Selective Estrogen Receptor Modulators. Antonio Cano

Chapter 1. Molecular Mechanisms of Estrogen Action in Target Tissues

• B. Nicolas Diaz Chico

• Domingo Navarro Bosch

• Juan C. Diaz Chico

• Eduardo Escrich Escriche



The majority of signals that govern cell operations have their origin in the plasma membrane. They proceed from membrane receptors that respond to substances of diverse origins. An important part of these signals arrives at the cells via blood circulation, as it is the case of endocrine signals transmitted by hormones. Another essential group of signals originates in the neighborhood of the cells, or even in the cell itself. This is the case of paracrine and autocrine signals transmitted by an extensive assemblage of growth and differentiation factors.

The interaction of external signals with membrane receptors generates second messengers. These messengers either modify the cell concentration of ions or metabolites or alter the functional state of a chain of several molecules that act as intermediaries. These intermediaries may modify the intensity of determined biochemical reactions or, in other cases, are integrated into the machinery of gene transcription and alter the expression of specific genes. The consequences of these activities can lead to the induction of cell division.

An important group of endocrine signals does not require membrane receptors, second messengers, or intermediaries in signaling chains. They proceed from substances that seem to penetrate into the interior of cells without difficulty, where they join with intracellular receptors, and through which they act on the cell genome. These substances are small liposoluble molecules, of which several are of a hormonal nature: the steroid hormones - androgens, estrogens, progestagens, glucocorticoids, and mineral corticoids - the thyroid hormones, and vitamin D3. There are also nonhormonal substances, such as retinoic acid, prostaglandin J2, or fatty acids, which utilize intracellular receptors and exert powerful genomic effects. All these substances share common mechanisms of action through soluble intracellular proteins that are members of the nuclear hormone receptor family (Evans 1988; Vaseduvan et al. 2002).

Once nuclear hormone receptors are bound to their hormone, they are capable of being integrated directly into the machinery that regulates the transcription of specific genes. This action is more direct, and apparently more primitive, than that originating in membrane receptors. By controlling gene expression, the hormones regulate more the abundance of determinate, specific proteins rather than their biochemical activity. Those hormone-receptor complexes are also efficient regulators of cell proliferation.

Nuclear hormone receptors accumulate several functions in a single molecule. They are capable of recognizing and binding small molecules like steroids with high affinity and specificity. These hormone-receptor complexes are capable of recognizing and joining with specific sequences of DNA present only in genes that are the object of hormonal regulation. They are capable, in short, of interacting with other proteins - coactivators or corepressors - that participate in the regulation of the machinery of gene transcription and of initiating or modifying the expression of specific genes. The meeting of all these functions, and others not mentioned, in a single molecule make these receptors an extremely elaborate product from the point of view of evolution.

This chapter reviews the main characteristics of two of the better known members of the nuclear hormone receptor family: estrogen receptors a and в (ERa and ERe ). First, the different functional regions harbored by the molecule of the receptor are described. These properties will be used to describe the cellular, molecular, and other consequences that derive from the interactions of receptors with their own hormone, other proteins, or DNA.

The interaction of estrogen receptors with signaling systems of the cell membrane that respond to growth factors and mediate nongenomic, fast actions of estrogens will be reviewed as well. These mechanisms have a growing importance in the comprehension of phenomena like the induction of endothelial NOS (nitric oxide synthase) by estrogens (Rubanyi et al. 2002).


General Aspects

The hypothesis that hormones act through cell receptors is as old as the concept of the hormone. Nevertheless, and as usually occurs in science, hormone receptors were only discovered when the required technology became available.


The Discovery of Hormone Receptors for Steroid Hormones

Hormone receptors for steroids were discovered in the early 1960s, when the technology to radioactively mark steroids became available. By obtaining tritium-labeled estradiol, Jensen could show the existence of an intracellular protein component that bound specifically to this hormone and that was called the estradiol receptor (ER).

Shortly thereafter, O’Malley obtained an autoradiography image in which an accumulation of estradiol was observed within the nuclei of cells from chicken oviduct (O’Malley et al. 1974). It had been known that estradiol significantly altered the synthesis of RNA within a few minutes. With these scant initial data, the theory that steroids act through intracellular receptors, by means of which they carry out the regulation of specific genes, was established (Toft et al. 1966). The use of radioactively marked compounds permitted the discovery of receptors for the other steroid hormones, vitamin D3, and the thyroid hormones (Bouillon et al. 1995; Navarro et al. 2002; Evans 1989; Evans 1988). For many years, the only available technology to determine concentrations of receptors and to study their properties was based on the use of radioactive hormones.

It was not until 1995 that news emerged of the existence of a second type of ER, the ERβ, described almost simultaneously in two European laboratories (Nilsson et al. 2001; Kuiper et al. 1996). Since then, the original ER has been called ERa. Both receptors are independent biological entities, encoded by different genes that respond to the same denomination. Both genes have different patterns of tissue expression, with exclusive expression in some cell types and joint expression in others (Palmieri et al. 2002; Kuiper et al. 1996; Krege et al. 1998). ERa dominates in the reproductive tract, while in other tissues, especially the nervous, digestive, and ovary tissues, ERβ dominates (Nilsson et al. 2001; Krege et al. 1998; Couse et al. 1999a; Couse et al. 1999b).


Nuclear Hormone Receptors?

The intracellular distribution of steroid hormone receptors has long been the object of controversy. The first theoretical formulation on the intracellular location of the ERs was elaborated by Jensen in 1968 and is known as the “two-step theory.” Its execution was based entirely on biochemical observations obtained by means of tritium-marked estradiol. The ERs, in cells not exposed to hormones, are found abundantly in the soluble cell fraction, or cytosol (Fig. 1.1). Treatment with hormones confines the receptors to the particulated or nuclear fraction and causes their disappearance from the cytosol. The two-step theory established that the receptor is found in the cytoplasm naturally and upon the arrival of a hormone it is transformed into a complex hormone-receptor (first step) capable of translocating itself to the nucleus and of modifying gene expression (second step).

In the 1980s, Jensen and others obtained monoclonal antibodies against several of the nuclear hormone receptors (Di'az-Chico et al. 1988; Jordan et al. 1990). These antibodies permitted the introduction of immunohistochemical techniques in the study of receptors. Consequently, King and Greene verified

Fig. 1.1. General mechanism of action of steroid hormones. Steroid hormones cross through the plasmatic membrane without apparent difficulty favored by gradient. Some, which can be considered prohormones, are metabolized and transformed into more active products. This is the case with testosterone, which becomes dihydrotestosterone (DHT) in the target tissues of androgens, through the 5-alfa-reductase enzyme. The hormone binds to the receptor, a soluble protein of the cellular cytosol that, in the absence of hormone, is found associated with other proteins (hsp90 and others) that maintain the receptor in an inactive state. The hormone-receptor bond causes the other proteins to separate and a homodimer to be formed. The homodimer is the activated form of the receptor since it is capable of recognizing the genes that depend on that steroid hormone as well as of activating its expression, which leads to the synthesis of specific proteins that there were receptors in the nuclei of cells sensitive to estrogens, regardless of whether or not the cells had been exposed to hormones (King et al. 1984).

