Dynamics Of Estrogen Receptor Activity Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

In recent years several studies demonstrated the presence of estrogen receptors in mammalian tissues and significantly improved our understanding of their ability to control biological processes in reproductive as well as non reproductive organs. Considering the manifold mechanisms and organs that are involved in estrogen action and the implication of estrogens in human female physiology, innovative approaches are required to shed light on the widespread activities of estrogen receptors in woman physiology. This is particularly relevant for the definition of novel, more efficacious hormonal replacement therapies or for the evaluation of the risk associated with the exposure to endocrine disruptors. The introduction of genetic engineering and the development and application of in vivo imaging techniques offer new tools for pre-clinical studies. The generation of the ERE-Luc mouse, a reporter animal developed for in vivo studies of the estrogen receptor activity, allows assessing the activity state of the ER signaling pathway in all target tissues and organs at once, under physiological stimuli or as a result of a pharmacological treatment. This review summarizes the main steps in the generation and appraisal of the estrogen receptor reporter mouse ERE-Luc, designed for in vivo molecular imaging studies, and describes examples demonstrating the suitability of the ERE-Luc model for drug development and for the investigation of the effects of endogenous, environmental, and dietary estrogens in vivo.


â-ºReporter systems are novel tools to monitor complex biological events.

â-º ERE-Luc mouse model allows a systemic view of the transcriptional effects of estrogens.

â-º ERE-Luc mouse is a powerful tool for drug development.

â-º ERE-Luc mouse model is critical to study action and tissue specificity of phytoestrogens.

Keywords: Reporter mouse; Estrogen Receptor; SERM; Drug development; Phytoestrogens

1. Introduction

In the recent years, many reports have revealed the complexity of ER physiological activities and the multiplicity of estrogen targets. For this reason is critical to define whether and to what degree new ER ligands, proposed for pharmacological treatments, are activating or repressing ERs in any given tissue. To this aim, cells transfected with reporter genes have been extensively employed to characterize the activity of new ER ligands. However, the application of reporter gene technology to animal engineering offers now the opportunity to take advantage of reporter tools to investigate a specific biochemical process in the whole animal, thus extending the study of ERs activity in the spatio-temporal dimension. In order to follow the transcription processes in a physiological context, several reporter animals have been generated; nevertheless, only recently the advancement in reporter gene technology and in transgenesis allowed to generate powerful tools designed to follow more and more complex biological events in vivo.

Following this line of thoughts, the ERE-Luc model was generated by integrating into the mouse genome a luciferase reporter gene whose expression is controlled by transcriptionally activated ERs. In this transgenic animal stimuli inducing receptor activation lead to a proportional increase in luciferase synthesis; thus, the reporter expression represents a biosensor allowing the measurement of the state of receptor activity. This unique feature makes possible to assess the activation state of the ER signaling pathway in all target tissues and organs concurrently, allowing the investigation of the outcomes of physiological stimuli as well as of pharmacological treatments. For this reason the ERE-Luc model can be considered as a prototype of reporter mouse effective for SERMs development , to identify the activities of endocrine disrupter , to study cross-talk among membrane receptor and ERs and to discover novel estrogen target tissues . Above all, the ERE-Luc emerged as an invaluable tool to unveil the physiology of ER activation, thus shedding light onto a novel mechanism that is paving the way to more selective therapies to treat the menopause-associated disorders .

2. Reporter mice - a new way to look at drug action

The notable advances in molecular genetics achieved in the past decade as well as the possibility to manipulate cells in order to investigate drug activity using reporter genes led to significant improvements in drug development strategies. Nowadays, eukaryotic cells transfected with reporter genes (genes that encode an easily detectable protein) are extensively used to study cis-regulatory sequences or trans-acting factors modulating the transcriptional activity of specific promoters.

The application of reporter gene technology to animal genomes allows studying the activity of the selected promoters within a living organism, thus providing, with the aid of imaging technology, the opportunity to observe and quantify in real time the dynamics of transcription of simple or complex promoters.

