Phytoestrogens: The Pediatric Perspective

Phytoestrogens are plant-derived compounds that possess estrogenic activity in animals and therefore have a potential beneficial role in estrogen deficient conditions. The most extensively studied phytoestrogens are isoflavones, which are abundantly present in soy as genistein and daidzein. The structural similarity of isoflavones to the mammalian estrogen is probably the basis for their hormonal effects. A common structural characteristic of many of these compounds is the presence of a phenolic ring, a prerequisite for the binding to estrogen receptors (Fig. 1). Competitive binding of unlabeled competitors and displacement of bound-labeled (H3) estradiol have been used to verify the ability of isoflavonoids to bind to estrogen receptors (1). Although the estrogenic potency of phytoestrogens is significant, especially for the recently discovered estrogen receptor β(2), these compounds maintain both receptor agonist and antagonist activity; hence, their effect may be appropriately described as estrogen-receptor modulation (3).FIG. 1.: Chemical structure of soy isoflavones (A) and 17β-estradiol (B). Daidzein R1=R2=R3=R4=OH; genistein R1=R3=H, R2=R4=OH; glycitein R1=OCH3, R2=H, R4=OH; biochanin A R1=R3=H, R2=OH, R4=OCH3; formonentin R1=R2=R3=H, R4=OH.β.The role of estrogens in certain diseases, such as cardiovascular disease, breast cancer, and osteoporosis, has prompted extensive research on the role of isoflavones in these conditions, especially among postmenopausal women. Children are exposed to isoflavones products from infancy to adulthood in a large variety of food. However, the main source of isoflavones in the pediatric age group is from soy-based infant formulas, which account for 7% of infant formula sales in the United Kingdom, 13% in New Zealand and 10% to 20% in the United States (3). Although isoflavones are therefore abundantly common in the diet of soy-fed infants, data on the estrogenic effects of these formulas among infants and children are very limited. EXPOSURE TO PHYTOESTROGENS FROM SOY-BASED INFANT FORMULAS As determined by high-performance liquid chromatography, the isoflavone content is quantitatively and qualitatively similar in five different soy-based infant formulas common in the United States (4). The principal isoflavones in soy formulas are daidzein and genistein, either as the aglycone (unconjugated form) or as different types of glycoside conjugates. Conjugates of genistein predominate in soy-based formulas and account for more than 65% of the total amount of isoflavones (4). Total isoflavone concentrations in commercially manufactured formulas ranges from 32 to 47 mg/L, four orders of magnitude higher than the isoflavone concentrations in samples of human breast milk (5.6 ± 4.4 μg/L) (4). Overall, these concentrations are similar to values first reported in 1987 (5), indicating the relative consistency of the isoflavone content of soy-based formulas over time. Somewhat lower levels of isoflavones have been detected in soy-based formulas manufactured in Australia (17.2–21.9 mg/L). By comparison, casein-based infant formulas contained negligible levels of isoflavones, in the range of 0.001 mg/L to 0.03 mg/L (6). In a study from New Zealand, where high-performance liquid chromatography detection limit was 0.05 mg/L, no phytoestrogens were detected in dairy-based infant formulas or in human breast milk, irrespective of the mother's diet (7). Based on typical consumption of soy-based infant formula and average body weight, it has been calculated that the daily isoflavone intake may increase from 22.5 mg to 24.8 mg during the first week of life to 41.0 mg to 45.0 mg at 4 months of age (4). These levels correspond to a daily isoflavone intake of 5.7 mg/kg to 11.9 mg/kg body weight in an American study (4), compared with 3.2 mg/kg body weight in studies from New Zealand (7,8). In these latter studies, it was found that supplementing the diet of a 4-month-old infant with a single daily serving of cereal can increase the isoflavone intake by more than 25%(7). When adjusted for body weight, these studies show that infants exclusively fed soy-based formulas are exposed to a daily intake of isoflavones that is four-to 13-fold higher than the 0.7 mg/kg intake shown to exert significant physiologic effects on the hormonal regulation of women's menstrual cycles (9). BIOAVAILABILITY OF ISOFLAVONES IN INFANTS There is much controversy as to whether natural flavonoid glycosides can be absorbed by the gastrointestinal tract or whether and how they are hydrolyzed in the small intestine before absorption (10). It has been generally agreed that isoflavone–glycosides cannot be absorbed from the small intestine and that cleavage of the glycosidic bond occurs when the compound reaches the established flora of the large intestine. The bacterial cleavage of the central heterocyclic ring may enhance the estrogenic activity of natural flavonoid glycosides via the formation of an active estrogenic compound named equol (11). The absence of equol from the urine of infants fed soy-based formulas (12) highlights the essential role that intestinal bacteria play in its formation. The lack of these bacteria in infants, at least during the first few months of life, may have a major impact on the estrogenic effects of the flavonoid glycosides. Recent studies in rats have suggested an alternative absorption