Simultaneously, Gorski’s group (Welshons et al. 1984), utilizing a type of cell fractionation that permits separating the cytoplasm from the nucleus, was also able to detect the presence of nuclear ER, even if the cell had not been exposed to hormones. These findings led to a different theoretical formulation, according to which the native receptors would be found in the cell nucleus, to which the hormone would accede directly.

At present, the two-step theory is still accepted, but it leaves out the question of receptor location within the cell so as to be able to cover all members of a family. The receptor, in the absence of hormone, is found associated with other proteins (hsp90, p59, and perhaps others) and very weakly bound to cell structures (nuclear or cytoplasmatic). The arrival of hormones transforms the receptor, freeing it from other proteins, giving it a greater affinity for nuclear structures, and causing it to achieve an active state as a transcription factor (Beato et al. 1996; Beato 1989). The difference is that the receptors not bound to estradiol are soluble, and they can be extracted also from the nucleus during homogenization: they are “cytosolic”, not “cytoplasmatic”.


Structure of Estrogen Receptors

Knowledge of the molecular structure of ERs was initiated from the cloning of the complementary DNA for the messenger RNA (mRNA) that encodes ERa. This was possible thanks to the explosive evolution of the recombinant DNA technology during the 1970s and to the production of monoclonal antibodies against ERa. In 1986 the sequence of amino acids of ERa was published by the Chambon group (Green et al. 1986a; Green et al. 1986b).

ERfî was cloned by chance, since it was found through the use of probes aimed at hybridizing with the most conserved part of the nuclear receptors - the DNA binding domain (DBD) - trying to find related sequences. This procedure has expanded the family of nuclear receptors to more than 100 members. Thus, the discovery of ERβ in rat prostate by the Gustafsson group was a surprise (Kuiper et al. 1996; Nilsson et al. 2001).


Primary Structure of Estrogen Receptors

The receptor molecule, that is to say the protein that interacts directly with the hormone, is formed of a single polypeptide chain in not only ERa and ERβ but also in the remainder of the known nuclear receptors (Evans 1989; Evans 1988; Krege et al. 1998; McDonnell et al. 2002; McEwen et al. 1999; Nilsson et al. 2001).

The idea that nuclear receptors belong to the same molecular family arose upon discovery of the considerable homology in the amino acid sequences among the receptors (Evans 1989). These homologous sequences affect six regions of the respective molecules, labeled with the letters A to F. The functions assigned to each homologous region were deduced from the comparison with the known functions of amino acid sequences in other proteins. The final confirmation was obtained from the analysis of the alterations in the function of the receptors that were produced after their structures were altered by means of controlled mutations.

Human ERa has 565 amino acids, greater therefore than the dominant isoform of ERβ, which has 530 amino acids (Kuiper et al. 1996), the isoforms are products of the same gene and are generated by alternate processing of mRNA. Nevertheless, the structure in the domains of both types of ER reflect the general pattern described, except that ERβ lacks the carboxyterminal F region (Kumar et al. 1987; Nilsson et al. 2001). Both ERa and ERβ, as well as their isoforms, can have different functions, inasmuch as they can activate different genes and even carry out antagonistic functions (Fuqua et al. 1992; Fuqua 2001; Pettersson et al. 2000). This is a field of extraordinary activity, where research has tried to find the different degrees of implication of every hormone and isoform in carcinogenesis and in tumor response to hormone treatment (Palmieri et al. 2002).


Activity Domains in the Molecules of Estrogen Receptors

ERs are transcription factors that are activated by means of a high affinity reaction (Kdbetween 0.01 and 1 nM) with a ligand (hormone or antihormone). The reaction transforms the receptor from a native state that is genetically inactive to an activated state capable of identifying the genes susceptible of responding specifically to each receptor. The functional organization of the ER is carried out through particular structures of the receptor molecule. These are formed by means of a series of folds of the molecule that permit reaching the (tertiary) spatial structure adequate for carrying out each function. Each one of the structures of the molecule that can carry out one of the particular functions is called a molecular domain. Molecular domains are not always formed by consecutive sequences of amino acids. They can be formed by several short sequences of amino acids, separated from each other by amino acids that are not part of a given domain.

ERs have domains responsible for nuclear location, hormone binding, dimerization, DNA binding, and transcription activation (Figs. 1.2 and 1.3) (Beato et al. 1996; Beato 1989; Fawell et al. 1990; Hall et al. 1999; Kumar et al. 1987).


Genetic Encoding of Estrogen Receptors

As in any another protein, the synthesis of an ER begins with the transcription of the gene that encodes it in RNA as a primary transcript (Chin 1995; Pong- likitmongkol et al. 1988). This forms a very long strand that is then processed to give mature mRNA. The maturation process includes the elimination of the exons and the modification of the ends of the RNA strand. The genes of the ER include 8 exons, some of them very large (> 26 kilobases). The mature mRNA has an open reading frame that encodes for the sequence of amino acids of the receptor, flanked on both ends by long sequences of nucleotides that are not translated.

The elimination of exons during RNA maturing requires two cuts and a coupling for each exon. Occasionally errors are produced, providing a source of receptor variants. The receptors thus formed can be truncated, that is, they lack the amino acids encoded by some of the exons. Additionally, receptor molecules can appear that have in duplicate amino acids encoded by some of the exons. These products of alternate mRNA maturation, called isoforms, exist for both types of ER and are produced in normal tissue, but their functional significance is not known (Fuqua et al. 1992; Fuqua 2001; McGuire et al. 1991; Scott et al. 1991).

Fig. 1.2. Structure and domains of the estrogen receptor. ERs have a structure that couple several functions in a single protein. The aminoterminal region of the receptor contains a binding domain to DNA, with two zinc fingers that confer upon the receptor the capacity to recognize short sequences of DNA, called estrogen response elements (ERE), in the promoter region of the estrogen-dependent genes. That region also contains a transcription activator region or transactivator, TAF1, which binds with nuclear transcription factors to complete the gene transcription machinery. The carboxyterminal region of the receptor contains a large binding domain to the hormone that occupies more than half the molecule and through which the receptor interacts with estrogens and antiestrogens. That same zone contains another transactivator region, TAF2, which only becomes activated in the presence of estrogen, a zone of dimerization, and another for binding to hsp90. Between the hormone and the DNA binding domains there is the hinge region, which contains a short sequence of basic amino acids that confer nucleophylia to the receptor and one of the zones that participates in the dimerization of the receptor

Fig. 1.3. Comparison of alpha and beta estrogen receptors. The alpha and beta ERs are products of different genes, but they maintain a similar structure. The figure shows that both receptors share different degrees of homology in their amino acid sequences, the highest one being that corresponding to the DNA binding domain. The beta receptor lacks the carboxyterminal zone, called zone F, which is absent in other members of the nuclear receptor family and is considered a peculiarity of the alpha ER

Alternative mRNA maturation is frequent in tumor tissue expressing ER. In some cases it would give rise to truncated receptors that would maintain the capacity to bind hormones but would have lost their capacity as a transcription factor. Additionally, truncated receptors would be produced that would lack the capacity to bind hormones but would conserve intact their capacity to interact with DNA. In this case, the truncated receptors can become tumorigenic by stimulating the proliferation of cells uncontrolled by hormones. These receptor variants havebeen theobjectofexhaustivestudy at thelevel of mRNA in tumors of the breast, mainly estrogen-dependent tumors (Clemons et al. 2001; Garcia et al. 1988; Palmieri et al. 2002), but tests for the existence of receptor protein with these characteristics have not corroborated the expectation created by their theoretical interest.