A key step in the generation of a new reporter animal model is the choice of the candidate reporter gene, that must satisfy several prerequisites in order to be used for molecular imaging studies: (i) they must be detectable by in vivo studies: their expression indeed must provide directly in vivo information about the location, the magnitude and the persistence of gene expression; (ii) their expression must match the expression of the gene endogenously controlled by the investigated promoter; (iii) the turnover rate of the reporter proteins must be short, in order to allow a rapid evaluation of changes that occurs in a short time in the studied molecular process.

The introduction of reporter genes into animal genomes has a huge impact on preclinical experimentation, thus providing the opportunity to appreciate physiological and pathological mechanisms by providing measurable endpoints useful for the assessment of drug efficacy in all tissues in vivo. In order to be exploited for drug development, a mouse model must satisfy a number of requirements that are crucial for the correct evaluation of drug candidates, thus allowing: (i) rapid evaluation of all the organs where a specific compound is active; (ii) assessment of the response to the drug after repeated administration; (iii) measurement of the response in each tissue, dependent on the route of administration; (iv) evaluation of the smallest quantity of drug needed to produce the pharmacological response independently of its plasma levels; (v) identification of active metabolites and their profile of action; and (vi) detection of sites of drug accumulation during chronic administration.

As result of the generation of novel tools to monitor more and more complex biological events in vivo, granted by the recent progresses in reporter gene technology and in transgenesis, several reporter animals have been generated to study transcription processes in a physiological context. Notwithstanding, if we take in consideration the abovementioned points, only few of the reporter mice generated up to now are suitable for drug development applications. Among the great number of mice generated to study the activity of ligands on specific receptors, only few, are designed to have a ubiquitous expression of the reporter and to guarantee that their signalling is free from influences resulting from position effects . In addition, several of these mouse models have reporters unsuitable for in vivo imaging analysis or have a too long turnover rate, making impossible assess drug action in time . Finally, other models rely on the activity of receptors missing functionally important domains . Even though these models can provide information on the capability of the receptor to recognize a ligand, still they are not appropriate to understand the physiological consequences of such an event, which is critical to evaluate the pharmacological potential of the investigated compound.

The ERE-Luc mouse is the first example of a reporter animal designed for in vivo studies of drug development: this model offers the opportunity to take advantage of reporter tools applied to the entire animal, thus expanding the analysis of ERs activation in the spatio-temporal dimension.

3. The generation of ERE-Luc Reporter mouse model

Estrogens control specific gene networks to modulate target cell activities by signaling through two nuclear receptors (estrogen receptors, i.e. ERs), ERalpha and ERbeta. Cell-based approaches developed in the last decade have extended our knowledge of the transcriptional regulation of ERs at the promoter of target genes, where the interplay between ERs and coactivators and corepressors provides the receptors with tissue specificity of action.

ER transcriptional activation is controlled by a multiplicity of mechanisms . In addition, growth factors stimulate unliganded ER to induce the transcription of selected target genes . Activated ERs may cross-talk with other signaling pathways by affecting the activity of other transcription factors like AP-1, STATs, NFkB, SP-1 or by networking with components of other transduction pathways such as Src , MAPK and G-proteins . Given that the estrogen action involves a multiplicity of mechanisms and organs, and considered the importance of estrogens in human female physiology, it is necessary to take advantage of innovative approaches to clarify the exact physiological relevance of these alternative mechanisms in each target tissue. Cell systems cannot give global information on the hormonal activity on an entire organism: a reporter animal system is the ideal model to investigate the physiology and pharmacology of estrogen and cognate receptor activity.

The ERE-Luc model was designed to satisfy the following parameters: (i) the reporter gene is expressed ubiquitously and responsive to ER-activating molecules in all the mouse cells; (ii) the reporter protein is not expressed by the mouse, can be easily evaluated by quantitative assays, its cellular localization by immunohistochemistry is easy and can be observed by non-invasive optical imaging technologies.

Optical imaging techniques are based on bioluminescence or fluorescence. The reporter genes used in bioluminescence imaging protocols, such as luciferase, generate photons proportionally to the level of expression in the presence of the appropriate substrate, i.e. luciferin.