mechanism. It has been postulated that flavonoids, either as glucosides or aglycones are glucuronidated and actively transferred across the jejunum and ileum and that this glucuronidation occurs without the need for gut flora. This mechanism would depend on the presence of glycosidase activity and uridine 5-diphosphate-glucoronyl transferase in the jejunum (10). Indeed, the high levels of intact isoflavones in serum and urine of soy-fed infants support this latter mechanism of absorption. In infants between 2 to 16 weeks of age who received soy-based formula, the mean percentage of the daily intake recovered in urine was 38 ± 4% for daidzein and 13 ± 3% for genistein. These results indicate that infants are able to efficiently digest, absorb, and excrete isoflavones from soy-based formulas (8) independent of the biotransformation of phytoestrogens by the gut microflora. This conclusion is further confirmed by measurements of isoflavones in plasma. Plasma total isoflavone concentrations ranged from 552 μg/L to 1,775 μg/L (mean, 980 μg/L) in infants fed soy-based formula (4). These levels are greater than values reported for adults (50–200 μg/L) consuming similar amounts of soy-derived isoflavones (5). By comparison, circulating concentrations of isoflavones in infants fed breast milk or cow milk formulas were less than 1 in 200 and 1 in 100, respectively, of the levels observed in soy-fed infants (4). Based on the studies of Setchell et al. (4,5), the total plasma isoflavone concentration in infants fed soy-based formula is 13,000 to 22,000 times higher than the plasma concentration of estradiol in early life, which is 40 pg/mL to 80 pg/mL. However, several in vivo studies have shown that the biologic effects of these high concentrations of isoflavones are mitigated by low bioactivity, which results from their low affinity to estrogen receptors. Formonoetin, an isoflavonoid phytoestrogen, is 40,000 weaker than 17β-estradiol in enhancing mammary gland proliferation in female mice (13), whereas genistein is only 1,000-fold weaker than estradiol in regard to proliferative effect on mammary glands in infant mice (14). These observations corroborate an in vitro study in which formononetin (13) and genistein (15) had relative binding affinities for the estrogen receptor that were 15,000-fold and 100-fold lower than that of 17β-estradiol, respectively. Formononetin is not found in soy-based formulas, but rather, in animal fodder and in certain human foodstuffs and is mentioned here as an example of the low estrogenic activity of phytoestrogens. Genistein, which maintains much higher potency and binding affinity than formononetin, is the main isoflavone in soy-based formulas. Therefore, even allowing for their weak estrogenic activity, it is difficult to believe that soy-derived isoflavones, circulating at these high concentrations, are biologically inert in infants, particularly given their weak binding to serum proteins (5). TABLE 1.TABLE 1: Isoflavone content, plasma concentration, and daily intake in infants fed human milk, cow milk–based and soy-based infant formulasSOY-BASED FORMULAS AND GROWTH Because sex steroids are known to promote skeletal maturation and growth in children, phytoestrogens in soy-based formulas may potentially induce similar effects in growing children. Several studies have focused on the contribution of soy-based formulas to growth, but none have specifically addressed the role of soy-derived phytoestrogens in this regard. In most studies, infants fed soy-based formulas have been compared to infants fed either human milk or cow milk–based formulas. In most of the studies, height (16–18), weight (16–20), and head circumference (16–18) attainments over a period of 4 to 12 months were similar between the groups. Conversely, some early studies have shown that the use of soy formulas may be associated with inadequate growth rates in term (21,22) and preterm (23) infants. These observations may possibly be attributed to the lower content of several growth-promoting elements in the soy formulas used in these studies, such as methionine (24), calcium, phosphorous, magnesium, and zinc (25). In only one study was the rate of somatic growth greater among infants fed soy-based formula (ProSobee; Mead Johnson, Evansville, IN) than in infants fed breast milk or cow milk–based formulas (26). In premature infants, there are conflicting data on the impact of soy-based formula on growth. One early study by Naude et al. (27) found reduced weight and length gain in premature infants weighing 1,500 g to 1,800 g, even with improved, methionine-supplemented formulas. In other studies of infants with birth weights less than 1,500 g, adequate (28) and reduced growth rates (29) both have been observed. It is the American Academy of Pediatrics position that soy-based formulas are not recommended for preterm infants weighing less than 1,800 grams (30). PHYTOESTROGENS AND SEXUAL DEVELOPMENT Phytoestrogens, mainly isoflavones, have been found to exert estrogenic effects on several target organs: the reproductive system, the mammary glands, the hypothalamus, and the pituitary. Most human studies have been conducted in mature women, and the implications of the data for infants and children therefore are limited. Furthermore, in some studies, isoflavones were added to experimental diets; hence, the effects observed may differ from those observed in response to