Native Receptor

In cell cultures deprived of hormones or in target organs of ovariectomized animals, ERs are found in a state known as “native”, characterized by their association with several proteins (Redeuilh et al. 1987). The best known are hsp90 and p59.

hsp90 is a chaperone protein that accompanies the ER from receptor synthesis and is indispensable in the acquisition by the receptor of the appropriate three-dimensional conformation. This protein, induced by heat or by cell stress (Redeuilh et al. 1987), is ubiquitous and much more abundant than the group of nuclear receptors of a given cell. hsp90 continues bound to the receptor until the receptor itself binds to the hormone. At this point, the receptor loses affinity for hsp90 and undoes the bond between both molecules.

p59 is an immunophyline, characterized by its specific binding with the immunosuppressant rapamicine (Ratajczak et al. 1993). It is unknown whether this property is related to the biology of the receptors. The native structure in vivo has been studied by means of substances that enter the cell and, once inside, establish covalent bonds between protein structures that were previously associated by noncovalent interactions (Segnitz et al. 1995). Studies carried out by the Ghering group have verified that the structure of glucocorticoid, estrogen, and progestin receptors is identical: a receptor molecule, two of hsp90 and one of p59. Other proteins, such as hsp70, have been identified in the native complexes of receptors in vitro (Smith et al. 1993). Nevertheless, the presence of these proteins has not been verified in vivo.

The formation of the structure of the native receptor depends essentially on the hsp90-receptor interaction. This is produced among specific sequences of both proteins in three dimensions; one of them is found in the DBD and the other in the ligand binding domain (LBD, domain E). When lacking a 24-amino-acid chain between the end of domain C, that of binding to DNA, and the beginning of domain D, the ER does not join with hsp90 and, therefore, does not form the hetero-oligomeric structure of the native receptor and remains a monomer. The binding of the ER with hsp90 is produced soon after the synthesis of the receptor and precedes the incorporation of the other proteins.


Estrogen Transforms the Native Receptor

The binding of hormone transforms the receptor in vivo, freeing it from the accompanying proteins. The transformation involves the conversion of the receptor into a more nucleophylic form, which can be extracted from the nuclei only with solutions of high salinity (0.4M KCl). The transformation of the receptor can be verified through centrifugation of cytosol in a sucrose gradient of density. The native receptor complex has a coefficient of sedimentation of 8S, which changes to 4S when the hormone transforms the receptor (Fig. 1.5). This change of coefficient of sedimentation reflects the rupture of the heterooligomer of the native receptor (8S), which frees the receptor monomer (4S) from its bond with hsp90 and p59 (Redeuilh et al. 1987; Navarro et al. 1998).

From what has been discovered, it can be deduced that part of the functions of ERs remains hidden in the native state. Interaction with hormone causes this structure to come apart. This process of activation or transformation permits the receptor to exhibit all the potential of interaction with DNA and makes the receptor exhibit the properties that were hidden by the proteins that accompanied it in the 8S form (McGuire et al. 1991).


Domain of Nuclear Location

All nuclear receptors have sequences known as domains of nuclear location (Picard et al. 1987). These sequences, rich in arginine and lysine, confer upon the many proteins that contain them the capacity to bind to nonhistone nuclear proteins. Receptors have up to four of these sequences, whose cooperation is necessary for nuclear location. When these sequences are exposed, the receptor tends to be located in the nucleus. When covered by other proteins, receptors are distributed throughout the cell.

In the case of the ER, when a region of 20 amino acids between 250 and 270 is missing, the receptor is located strictly in the cytoplasm. Domains of similar size and function have also been located in the receptors of glucocorticoids and of progesterone. The zone of nuclear location overlaps with one of the sequences for interacting with hsp90, which at the same time is next to the DBD. The coincidence of the three functions in a space so restricted implies that they are totally or partly incompatible sterically (Evans 1989; Gruber et al. 2002).


Hormone-Receptor Interaction

The recognition of each receptor by its respective hormone is a highly specific process occurring at the LBD. The small hormonal molecule enters a hydrophobic cavity of the receptor molecule, forming a high affinity bond.


Ligand Binding Domain

The binding domain for the hormone, or LBD, is situated in the carboxyl half of the receptor, the final portion of which is critical. For example, the deletion of 12 amino acids in the carboxyl end of the androgen receptor suppresses its capacity to bind hormone (O’Malley et al. 1974). The LBD has an amino acid composition that confers upon it a net hydrophobic character, suitable for interacting with organic molecules of low molecular weight, such as steroids.

In spite of the extensive homology in the key amino acids of the LBD (Fig. 1.3), each of the two ER isoforms has different affinities for natural and synthetic ligands (Table 1.1). This suggests that the responses are very different in tissues dominated by one or another receptor (Kuiper et al. 1996).

It is noteworthy that such a long portion of ERa is required (more than 220 aa.) to interact with a structure as small as a steroid. Nevertheless, the whole structure seems necessary since this domain includes a transcription activation function through which the receptor binds with the cofactors (coactivators and corepressors). It is considered that amino acids kept among different members of the nuclear receptor family form a hydrophobic cavity that lodges the hormone. Amino acids not preserved among different members of the family but preserved by the same receptor in different species can be important for discriminating among structurally similar hormones and provide specificity for the binding of each receptor with its hormone (Mester et al. 1995).

Table 1.1.

Relative binding affinity of ligands to estrogen receptors α and β




17-β Estradiol



17-α Estradiol




































Data from Kuiper et al. Endocrinology 138:863 (1997)

At the time the hormone is introduced into the LBD (Fig. 1.4), a conformational change is produced in the three-dimensional structure of the receptor, a change that is key to the subsequent steps in hormonal action. This change is produced by a few contacts (between 6 or 7 and 15) of the receptor’s amino acids with related groups from the hormone’s structure. Some basic amino acid residues, particularly from arginine, which are preserved virtually intact among receptors, are critical in the execution of this function (Quingley et al. 1995).

