Phytoestrogens are plant-derived compounds that demonstrate estrogenic responses that are very much weaker than the estrogenicity of 17β-estradiol. However, because some foods, including soy and soy products, contain comparatively large amounts of phytoestrogens, plasma concentrations may exceed the concentrations of endogenous estrogens by several orders of magnitude.1 Phytoestrogens are capable, therefore, of causing various effects in humans and animals, including potential cardiovascular benefits, prevention of hormone-dependent cancers and osteoporosis, and relief of menopausal complaints.2 Furthermore, these compounds have been shown to exhibit antioxidant and metal-chelating activities, as well as anticarcinogenic, immunomodulating, and antiviral effects.3
The activities associated with phytoestrogens vary according to the nature of the phytoestrogen studied and the types of cell line, tissue, and species, while, also, the responses being evaluated are variable. Individual phytoestrogens have different mechanisms of action based on the presence of estrogen receptor subtypes, endogenous estrogen concentrations, and cellular genetic make-up.4 Although numerous articles have been published on the health benefits of phytoestrogens, little is known about their retention and distribution within the human body. This study was carried out with the main aim of quantifying soy-derived phytoestrogens in human breast tissue after exposure and comparing these with concentrations in serum and urine of the same subjects. Several methods exist for the analysis of soy-derived phytoestrogens in various biological matrices, such as serum, urine, breast milk, and prostatic fluid.5,6 We recently developed a new and reliable method for quantification of genistein, daidzein, and equol in human breast tissue homogenate.7 The method was applied to study the tissue distribution of these compounds after ingestion, and the results are presented here. Upon analyzing human breast tissue samples from the recruited subjects, significant compositional differences were evident, some resulting from variations in age, stage of the menstrual cycle, and use of exogenous hormones. Therefore, markers for epithelial content, cellularity, and blood vessel and total fat content were applied.
MATERIALS AND METHODS
A randomized, double-blind, placebo-controlled study was undertaken to compare the concentrations of genistein, daidzein, and equol in subjects ingesting either a soy-based food supplement or a placebo tablet for 5 consecutive days before aesthetic breast surgery. The soy-based food supplements (EB 605; Eko-Bio, Eede, The Netherlands) contained 100 mg of genistein/genistin, 37 mg of daidzein/daidzin, and 15 mg of glycitein/glycitin (more than 90% as glycosides). Because glycitein has negligible or no estrogenic activity, this compound was not investigated further. Subject information, such as age, number of children, breastfeeding, exogenous hormone use, inclusion of soy in the diet, and intake of vegetables (high: 3, medium: 2, low: 1), was collected upon recruitment. The study was given ethical approval by the Ethics Committee of the Ghent University Hospital, and a written consent form was obtained from each subject.
The randomization was carried out by distributing random codes (based on an online randomizator at http://www.random.org/sform.html) to 50 plastic boxes containing 5 dosages of either a soy-based food supplement or placebo. After randomization, the codes were arranged in a list according to increasing numbers. The boxes were then distributed chronologically to the recruited patients, who thus received either a soy-based food supplement or placebo, completely randomized.
The study was double-blind; neither the surgeon nor the subjects or the experimentalists knew the codes of randomization. Only the professors who initiated the study (H.T.D. and M.E.B.) knew the codes. These were only revealed after full presentation of all results.
Subjects were recruited upon contacting the Department of Plastic and Reconstructive Surgery at the Ghent University Hospital for the purpose of aesthetic breast reduction during the period of 28 months starting on April 18, 2000. No exclusion criteria were used. A total of 34 adults with a mean age of 36.5 years (range 20–70 years) were included in the study. The recruitment process involved careful explanation of the study aims and 94.4% (34 of 36) of the contacted subjects agreed to participate. The subjects received either the soy-based food supplement or placebo, based on the randomized list.
During surgery, samples of breast tissue, blood, and urine were collected. The breast tissue and urine samples were placed in sterile containers, while blood was drawn into sterile Venoject II tubes (4 mL; Autosep, Terumo Europe, Leuven, Belgium). Whole blood was centrifuged with an Ecco-Praxa 2 (Vel, Leuven, Belgium) at 3,000g for 10 minutes upon collection, and serum was frozen at −20°C before analysis. Urine samples were also frozen at −20°C. Breast tissue samples were either frozen at −80°C or immediately subjected to histological evaluation.
