Introduction
Introduction
Phytoestrogens are polyphenolic non-steroidal plant compounds with estrogen-like biological
activity. The estrogenic properties of certain plants have been recognized for more
than fifty years. In the mid-1940 s, an infertility syndrome in sheep had been attributed
to the ingestion of clover containing high levels of the isoflavones formononetin
(1) and biochanin A (2) [1], [2]. More recently, an increasing number of epidemiological and experimental studies
has suggested that the consumption of phytoestrogen-rich diets may have protective
effects on estrogen-related conditions, such as menopausal symptoms [3], and estrogen-related diseases, such as prostate [4] and breast cancers [5], osteoporosis [6], and cardiovascular diseases (CVD) [7]. However, concerns have been raised about the potential dangers of consuming high
levels of these compounds [8]. Consequently, phytoestrogens are currently under active investigation for their
role on human health.
This review will discuss the recent literature on phytoestrogens, focusing on their
biological effects and biotransformation, as well as on their epidemiological and
experimental studies in animals and humans. In this way, a previous review in this
journal on the in vitro evaluation methods of phytoestrogens will be complemented [9].
Chemistry, Dietary Sources, and Biotransformation of Phytoestrogens
Chemistry, Dietary Sources, and Biotransformation of Phytoestrogens
Many structurally diverse compounds, originating from both industrial and natural
sources, have been reported to possess estrogenic activity (Fig. [1]) [10]. Humans are exposed through the food chain to a variety of xenobiotic estrogen-like
chemicals, such as DDT, polychlorinated biphenyls (PCBs), and diethylstilboestrol
(DES) (3). Recently, much attention has been focused on these xenoestrogens for their long-term
effects on the endocrine system. In addition, pharmaceutical estrogens, such as ethinylestradiol,
can also be classified as synthetic estrogen-like compounds.
Without discussing the ovarian steroids, most of the natural estrogen-like compounds
are produced by plants. These phytoestrogens are a diverse group of polyphenolic non-steroidal
plant compounds that bind to human estrogen receptors (ERs) and exert the characteristics
of endogenous steroidal estrogens. Based on their chemical structure, phytoestrogens
can be classified into four main groups, i. e., isoflavonoids, flavonoids, stilbenes,
and lignans (Fig. [1]), while β-resorcyclic acid lactones, which are produced by molds that contaminate
cereal crops, are classified as mycoestrogens. To be complete, some terpenoids and
saponins have also been reported to exert similar effects, although the number of
publications on these compounds as phytoestrogens is rather limited. A striking example
are the triterpenoids present in Cimicifuga racemosa extracts, which are considered to be at least partly responsible for the SERM activities
of these extracts [11].
Fig. 1 Classification scheme of dietary estrogens.
Isoflavonoids
Isoflavones are the most studied group of phytoestrogens and are found almost exclusively
in the family Leguminosae [12]. Soybeans are a very rich source of isoflavones and contain approximately 2 grams
of isoflavones per kilogram fresh weight [13]. However, it must be emphasized that the isoflavone content of soy products can
greatly vary between different soybean varieties [14] and through soybean processing [15]. Consequently, not all soy protein sources are equal with respect to their isoflavone
content and this should be taken into account when conducting epidemiological and
nutritional studies. For example, soy protein concentrates from which meat analogues
are made, have low concentrations of isoflavones if prepared by water extraction,
and almost no isoflavones are present if prepared by alcohol extraction [16]. Furthermore, soybean oil contains no isoflavones and soy sauce has little or no
isoflavones [17].
A large number of isoflavones has been identified from plants, with daidzein (4) and genistein (5) as the principal isoflavones. They occur in plants as the inactive glycosides daidzin
(6) and genistin (7) and their respectively 4′-methyl ether derivatives, formononetin and biochanin A
(Fig. [2]). Despite the high stability of the β-glycosides genistin and daidzin during processing
[18], these precursors can be metabolized in the digestive tract by the enzymes of the
normal microflora to their corresponding aglycones, genistein and daidzein [19], [20]. The gastrointestinal microflora can further metabolize daidzein to the potent estrogen
equol (8), but this biotransformation has a high inter-individual variability [21]. Another metabolite produced by daidzein is O-demethylangolensin (O-DMA) (9) [22]. Metabolism of genistein by the microflora yields the end-products 2-(4-hydroxy-phenylpropanoic
acid (10) and 1,3,5-trihydroxybenzene (11) [23].