The LBD harbors a zone of interaction with hsp90. When the hormone binds with the corresponding domain in the receptor, the protein changes its conformation, losing its affinity for hsp90. As a result, the receptor loses its affinity for hsp90.

As previously noted, the LBD of the receptor presents a series of functions that are not very well delimited such as those of dimerization with another receptor, nuclear translocation, and activation of the ligand-dependent gene transcription. As was just mentioned, the interaction of a ligand with its receptor has as its immediate consequence the conformational change of the molecule, a change that also determines the molecule’s functionality. The importance of this point is that the stated conformational change is predetermined by the chemical nature of the ligand and the form in which it interacts with the receptor.

This can be verified easily if one analyzes the changes that take place in this zone when both the ERa and ERβ bind to agonist or antagonist ligands. By means of crystallography studies, it has been verified that the binding of agonists such as estradiol or diethylethylbestrol (DES), or even partial antagonists like raloxifene or 4-OH tamoxifen, induces different three-dimensional conformations in both isoforms, affecting the spatial disposition of the LBD zone and, therefore, the functionality of the molecule. This could explain why a drug such as tamoxifen behaves like a partial agonist in the case of ERα and, in contrast, like a complete antagonist when interacting with ERβ, or why the phytoestrogen agonist genistein has 30 times more affinity for ERβ than for ERa, when the homology between both isoforms in terms of their tertiary structure is very high (Barkhem et al. 1998).

Fig. 1.4. Hormone-binding domain. Estrogen receptors concentrate in the extensive hormone-binding domain (> 300 amino acids) several functions with a common characteristic: they only manifest themselves when the receptor has been bound to the hormone and a change in its three-dimensional structure has been produced. The hormone-binding domain forms a bag-shaped structure, hydrophobic in nature, that lodges the hormone. The interaction occurs between specific atoms of the hormone and residues of specific amino acids of the receptor that at times are very distant in the primary sequence but that are next to each other in the three-dimensional structure of the receptor. Estrogens interact with a few concrete amino acids of this domain, while antiestrogens and SERM (selective estrogen receptor modulators) are in contact with some of the same amino acids as well as with some others. The result is a structural folding different for the receptor as a function of the ligand with which it interacts


Structure of Receptor and Hormonal Antagonism

As was previously established, the spatial structure of the receptor domains is altered by interaction with the hormone, with DNA, with other proteins, and by the state of the receptor phosphorylation. Different states of folding suppose that the receptor exhibits different surfaces that permit it to gain or to lose affinity for DNA sequences or for proteins, as they are components of the native receptor or of the transcriptional machinery. The different properties that characterize the receptor are fully manifested only when the adequate spatial distribution of the molecule is reached (Brzozowski et al. 1997; Castellano-Di'az et al. 1989; Edwards et al. 1995; Edwards et al. 2002; Jordan et al. 1990).

Fig. 1.5. Activation of the native receptor by the hormone. The hormone-receptor interaction determines a very strong bond that attracts distant amino acid residues, which alters the three-dimensional structure of the receptor. As a consequence, the receptor loses its affinity for the proteins that were originally close but that no longer find their zones of contact with the receptor. Simultaneously, the receptor reorganizes other hormone-dependent zones: it acquires dimerization capacity and exhibits a capacity to bind to DNA and to transcription factors. The interaction with antiestrogens also produces a conformational change, which can give rise or not to the formation of dimers, in any case with a different conformation

The interactions of the receptor with structural analogs of the natural hormone give rise to conformation changes that can be similar, slightly different, or totally different from those produced by interaction with its natural ligand. A structural analog that produces a folding of the receptor that is very different from the normal one will give rise to nonproductive configurations from a transcriptional point of view. This would occur in the case of hormonal antagonists or antihormones that would blockade the receptors into a state of incapacity to induce gene expression. As a consequence, they would not manifest the physiological effects of the hormone (Gruber et al. 2002; MacGregor et al. 1998; McDonnell et al. 1994; Nilsson et al. 2001; Shiau et al. 1998; Wakeling 1993).

On the contrary, if the structural analog structure produces a folding sufficiently similar to the normal one, it can give rise to interactions with diverse degrees of transcription capacity. This is the case of analogs that function like partial agonists. In this case, the context of the gene promoter plays an important role, so that the complexes formed are capable of inducing the transcription of some genes but not of others. The cell context is also important since different cell lineages contain distinct transcription factors. This way, the same product behaves like an agonist in some cells and in others like an antagonist. Everything depends on whether or not the configurations attained are capable of interacting with the present transcription factors. Tamoxifen is the paradigm par excellence of the partial estrogen agonist. It functions like an antiestrogen in human breast cancer and as estrogen in the liver (MacGregor et al. 1998; Shiau et al. 1998; Tzukerman et al. 1994).

It is precisely in the so-called transcription activation zone-1 (TAF1) where we find significant differences between both kinds of ER. If in ERα this function represents a fundamental role in the specific activation of various genes in experiments carried out with cell lines, it has been verified that, in the same conditions, the TAF1 of ERβ practically does not intervene in such processes (Cowley et al. 1999). Similarly, the interaction of both receptors with specific ligands presents certain similarities and differences. Thus, synthetic antiestrogens such as tamoxifen, raloxifene, and ICI 164.384 present partial estrogen activity when they are bound to ERa, since they manage to induce the gene transcription mediated by this receptor. Meanwhile, they are pure antagonists for ERβ (McDonnell et al. 1995). The contrast may be explained by differences found in the TAF1 zone for both receptors. It is known that two different zones exist inside the TAF1 in the alpha isoform, both of which are necessary for the agonism with estradiol and for the partial agonism with tamoxifen, while in the beta form this dual function of the AF1 zone has not been detected (McInerney et al. 1998). Therefore, it is fitting to conclude that, based on these observations, the exact function of the TAF1 zone in ERβ, as opposed to that in the alpha isoform, still remains without clarification.


Receptor Folding in Separate Domains

A truncated ER lacking domains A, B, and C still grasps estradiol with high affinity. This indicates that these regions do not participate in the binding to hormone. It also indicates that domains D, E, and F fold themselves autonomously, reaching the necessary configuration to bind with the hormone. It is presumed that region D, which connects the LBD with the DBD, functions like a hinge pin, keeping apart two areas functionally separated in the protein. Region C tends to form a partly autonomous, very compact structure with regard to domains A and B. It is possible to idealize the receptor as an assembly of three separate structures of folding, where the LBD would be able to rotate with ample freedom with regard to the other two structures (Evans et al. 1988).

The simultaneous presence of several functions in the same zone of the molecule is something that should not surprise. The LBD is sufficiently extensive so that different amino acids participate in distinct functions. When these functions reside in neighboring sequences that overlap, functions demonstrate themselves successively: the receptor binds the hormone, then it loses affinity for hsp90, soon after that it gains affinity for another receptor and dimerizes, and finally it earns affinity for other cofactors of transcription. Each step is necessary so that the following can be taken in the process of activation.