The procedure for pretreatment of the breast tissue, serum, and urine samples was described in detail by Maubach et al.7 Briefly, breast tissue was homogenized in Tris buffer (0.222 g/mL), lipase and protease were added, and the sample was mixed and incubated for 1 hour at 37°C. Then, acetate buffer (pH 5.0, 5 mL) and β-glucuronidase/arylsulfatase (60 μL) were added, followed by incubation for 3 hours at 37°C and addition of a methanolic solution of the internal standard and hexane, respectively. The methanol-buffer layer was diluted with water and applied to a preconditioned solid-phase (C18-silica) extraction column. The compounds of interest were eluted with methanol and, after the usual work-up, analyzed. Urine and serum were added to acetate buffer (pH 5.0, 7 mL) and β-glucuronidase/arylsulfatase (30 μL), and the samples were mixed and incubated for 3 hours at 37°C, followed by addition of the internal standard in ethyl acetate and work-up.
Urine samples were analyzed by high-performance liquid chromatography (HPLC) using a Waters 2690 Separations HPLC module (Waters Corporation, Brussels, Belgium) equipped with a 996 photodiode array detector, as described by Maubach et al.7 Chromatographic processing was done using the Waters Millennium 3.2 software (Waters Corporation). The reversed-phase silica column was an XTerra MS C18 (5 μm; 4.6 mm × 250 mm; Waters Corporation). Analyses were carried out in isocratic conditions with an eluent mixture consisting of 40% acetonitrile/methanol, 20/80 vol/vol in 0.05% formic acid in water. The flow rate was 1.5 mL/min, and the column temperature was 40°C. The run time was set to 20 minutes, sufficient for all compounds of interest to be eluted. The injection volume was 30 μL, and all samples were analyzed in triplicate. Ultraviolet detection was done at a wavelength corresponding to the absorption maximum (highest intensity) of each analyte: 261 nm for genistein, 249 nm for daidzein, and 230 nm for equol. The internal standard (4-hydroxybenzophenone) was detected simultaneously with the analytes and at the same wavelength as described above for each compound. The concentrations of analytes were established by dividing the peak area of each analyte by the peak area of the internal standard and multiplying by the concentration of the internal standard used.
Samples of breast tissue homogenate and of serum were analyzed by HPLC–mass spectroscopy using an Agilent 1100 Series HPLC system (Agilent Technologies, Palo Alto, CA) coupled to an Agilent 1100 Series MSD (UV) in the negative ionization mode. The reversed-phase silica column was a Phenomenex Luna C182 (5 μm; 4.6 mm × 250 mm) (Bester, Amsterdam, The Netherlands), the injection volume was 25 μL, and the eluent mixture was composed of 5 mmol/L ammonium acetate (solvent A) and acetonitrile (solvent B). Gradient profile was as follows: 0–5 minutes 35% B in A; 6–27 minutes 26% B in A; 28–35 minutes 35% B in A; 36–41 minutes 100% B in A; and 42–45 minutes 35% B in A. Ionization was effected by atmospheric pressure chemical ionization, and ions were monitored at m/z 253 (0–4 minutes; gain 5), m/z 269 (4–9.5 minutes; gain 10), m/z 241 (9.5–16 minutes; gain 50), and m/z 198 (16–28 minutes; gain 2). The HPLC–mass spectroscopy settings were as follows: drying gas flow 13 L/min; nebulizer pressure 50 psi; drying gas temperature 350°C; capillary voltage 3500 V; fragmentor setting 70 eV; and single ion monitoring resolution, low.
The quality of the breast tissue homogenate samples was evaluated by assessing markers for epithelial content (episialin), total cellularity (lactate dehydrogenase, LDH), and blood vessels (albumin). The episialin antigen concentrations in breast and abdominal tissue samples were determined using a commercially available 2-step sequential chemiluminescent immunometric assay (Immulite 2000; Diagnostic Products Corporation, Los Angeles, CA). The LDH concentrations were determined by an enzymatic assay (COBAS INTEGRA 700; Roche Diagnostics, Basel, Switzerland). The albumin concentrations were measured by an immunoturbidimetric assay (COBAS INTEGRA 700; Roche Diagnostics).