Recently, several caco-2 cell line [24], [25], [26] and animal [27], [28] studies have examined the intestinal absorption of isoflavone glycosides. The results
obtained from the caco-2 cell line, which is used as an in vitro model of the human intestinal epithelium, suggested that isoflavone aglycones are
taken up into enterocytes more efficiently than their corresponding glycosides. Similar
results were obtained in isolated rat intestinal perfusion models [27], [28], but the question still remains whether isoflavone glycosides can be absorbed intact
from the human intestinal tract. A recent study showed that isoflavone glycosides
were not absorbed intact across the enterocyte of healthy adults and thus hydrolysis
of the sugar moiety was required for the absorption of isoflavone glycosides [29]. It was also suggested that isoflavone aglycones are absorbed through non-ionic
passive diffusion from the jejunum. After absorption, the isoflavones are readily
conjugated with glucuronic acid and to a lesser extent with sulfates, and are then
excreted in urine [30].
Coumestans, with coumestrol (12) and 4′-O-methylcoumestrol (13) as the major members, exhibit a close structural similarity to isoflavones (Fig.
[3]) [31]. Their main dietary sources are alfalfa, soybean, and clover sprouts. Little is
known about their metabolism in humans; however, their metabolism in a variety of
animals is extensively reviewed elsewhere [32].
Fig. 2 Formation and biotransformation of the isoflavones genistein and daidzein.
Fig. 3 The chemical structures of coumestrol and 4′-O-methylcoumestrol.
Flavonoids
The female flowers of hops (Humulus lupulus L.) are used in the brewing industry to add flavor and bitterness to beer. Recently,
several prenylated flavonoids have been identified in hops and beer, including the
flavanones 8-prenylnaringenin (14), 6-prenylnaringenin (15), and isoxanthohumol (16), and the chalcone xanthohumol (17) (Fig. [4]) [33], [34], [35]. 8-Prenylnaringenin, also named hopein, has been characterized as a very potent
phytoestrogen [35], [36] and large quantities of this phytoestrogen are now available through synthesis,
starting from the commercially available naringenin [37].
Fig. 4 Chemical structures of several prenylated flavonoids and the stilbene trans-resveratrol.
Resveratrol
Resveratrol (3,5,4′-trihydroxystilbene) (18), which exists as cis- and trans-isomers, is a secondary plant metabolite belonging to the class of stilbenes (C6-C2-C6) (Fig. [4]). This phytoalexin is synthesized by plants, including grapevines (Vitis vinifera), in response to injury and fungal attack [38]. The extraction of resveratrol from natural sources is time-consuming and yields
low amounts of the compound. Therefore, research on its biological properties really
started when trans-resveratrol was synthesized [39].
In contrast to the flavonoids, resveratrol is not widely distributed in the plant
kingdom. The compound can be found mainly in grapes, peanuts, and pines. Therefore,
red wine is one of the major dietary sources of resveratrol. Red wine contains much
greater amounts of resveratrol than white wine, since resveratrol is concentrated
in the grape skin and the manufacturing process of red wine includes prolonged contact
with grape skins. Several scientists are claiming that resveratrol is the wine component
responsible for the ”French Paradox”, i. e., the low incidence of heart diseases among
the French people, who eat a relatively high-fat diet [40]. However, there is no consensus on the so-called French Paradox, since other scientists
believe that proanthocyanidins are at least partly responsible for it.
Until now, data on the absorption and metabolism of resveratrol are still scarce.