Dimerization Domains

The formation of dimeric structures - homodimeric in the case of the ER, heterodimeric in the case of the thyroid hormone with a retinol receptor - is very common among the proteins that regulate gene transcription. The dimers embed themselves in the greater furrow of the DNA double helix and in this way facilitate the interaction between specific amino acids and nucleotides. It has been shown that the receptor dimers, and those of other regulating proteins, produce angulation of the double helix. This process facilitates the fixation of other components of the transcription machinery and the initiation of the transcription (Lee et al. 1989). The formation of dimers plays a central role in the recognition of genes regulated by the hormones.

Dimerization is a process necessary for the receptor to carry out its interaction with DNA and to initiate the response to the hormone. Dimerization occurs when the receptor monomer has freed itself of hsp90 and the other accompanying proteins forming the structure of the native receptor. Moreover, binding to the hormone provides the receptor with the necessary threedimensional structure to produce the interaction between the two receptor monomers (Kuiper et al. 1996; O’Malley 1990).

At least three regions of the receptor participate in the process of dimer formation. One of them is unspecific and is made up of the sequences of hydrophobic amino acids of the LBD. These form hydrophobic contact surfaces that facilitate, in a general way, the interactions among proteins. The other two are specific sequences of amino acids. One of them is situated immediately after the DBD. It is comprised of a group of some 20 amino acids, and its capacity to intervene in the dimer is independent of binding to the hormone. The other dimerization region is found inside the LBD. It is poorly located, and it is possible that noncontiguous sequences of amino acids participate in it. It is exhibited only when the receptor has been already bound to a hormone.

The formation of ER dimers can be favored once the first monomer has bound to the DNA, since this presents positive cooperation in binding the next monomer. In any case, DNA binding creates a greater compaction of the dimer that results in a subsequent spatial restructuring of the receptor molecules.


Receptor-Genome Interaction

Cell differentiation during the embryonic period has as consequences that the majority of genes remain definitively silenced and that only a reduced number can be expressed in each cell (Beato et al. 1996; Beato 1989). The last group of genes constitutes the patrimony of each differentiated cell lineage and includes two subgroups: the genes that are expressed constitutively and those that are inducible and/or repressible. The latter are the object of regulation by factors internal or external to the cell, for example the hormones. It is the task of the nuclear hormone receptors to recognize which genes are susceptible to respond to a specific hormone.


Specific DNA Sequences for the Hormone Response

The identification of the few genes regulated by hormones, among the multitude of the genes that are expressed in each cell, is a first-order problem. What makes the identification possible is the existence of some short specific sequences of DNA, situated in the promoter region of each gene, that are recognized by the dimer of the hormone receptor. These sequences are called hormone response elements (HRE) (Seiler-Tuyns et al. 1986; Tzukerman et al. 1994).

The genes that respond to a specific hormone contain identical HRE (Fig. 1.6). Normally, it is a matter of short nucleotide sequences: pentamers or hexamers. In the case of the ER, the sequences are found repeated in inverse order in the same strand of DNA (palindromic, or symmetrically legible sequences: 5'GGACA-nnn-ACAGG 3'; n is any nucleotide). In the case of the thyroid hormones and retinoic acid, the HRE at times are presented like two repeated sequences in the same order (direct repetition: GGACA-GGACA).

Generally, between the two halves of the palindrome (or of the direct repetition) there are from one to five nucleotide spacers whose sequence varies from one gene to another. The sequence in which these nucleotides are found is irrelevant, as they do not directly participate in the dimer-DNA interaction. It is very important, however, that the number of nucleotide spacers be fixed to allow for correct binding to its corresponding receptor dimer.

Note that there is a great similitude among all the known HRE for nuclear receptors. Two subgroups have been established in which the sequences are practically identical: the subgroup of the glucocorticoid receptor, which utilizes the pentamer sequence GGACA and also includes the progesterone, mineralocorticoid, and androgen receptors (Seiler-Tuyns et al. 1986) and the subgroup of the ER, which utilizes the pentamer GGTCA and also includes the receptors for vitamin D3, thyroid hormones, and retinoic acid (Bouillon et al. 1995; Tora et al. 1988). The first two utilize the palindromic system, and the last two can utilize the palindrome or the direct repetition, depending on the receptor subclass (а, β, or у).

Fig. 1.6. Binding domain to DNA. ERs contain two structures called zinc fingers, typical of proteins that interact with DNA. One zinc atom forms four links of coordination with four cysteine residues of the protein structure, which occupy nearby positions, thus leaving aloop ofsome 15 to 22 aminoacids. The zinc fingers of the receptor are capable ofinteracting with specific sequences of DNA, the hormone response elements, with which they establish hydrogen bridges and form stable structures

The structure in palindrome or in direct repetition and the size of the spacing sequence between the two pentamers are the critical variables in establishing the specificity of response to each one of the receptors that share the same pentamer. For example, the elements of estrogen and thyroid response can share the same palindrome and be differentiated only by the number of nucleotide spacers: three for the first one, none for the second.

HRE are not always a perfect palindrome, nor are pentamers repeated perfectly. It is frequently sufficient that one of the pentamers be the one that corresponds to an HRE. In the second pentamer, there can be a different nucleotide without altering the interaction with the receptor in a noticeable way.


DNA-Binding Domain

The interaction of the receptor with the HRE is produced once the dimer of the receptor has been formed. Given that the majority of HRE are palindromic, the interaction requires that the dimer be formed symmetrically facing both receptor monomers.

The DBD resides in a zone rich in cysteine in domain C of the receptor. This region is characterized by an interaction among four cysteines next to an atom of zinc (Fig. 1.6). The zinc atom stabilizes the structure by means of four coordination links to form what are called zinc fingers. The nuclear receptors form two zinc fingers per molecule. The zinc fingers contain a quite constant number of amino acids (18 to 22), and the space between the fingers is filled by a group of amino acids that varies from one type of receptor to another (Evans et al. 1988; Klug et al. 1987).

The zinc fingers are common structures among the transcription factors. Nevertheless, the coordination with zinc is more frequently produced between two histidine residues and two neighboring cysteines than when it is among four cysteine residues, as occurs in the nuclear hormone receptors. The zinc fingers provide an optimum architecture for the mutual recognition between specific sequences of amino acids and nucleotides. In the case of the nuclear receptors, the interaction occurs between particular amino acids of the DBD and guanine residues of the DNA sequence (Fig. 1.7).

In the recognition of each pentamer of the HRE participate two groups of amino acids, one from each zinc finger, perfectly preserved along evolution. The first, or proximal, group is situated in the nodule of the zinc finger and participates with three amino acids. The distal group participates with three other amino acids and seems to be essential to the recognition of the segment

Fig. 1.7. Interaction of receptor with hormone response element. The hormone response elements are located in the promoter region of genes regulated by hormones with nuclear receptors. They are constituted of sequences from 13 to 15 nucleotides. The elements of estrogen response are formed by two semi-elements, which are sequences of 5 nucleotides, and a spacer of 3 unspecific nucleotides (n). The interaction is produced so that the section of the zinc fingers of the receptor lodges in the main furrow of the DNA double helix spacer between the two pentamers of the palindrome (Freedman et al. 1988; Filardo 2002).