a nonmodified soy diet. Isoflavones in a soy diet have been found to change the estrogenic milieu by decreasing ovarian-derived hormones (17β-estradiol and progesterone) (31) and by increasing sex hormone–binding globulin (SHBG) in women who ingested soy milk daily for 10 weeks (32). At the hypothalamic–pituitary–gonadal axis, isoflavones exert significant physiologic effects on the hormonal regulation of women's menstrual cycles (9). Finally, an isoflavone-enriched diet administered to premenopausal women for 14 days has been found to induce an increase of the proliferation rate of breast lobular epithelium (33). Some of these results have been challenged by studies that have failed to show any effect of isoflavones on the sex hormone profile (34) or on vaginal cytology and the endometrium (35) in postmenopausal women. Interestingly, in the studies by Baird et al. (34) and Duncan et al. (35), the daily isoflavone doses (165 mg/d (34) and up to 132 mg/d (35)) were significantly higher than corresponding doses of isoflavones in the studies by Cassidy et al. (9), Pino et al. (32), and McMichael-Phillips et al. (33) (65 mg/d (32) and 45 mg/d (9,33)), in which hormonal and histologic changes were observed. The differences in results illustrate the complexity of the action of phytoestrogens, which may be related in part to the biphasic effects of these compounds (36,37). In only one study on children was a significant positive association found between premature thelarche and consumption of soy-based formula, in addition to other environmental factors (consumption of fresh chicken meat, maternal ovarian cyst) (38). This study was conducted in Puerto Rico in response to a threefold increase in the prevalence of premature thelarche between 1978 and 1981. In spite of the apparent association, however, these environmental factors were probably not sufficient to account for the increased incidence of premature thelarche because in more than 50% of the cases, there was no exposure to any of these additional factors. When assessing the effect of isoflavones on sexual development in infants and children, we must rely mainly on data obtained in animal studies. Exposure of pregnant mice to genistein increases the density of terminal end buds in the mammary glands of the female offspring (14). By comparison, a similar effect was induced by a 1,000-fold lower dose of estradiol. In another study, Awoniyi et al. (39) found that intrauterine (from day 17 of gestation) and postpartum (until the end of lactation) exposure to genistein induced a reduction in the size of the ovaries and uterus in female pups and a reduction in estradiol and progesterone. In mature animals, by contrast, genistein administration induced an increase rather than a reduction in uterine weight in rats (15) and mice (40). In a study in which rats were exposed to genistein in utero and postnatally until puberty, a 0.1% genistein diet (but not a soy-free diet or a 0.02% genistein diet) was associated with a markedly increased (2.3-fold) uterine-to-body-weight ratio (41). The effect of phytoestrogens on sexual maturation and onset of puberty has been evaluated in studies in which rats have been exposed to the isoflavone genistein during critical periods of central nervous system development. This critical period is related to certain developmental milestones in the ontogeny of gonadotropin synthesis and secretion in the human fetus: the presence of gonadotropin-releasing hormone (GnRH) neurons in the fetal hypothanamus (9 weeks of gestation) and functional connections of these neurons with the portal system by 16 weeks of gestation (42). In one study (43), newborn female rats that had been exposed to genistein in utero for 5 days had delayed onset of puberty. This delay was not associated with decreased responsiveness of the pituitary gland to GnRH challenge. The daily dose of genistein in this study was similar to the estimated daily intake of genistein (30.1 mg/d) among Japanese middle-aged women (44). However, when calculated on a body weight basis, the genistein amount administered to the rats was higher by two to three orders of magnitude than the corresponding amounts in humans. Two previous studies from the same group have shown an altered pituitary response to GnRH in rats exposed to genistein in the postnatal period (1–10 days of life): low-dose genistein (0.01 mg/d) enhanced whereas higher doses (0.1 to 1.0 mg/d) attenuated GnRH-stimulated secretion of luteinizing hormone (36,37). The genistein doses in these latter studies are comparable (when corrected for weight) with the daily genistein intake of infants consuming soy-based formula in the first week of life (i.e., 15–16.6 mg/d) (4,5). In a differently designed study by Casanova et al. (41) rats were continuously exposed to genistein in utero and after birth until puberty. When exposed to a diet containing 0.1% genistein, female rats showed a significant acceleration of puberty, whereas other groups of rats that were exposed to a lower percentage of genistein showed no sexual development acceleration compared to a control group maintained on soy-free diet (41). From the animal studies, it is clear that several key factors modify the effects of isoflavones on sexual development, namely, the time of exposure (in utero, postnatal, or both), the dose of isoflavone administered to the animals, and the age of the animals (pup