In a rat intestinal perfusion model, it was demonstrated that the majority of the
absorbed resveratrol was conjugated to yield resveratrol glucuronide [41]. In a study on rats, resveratrol was bioavailable when administered in a solution
of hydroxypropyl-β-cyclodextrin [42], but underwent extensive first-pass glucuronidation. Recently, it was shown that
resveratrol is absorbed much more efficiently than (+)-catechin and quercetin in humans
after oral consumption [43]. Nevertheless, further studies are needed to confirm these results.
Lignans
Lignans, i. e., a group of dimeric phenylpropanoids, are mainly found in oilseeds,
such as flaxseed, but they are also present in whole cereals, grains, vegetables,
and fruits [44]. Matairesinol (19) and secoisolariciresinol (20) have been identified as two primary plant precursors of mammalian lignans (Fig.
[5]). They are converted after ingestion by intestinal bacterial flora to the biologically
active metabolites enterolactone (21) and enterodiol (22), respectively [45]. Both the parent compounds and the metabolites are measurable in various body fluids,
such as urine, feces, and plasma [46], [47].
Fig. 5 Metabolism of lignans by the normal microflora.
Estrogen Receptors and Selective Estrogen Receptor Modulators (SERMs)
Estrogen Receptors and Selective Estrogen Receptor Modulators (SERMs)
Estrogens are key regulators in a wide variety of target tissues, such as the male
and female reproductive systems, bone tissue, and the cardiovascular and central nervous
systems [48]. Estrogens are used for prevention and treatment of postmenopausal symptoms and
as contraceptives, while estrogen antagonists are used in the treatment of hormone-dependent
breast cancers. Consequently, it was believed that the administration of the estrogen
antagonist tamoxifen (23) in breast cancer patients would lead to a decrease in bone mineral density (Fig.
[6]). However, in a 24-month placebo-controlled study in breast cancer patients, the
opposite was found [49]. The study indicated that tamoxifen could act as an agonist in bone and as an antagonist
in the breast, and it is therefore termed as a selective estrogen receptor modulator
or SERM. SERMs, such as tamoxifen (23), raloxifene (24), and faslodex (ICI 182,780) (25), are compounds that bind to estrogen receptors and modulate agonist or antagonist
responses depending on the target tissue (Table [1]) [48], [50], [51]. Unfortunately, tamoxifen also exhibits stimulatory effects on the endometrium,
but raloxifene, an SERM approved for osteoporosis prevention, does not stimulate the
endometrium [52]. The biological effects of SERMs were better understood after the finding in 1996
of a second subtype of ER [53], [54]. So, to date, two estrogen receptors (ER) have been identified, i. e., ERα and ERβ, which have eight exons encoding for six functional domains, designated A-F. Both
ER subtypes differ in the ligand independent transactivation domain AF1 at the amino
terminus and the ligand binding domain at the carboxy terminus [55], as well as in the tissue distribution [56], [57]. Studies of their tissue distribution and/or their relative levels indicated that
ERα has moderate to high expression in uterus, testis, ovary, and kidney, while ERβ is expressed mainly in prostate, uterus, ovary, testis, bone, lung, and brain [56], [57].
Animal estrogens are exclusively steroidal compounds, with 17β-estradiol (26) as the principal physiological estrogen in most species, including humans. 17β-Estradiol
contains a phenolic group at position 3 and a secondary alcohol group at position
17 of a steroidal skeleton, separated from each other by a hydrophobic rather inflexible
structure of about 1.2 nm. As shown in Fig. [7], phytoestrogens are polyphenolic non-steroidal plant compounds that are structurally
similar to 17β-estradiol and thus may act as estrogen agonists or antagonists. In
this review, the SERM activity of phytoestrogens will be limited to isoflavonoids,
prenylated flavonoids, and the stilbene resveratrol.
Fig. 6 Chemical structures of the synthetic SERMs tamoxifen, raloxifene, and faslodex.
