenicity. Specific antisera were obtained by hyperimmunization of 15 rabbits as follows: five rabbits were inoculated with SBLF, five were inoculated with an isolation of soybean, and five were inoculated with a soybean antigenic extract. The stimulation was done with an emulsion in equal parts of antigen and Freund’s incomplete adjuvant. The presence of the major immunoglobulins (Ig), anti-SBLF and antisoybean was investigated in sera of 213 children by double diffusion gel and crossover immunoelectrophoresis techniques. The levels of total serum IgE and the presence of IgE antisoybean were determined by radioallergosorbent test. The allergenicity of SBLF and soybean was evaluated by skin testing (skin prick) in 202 atopic children. The group of rabbits stimulated with SBLF demonstrated an immune response slightly lower than that of the other groups. The presence of IgG, IgM and IgD anti-SBLF or antisoybean was not observed in children’s sera. IgE antisoybean was detected in five sera. Eight atopic children showed a positive soybean cutaneous test. The prick test with SBLF was negative in all children. The immunogenicity of SBLF was slightly lower than that of the soybean extract; SBLF shares two antigenically related components with soybean; IgE anti-SBLF was not observed in the children. It would be due to the industrial elaboration of the SBLF that includes thermal denaturalization of trypsin inhibitors that are strong allergenic factors.

Biological Effects of Soybean Isoflavones on Nutritional Status in the Rat Model. G. Sarwar, M. R. L’Abbé and S. Brooks. Nutrition Research Division, Health Protection Branch, Health Canada, Ottawa, ON, Canada.

The isoflavones, genistein and daidzein, and their glycosides, found in high concentrations in soybeans and soy-protein foods, may have beneficial effects in the prevention of many hormone-dependent diseases. However, these bioactive phytoestrogens possess a wide variety of hormonal and nonhormonal activities. It has been suggested that potential adverse effects may occur in infants fed soy formulas (Setchell 1997 ). The daily exposure of infants to isoflavones in soy-based formulas was calculated to be 6- to 11-fold higher on a weight basis than the dose that has hormonal effects in adults consuming soy foods. Circulating concentrations of isoflavones in infants fed soy-based formulas have been reported to be 13,000–22,000 times higher than plasma estradiol concentrations in early life that may exert biological effects (Setchell 1997 ). However, The no observed adverse effect level (NOAEL) of soy isoflavones is not known. To address this, a 16-wk feeding study was conducted to investigate the biological effects of the addition of graded levels (0, 50, 100, 200 and 400 mg/kg diet) of soybean isoflavones to a casein control diet on growth, protein, mineral and hormone status and reproductive endpoints in male and female rats. An isoflavone rich extract (Novasoy) and a soy-based infant formula were used as the source of dietary isoflavones. Isoflavones in foods and in plasma samples were determined by the methods of Wang and Murphy (1994) and King et al. (1996) , respectively, using Waters HPLC linear gradient with UV detector monitored at 254 nm. The commercial powder forms of soy-based infant formulas sold in Canada contained 178–292 µg/g of total isoflavones (daidzin, genistin, daidzein and genistein), whereas Novasoy contained 300 mg/g of the total isoflavones. The rat feeding study started on August 20, 1999, and will be completed on December 10, 1999. Interim analyses (after 6 wk of study) revealed an adverse effect on body weight of male rats fed the highest two levels of isoflavones, 200 mg/kg from formula and 400 mg/kg from Novasoy. There was a dose-related increase (up to fivefold in males and up to threefold in females) in the levels of plasma isoflavones. The plasma data showed that the absorption of isoflavones from formula was lower than that from Novasoy. There was also a dose-related increase in the length of the estrous cycle in female rats. The cycle in rats fed the infant formula diet (providing 200 mg/kg of isoflavones) was shorter, however, than that in those fed the Novasoy diet providing the same amount of isoflavones. This suggests differences in the potency of endogenous and extracted isoflavones.