The total fat content in the breast tissue homogenate samples was measured by a modified version of the original Rose-Gottlieb method.8 Breast tissue homogenate (0.5 g) was weighed directly into 50-mL tubes in duplicate, and hot water (6 mL) was added, followed by vortex-mixing for 1 minute. After cooling, aqueous ammonia (25%, 1 mL) was added, the solution was vortex-mixed for 1 minute, and methanol (7.5 mL) was added, followed by vortex-mixing for 1 minute. Addition of diethyl ether (17 mL) and petroleum ether (17 mL), phase separation, and withdrawal of an aliquot from the upper phase (25 mL) into a flask were followed by drying (0.5 hours in a drying cabinet, 0.5 hours in a desiccator). The flasks were weighed until constant weight and the percentage of fat were calculated. All solvents were purchased from Biosolve (Valkenswaard, The Netherlands).
The concentrations of genistein, daidzein, and equol in breast tissue homogenate, serum, and urine in the 2 treatment groups were investigated using the nonparametric Mann-Whitney U test (1-tailed). Correlations were carried out using the Spearman ρ correlation coefficient. The tests were performed using SPSS 11 (SPSS Inc, Chicago, IL).
The active group consisted of 15 subjects who ingested a soy-based food supplement, while 19 subjects ingested a placebo tablet, all during 5 consecutive evenings preceding aesthetic breast surgery. In the final data analysis, only 30 subjects were included, because 4 persons within the active group were excluded because of a lack of detectable concentrations of soy-derived phytoestrogens in their urine or postponed operations. Two subjects received a triple dose of the soy-based food supplement for 5 days before surgery. Consequently, the 9 subjects in the active group, receiving a single dose, had a median age of 39 (range 22–49) years, whereas the 19 subjects in the placebo group had a median age of 34 (range 20–70) years. Reproductive history revealed that all subjects in the active group had children, whereas only 47.4% of the placebo subjects had children, and breastfeeding was, therefore, more frequent in the active group. The use of exogenous hormones was extensive in both groups (44.4% and 31.6% in active and placebo subjects, respectively). Soy intake was similar within the 2 treatment groups, with 22.2% of the subjects in the active group and 15.8% of the subjects in the placebo group stating that they included soy in their overall diets. Furthermore, no statistically significant differences between the 2 groups were found for their reported vegetable intake. No exclusion criteria were used in the recruitment, although the subjects were undergoing elective surgery and not taking antibiotics at the time of surgery.
Mass detection was used to quantify the concentrations of genistein, daidzein, and equol in human breast tissue homogenate and serum having system limits of quantitation of 12.5, 6.25, and 12.5 nmol/L for genistein, daidzein, and equol, respectively. These limits corresponded to absolute limits of quantitation of 0.19, 0.10, and 0.19 nmol/L (relative standard deviation within 20%) in breast tissue and 3.75, 1.88, and 3.75 nmol/L in serum for genistein, daidzein, and equol, respectively (concentration factors of 66.67 for breast tissue and 3.33 for serum). Mass detection offered enhanced sensitivity factors of 4.5, 7.9, and 23.7 for genistein, daidzein, and equol, respectively, when compared with ultraviolet detection. Satisfactory linearity was noted for each phytoestrogen (R 2 > 0.99), and adequate separation was achieved.
The median, minimum, and maximum concentrations of genistein, daidzein, and equol quantified in breast tissue homogenate, serum, and urine are shown in Table 1. Additionally, individual concentrations in the respective matrices were compared between the 2 groups. Although concentrations of all soy-derived phytoestrogens were significantly higher in urine of the active subjects than in that of the placebo subjects (P ≤ .05), the concentration of only genistein was significantly higher in serum of subjects belonging to the active group (P < .05). No significant differences in the concentrations of the analytes were evident for breast tissue homogenate. To investigate whether a triple dose of the soy-based food supplement would result in increased concentrations of genistein, daidzein, and equol in breast tissue homogenate, 2 subjects completed a 5-day course before surgery. No significant increase in soy-derived phytoestrogen concentrations was found in either breast tissue homogenate, serum, or urine (data not shown).