Table 1 Comparison between the biological effects of 17β-estradiol and some SERMs on selected
target tissues
|
Bone |
Breast |
Uterus |
Vasculature |
| 17β-Estradiol (26) |
+ |
+ |
+ |
+ |
| Tamoxifen (23) |
+ |
- |
(partial) + |
+ |
| Raloxifene (24) |
+ |
- |
- |
+ |
| Genistein (5) |
+ |
+/- |
No effect |
+ |
| +, agonist activity; -, antagonist activity. |
| According to Hall et al. [48] and Tikkanen et al. [51]. |
Fig. 7 A comparison of the chemical structures of 17β-estradiol, diethylstilbestrol, and
some phytoestrogens.
Biological Effects of Isoflavonoids
Biological Effects of Isoflavonoids
Isoflavonoids as SERMs
A great number of isoflavonoids has been tested in a competition binding assay to
assess their relative binding affinities [58]. The estrogen receptor relative binding affinities of the isoflavonoids tested decreased
in the following order: 17β-estradiol (control) > coumestrol > genistein > equol >
daidzein > biochanin A. In this assay the binding of a compound to a receptor is determined,
but it cannot distinguish between agonistic and antagonistic activity [9]. In addition, the study mentioned above did not make a difference between ERα and ERβ binding affinities.
Besides the receptor binding assay, there are several in vitro test systems, including cell-proliferation assays and gene reporter assays, to evaluate
the estrogenic activity of natural compounds [9]. In the so-called ”E-screen” the ability of a compound to stimulate the proliferation
of human estrogen-dependent breast cancer cell lines, such as MCF-7 and T47-D, is
measured. In a reporter gene assay, the capability of a compound to activate the transcription
of an estrogen-sensitive promotor is analyzed. Several studies have used these assays
together with the competition binding assay to compare the estrogenic activity of
isoflavonoids on ERα and ERβ [56], [59], [60], [61], [62]. They clearly demonstrated that coumestrol was the most active isoflavonoid and
bound almost as strongly as 17β-estradiol to both ERα and ERβ, but genistein induced transcription as strongly as coumestrol [61]. The isoflavones tested, including genistein, daidzein, and equol, exhibited a greater
binding affinity to ERβ than to ERα [56], [59], [60], [61], [62]. However, the concentration required for induction was almost the same for both
ERs and was much higher than expected from the binding affinity [60]. Isoflavone glycosides, such as daidzin, glycitin, and genistin, bound weakly to
both receptors and estrogen receptor-dependent transcriptional expression was poor
[60]. Interestingly, genistin stimulated the growth of MCF-7 cells more strongly than
genistein. Formononetin and biochanin A exhibited a significantly lower binding affinity
and transcription induction compared to their non-methylated forms daidzein and genistein,
respectively [59], [61]. Consequently, metabolization of formononetin and biochanin A by microflora is necessary
to obtain phytoestrogenic activity.
A recent study on MCF-7 cell lines showed that 17β-estradiol and coumestrol strongly
increased the progesterone receptor (PR) mRNA expression and slightly down-regulated
the ERα mRNA expression [63]. Genistein and SERMs such as raloxifene and faslodex strongly decreased the ERα protein levels. The study concluded that coumestrol exerts molecular properties which
are very similar to those of 17β-estradiol, whereas the molecular properties of genistein
are comparable to those of the SERMs raloxifene and faslodex.
In conclusion, isoflavones have a relatively greater binding affinity for ERβ than for ERα, but are 102 to 105 times less active than steroidal estrogens. They are, however, frequently present
in the human body in much higher quantities than endogenously produced estrogens.
Additionally, methylation or glycosidation of isoflavones generally decreased their
binding affinity to ER and their estrogen-dependent transcription expression.
Other biological effects of isoflavonoids
Some isoflavonoids are able to inhibit several key enzymes in estrogen and androgen
biosynthesis, including 5α-reductase [64], 17β-hydroxysteroid oxidoreductase [65], and aromatase [66], and can stimulate the synthesis of sex hormone-binding globulins (SHBG) [67]. Furthermore, several isoflavonoids have been reported to exert other non-hormonal
effects in vitro, including inhibition of tyrosine kinases [68], DNA topoisomerases I and II [69], and anti-angiogenesis [70] and antioxidant activity [71]. In addition to the SERM activity of some isoflavonoids, all these non-hormonal
effects may contribute to their potential preventive effects against certain types
of cancer. However, it must be emphasized that many of these non-hormonal effects
have been shown with very high concentrations, which can hardly be obtained in vivo [72].