Transfer of Isoflavones from Mother to Suckling Pups through Milk in Rats. Yoko Nakashima, Yoshiko Ishimi,* Keizo Umegaki* and Sachie Ikegami.* Seitoku University, Iwase, Matsudo-city, Chiba, Japan and *The National Institute of Health and Nutrition, Toyama, Shinjuku-ku, Tokyo, Japan.

There is a general agreement that soy isoflavones can be beneficial to health in adults. However, isoflavones have acted as endocrine-disrupting chemicals. Therefore, some consideration should be given to the possibility that soy foods may adversely affect the reproductive system in infants. On d 5 of pregnancy, we fed Sprague-Dawley rats one of three diets containing daidzein and genistein and having a total isoflavone concentration of 0, 0.5 and 1.0 g/kg. Although the body weight and serum triiodothyronine concentration of dams in the 1.0-g/kg group were lower than those in 0-g/kg group, no significant differences were observed among the suckling pups in the three groups. We detected both daidzein and genistein in the blood of dams and pups and the stomach contents (milk) of pups. However, small amounts of isoflavones transferred from mother to suckling pups.

No Estrogenic Hormonal Effects in Children Fed Soy Formula Long Term. L. Businco, G. Bruno, P. G. Giampietro, G. Furcolo and F. Lucaroni. Allergy and Clinical Immunology Division, Department of Pediatrics, University ‘La Sapienza,‘ Rome, Italy.

Phytoestrogens are present in high concentration in soybeans. Therefore, hormonal effects may occur in infants fed soy formula (SF). The aim of this study was to evaluate the hormonal effects of long-term SF feeding in a selected group of children who were fed SF (Isomil) for more than 6 mo early in life. We enrolled 50 children (30 boys and 20 girls), median age 29 mo (range 7 mo–12 y). Every child had a physical examination with particular attention paid to signs and symptoms of precocious puberty. In addition, the following were evaluated: bone density (mineralometry with double photon absorptiometry) in children aged 3 y; bone age (X-ray of carpus and metacarpus according to Tanner) in all of the children; bone metabolism urinary markers, i.e., urinary deoxypyridinoline (by chemiluminescence), calcium, creatine, phosphorus (by colorimetry), in all the children; and serum levels of bone alkaline phosphatase (by immunoenzymeassay), osteocalcin, 17-ß-estradiol (by RIA) and parathyroid hormone (by chemiluminescence) in all the children. The same variables were evaluated in 20 normal children of the same age and sex (controls) who were not fed any soy. No abnormalities were seen in the evaluated variables. None of the enrolled children presented signs or symptoms of precocious puberty; weight and height were within the normal range; and bone age was in the normal range. Serum levels of bone alkaline phosphatase, osteocalcin, 17-ß-estradiol and parathyroid hormone and the levels of the urinary markers of bone metabolism (deoxypyridinoline, calcium, creatine and phosphorus) were in the normal range. Of the 20 children aged 3 y, 8 (40%) had mineralometry values that were in the range for sex and age. The preliminary data of the present study suggest that prolonged SF feeding does not induce estrogenic effects for the evaluated variables. To confirm these preliminary data, further studies, which are in progress in our department, on a larger cohort of children fed SF for a prolonged period are required.

Effects of Dietary Genistein Exposure during Development on CD (Sprague-Dawley) Rats. K. Barry Delclos, Thomas J. Bucci,* Larry G. Lomax,* John R. Latendresse,* Alan Warbritton,* Constance C. Weis and Retha R. Newbold. National Center for Toxicological Research, Jefferson, AR; *Pathology Associates International, Jefferson, AR; and National Institute of Environmental Health Sciences, Research Triangle Park, NC.