As Table 2 shows, equol concentrations in urine correlated with daidzein concentrations in breast tissue homogenate (P < .05), whereas genistein concentrations in serum were found to correlate with equol concentration in breast tissue homogenate (P < .05). Urine and serum concentrations correlated significantly for daidzein (P < .01) and genistein (P < .01), which was also the case for daidzein concentrations in urine with equol concentrations in urine (P < .01) and genistein concentrations in serum (P < .01) and urine (P < .01). Daidzein concentrations in serum correlated significantly with equol concentrations in serum (P < .01) and genistein concentrations in serum (P < .01) and urine (P < .01). There was also a correlation between the equol and genistein concentrations in serum (P < .05).
In an attempt to control for the heterogeneity of the breast tissue samples, markers for epithelial content (episialin), cellularity (LDH), blood vessel content (albumin), and total fat were assayed. Both groups of subjects showed great variations for each parameter, and no significant differences were observed. Furthermore, because it is not known whether phytoestrogens would localize in the epithelial cells due to the estrogen receptor content, in the adipose tissue due to its hydrophobic nature, or in the blood vessels supplying the tissue, it was of interest to detect correlations of the phytoestrogens quantified in breast tissue homogenate with the values determined for the markers. Episialin correlated significantly with daidzein in breast tissue homogenate samples (P < .05, Spearman ρ correlation coefficient of −0.580), but no other statistical correlations were found (data not shown).
Urine concentrations of genistein, daidzein, and equol served to control effective intake of soy tablets.9,10 Thus, 4 subjects were excluded from the analysis of the study, because the compounds were not detected in urine, indicating noncompliance.11 The concentrations of the analytes were all in the low nanomolar range in breast tissue homogenate samples, and no significant differences between the 2 treatment groups were apparent. The concentrations of daidzein and equol were highest, whereas those of genistein were very low or even undetectable. The concentrations in the breast tissue homogenates were generally about 100-fold lower than those in the corresponding serum samples, suggesting rapid clearance, metabolic breakdown at the site, or exclusion of these compounds from the tissue compartment. Serum concentrations of all analytes were higher in the soy-supplemented group than in the placebo group, whereas values were generally much lower than expected. Thus, for a daily ingested dose of 146 μmol daidzein, concentrations from 0 to 0.74 μmol/L serum were traced in the active group. Because the tablets were ingested every evening, the time of sampling before or during surgery would have exceeded 12 hours after the intake. Although the half-life times of the compounds in blood and urine have been reported to be 7–8 hours for genistein and daidzein12 and 8–9 hours for equol,13 their tissue distribution and their fate at tissue sites are unknown. Great interindividual variability was apparent in all matrices, and we have previously demonstrated higher soy-derived phytoestrogen concentrations in a subject ingesting the same soy-based food supplement.7 Wide ranges of isoflavone levels in response to soy intake have been noted by other investigators, and it was suggested that the composition of the microflora in the gut, required for hydrolysis of isoflavone glycosides and for isoflavone metabolism, leads to this variability among individuals.14 An investigation of several time-points of phytoestrogen administration before tissue sampling could help in understanding this matter. It is, furthermore, feasible that tissue concentrations of the compounds show marked qualitative and quantitative differences depending on the type of soy-based food or supplement ingested.15
The urine concentrations found in this study (between undetected and 133 μmol/L) showed pronounced interindividual variability, but they were in agreement with values reported for various soy isoflavone intakes.12,16 Interestingly, the ingestion of a triple dose of the soy-based food supplements in 2 subjects did not result in significantly increased concentrations of these compounds in breast tissue homogenate, serum, or urine. Setchell et al12 found no advantage in consuming high amounts of isoflavones in single doses, because the absorption appeared to be rate-limiting and saturable. Furthermore, it was suggested that, on the basis of elimination half-life times of daidzein and genistein, steady-state plasma concentrations would be more readily maintained by repeated ingestion of isoflavones throughout the day than by ingestion of just a single daily dose. It is also difficult to establish whether ingestion of a soy extract for 5 days is sufficient for reaching a steady-state level, but the low serum concentrations observed in this study would indicate that it is not.