Biological Effects of Prenylated Flavonoids
Biological Effects of Prenylated Flavonoids
Prenylated flavonoids as SERMs
Several prenylated flavonoids have been studied for their estrogenic activity [35], [36], [73], [74]. All these studies concluded that 8-prenylnaringenin exerts in vitro a very high estrogenic activity. In a competition binding assay, it was demonstrated
that 8-prenylnaringenin competed strongly with 17β-estradiol for binding to both ERα and ERβ. The relative binding affinities were significantly higher than the most active isoflavonoids
coumestrol and genistein [36]. No significant difference in relative binding affinity was observed between 2(S)- and 2(R)-enantiomers of 8-prenylnaringenin [73]. Movement of the prenyl unit from position 8 to 6 resulted in loss of the activity
[36], [73]. Xanthohumol and isoxanthohumol showed no affinity for both ERs [36].
In an MCF-7 cell line proliferation assay, the estrogenic activity of 8-prenylnaringenin
was found to be 25 times higher than that of the isoflavone genistein [73]. It was suggested that the high activity of 8-prenylnaringenin is related to its
lipophilicity, resulting in a higher permeability of cellular membranes compared to
genistein. In an estrogen-inducible yeast (Saccharomyces cerevisiae) assay the estrogenic potency of a series of isoflavonoids and prenylated flavonoids
decreased in the following order: 17β-estradiol (control) > 8-prenylnaringenin > coumestrol
> genistein > daidzein >> 6-prenylnaringenin [35].
Other biological effects of prenylated flavonoids
Prenylated flavonoids are less studied than isoflavonoids, but two studies attracted
our attention. First, it was found that 8-prenylnaringenin up-regulated the function
of the E-cadherin/catenin complex in human mammary carcinoma cells [75]. Down-regulation of elements of the E-cadherin/catenin complex at transcriptional
or posttranslational levels is a common feature of carcinoma cells. Nevertheless,
further studies are required to demonstrate its potential anti-cancer activity in vivo. Second, prenylation of naringenin antagonized the pro-oxidant effect of naringenin
on LDL oxidation, but the antioxidant activity of the prenylated flavanones is still
lower than that of the flavonol quercetin [76].
Biological Effects of Resveratrol
Biological Effects of Resveratrol
Resveratrol as SERM
trans-Resveratrol is structurally similar to DES and binds equally to both ERα and ERβ [77]. The latter finding contrasts with other phytoestrogens, such as genistein and coumestrol,
which have a higher binding affinity for ERβ than for ERα. Although resveratrol can exist as a cis- or trans-isomer, the trans-isomer exerted a higher activity in estrogen-dependent human breast cancer cell lines
[78]. At concentrations of 10 and 25 μM it increased the in vitro growth of MCF-7 cell lines, whereas at concentrations of 0.1 and 1 μM it had no effect.
At a concentration of 10 μM resveratrol inhibited binding of 17β-estradiol to ER and
activated transcription of estrogen-response reporter genes transfected into human
breast cancer cells [79]. In another study, resveratrol antagonized 17β-estradiol-stimulated growth and inhibited
transcription of PR in MCF-7 cells [80]. These results suggest that resveratrol acts as a mixed estrogen agonist/antagonist,
which was examined more in detail in two other studies [77], [81]. It was reported that in the absence of 17β-estradiol, resveratrol weakly induced
ER-dependent transcriptional events in some mammary tumor cell lines, whereas down-regulation
was observed when resveratrol was co-administered with 17β-estradiol [81]. In mouse mammary glands, grown in culture, resveratrol inhibited the formation
of 7,12-dimethylbenz[a]anthracene-induced and 17β-estradiol-promoted atypical ductal hyperplasia. In another
study, it was shown that resveratrol exhibited 17β-estradiol antagonist activity for
ERα with select estrogen response elements (EREs), while resveratrol showed no 17β-estradiol
antagonist activity with ERβ [77]. These results indicated that resveratrol differentially affects the transcriptional
activity of ERα and ERβ in an ERE sequence-dependent manner.
