On the Milan soy safety symposium of May 2009 (Mai 13-14) the available facts on isoflavones were discussed in the context of safety of application in menopausal women.
Prof. Aidin Cassidy (University of East Anglia, United Kingdom) reported on the basics of the isoflavone effects (Cassidy 2009). Among the isoflavones in soy the glycosides genistin and daidzin dominate over glycitin, which is present in only small quantities (Mortensen et al. 2009). In the gastrointestinal tract the glycosides are cleaved to genistein, daidzein and glycitein, but are also partially transformed to other forms such as equol – however, only approximately 30 % of Europeans seem to be capable of producing equol. The original aglycones as well as their (active) metabolites are absorbed and undergo an enterohepatic cycling. Elimination finally occurs with the urines after binding to glucuronic acid (Mortensen et al. 2009).
Once inside the organism the isoflavones can trigger estrogen-receptor mediated effects – preferentially at the estrogen receptor beta (ER-β) (Harris et al. 2005). Compared with estradiol genistein has only an affinity of 0.003 % at ER-α, but a relative affinity of approximately 0.1 % at ER-β. For daidzein the relative affinities at the ER-α and ER-β are 0.003 % and 0.03 %, respectively. The effects become clinically plausible when it is considered that the blood levels of both isoflavones after consumption of usual quantities of soy food are approximately 10,000 times higher than those of estradiol (Mortensen et al. 2009; Setchell et al. 1997) – especially the property as a “selective estrogen receptor modulator” (SERM). The effect depends on various factors, mainly the number and distribution of estrogen receptors – especially ER- β – at the target organs (Hartman et al. 2009). This leads to complex conditions in the human organism, which are not adequately mirrored in cell cultures and in the animal models from which the risk is extrapolated. In addition the dose of isoflavones applied in studies presumably demonstrating a risk was in a range far beyond what can be reached by consumption of soy food or supplements (Anon. 2008; Setchell 2006). The relevance of such studies must therefore be questioned.
Fundamental pharmacokinetic differences between animal models and humans
Prof. Kenneth Setchell (Cincinatti Children’s Hospital Medical Center, USA) compared the facts on absorption and metabolism of isoflavones with the experimental safety studies (Setchell 2009). In order to be absorbed isoflavones from soy must first be cleaved by glucosidases in the gastrointestinal tract. They are then partly directly absorbed, but also partly transformed to active metabolites such as equol, which are subsequently absorbed (Cassidy et al. 2006). The pharmacokinetics of the isoflavones, especially of genistein, daidzein and equol, are well-examined. Blood levels may be 10,000 times higher than those of estrogen, and may therefore reach levels where an activation of ER-β is possible (Mortensen et al. 2009; Setchell et al. 1997). However, when discussing blood levels it must not be forgotten that only the fraction of free isoflavone not bound to albumin is available for an effect. The proportion of unbound isoflavones is approximately 50 % for genistein and for equol. The type of isoflavone source is obviously an important factor: after ingestion of the soy food Tempeh higher levels of free genistein are reached as with the supplementation of isolated genistein (Gardner et al. 2009).
The ability of the organism to produce equol from daidzein and the re-enforcement of the effects of this isoflavone also plays a role. Newborns and approximately 70 % of adults are unable to produce equol by transformation of daidzein by the intestinal flora (Cao et al. 2009; Setchell et al. 2006). Equol has a strong effect on ER-β, and only weakly binds to ER-α. Synthetic equol could be disadvantageous, as it consist of the natural S-(-)-Equol and of R-(+)-Equol which does not occur in soy. R-(+)-Equol is better absorbed than S(-)-Equol, but has no affinity to ER-β (Setchell et al. 2002; Setchell et al. 2006).
Concluding his presentation Setchell stated that the simple discussion of blood levels in animal cancer models without an exact determination of the effective fraction of isoflavones, the unbound compounds, may lead to misunderstandings. The presumed risk of a cancer-inducing effect of soy is based on artificial conditions which are not reproducible in the human organism.
Further reports from the Milan 2009 Soy Safety Symposium:
2. Facts Related to Bioavailability
3. Lack of relevance of animal models for an extrapolation of risks of isoflavones
4. Isoflavones protect „menopausal” mice from breast cancer
5. Breast cancer risk is increased by synthetic gestagens
6. Breast tissue density remains unaltered with soy
7. Clinical studies demonstrate safety of soy in the breast
8. Study in more than 5,000 breast cancer patients: First positive tendencies with soy!
9. No effects of isoflavones on the endometrium
10. Isoflavones also safe at the thyroid gland
11. Backgrounds on Menopausal Hot Flushes
12. Clinical safety of isoflavone-containing preparations
13. Clinical effects of isoflavones against menopausal hot flushes
References
Cassidy A (2009). Overview of isoflavones: Chemistry and interactions with estrogen receptors. Symposium on Evaluating the Efficacy and Safety of Isoflavones for Postmenopausal Women, 13-14 May. Milan (Italy): Council for Responsible Nutrition.
Setchell KD (2009). Overview of isoflavone physiology and pharmacokinetics. Symposium on Evaluating the Efficacy and Safety of Isoflavones for Postmenopausal Women, 13-14 May. Milan (Italy): Council for Responsible Nutrition.