The main limitation when bioluminescent reporters are used is that the light signal is absorbed and scattered in the tissue volume intersected by the low energy photons in the path from the emission site to the detector system. Red light is more efficiently transmitted through tissues than light with lower wavelengths, therefore efforts are being made to generate reporters producing photons with wavelengths above 600 nm. Luciferase activity from Photinus pyralis produces a broad spectrum peaking at 560 nm with a wide fraction above 600 nm, whereas wild-type green fluorescent protein (GFP) from Aequorea victoria, upon excitation, emits light with a peak at 509 nm. In order to shift the emission closer to infrared wavelengths, recently GFP and luciferase mutant variants have been created . Nevertheless, GFP needs to be excited by an external light source. As consequence, the effect of tissue scattering and absorption is doubled with respect to bioluminescent probes that directly emit photons, thus limiting the use of GFP to very small organisms. In addition, the luciferase gene offers a number of notable advantages: (i) encodes an insect enzyme absent in mammalians, thus excluding the background caused by endogenous product; (ii) is easily measured using highly sensitive enzymatic assay; (iii) allows the use of specific, commercially available antibodies for localization studies; (iv) enable optical imaging in live animals; (v) most importantly, provides a dynamic view of the state of ERs activity since it has a very short turn-over rate and does not accumulate. On this basis, the firefly luciferase gene was preferred as a reporter gene in the generation of this first animal model.

In the phase of selection of the estrogen-responsive promoter, the strategy of using strong promoters was discarded because of their high activity, which would have covered hormonal response or established new estrogen-independent elements for transcription. Therefore, it was pursued the idea to use a promoter ensuring a low basal state of reporter expression to emphasize the ubiquitous capacity of the reporter to be expressed in all tissues and the hormone-dependent transcription.

To this aim, using deletion mutants of the minimal promoter from the thymidine kinase (tk) gene from herpes simplex virus or the basal TATA box linked to a variable number of estrogen receptor-responsive elements (ERE), several different estrogen-inducible promoters were generated and tested in transient transfection assays in different cell lines. A low basal transcription and the highest estrogen-induced reporter expression was obtained with the combination of 2 palindromic EREs spaced 8 bp, and located at 55 bp from the constitutive tk promoter . In order to avoid position effects and to guarantee the ubiquitous expression of the transgene, we took advantage of the insulator strategy using insulators from the chicken genome : the β-globin hypersensitive site 4 (HS4) and the matrix attachment region (MAR) . Analyzing luciferase expression in clones obtained in stable transfection assays carried out with pERE, pMAR and pHS4 plasmids, we observed that the use of insulator sequences raised the number of luciferase expressing clones from 19 to up to 70%, thus improving the inducibility of luciferase expression by estrogen treatment. Subsequent studies demonstrated that the generated mouse has ubiquitous, regulated expression of the reporter

This model has a much wider potential than the classic preclinical investigation models, since it allows to: (i) observe in which organs or tissues the studied compound is active; (ii) define the effects of a given compound and possible changes in response after repeated administration; (iii) assess precisely the response of each tissue to the administration of a selected compound independently of the route of administration; (iv) evaluate the minimal drug concentration needed to stimulate ERs independently of the drug's plasma levels; (v) observe the production of active metabolites as well as their profile of action; (vi) confirm possible sites of drug accumulation during chronic administration; and (vii) it can considerably shorten the time and decrease the number of animals necessary to develop a specific compound thanks to the combination of pharmacological and toxicological studies.

4. ERE-Luc mouse as a tool to study the dynamics of Estrogen Receptor activity

Among the huge number of studies on the molecular mechanism of estrogen receptor action, only very few have been aimed to the characterization of the receptor activation in the spatio-temporal dimension. The use of molecular imaging technologies is smoothing the progress of these studies, offering the possibility to follow biological processes on a recurring basis in the same individual. Biochemical, immunohistochemical as well as pharmacological criteria have been applied to the ERE-Luc mouse to verify that the luciferase expression can be adopted as a measure of the state of ER activity , thus establishing that, by measuring the photons emitted by the luciferase/luciferin enzymatic reaction within the estrogen target tissue, the state of ER activation can be visualized and quantitated. Photons are detected and assessed using a cool-charged coupled device camera (CCD-camera). Using a computer-assisted analysis is then possible to localize the photon emission with high sensitivity and with a resolution of 3-5 mm by merging the luminescence detected with the CCD and a light picture of the animal . This non-invasive procedure can be applied several times to the same individual without its sacrifice; thus, it is possible to monitor the modulation in receptor activation in different organs during all the main physiological changes taking place in life, from embryonic state to adulthood.