Dose-response studies revealed that orally administered resveratrol had minimal in vivo effects on estrogen target tissues in growing Sprague-Dawley rats, including no effects
on uterine growth, body weight, serum cholesterol, and radial bone growth [82]. In contrast, resveratrol antagonized the serum cholesterol-lowering effect of 17β-estradiol.
It was concluded that resveratrol has little or no estrogenic activity on estrogen
target tissues and may even be an estrogen antagonist [82].
Other biological effects of resveratrol
Resveratrol exerts a wide variety of biological effects, including inhibition of platelet
aggregation [83], modulation of lipoprotein metabolism [84], and antioxidant activity [39], [85]. The inhibitory activity of resveratrol on Cu2+-catalyzed oxidation of low-density lipoproteins (LDL) has been related to its high
Cu2+ chelating capacity [39]. Furthermore, it has been demonstrated that resveratrol inhibits membrane lipid
peroxidation mainly by scavenging peroxyl radicals within the membrane [85]. Although it is less active than the chain-breaking antioxidant α-tocopherol, its
capacity of spontaneously entering the lipid environment may allow resveratrol to
exert a significant antioxidant activity in vivo. The biological effects mentioned above suggest a possible role for resveratrol in
the prevention of atherosclerosis and CVD [40].
Resveratrol has also been reported to promote apoptosis of human tumor cells [86] and to induce the expression of the tumor suppressor p53 [87], suggesting a potential anti-cancer activity.
Effects of Phytoestrogens in Humans
Effects of Phytoestrogens in Humans
Recently, several epidemiological and experimental studies in animals and humans have
suggested that the consumption of foods rich in phytoestrogens may have protective
effects on estrogen-related conditions, such as menopausal symptoms, and estrogen-related
diseases, such as prostate and breast cancers, osteoporosis, and CVD. In this review,
the latest developments on the potential protective properties of phytoestrogens and
phytoestrogen-containing foods against hormone-dependent breast and prostate cancers
and CVD, and as estrogen replacement therapy (ERT) for postmenopausal women will be
discussed more in detail. Additionally, the potential dangers of consuming high levels
of these compounds will be discussed.
Phytoestrogens and cancers
Breast cancer
Epidemiological studies have indicated that the incidence and mortality of breast
cancer in the Western world is much higher compared to Asian countries. When Asian
people emigrated to the USA, it was shown that the first-generation female migrants
had a lower risk of breast cancer, but the protection was lost in the second generation
with an increasingly Western diet [88]. It was therefore suggested that certain phytochemicals present in Asian diets can
affect cancer incidence. Until now, most of the research has been focused on phytoestrogens
and more particularly on isoflavones. First, the average daily dietary intake of soy
and isoflavones in Asian populations has been estimated to be respectively 50 g/day
and 30 mg/day, while in the Western populations the intake is limited to respectively
1 g/day and 1 mg/day [89], [90]. Second, these phytochemicals display estrogen-like activity and high affinity binding
to the ERs, suggesting a role in hormone-dependent diseases, such as breast cancer.
Nowadays, there is an increasing number of human (and animal) studies demonstrating
that a high soy intake during childhood is associated with a reduced breast cancer
risk [91], [92], [93], [94]. However, there is no convincing evidence to suggest that soy or isoflavone consumption
in Western countries during adult life is protective against breast cancer [91]. Soy consumption before puberty may have the same risk-lowering effect as an early
pregnancy. It is suggested that phytoestrogens promote cell differentiation in the
mammary gland, resulting in enhancement of mammary gland maturation [94]. Further studies must confirm these results and must provide evidence that the isoflavones
present in soy are responsible for the health effects of soy.