The relatively high levels of equol in all matrices require critical comments. Setchell et al13 have distinguished equol and nonequol producers as persons having plasma equol concentrations of above 83 nmol/L and below 40 nmol/L, respectively. This distinction was also made in urine, and equol excreters would have urine levels higher than 1 μmol/L. In the present study, approximately 28% of the subjects had serum concentrations above 83 nmol/L, and nearly 70% of the subjects had urine concentrations above 1 μmol/L. The proportions of subjects able to produce equol according to the serum concentrations are close to the 30–40% reported previously.9,14 The discrepancy between serum and urine may perhaps be due to the low concentrations observed in serum, when compared with those in urine.
The concentrations of genistein, daidzein, and equol in both the breast tissue homogenate and urine of the placebo subjects suggest some dietary background, and similar observations have been made by others.17,18 The increased use of soy products as fillers in certain foods may have contributed to this effect.9,10 Furthermore, daidzein and equol have been demonstrated to be present in cow’s milk and certain vegetables, albeit in low concentrations, and equol excretion has been associated with meat intake.19
The phytoestrogen concentrations in breast tissue homogenate are well below those that have been reported to exert growth-suppressing activities on malignant breast cells. Furthermore, the concentrations of daidzein and equol (10–15 nmol/L) did not increase substantially after a triple-dose ingestion over 5 days, and even the highest concentrations observed (approximately 55 nmol/L) would not be sufficient to elicit biological effects. An increase in phytoestrogen concentrations in breast tissue with increasing intake of a soy-based food supplement, although not found in this study, could have important implications with respect to reaching optimal concentrations at a specific tissue site.
Significant correlations were detected between serum and urine concentrations of daidzein and genistein. Moreover, several correlations were observed between genistein, daidzein, and equol in serum and urine. Adlercreutz et al20 also demonstrated correlations between plasma and urine concentrations of isoflavones. However, in the present study, correlations between breast tissue homogenate concentrations of soy-derived phytoestrogens on the one hand and serum and urine parameters on the other hand were of no or low significance. From determinations of lactate dehydrogenase, episialin, albumin, and total fat contents in breast tissue homogenate, only the correlation between episialin and daidzein concentrations was significant. Because episialin was used as a marker for the epithelial content of the tissue, it could be derived that daidzein mainly localizes in this part of the tissue. However, one would then expect similar behavior for genistein.
To summarize, mass detection was applied for the quantitation of soy-derived phytoestrogens (genistein, daidzein, and equol) in breast tissue homogenate and serum samples to increase the sensitivity of detection, whereas ultraviolet detection was used for urine samples. Comparison of the 3 matrices was made to elucidate the tissue distribution of the compounds. Concentrations were in the low nanomolar range in breast tissue homogenate, but they were generally a hundred-fold higher in the corresponding serum and urine samples. Daidzein and equol predominated over genistein in the breast tissue samples, whereas relatively high background concentrations of these were also found in placebo subjects. Serum concentrations may not be accurate determinants of tissue exposure, and more studies should be undertaken to examine the tissue distribution of phytoestrogens upon ingestion. Furthermore, insight into the localization of these compounds in the various tissues could help our understanding of tissue distribution.
1. Cassidy A, Faughnan M. Phytoestrogens through the life cycle. Proc Nutr Soc 2000;59:489–96.
2. Wuttke W, Jarry H, Westphalen S, Christoffel V, Seitlova-Wuttke D. Phytoestrogens for hormone replacement therapy? J Steroid Biochem Mol Biol 2002;83:133–47.
3. Ren M, Kuhn G, Wegner J, Chen J. Isoflavones, substances with multi-biological and clinical properties. Eur J Nutr 2001;40:135–46.