The interaction between the receptor dimer and DNA is produced in an orderly manner. First, the dimer is placed in the main furrow of the double helix, and the first monomer interacts with the first pentamer of the HRE in the main furrow of the double helix. Later the second molecule of the receptor dimer binds to the second pentamer. The distance between both pentamers is minimum: from zero to five nucleotides, depending on the type of receptor. This implies that the dimer assures a sufficiently compact and symmetrical structure among both receptor molecules, so that a similar intimacy can be produced in the association with the palindrome.


Recognition of Hormone Response Element

The native receptors of steroid hormones, in the absence of hormone, have little affinity for nuclear structures. Contact with the hormone augments this affinity to an extraordinary degree. This fact does not necessarily show that the hormone has caused the receptors to enter into contact with specific DNA sequences in the genes that respond to the hormone in question (Beato et al. 1996; Beato 1989). It has been calculated that in a normal target cell there are only a few genes that can respond to a hormone. Each target cell contains from 1000 to 10,000 receptors, which become activated in increasing number, in function with the concentration that the hormone reaches in the cell. It is not possible, therefore, for all the activated receptors to find specific genes with which to interact.

The excess of the activated receptor, at least in the case of the progesterone, binds with acid proteins that function like acceptors. Two possible functions are attributed to these proteins that have not been confirmed: that of being an active receptor reservoir and that of being responsible for directing the excess of the receptor toward degradation (Filardo 2002; Gruber et al. 2002).

The dimer of the hormone-receptor complex should scrutinize an infinity of sequences before finding its HRE. The role of the hormone in the recognition of the HRE seems to be that of dramatically increasing the velocity of DNA sequence recognition, that is to say, it binds and disconnects more quickly to sequences of nonspecific DNA. When it finds the sequence of its HRE, a bond of affinity is formed that is similar to that of hormone-receptor interaction (Kd in the nM range).

The state of the chromatin has influence as well on the velocity with which is produced the recognition between the receptor dimer and the HRE sequences. The inactive heterochromatin has methylated histones, so that a different compactment from that of the nucleosomes is produced that is characteristic of genetic inactivation. The active chromatin presents different degrees of acetylation, and those regions are more relaxed and permit the dimer to move more freely to scrutinize until it finds the HRE. These are found in accessible places of the nucleosome so it is not necessary for the dimer to travel the entire length of the DNA strand.

The normal form of interaction between receptor and DNA requires the hormone to have broken the native structure of the receptor and the dimer to have been formed. That is to say, the receptor-DNA interaction comes after the hormone-receptor interaction. Nevertheless, situations have been described in vitro in which the receptor is able to be previously associated to the HRE. This situation occurs in vivo for the thyroid hormone receptors, in which case it seems that the hormone-free dimer acts as an expression repressor of genes dependent on these hormones (Evans et al. 1988). The arrival of the hormone activates the dimer in situ and inverts its role as regulator.


Role of Receptor-Hormone Response Element Complex

The sequences of the HRE are situated in the promoter region of the gene. In a zone close by (less than 100 nucleotides away) and always between the HRE and the point of initiation of the transcription, there is a sequence rich in thymidine and adenosine (TATA, or its equivalent) on which the RNA polymerase II attaches itself (Beato et al. 1996; Beato 1989; Chin 1995; Gruber et al. 2002; Nilsson et al. 2001).

Once the interaction receptor dimer-HRE of the DNA occurs, a very fast progression of events is produced (Fig. 1.8). The receptor dimer causes a curvature in the structure of the double helix in the neighborhood of the region next to the starting point of the gene transcription. This curvature implies a structural change that permits the RNA polymerase II to accede to the TATA-rich sequence of DNA. The RNA polymerase II recruits some transcription factors and forms the transcription preinitiator complex on the sequence of TATA (or its equivalent one) (Klug et al. 1987; Nilsson et al. 2001).

The receptor dimer, associated with the HRE, will attract other transcription factors, with which the protein-protein interaction is produced. Finally, they will come together with the RNA polymerase II and the remaining transcription factors that formed the preinitiator complex to complete the machinery of gene transcription. The role of the receptor dimer is, therefore, that of assuring the correct anchorage of the transcription factors in the promoter region of the gene so that the functional assembly of the machinery of gene transcription is produced.

Fig. 1.8. Activation of gene expression. The interaction of the receptor dimer with the estrogen response element, located in the promoter region of the genes that are estrogen activatable, enables the recruitment of cofactors (coactivators) of transcription. The dimer of the receptor establishes a connection through coactivators with the basic transcription machinery that is associated with the promoter region of the gene in the TATA region (or similar region, depending on the gene). The consequence is that the transcription machinery becomes activated by one or two orders of magnitude and multiple copies of the mRNA of the gene begin to appear. In the case of a pure estrogen, like estradiol or DES, the activation will affect all of the estrogen-dependent genes and in all the estrogen target cells, although in varying intensity depending on the gene since not all are equally sensitive to estrogen

The structure of the chromatin and their state of acetylation are important at the moment of initiating the gene transcription. Indeed, some of the transcription factors recruited by the receptor dimer have histone-acetyltransferase activity that permits the gene transcription after diminishing the condensation of the chromatin (Gruber et al. 2002; Nilsson et al. 2001; Vigushin et al. 2002).

Well-documented cases exist where the estrogens inhibit the expression of some genes. These are usually transcribed by means of the constitutive activity of powerful promoters. The inhibition is a result of the steric interposition of the receptor dimer in the development region of the gene, which thereby recruits corepressors that interrupt the prior instigator effect in the absence of receptor (McKenna et al. 1999; Mester et al. 1995; Smith et al. 1997; Tora et al. 1989).


Hormonal Regulation of Gene Transcription

The final phase of action of the hormones that utilize nuclear hormone receptors lies in the modification of the gene transcription. In spite of the enormous effort expended, it is a process that remains only partly understood. This is due to its extraordinary complexity and to the multiple varieties that show up as gene or tissue specific.


Domains of Transcription Activation (Transactivators)

The participation of the nuclear receptors in the machinery of gene transcription takes place by means of specific domains of the molecule known as transactivators (abbreviation for transcription activators). These are made up of sequences of amino acids that interact by means of protein-protein contacts with other transcription factors. The artificial alteration of these sequences has as a consequence the inability of the hormone to induce gene expression (Beato et al. 1996; Klug et al. 1987; Lones et al. 1995).