The plant-lignan glycosides matairesinol and secoisolariciresinol are converted after
ingestion by intestinal bacterial flora to the biologically active enterolactone and
enterodiol, respectively. In several human studies a very low plasma enterolactone
concentration was associated with an increased breast cancer risk [91], [95]. Consumption of fiber-rich whole-grain bread may stimulate the production of enterolactone,
but in rats and humans, an increase in dietary fat intake decreases the urinary excretion
of lignans, despite constant grain-fiber intake. Although obesity is negatively associated
with plasma enterolactone in women, the effect of fat intake on breast-cancer risk
may be indirect, via production of mammalian lignans [91]. This emphasizes the importance of the gut microflora, but the question still remains
if the mammalian lignan enterolactone is protective or is just a biomarker of a healthy
diet.
Prostate cancer
Asian men have a lower incidence of prostate cancer compared to men from Western countries.
As discussed above for breast cancer, it was also suggested for prostate cancer that
soy intake could be a protective diet factor. Several recent human studies support
the hypothesis that soy intake prevents prostate cancer [91], [96]. Although the recent studies are encouraging, it is still premature to make recommendations
on phytoestrogen intake and prostate cancer prevention or management.
The mechanism of phytoestrogen action is still unknown, but most studies suggest that
the protective effects could be related to a reduction in androgen production, e.
g., through inhibition of 5α-reductase. A recent study has shown an inverse association
of soy product intake with serum androgen and estrogen concentrations in Japanese
men [97]. Another possible mechanism of action is the binding of phytoestrogens, such as
genistein, with the ERβ, which is the predominant ER in the prostate.
Colon cancer
In contrast to the breast and prostate cancers, colon cancer does not have a strong
association with hormone status [98]. In a recent review article, it was stated that soy and isoflavonoids do not seem
to protect against colon cancer, but lignans or lignan-rich food can inhibit colon-cancer
development in animal models [91]. Therefore, the relationship between intake of lignans and colonic cancer risk warrants
further investigation [99].
Phytoestrogens and CVD
Recently, double-blind clinical trials have shown that consumption of soy protein
compared to other proteins such as casein can lower total and LDL-cholesterol levels
[100], [101], [102], [103]. The effect is variable, but is generally greater in hypercholesterolemic than in
normocholesterolemic subjects. In contrast, several double-blind, placebo-controlled
clinical trials using isoflavone supplements alone have not shown a beneficial effect
on serum lipids [104], [105]. A recent consensus paper indicated that both soy protein and isoflavones may be
needed for lowering serum cholesterol concentrations [106]. Soybeans are an excellent source of proteins since, in contrast to animal proteins,
soybeans contain no cholesterol. In addition, soybeans are low in saturated fat. Soy
isoflavones may exert its effect by up-regulating LDL-receptor activity [107]. In conclusion, although soy proteins may reduce the lipid values, it is now essential
to start some clinical studies to investigate the effect of soy on CVD prevention.
Phytoestrogens as estrogen replacement therapy (ERT)
ERT is recommended for postmenopausal women to prevent menopausal symptoms, osteoporosis,
and CVD [103]. Despite these benefits, however, there are still concerns that ERT may cause cancer
of the breast. Consequently, there is a growing interest among patients and researchers
in phytoestrogens as an alternative to the conventional ERT.
The best results for osteoporosis prevention were obtained for ipriflavone (7-isopropoxyisoflavone,
Fig. [8]) (27), suggesting that it is a useful and safe alternative to ERT in treating existing
low bone mass or osteoporosis in postmenopausal women [108], [109]. Nevertheless, one study questioned the efficacy and safety of ipriflavone for prevention
of postmenopausal bone loss. [110]. Ipriflavone is a synthetic isoflavone, derived from daidzein. It does not seem
to act through direct estrogen receptor activity and is therefore not strictly a phytoestrogen.