Many of the reported beneficial health effects of soy products have been suggested to be due, at least in part, to the soy isoflavones. Genistein is a major soy isoflavone that interacts with the estrogen receptor and multiple other targets; it has been investigated for potential benefit in preventing cancers at multiple sites and alleviating or preventing adverse effects associated with menopause. A study to find dose range was conducted as a prelude to an on-going multigeneration bioassay to assess toxicities potentially associated with genistein consumption. Genistein was administered in a soy- and alfalfa-free diet at 0, 5, 25, 100, 250, 625 and 1250 ppm to dams from gestation d 7. Litters were standardized to 4 pups per sex on postnatal d 2. Exposure of pups continued through lactation and after weaning until postnatal d 50. Body weight and food consumption of the dams showed a decreasing trend with a significant reduction at the highest dose. Similarly, pups of both sexes in the high dose group showed significantly decreased body weights. Absolute organ weights, relative organ weights or both showed relatively minor dose-related changes in terms of main dose effects, trends or difference between specific treatment groups and controls. In males, affected organs were liver, thymus, pituitary, preputial gland and ventral prostate. In females, affected organs were thymus, pituitary, uterus and vagina. Bone measures in both sexes showed treatment-related effects. Histopathologic examination of female pups revealed ductal and alveolar hyperplasia and hypertrophy of the mammary glands at 625 and 1250 ppm; similar effects were observed in males at 25 ppm. Estrous cycle morphologic abnormalities and abnormal ovarian antral follicles were observed at 1250 ppm. In males, degeneration of the seminiferous tubules was observed at 1250 ppm. There was an apparent deficit of sperm in epididymal sections at 625 and 1250 ppm, although testicular spermatid head counts and epididymal spermatozoa counts did not show significant differences from controls at these doses. Dietary genistein thus produced potentially adverse effects in multiple estrogen-sensitive tissues in males and females that are generally consistent with its estrogenic activity and occur within exposure ranges, as determined by serum levels, achievable in humans.

Inactivation of Thyroid Peroxidase by Genistein and Daidzein In Vitro and In Vivo: Mechanism for Antithyroid Activity of Soy. Daniel R. Doerge, Hebron C. Chang and Mona I. Churchwell. National Center for Toxicological Research, Jefferson, AR.

The association between soybean consumption and goiter in animals and humans has a long history. Current evidence for the beneficial effects of soy requires a full understanding of potential adverse effects as well. We previously identified the isoflavones, genistein and daidzein, as the only antithyroid constituents present in soy (Divi et al. 1997 ). Genistein and daidzein caused time- and H₂O₂-dependent irreversible inactivation of bovine lactoperoxidase (LPO) and porcine thyroid peroxidase (TPO), the enzyme that synthesizes thyroid hormones. The inactivation kinetics were consistent with a suicide mechanism; the apparent dissociation constants and partition ratios for genistein and daidzein were 0.2 and 0.5 µmol/L and 1 and 3, respectively. Radiolabeled genistein became covalently bonded to LPO concomitant to loss of enzyme activity (~3 mol genistein/mol enzyme inactivated). Minimal effects were seen on the prosthetic heme content. These data are consistent with potent mechanism-based inactivation of LPO and TPO in which radical products derived from oxidation of genistein and daidzein inactivate the peroxidases by covalent binding to critical amino acid residues. Antithyroid effects were also investigated in Sprague-Dawley rats fed a diet supplemented with genistein aglycone (0–500 µg/g dry food) through postnatal d 140 after exposure to genistein in utero and in maternal milk. Blood and thyroid glands from male and female rats were snap-frozen upon collection. Both aglycone and conjugated forms of genistein were quantified in blood and thyroid glands by using isotope dilution liquid chromatography–electrospray mass spectrometry. A dose-dependent increase was observed in both compartments. In blood, the aglycone content was 1–2% of total genistein and in thyroid it was 18–28%. Furthermore, TPO activity was depressed in a dose-dependent manner in the thyroids of both male and female rats. These data show that genistein is present in the thyroid at concentrations equivalent to those causing enzyme inactivation in vitro and that this is sufficient to inactivate rat TPO in vivo. The consequences of TPO inactivation await confirmation by histopathologic evaluation, but the results are consistent with the reported goitrogenic effects of soy and underscore the potential for thyroid toxicity from isoflavones at the levels actually observed in human blood.

