The molecular basis of oestrogen action involves the binding of ligand to intracellular receptors (ER) which then dimerise, bind to specific nucleotide sequences in the DNA (oestrogen response elements) and influence patterns of gene expression. A major mechanism of the action of phytoestrogens in target cells is through their ability to interact with oestrogen receptors and so exert effects directly on the genome. However, the identification of a second oestrogen receptor (ER£]), which can bind phytoestrogens more strongly than the classical oestrogen receptor (ER£\) and which has different tissue distributions from ER£\, raises important unanswered questions about the physiological role of the two receptors in phytoestrogen action. Studies in vitro of phytoestrogen action to date have used a wide variety of disparate assays based in cells with differing receptor contents and differing cellular context (Woods, 2002) which makes inter-study comparisons difficult. None have addressed the question of whether there are specific differences in the action of the phytoestrogens through the two receptors ER£\ and ER£], and therefore whether the differing levels of receptors or differing ratios of the two receptors in cells could result in different patterns of gene expression and different tissue responses.
The objectives of this project were to determine the effects of overexpression of ER£] on phytoestrogen action in oestrogen-responsive cells from different tissues.
The approach taken was to select clones of oestrogen-sensitive cells which stably overexpress ER£] and to investigate the effects of this overexpression on the ability of phytoestrogens to regulate oestrogen-responsive gene expression.
<ul> <LI> This was carried out in cell lines from breast and uterus, both of which are oestrogen-responsive tissues. MCF7 human breast cancer cells show a well documented responsiveness to oestrogen but possess endogenously only detectable ER£\. Hec1A human uterine cells show more limited oestrogen responsiveness and have no detectable endogenous ER
<LI> Two isoforms of ER£] were used in these studies, ER£]long which is the native human ER£] and ER£]short which was the first cloned isoform of human ER£]ƒnand codes for a functional protein but shortened by 53 amino acids at the N-terminal end.
<LI> Effects on gene expression were monitored using reporter genes either transiently transfected (ERE-LUC) (dual luciferase assay) or stably integrated (ERE-CAT).
<LI> Dose-responses on gene expression were monitored for six phytoestrogens: genistein, daidzein, equol, coumestrol, 8-prenylnaringenin, resveratrol.
Outcome/Key Results Obtained: <BR>
Two clones of MCF7 human breast cancer cells were selected and characterised for overexpression of ER£]short and two clones for ER£]long. Overexpression of ER£] at an mRNA level was demonstrated by semiquantitative RT-PCR and confirmed using real-time PCR. Overexpression at a protein level was characterised by Western immunoblotting. The highest level of ER£] protein overexpression was 4.3-fold but all clones with raised ER£] levels also had raised ER£\ levels. Use of full-length human ER£\ and human ER£] protein standards enabled quantitation of the ratio of absolute levels of the two ERs within the MCF7 cells. The extent of overexpression of ER£] in terms of overall ER protein content within the MCF7 cells was limited, reducing at best the ER£\ : ER£] ratio from 19:1 in the untransfected cells to 10:1.
Using a stably integrated ERE-CAT reporter gene, overexpression of ER£]long was found to reduce oestrogen induction of CAT activity. Overexpression of ER£]long also reduced induction of CAT activity by daidzein, equol, coumestrol and 8-prenylnaringenin. However, by contrast, overexpression of ER£]long was found to enhance CAT induction with high concentrations of genistein (10-6M, 10-5M) and resveratrol (10-5M).
<P>Assessment of phytoestrogen action using the ERE-LUC reporter gene was compromised by the varied superagonist activity in which genistein and equol were able to induce reporter gene expression to a greater extent than oestradiol in MCF7cells. Superagonist effects were found also for equol, genistein, coumestrol and 8-prenylnaringenin in cells overexpressing ER£]. Such superagonist effects were not seen when using the ERE-CAT reporter gene.
<P>The reasons why such a high level of overexpression at an mRNA level was not translated into high overexpression at a protein level remains unknown. However, even this level of ER£] protein overexpression in the MCF7 cells was detrimental to cell growth and these cells were more difficult to maintain and slower to grow than the parental untransfected cells. It is thus possible that cells with higher ER£] protein levels simply died at the early stages of selection and were never identified. A good future approach would be to reduce the levels of ER£\ in these clones using siRNA technology and to do this in the short term (4-7 days) to avoid the long term adverse effects on the cells.
<P> Nine clones of Hec1A human uterine cells were successfully identified as overexpressing ER£]long at an mRNA level but none of these clones possessed any detectable functional ER£] protein. The aim of overexpressing ER£] protein in these cells therefore failed for technical reasons which could relate again to the detrimental effect of ER£]ƒn on cell growth.
Future work: Should aim to transfect the ER£] on an inducible expression vector to avoid long term constitutive overexpression causing adverse effects on the cells.
What it means and why it¡¦s important:
Today there is a wide consumer interest in phytoestrogens for their potential health benefits in countering menopausal symptoms and in lowering the incidence of oestrogen-dependent diseases including breast cancer. An understanding of the mechanism of action of these compounds in relevant cell types is central therefore to providing accurate consumer advice in relation to consumption of phytoestrogens.
Find more about this project and other FSA food safety-related projects at the <a href="http://www.food.gov.uk/science/research/" target="_blank">Food Standards Agency Research webpage</a>.