The application of this technique allowed to challenge the well-established knowledge on estrogen action in the control of reproductive-related functions. By detecting the photon emission from a cycling ERE-Luc animal during estrous cycle we surprisingly observed no correlation between the serum level of 17β-estradiol and the ER activation in tissues not directly connected to the reproductive function (e.g. thymus, intestine, aorta, bone, areas of the brain like cortex, hippocampus, etc.). The highest receptor activity in these "non-reproductive tissues" has been observed at diestrus phase, when levels of estrogens in serum are lowest. This unexpected finding has been confirmed by evaluating in tissue extracts the activity of luciferase gene product coupled with the expression of the progesterone receptor, a natural estrogen target gene.

In order to define the dynamics of estrogen receptor activity as function of the estrogen synthesis in the ovaries, we evaluated the oscillation of ER activity in intact and ovariectomized (OVX) mice . To this aim, we measured daily luciferase-dependent photon emission in intact and OVX mice for 21 d. We observed that in the cycling mice (Fig. 1A) luciferase activity oscillated with time and the amplitude of the oscillation was different in each body area and the frequency of oscillation resulted to be tetradian (lasting 4 days, average in all tissues 4.4d) in most body areas. This result was predictable, considering that the length of estrous cycle in mice is 4-5 d; however, ER oscillations did not occur synchronously in all the body areas. In addition, quite unexpectedly ER activity fluctuated also after ablation of the ovaries with an oscillation period of about 4 d (Fig. 1B) with an amplitude lower than in intact mice and with altered phasing among organs: for instance, oscillation in the hepatic and genital areas was seldom in phase.

These results show that in intact mice ER activity oscillates with a frequency that is compatible with the estrous cycle, but the master regulator of ER oscillatory activity cannot be 17β-estradiol (E2), because the oscillations were asynchronous and persisted after ovariectomy. The asynchrony of ER activity in the different body areas of intact mice was verified by quantitative real-time PCR analysis of mRNAs encoded by endogenous ER target genes , thus confirming the reliability of luciferase as indicator of ER transcriptional activity. Therefore, luciferase can be used as a surrogate target gene in studies on the effect of selected HT.

The distinct regulation of ERs in reproductive and non-reproductive tissues can be ascribed to a differential functionality of ligand-bound vs. unliganded-activated receptors, thus inspiring the intriguing hypothesis that hormone-dependent ER functions were acquired as result of the development of reproductive functions only late in the phylogenetic evolution. Before the acquisition of these functions, the ERs were orphan receptors able to modulate in the different organs the response to membrane receptor signaling pathways. As demonstrated by our imaging studies, ER in liver is very active and may be activated by factors other than estrogens. Indeed, in the past we observed that liver ER was significantly activated after food ingestion even when the synthetic diet utilized to feed the mice was deprived of known estrogenic compounds . It took us several experiments to realize that, among all macronutrients, amino acids (aa) were responsible for a transcriptional activation of liver ER strictly associated and necessary for the proper progression of the estrous cycle . Furthermore, by genome-wide experiments we observed that regulation of energy metabolism by ERα is tightly linked to reproductive functions as indicated by several findings: (i) in adult mice the expression of genes involved in fatty acid and cholesterol synthesis oscillated synchronously with the estrous cycle (possibly in function of a potential egg fertilization); (ii) in prepuberal female mice the expression of the two sets of genes had an opposite trend (the high cholesterol synthesis being probably due to the requirements of a growing organism); (iii) in the later stages of pregnancy, characterized by high circulating estrogen levels , there was a impressive decrease in the expression of all enzymes; and, most relevant, (iv) in the absence of a cycle due to age or surgical menopause, the expression of most of the genes in study lost the oscillatory pattern, and the mRNAs levels were generally significantly higher than at proestrus. It's noteworthy that a long-term dysregulation of the estrous cycle and tetradian ERα oscillatory activity (e.g. ovariectomy, aging, alteration of IGF-1 signaling) resulted also in accumulation of fat deposits in liver. Experiments on liver Igf1-/- mice and liver specific ERα KO mice showed that the amount of deposited lipid is inversely proportional to the synchrony between the estrous cycle and production of mRNAs for fatty acid/cholesterol synthesis. These results underline the importance of the tetradian activity of hepatic ERα to poise the receptor to adjust to the different energy requirements associated to the reproductive stage. These findings are particularly relevant to understand the etiology of metabolic disorders associated with impaired ovarian functions (e.g. PCOS) or the cessation of the reproductive cycle , suggesting to revise current rationales in the treatment of the post-menopause, taken into due consideration the periodic nature of ER signaling for a more efficacious use of estrogens.