4. Haddad N, Fuqua J. Phytoestrogens: effects on the reproductive system. Endocrinologist 2001;11:498–505.
5. Franke AA, Custer LJ, Tanaka Y. Isoflavones in human breast milk and other biological fluids. Am J Clin Nutr 1998;68 suppl:1466S–73S.
6. Morton MS, Chan PS, Cheng C, Blacklock N, Matos-Ferreira A, Abranches-Monteiro L, et al. Lignans and isoflavonoids in plasma and prostatic fluid in men: samples from Portugal, Hong Kong, and the United Kingdom. Prostate 1997;32:122–8.
7. Maubach J, Bracke ME, Heyerick A, Depypere HT, Serreyn RF, Mareel MM, et al. Quantitation of soy-derived phytoestrogens in human breast tissue and biological fluids by high-performance liquid chromatography. J Chromatogr B Analyt Technol Biomed Life Sci 2003;784:137–44.
8. Evers JM, Wightman LM, Crawford RA, Contarini G, Coors U, Farrington DS, et al. A precise method to measure the total fat content of spreadable fats. Int Dairy J 2000;10:815–27.
9. Lampe JW. Isoflavonoid and lignan phytoestrogens as dietary biomarkers. J Nutr 2003;133 suppl:956S–64S.
10. Hargreaves DF, Potten CS, Harding C, Shaw LE, Morton MS, Roberts SA, et al. Two-week dietary soy supplementation has an estrogenic effect on normal premenopausal breast. J Clin Endocrinol Metab 1999;84:4017–24.
11. Setchell KD, Zimmer-Nechemias L, Cai J, Heubi JE. Isoflavone content of infant formulas and the metabolic fate of these phytoestrogens in early life. Am J Clin Nutr 1998;68(suppl):1453S–61S.
12. Setchell K, Faughnan M, Avades T, Zimmer-Neschemias L, Brown N, Wolfe B, et al. Comparing the pharmacokinetics of daidzein and genistein with the use of 13C-labeled tracers in premenopausal women. Am J Clin Nutr 2003;77:411–9.
13. Setchell KD, Brown NM, Zimmer-Nechemias L, Brashear WT, Wolfe BE, Kirschner AS, et al. Evidence for lack of absorption of soy isoflavone glycosides in humans, supporting the crucial role of intestinal metabolism for bioavailability. Am J Clin Nutr 2002;76:447–53.
14. Xu X, Harris KS, Wang HJ, Murphy PA, Hendrich S. Bioavailability of soybean isoflavones depends upon gut microflora in women. J Nutr 1995;125:2307–15.
15. Setchell KD, Brown NM, Desai P, Zimmer-Nechemias L, Wolfe BE, Brashear WT, et al. Bioavailability of pure isoflavones in healthy humans and analysis of commercial soy isoflavone supplements. J Nutr 2001;131 suppl:1362S–75S.
16. Busby MG, Jeffcoat A, Bloedon LT, Koch MA, Black T, Dix KJ, et al. Clinical characteristics and pharmacokinetics of purified soy isoflavones: single-dose administration to healthy men. Am J Clin Nutr 2002;75:126–36.
17. Morton M, Arisaka O, Miyake N, Morgan L, Evans B. Phytoestrogen concentrations in serum from Japanese men and women over forty years of age. J Nutr 2002;132:3168–71.
18. Pumford S, Morton M, Turkes A, Griffiths K. Determination of the isoflavonoids genistein and daidzein in biological samples by gas chromatography–mass spectrometry. Ann Clin Biochem 2002;39:281–92.
19. Lampe JW, Gustafsson DR, Hutchins AM, Martini MC, Li S, Wähälä K, et al. Urinary isoflavonoid and lignan excretion on a Western diet: relation to soy, vegetable, and fruit intake. Cancer Epidemiol Biomarkers Prev 1999;8:699–707.
20. Adlercreutz H, Fotsis T, Lampe J, Wähälä K, Makela T, Brunow G, et al. Quantitative determination of lignans and isoflavonoids in plasma of omnivorous and vegetarian women by isotope dilution gas chromatography-mass spectrometry. Scand J Clin Lab Invest Suppl 1993;215:5–18.