vs. mature). Well-designed, prospective studies in pregnant women and infants during the first year of life will illuminate the relevance of these animal studies to humans. An example of that approach was provided recently by Adlercreutz et al. (45), who measured phytoestrogens in maternal blood, cord plasma, and amniotic fluid in seven Japanese women at delivery of a neonate, showing that isoflavonoids are indeed transferred from the maternal to the fetal The in the age of onset of puberty over the the contribution of environmental factors as to this studies among soy and studies among may to the effect of phytoestrogens on the age of onset of puberty. PHYTOESTROGENS AND The effects of phytoestrogens on and density have been extensively evaluated in animal and in postmenopausal women. In genistein the of the that was markedly reduced after a soy diet in an effect that was related to the isoflavone content of the diet In studies, it has been found that the positive effect of soy on was enhanced formation rather than by or of In postmenopausal women, the of is the reduction in estrogen women who either soy with high content of isoflavones or the isoflavone have shown an increase in density and decreased of of these observations showing density in animal and in infants fed soy formulas, studies have with or no change in density in different studies. the effect of phytoestrogens on density in infants be from studies that evaluated the effect of soy-based infant formulas on These studies on were conducted 10 to and focused mainly on and content and their rather than effects of the soy-based formulas. However, on we can these soy formulas as a source of In a study infants between and weeks of age fed with breast milk, cow milk formula, or soy formula, et al. found that infants given soy formula had less skeletal maturation and at months of age than the other two groups. this was not significant at months of These were by observations of decreased and reduced and reduced content in the first few months of life among infants who were fed soy-based formulas, compared to infants fed human milk or cow milk–based formulas The between the groups was after months of age in with the study of et al. Some studies have soy-based formulas as the of in infants fed soy formulas In one study, which for 12 a group of term infants who were fed soy-based formula had significantly lower at and 12 months of age compared to a cow milk–based formula group However, the values in the soy formula group were similar to values obtained in infants human milk with This that soy-based formulas may adequate in infants because human milk is the for formulas. The effect of formulas on has been attributed to factors such as absence of a that intestinal absorption of or the presence of in which absorption of and from the diet. In the of and of soy-based formulas used in early studies was with of only 50% to from the Indeed, the of and for density was in two studies among infants. et al. used a soy formula in which was by a system and found that was similar in infants fed human milk or soy-based formulas over a period of 12 et al. used a soy-based formula with an and found that of infants at 16 and to weeks of age was higher in the group fed soy-based formula than in the group fed human In in such as content and between the soy formulas used in studies ability to the effects of soy-derived phytoestrogens on PHYTOESTROGENS AND The effect of soy has been for many In a of 38 of of soy was associated with a significant reduction in serum total and in with control groups There was no significant change in serum concentration of in these studies. The of reduction in serum and concentration was on the serum In 4 of the 38 studies, the were children, but results in these studies not differ significantly from results observed among The children studied had and they to a to 2 soy diet by a reduction in serum total of to in of in of and in total to ratio of in these studies were not however, and there was no effect of soy diet on total and to ratio in one of the studies and no effect on in two of the studies it is a common conclusion of these studies that soy diet may an contribution to the diet for in The mechanism of the effect of soy is not but phytoestrogens may play a similar to the effect of estrogen in postmenopausal women In women, a diet ± significantly the profile by serum total to and the ratio of to In a study by et al. a soy diet significantly reduced plasma concentrations of total and in compared with the effect of an diet. Furthermore, in a of with levels in the 50% of the the effect of the profile was more with a effect of increasing amounts of isoflavone on the in total and In a similar studies in have shown that a soy diet high in isoflavones soy diet) was associated with a significant reduction in and total to whereas was significantly These data that the and effect is by the content of isoflavones in soy In many infants, soy-based formulas the source of for the first months of life, a critical period for growth, and sexual The of phytoestrogens in these formulas that have been found to exert effects in animals and women is by the low affinity of soy phytoestrogens to estrogen receptors. or for infants consuming phytoestrogens in soy-based formulas have not been in However, more research is in regard to the and effects of soy-based formulas and other of phytoestrogens in the pediatric age