5. SERM classification exploiting the ERE-Luc mouse

The reported experiments raised doubt on whether the existing approaches in hormone administration are correctly mimicking the physiological activation of the receptors in terms of asynchronous activation in reproductive versus non-reproductive tissues. Using ERE-Luc mice, we demonstrated that as consequence of prolonged treatment with natural and synthetic ligands, the transcriptional activity mediated by ER oscillates with pulses characterized by frequency and amplitude closely related to the nature of estrogenic compound investigated, its dosage, and the organ in study. We investigated the effects of prolonged hormone replacement therapy on ER signaling by whole-body in vivo optical imaging. Doses calculated by the allometric approach to be equivalent to those used in humans of estrogens and selective ER modulators (17β-estradiol (E2), conjugated estrogens (CE, Premarin), bazedoxifene (BZA), lasofoxifene (LAS), ospemifene (OSP), raloxifene (RAL), and tamoxifen (TAM) were administered daily for 21 days. During this study, photon emission was assessed once a day in defined body areas using a segmentation algorithm. ER activity was also measured in cycling and ovariectomized mice as controls. The experiment showed the oscillation in time of the ER-dependent transcriptional activity, and revealed the strict association of frequency and amplitude of the transcription pulses between the target organ and the estrogenic compound. In all of the studied body areas, each compound showed a different profile of activity. In mice treated with E2, we observed an increment in luciferase activity over time in the skeletal and genital areas; in contrast, photon emission in the hepatic area increased rapidly after E2 administration and decreased with time. LAS administration led to little to no change in hepatic and skeletal areas, whereas photon emission in genital area was higher than in controls just before the end of the treatment. In the OVX mice ER activity did not change noticeably in all anatomical areas during the treatments. The analysis of the photon emission evidenced a fluctuation over successive days of exposure in all the observed mice, including the OVX. Additional analysis of the evolution of the bioluminescence profile with time evidenced that with some of the drugs, the oscillatory pattern had a specific frequency and amplitude (e.g. LAS in the genital area). The comparison of the outcomes of the diverse treatments in each experimental group showed that the oscillation was a distinctive response to the specific ligand in the different tissues examined.

After the analysis of bioluminescence profiles generated in the studied areas, five independent parameters (peak number, peak amplitude, peak period, AUC, and potency), namely descriptors, were used to describe unique features of the drug effect. The selected descriptors were inserted in a clustering algorithm in order to define the extent to which the drugs act differently or similarly with each other. Most important, the comparison of the sets of parameters of drug-treated mice versus the OVX or cycling animals, allowed to assess the effectiveness of each drug in replacing the endogenous hormones, thus representing the best replacement therapy.

By using all the descriptors portraying the effect of SERMs on ER in the various anatomical areas (genital, skeletal, hepatic, abdominal, and thymic) of the ERE-Luc mouse, we created a single phenogram to classify the drugs according to their spatio-temporal activity. This analysis includes the distinct attributes of drug action in time in each investigated organ, and may be considered as a multivariate fingerprint of drug efficacy (Fig. 2).