	REFERENCES

TOP Isoflavone quantification Isoflavone absorption and... Osteoporosis Cancer Heart disease and lipids Other topics Isoflavone intake Menopause symptom relief Safety issues REFERENCES

 

1. Day A. J., DuPont M. S., Ridley S, Rhodes M., Rhodes M.J.C., Morgan M.R.A., Willamson G. Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and liver ß-glucosidase activity. FEBS Lett. 1998;436:71-75[Medline]

2. Ingram D., Sanders K., Kolybaba M., Lopez D. Case-control study of phyto-estrogens and breast cancer. Lancet 1997;350:990-994[Medline]

3. Folsom A. R., Kaye S. A., Potter J. D., Prineas R. J Association of incident carcinoma of the endometrium with body weight and fat distribution in older women: early findings of the Iowa Women’s Health Study. Cancer Res 1989;49:6828-6831[Abstract/Free Full Text]

4. Foth D., Cline M. J. Effects of mammalian and plant estrogens on mammary glands and uteri of macaques. Am. J. Clin. Nutr. 1998;68(suppl.):1413S-1417S[Abstract]

5. Goodman M. T., Wilkins L. R., Hankin J. H., Lyu L., Wu A. H. Association of soy and fiber consumption with the risk of endometrial cancer. Am. J. Epidemiol. 1997;146:294-306[Abstract/Free Full Text]

6. Hodgson J. M., Puddey I. B., Beilin L. J., et al Effects of isoflavonoids on blood pressure in subjects with high-normal ambulatory blood pressure levels—a randomized controlled trial. Am. J. Hypertens. 1999;12:47-53[Medline]

7. King R. A., Smith R. M., Krishnan R., et al Effects of enalapril and hydralazine treatment and withdrawal upon cardiovascular hypertrophy in stroke-prone spontaneously hypertensive rats. J. Hypertens. 1992;10:919-928[Medline]

8. Nestel P. J., Pomeroy S., Kay S., et al Isoflavones from red clover improve systemic arterial compliance but not plasma lipids in menopausal women. J. Clin. Endocrinol. Metab. 1999;84:895-898[Abstract/Free Full Text]

9. Nestel P. J., Yamashita T., Sasahara T., et al Soy isoflavones improve systemic arterial compliance but not plasma lipids in menopausal and perimenopausal women. Aterioscler. Thromb. Vasc. Biol. 1997;17:3392-3398[Abstract/Free Full Text]

10. Toma C., Jensen P. E., Prieto , et al Effects of tyrosine kinase inhibitors on the contractility of rat mesenteric resistance arteries. Br. J. Pharmacol. 1995;114:1266-1272[Medline]

11. Lovati M. R., Manzoni C., Gianazza E., Sirtori C. R. Soybean protein products as regulators of liver low-density lipoprotein receptors. I. Identification of active ß-conglycinin subunits. J. Agric. Food Chem. 1998;46:2474-2480

12. Manzoni C., Lovati M. R., Gianazza E., Morita Y., Sirtori C. R. Soybean protein products as regulators of liver low-density lipoprotein receptors. II. -’ rich commercial soy concentrate and ’ deficient mutant differently affect low-density lipoprotein receptor activation. J. Agric. Food Chem. 1998;46:2481-2484

13. Sirtori C. R., Lovati M. R., Manzoni C., Gianazza E., Della Mura N., Staels G., Auqerx J. Reduction of serum cholesterol by soy proteins: clinical experience and potential molecular mechanism. Nutr. Metab. Card. Dis. 1998;8:334-338

14. King R. A., Broadbent J. L., Head R. J. Absorption and excretion of the soy isoflavone genistein in rats. J. Nutr. 1996;126:176-182

15. Setchell K.D.R., Zimmer-Nechemias L., Cai J., Heubi J. E. Exposure of infants to phyto-oestrogens from soy-based infant formula. Lancet 1997;350:23-27[Medline]

16. Wang H.-J., Murphy P. A. Isoflavone content in commercial soybean foods. J. Agric. Food Chem. 1994;42:1666-1673

17. Divi R.L., Chang H. C., Doerge D. R. Anti-thyroid isoflavones from soybean: isolation, characterization, and mechanisms of action. Biochem. Pharmacol. 1997;54:1087-1096[Medline]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content PubMed PubMed Citation