At least seven proteins, besides the RNA-polymerase II, participate in the transcription machinery. The initiation of the transcription occurs when the transcriptional complex in the promoter region of the gene has been stabilized. The receptor dimer forms a complex of high affinity with the sequence of the HRE. This binding provides a firm base for the anchorage and stabilization of the transcriptional complex. The dimeric structure of the receptor acquires affinity to attract different coactivators that bring together the proteins of the transcriptional complex (Fig. 1.9).

As mentioned in a previous section of this chapter, there are two fully identified transactivator domains in the family of nuclear hormone receptors. One of them resides in the region preceding the DBD and is independent of the binding to the hormone. It is called TAF1 (trans-activation factor 1) and is transcriptionally active in the absence of the LBD (Tora et al. 1989). TAF1 is regulated by means of phosphorylation and can form part of signal transmission systems from the cell membrane. These can recruit the ER to activate some genes that have elements of estrogen response in the absence of estrogens (Filardo 2002; Lee et al. 2002; Osborne et al. 2001; Segars et al. 2002).

The other transactivator domain, TAF2, is found immersed in the LBD and acts only when the hormone-receptor complex is formed. A sequence of 15 well-conserved amino acids from the different members of the family of nuclear receptors, and situated very close to the carboxyl end of the receptors, participates in it (Gruber et al. 2002; Nilsson et al. 2001).

The transactivation domains only make their accessibility evident in the dimer bound to the HRE. It is very probable that, in this way, the spatial structure (tertiary) optimizes itself so that the contact surfaces between the receptor and the other cofactors of transcription are formed.

Fig. 1.9. Coactivators, corepressors, and the binding domain to the hormone. It has been possible to express the hormone binding domain in bacteria and to obtain great quantities in apure state. This has made it possible to analyze its crystalline structure bound to ligands and to determine how this influences the recruitment of coactivators and corepressors. Several parts of the domain experience changes that justify its activity, but the most important one is helix 12. In the absence of estrogen, helix 12 leaves a hydrophobic cavity uncovered where corepressors containing zipper sequences of leucines lodge. In the presence of estrogens, helix 12 blocks that cavity and the corepressor does not fit. At the same place (or in the neighborhood) a new site is created that interacts with the complementary domains of the coactivators, thus initiating the bridge that connects with the transcription machinery, which is finally activated


Intermediary Transcription Cofactors

Among the proteins that form part of the transcription machinery are found some cell factors that are produced in limited quantities. They are called cofactors of transcription (NCoA, for nuclear-receptor coactivator; NCoI, for nuclear-receptor coinhibitor), formerly known as transcription intermediary factors (TIF) (McDonnell et al. 2002; McKenna et al. 1999). They constitute one of the classes of proteins that form part of the transcription machinery. These proteins are utilized by diverse types of intensifiers, that is to say, by sequences of DNA that anchor transcription factors, of which HRE are a particular case (Gruber et al. 2002; Mester et al. 1995). They do not interact directly with the DNA, but they do with the receptors and with the other elements of the transcription apparatus (Fig. 1.9).

The participation of the different cofactors that form part of the transcription machinery is not homogeneous. Some, like p160, can interact with both transcription activator domains of the receptor, TAF1 and TAF2, even though they utilize different p160 domains. Others, like CBP/p300, do not enter into contact with the receptor, but do with other coactivators, such as p160. It seems clear that each protein-protein contact causes conformational changes in them, so that new affinities for other coactivator proteins arise (Fig. 1.10). The final result of those contacts is the assembly of a transcription machinery that functions at full steam (McKenna et al. 1999).

In the moments prior to initiation of the transcription, a true rivalry is established by the transcription coactivators. The affinity with which the receptors are capable of interacting with the coactivators is decisive at the moment of including these in the transcription machinery and inducing the gene expression. If at the same time other machineries are themselves configuring transcriptions that capture these intermediary factors more efficiently than do the receptors, then the hormone-regulated transcription of the gene does not occur. In this case, the expression of such genes will have to wait for more favorable conditions for the transcription to appear.

Fig. 1.10. Crystallography of receptor bound to estradiol or SERMS. A Crystallographic structures of binding domain to hormone of estrogen receptor alpha bound to estradiol (E2) and to 4-OH-tamoxifen. Gray: Parts of domain that do not experience changes bound to ligand. Green: Changes induced by estradiol. Red: Changes induced by 4-OH-tamoxifen. Notice that helix 12 of the domain bound to 4-OH-tamoxifen occupies a position salient and perpendicular to that occupied by the same helix in the case of binding to estradiol. B Changes induced by estradiol (green) and by raloxifene (yellow)inthe crystallinestructure of the binding domain to the hormone. Raloxifene causes a change in the position of helix 12 that is different from that of estradiol, although it is not as dramatic as in the case of 4-OH-tamoxifen. In each cell type, interactions by coactivators or corepressors with the new structures of the domain formed will be produced, a process that will depend on which of those coregulators are present


Interaction of Receptor with Transcription Cofactors

In the absence of hormone, the three-dimensional configuration of the receptor favors binding to corepressors present in the cell nucleus. The interaction is produced at the level of zipper-type sequences of leucines (-L-X-X-L-), present in the corepressors, with the LBD of the receptor. This has a structure that is complementary to the leucine zipper, which remains accessible while the receptor itself does not bind to the hormone (Gruber et al. 2002; Nilsson et al. 2001).

The spatial conformation that the ligand-receptor acquires, particularly the spatial disposition that helix 12 of the LBD attains when it binds to estradiol, is key for the subsequent recruitment of the transcription cofactors (Fig. 1.9). Indeed, the arrival of estradiol restructures the entire domain, making helix 12 rotate and close the hole where the leucine zipper sequence of the corepressor had been lodged before (Fig. 1.10). Consequently, both molecules, corepressor and receptor, lose their affinity and their bond is undone. Another structure capable of interacting with gene transcription coactivators is formed at the same place on the receptor (MacGregor et al. 1998; McDonnell et al. 2002).

Hormone agonists share various contacts with the amino acids of the LBD, although these are of variable intensity. The result of these interactions is a ranking of hormonal power the different agonists display (Cosman et al. 1999; Kelly et al. 1999; Jordan 2001; McDonnell 1999; Shang et al. 2002). The antagonists interact with a part of the same amino acids as the agonists, but these interactions include other contacts (Chan 2002; Jordan 2002; Riggs et al. 2003). From this interaction of the antagonists a structure of the receptor is created that varies as a function of the ligand, and this is reflected in the resulting crystallographic aspect of the LBD when occupied by different agonist or antagonist compounds (Fig. 1.10).

When bound with pure antagonists, the configuration of the LBD is such that helix 12 does not rotate to undo the binding site of the corepressor (Jordan 2002; Riggs et al. 2003) (Fig. 1.11). On the contrary, the binding of an agonist to the receptor causes the production of a structure that is more similar to that formed upon binding to estradiol. Nevertheless, each agonist creates a different tertiary structure that, consequently, presents slight variations as to the spatial conformation in which the coactivators need to lodge.