An example showing the feasibility of this approach is provided by our experiment aimed to confirm the relevance of ER rhythmic oscillations on lipid accumulation . In details, we ovariectomized ERE-Luc mice of 2 months of age for 4-5 weeks prior initiating a 21-day treatment with vehicle, CE, BZA, TSEC (Tissue Selective Estrogen Compound) and RAL. The drugs were administered at a dosage previously shown to induce a proper oscillation of ER in liver and intestine. The measurement of luciferase activity in liver and intestine and the application of Fourier transform led us to measure the frequency of the oscillations induced by the treatments.

Next, we evaluated the synchrony of the peaks of ER activity in liver and intestine and evaluated the data obtained by agglomerative hierarchical clustering to identify the classes of compounds which better mimicked the ER oscillatory behavior characteristic of cycling mice. The results (Fig. 3A) clearly indicated that OVX changed significantly the oscillatory pattern of the receptor (as indicated by the Manhattan distance from the cycling mice) and that CE and RAL did not significantly improve the distance from the cycling controls: this indicated their inability to induce a physiological oscillation. Conversely, BZA and TSEC were found to cluster with cycling animals, thus indicating to be able to reinstate an oscillation more similar to cycling animals than to OVX. When we measured liver FFA content, we found that lipids were increased significantly in OVX mice treated with vehicle for 21 days (+ 84% vs. intact cycling mice); among the selected HRT only TSEC and BZA prevented liver FFA accumulation. When hormone replacement therapy was not reinstating the correct oscillatory pattern, with CE and RAL, lipid accumulation occurred (Fig. 3B).

Experience with Selective Estrogen Receptor Modulators (SERMs) has demonstrated the immense variability of the in vivo action of estrogenic molecules: the long-term treatments performed induced a state of activation of ER which was dependent on the tissue evaluated, the dosage utilized, and the time of treatment, challenging the ability to establish the parameters necessary to evaluate the extent of beneficial/negative effects associated with each estrogenic compound. We are confident that the dynamics of ER transcriptional activity are of great significance from the physiological point of view, therefore it is necessary to faithfully reproduce such distinctive features in hormone replacement therapy. To enhance the effectiveness of hormone replacement therapies, we should discover ligands able to mimic the effects in time of endogenous hormones. Clustering the treatments according to their effects on ER transcriptional activity in space and time, and evaluating the extent to which the treatment circumvents the effect of ovariectomy, thus mimicking the activity of ER in cycling mice may pave the way to the discover effective and safe treatments for the post-menopause. Moreover, the classification of environmental and dietary endocrine disrupters and the comparison of their activity to well-studied compounds will make easier the evaluation of their real toxicity.

6. ERE-Luc and phytoestrogens

Human beings, as consequence of their individual diet, are distinctively exposed to phytoestrogens, chemicals produced by plants that act like estrogens by binding with high affinity to estrogen receptors. Phytoestrogen-bound ERs may network with other nuclear factors and co-activators as well as regulate the transcription of specific genes . The isoflavone genistein is the most abundant of these compounds, and is commonly found in edible plants such as legumes . The intake of a soy-nuts serving or a single capsule of a soy-extract induce a peak of plasma isoflavone levels to a concentration supposed to have the capability to perturb endocrine signaling. Since ER expression is not restricted only to reproductive organs, the possible influence of phytoestrogens on a number of physiological functions can also cause side effects in non reproductive organs. Studies in adult rodents have linked phytoestrogen intake with altered bone development , obesity , alteration of thymic functions , myelotoxicity and changes in the sexually dimorphic behavior .

Notwithstanding, the clinical community has a less than negative position towards phytoestrogens: it is frequently reported that soy or isoflavones may foster beneficial effects against age-related pathologies. Several effects have been associated dietary estrogenic compounds, and the intricacy of ER physiological activities is well known. For this reason is mandatory to consider a more systemic approach to investigate the effects of protracted exposure to phytoestrogen-rich diets. The availability of the ERE-Luc reporter mouse provides a suitable tool to study the effects of acute or chronic exposure to specific diets on ER transcriptional activity.