However, approximately 10 % of the ingested dose is converted back to daidzein in
the body [111]. Clinical studies of the effects of phytoestrogens and phytoestrogen-containing
foods on bone health have produced mixed results. One clinical study indicated that
flaxseed supplementation had no effect on biomarkers of bone metabolism in postmenopausal
women [112], while two clinical studies administering isoflavone preparations to postmenopausal
women reported a bone loss prevention effect [113], [114].
Clinical studies of the effects of isoflavone-containing soy preparations on the incidence
of hot flashes in postmenopausal women provided mixed results. Some clinical studies
showed a significant but small reduction of hot flashes [115], [116], while one study administering a soy preparation to breast cancer survivors did
not improve their incidence of hot flashes [117]. As stated by Glazier and Bowman, it must be emphasized that a 15 % reduction in
hot flashes would mean a reduction of 1 hot flash per day in patients experiencing
12 hot flashes per day [103]. So, if supplementation of phytoestrogens will cause a statistically significant
improvement on the number of hot flashes, this does not mean that the patient’s quality
of life will improve.
Finally, a human study in postmenopausal women supplemented with isoflavone-containing
soy powder concluded that isoflavones do not have estrogenic effects on endometrial
tissue, in contrast to 17β-estradiol [118].
Fig. 8 Chemical structure of the synthetic ipriflavone or 7-isopropoxyisoflavone.
Potential adverse effects
Concerns on the potential adverse effects of phytoestrogens have been originated from
their structural similarity to DES. The best-known adverse effects are reported in
animals. In the mid-1940 s, an infertility syndrome in sheep had been attributed to
the ingestion of phytoestrogen-rich clover. Nevertheless, there are concerns about
the long-term effects of phytoestrogens given to infants and young children [72]. It was found that infants fed on soy milk formulas had plasma isoflavone levels
that are orders of magnitude greater than those of infants fed on human or cow’s milk
[119]. Further studies are needed to investigate if the soy formula-fed infants would
display later in life a greater risk to breast or prostate cancer.
In a recent study, s. c. injection of genistein in ovariectomized mice decreased thymic weight and thymic
and splenic CD4+ CD8- T cell numbers and resulted in immune suppression and lymphocytopenia [120]. The latter was also seen in a human study on ipriflavone for the prevention of
postmenopausal bone loss [110].
In conclusion, although the intake of phytoestrogens is higher in countries where
the incidence of estrogen-related cancers is lower, it is unwise to exclude adverse
effects of phytoestrogen supplementation. More in particular, supplementation with
very high doses of pure phytoestrogens and supplementation of infants and pregnant
women must be approached with caution.
Concluding Remarks
Concluding Remarks
In view of the current data, phytoestrogens are generally accepted as beneficial rather
than deleterious, particularly when consumed in food products. The consumption of
phytoestrogen-containing food products, especially soy products, may contribute to
a lower risk of developing CVD and prostate cancer in healthy people, while soy intake
during childhood is associated with a reduced breast cancer risk. Although a large
number of studies is encouraging, it is still premature to recommend specific amounts
of dietary phytoestrogens for prevention of chronic diseases. First, the dosage and
purity of commercial phytoestrogens and their possible adverse effects remain largely
unknown. Second, large population-based studies have not been carried out nor are
there randomized controlled clinical trials to standardize dosage and ensure safety
and efficacy. Third, at this moment, most of the studies were performed with food
products and not with a single phytoestrogenic compound. It is therefore a simplistic
view to extract the health effects of a high soy consumption to the potential health
effects of isoflavones present in soy. It may be that other components present in
these food products contribute to these effects. Additionally, the lifestyle factors
in Asian countries are quite different from these in Western countries and cannot
simply be reduced to the intake of one food product.
In conclusion, until more information on phytoestrogens is gathered and fully understood,
it is recommended to eat a balanced diet containing a wide variety of fruits, vegetables,
including soy, and whole grain products.
Acknowledgements
Acknowledgements
PC, TDB and SA are postdoctoral researchers of the Fund for Scientific Research (FWO-Flanders,
Belgium).
