Isoflavones are frequently called „phyto-estrogens“, and they are usually compared with the natural sexual hormone estradiol. Even in most recent reviews the term “phyto-estrogen” is equated with “acts like estrogen”. This equation already caused safety concerns based on the fact that estrogen has proliferation-enhancing effects which may contribute to the development of cancer. Such effects have simply been transferred to the isoflavones, regardless of the clinical realities (Duffy et al. 2007; Wuttke et al. 2007). Wuttke even uses highly dramatic words such as “alarming” in his review of in vitro-data, without taking the re-assuring data from clinical observations into account.
Physiological effects of isoflavones as ER-β agonists
In fact, isoflavones are known as selective estrogen receptor beta (ER-β) agonists, whereas the potential negative effects of estrogens are triggered by activation of the estrogen receptor alpha (ER-α), a system with completely different physiological tasks. Activation of ER-β is related to
- - The normalisation of cell proliferation at the breast, the uterus and (in men) the prostate;
- A reduction of hot flushes;
- Health benefits for the cardiovascular system, and
- An improvement of bone mineral turnover (Handa et al. 2008; Harris 2007; Heldring et al. 2007).
These very same effects have been attributed to soy rich diet and the supplementation with isoflavones. Health benefits in exactly the same conditions have been described for the dietary intake of soy isoflavones in epidemiologic studies. Examples of benefits are an improved cardiovascular function and the prevention of osteoporosis (Ikeda et al. 2006; Yang et al. 2005; Zhang et al. 2003; Zhang et al. 2005), and the alleviation of menopausal complaints (Nagata et al. 2001). These effects have been confirmed in clinical trials.
Affinity of isoflavones to estrogen receptors
The therapeutic prospects within a selective activation or ER-β have triggered a race for potential new drug substances. One of the structural parent compounds is genistein, the major isoflavone from soy (Sarkar et al. 2006). Selectivity of genistein at ER-β was already described shortly after the discovery of the new receptor (Kuiper et al. 1997), and has regularly been confirmed since then (e.g., (Choi et al. 2008)). At ER-α genistein possesses only 4 % of the affinity of estradiol, whereas the comparative affinity at ER-β is 87 % (Gruber et al. 2002). The following table highlights the differences regarding the concentrations required for an effect.
Table: Halfmaximal concentrations for an effect at the estrogen receptors. Data taken from (Harris et al. 2005).
| Compound | EC50 at ER-α | EC50 at ER-β |
| 17β-Estradiol | 6.4 × 10-11 M | 3.9 × 10-11 M |
| Genistein | 4.6 × 10-7 M | 3.4 × 10-9 M |
The difference found for genistein is over-additively increased by 218 % in the presence of even small amounts of estrogen. This situation of low estrogen levels is present in postmenopausal women (Harris et al. 2005). Isoflavone quantities typically ingested with soy food and supplements result in blood levels sufficiently high to induce ER-β, but not ER-α (Harris et al. 2005). Consequently, the clinical experience is that of reduced menopausal complaints, and the epidemiological observation of a decreased incidence of estrogen-dependent tumours (Harris 2006). In contrast, an increased risk of proliferation of hormone-dependent tissues has never been clinically observed respectively confirmed.
Blood levels of isoflavones
The effect of isoflavones at ER-β are by now widely accepted, although risk debates have been based on secondary sources such as Kulling & Watzel (2003), which wrongly and without citing a bibliographic source allege a higher affinity of soy isoflavones to ER-α than to ER-β (Kulling and Watzl 2003). A hypothetical risk was also extrapolated from the fact that isoflavones may activate ER-α next to ER-β, albeit in much higher concentrations than required for ER-β. However, such affinities are measured in vitro, whereas the physiological effect depends on the concentrations which can be reached in the human organism. In fact, an activation of ER-α through isoflavones has exclusively been observed in cell culture and in specific animal models in the absence of ER-β and estrogen. None of these studies has compared the tested concentrations with the levels encountered in the human organism after intake of soy preparations. In contrast, blood levels achievable after dietary intake of isoflavones are well-examined. Clearly the observed serum concentrations do not support a hypothetical activation of ER-α in the human organism, even with high dietary doses.
Activation of ER-α in addition to ER-β would require 10fold higher blood levels than the levels which can be reached even after dietary intake of extremely high quantities of isoflavones (for blood levels see (Choi et al. 2008; Guerini et al. 2005; Heldring et al. 2007; Kuiper et al. 1997; Kuiper et al. 1998; Mishra et al. 2006; Remer et al. 2005; Wuttke et al. 2007)). After several weeks of soy consumption typical serum levels of the isoflavones genistein and equol are 0.8-0.9 µM (Harris et al. 2005). After 12 months of intake of 54 mg of genistein/day the serum level was 0.73 µM (D'Anna et al. 2007). In another study with a 10-month application of 250 mg of soy extract (corresponding to 100 mg of isoflavone glycosides or to 60 mg of aglycones), 1.4 µM of genistein in the serum was reached – concentrations which are sufficient for an activation of ER-β, but not of ER-α.
Conclusions
The affinities of isoflavones to estrogen receptors and the distribution of the different subtypes of estrogen receptors in the human organism are discussed in multiple reviews (Heldring et al. 2007; Koehler et al. 2005; McCarty 2006). The preferential affinity of isoflavones to ER-β has been consistently demonstrated. This offers a plausible explanation of the effects observed for the dietary exposure to isoflavones, either with soy food or through supplements with soy or red clover.
References
Kobayashi, S. (1992). Trends in national nutritional survey of Japan. Nutr. Health 8 (2-3): 91-96.
Kulling, S. E. and Watzl, B. (2003). Phytoöstrogene. Ernährungs-Umschau 50: 234-239.