Recently, we investigated the effect of short-term and long-term consumption of isoflavones, pure or as elements of a specific diet, on ER activity . For 20 days, luciferase activity was assessed daily by bioluminescence imaging in ERE-Luc animals allowed to drink soymilk (water for controls) ad libitum. Luciferase activity noticeably increased in the first days of treatment in the chest area of the soymilk group and remained significantly higher than in controls until the end of the experiment. The level of photon emission (800 cts/s) was comparable with what observed after acute treatment with E2 (800-1200 cts/s). No significant changes were detected in limb, thymus, and genital area. Interestingly, in abdomen area of the soymilk group photon emission increased with time and peaked around day 15. These data make evident that in definite organs such as the liver, the chronic exposure to the estrogens enclosed in soymilk keeps ER activity to high levels. Then, we studied the effect of chronic administration of pure genistein using two doses 1 and 5 mg/kg/day of the compound, observing that the treatment did not modulate in ER activity in abdomen. Oscillations of photon emission were observed in this organ in both controls and treated mice: the origin of the changes related to the prolonged treatment with gavage is unknown.

The main finding of this study is that factors different from phytoestrogens (the food matrix) contribute to the global estrogenicity of soymilk. In order to explain the findings, we hypothesized three possible mechanisms : (i) differential receptor regulation: the treatment with free isoflavones cause a fast increase of their concentration in plasma, with simultaneous induction of a maximum in ER activity that, recurring in time, provoke a decrease in the responsiveness of the receptor to the ligand or in its transcriptional capacity. On the other hand, the prolonged consumption of soymilk may cause a steady status of ER activation without ER down-regulation due to phytoestrogens. Preliminary data did not evidenced a sharp reduction in hepatic ER protein after prolonged treatment with genistein, nevertheless it is likely that the receptor activity could be fainted in terms of interactions with co-regulators. (ii) Differential bioavailability: ER activity did not increase after both acute and chronic concomitant administration of genistein and daidzein. A possible reason could be a modulation of hepatic catabolic enzymes exerted by daidzein or its metabolites, with consequent catabolism/excretion of the administered compounds . It is possible that the soy matrix, by influencing the kinetics of absorption, distribution, and excretion of isoflavones, could synchronize their persistence in plasma or in tissues with the readiness of ER to be transcriptionally activated. (iii) Estrogenic compounds different from isoflavones present in soymilk: existing data indicate that soy proteins could exert a direct effect on liver, thus contributing to the modulation of LDL receptor activity (reviewed in Sirtori et al. ). Interestingly, is still generally acknowledged that hepatic LDL receptor is an estrogen-target gene . For these reasons, we may hypothesize that nutritional proteins control estrogen receptor activity, in a like manner to what has been reported for other nuclear factors with macro-nutrients such as fatty acids or glucose .

7. Conclusions

The main benefit of using reporter animals is that they supply measurable endpoints useful to evaluate the effects of natural and synthetic estrogens in all tissues of living animals. This approach changes entirely the study of ER physiological activities as well as preclinical experimentation thus providing the chance to unravel physiological and pathological mechanisms and to assess and compare in healthy animals and in models of specific diseases the effects of selected treatments.

The studies described in this review show the power of reporter mouse technology that allows: (i) comprehensive vision of the tissues affected by the investigated compound, useful for the identification of the organs potentially involved; (ii) precise evaluation of dosage and timing required to stimulate the receptor activity; (iii) repeated observations in a single individual disclose the sites of accumulation and activity, and the dynamics of response to the treatment.

The presented results further demonstrate the suitability of the ERE-Luc model to the investigation of the effects of endogenous, environmental, and dietary estrogens in vivo. ERs are expressed in most tissues in mammals, and estrogens have unpredictable tissue-specific actions on ERs: therefore, the ERE-Luc mouse model represents the most proper model to achieve a solid, systemic outlook of transcriptional effects of endogenous or exogenous estrogens. The relevance of the ERE-Luc mouse as a model to discover novel selective ER modulators (SERMs) and to outline their activity is clear; furthermore, the use of the ERE-Luc model revealed the broad action and tissue specificity of endocrine disrupters and xenobiotics such as phytoestrogens.


The work discussed in this article was supported by Pfizer Pharmaceutical Co. and the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 278850 (INMiND).