Coactivators in Cellular and Gene Promoter Context

Depending on the type of coactivator(s) present in each particular cell type, a productive or unproductive receptor-agonist-coactivator bond can be created. If the bond is productive, the compound behaves like an estrogen in that cell type; on the contrary, if the bond is unproductive, the compound can block the activity of a concurrent estrogen compound in the cell, and therefore the compound behaves like an antiestrogen (Jordan 2001; Riggs et al. 2003).

Fig. 1.11. Pure antiestrogen effect. The conformational change produced in the zone of binding to the hormone by a pure antiestrogen can be of a nature that incapacitates the receptor for dimerizing, making it fragile to the attack of intracellular proteases (Faslodex). It is also possible that the pure antiestrogen confers upon the receptor a conformation that incapacitates it from interacting with coactivators so that it cannot form the bridge with the transcription machinery. Finally, it is possible that the conformation acquired by the receptor recruits corepressors instead of coactivators, thus inhibiting the synthesis of mRNA. For a compound to be considered a pure antiestrogen, it must interfere with the estrogen-dependent gene expression in all cell types

An additional variable to consider is how the subsequent interaction of the coactivators with other proteins of the transcription machinery is affected. This interaction occurs in the context of the promoter of each particular gene.

The bond of the receptor dimer with the nucleotide sequence of the HRE in the promoter region of the gene is what directs the assembly of the proteins (up to 19) that yields the transcription machinery. The operation of the machinery depends on the continual, sequential reestablishment of protein-protein contacts. Each new interaction depends on whether the previous proteins had assembled themselves correctly in such a way that the protein under consideration does not bind unless the prior interactions have created the appropriate surface of contact.

Agonist and antagonists not only modify the three-dimensional structure of the receptor, they also modify the three-dimensional structure of the coactivator and help to create the contact surface with the following protein. It is easy to imagine that small variations in the conformation of the site of interaction between the receptor bound to the agonist and the coactivator can create spatial orientations that can be incorrect in the context of the promoter of each particular gene (Jordan 2001; Riggs et al. 2003).


Concept of SERM from Point of View of Coactivator

For any substance with potential estrogen activity it is necessary to consider whether the configuration that the receptor acquires upon binding is capable of interacting correctly with the coactivators present in the cell. It is also necessary to consider whether from an imperfect interaction between receptor and coactivator the capacity to activate can be deduced from the expression of some, several, or all the genes that have HRE.

The concept of SERM (Selective Estrogen Receptor Modulator) refers to compounds capable ofbinding to the ER and to have an extensive range of cell responses that go from the net estrogen to antiestrogen activity (McDonnell 1999; Riggs et al. 2003; Shang et al. 2002). From what has previously been presented, it can be deduced that depending on the cell context, there will be coactivators that are either capable or incapable of binding to each receptor-SERM complex or of doing it in such a way that these can activate some genes with determinate promoter conformation, but not others (Fig. 1.12).

What will occur with a SERM in a particular tissue is unpredictable. Its behavior depends on at least two factors:

- The availability of coactivators in that cell line that recognize the receptor- SERM complex that, at the same time, is subject to a regulation of its expression, to competition, as they may be in the process of being recruited by other receptors, etc.

- The context of the gene promoter being considered that has some specific conditions for accepting activation by particular conformations of the transcription machinery.


Structure of Chromatin and Hormone Response

The curvature effect of the double helix of DNA, caused by the binding of dimers of active receptors to the HRE sequences, has been obtained by means of experiments of transfection of lineal DNA structures to cells that previously did not express the gene under study. The reality of the cells in vivo must be much more complex (Nilsson et al. 2001; Vigushin et al. 2002). In the regulation of gene expression in vivo, the structure of the chromatin participates decisively. This represents the effect of cell differentiation on the accessibility of only determinate genes to induction by hormones.

Fig. 1.12. SERM or not SERM, that is the question. The spatial conformation that a compound with SERM activity confers to the estrogen receptor gives it the capacity to interact with determinate coactivators or corepressors, but not with others. Depending on the cell lineage, and therefore on the collection of coactivators and corepressors present in that cell, this will activate the expression of some estrogen-dependent genes and the repression of others. One gene induced by estrogens in several cell types containing different coactivators is activated by SERMs in some cell types but not in others. In this case, the promoter context of each gene plays a determining role in whether that gene is or is not activatable by a determinate SERM

The Beato group has studied in depth the influence of the nucleosome structure in response to glucocorticoids (Beato 1989). Nucleosomes are formed by segments of 120 nucleotides of the double helix of DNA that make two twists around an octamer of histone. There are 200 nucleotides between two consecutive nucleosomes, so that a gene normally has tens of nucleosomes.

The structure of the promoter region of some of the genes studied overlaps a nucleosome in such a way that the RNA polymerase II cannot get to its binding site. The interaction between the receptor dimers and the HRE causes the nucleosomes to have their structures altered, either by being displaced or by having their structure come partly undone. This change in the configuration permits the fixation of the RNA polymerase II, with which the transcription can be initiated.

The histones, which provide the nucleosomes with their structural base, are susceptible to acetylation. Receptors bound to antagonists, or even hormone-free ones, are available to bind with corepressors that recruit histone- deacetylase, a process that provokes the contraction of the nucleosomes and impedes gene transcription. In contrast, the agonist-receptor dimer attracts coactivators, some of which directly have acetyltransferase activity, or even recruit other coactivators that have that enzymatic activity. The result is the acetylation of the histones and the consequent relaxation of the nucleosome structure, which permits the transcription of the gene regulated by the hormone.


Specificity of Gene Transcription Induced by Hormones

The specific transcription of the genes depending on each hormone runs up against the evident similarities among HRE for each hormone. For example, HRE are identical for androgens and progesterone (Navarro et al. 2002). This seems to introduce a certain degree of confusion at the moment of assuring a correct hormone-specific gene transcription. In other words, since the resemblance among HRE does not offer guarantees of specificity in the response, other elements have to exist that guarantee it.

The fact that a receptor dimer identifies a HRE does not assure, by itself, the transcription of the gene. This is a necessary, but insufficient, condition. Once the dimer-HRE interaction has been produced, the machinery of transcription needs to be assembled, requiring the binding of other intermediary cofactors. Some of these are tissue specific, and others recognize only a particular receptor dimer, thus obviating others that could recognize the same HRE.


Multiple Regulation of Gene Expression

Various elements of response to regulatingt ranscription factors concur in a real promoter region of a gene. This reflects the complexity of situations influencing transcription in response to different signals. One of these, necessary but not totally sufficient for the induction of maximum transcription, is the receptor dimer. Others may be the protein that mediates the transcriptional response to cyclic AMP or the AP1 that recognizes fos-jun dimers (Gruber et al. 2002; Nilsson et al. 2001). Each one of these DNA sequences recognizes its own coactivators, and all can be simultaneously present in the promoter region of the gene, offering a variety of possible interactions.

The situation is still more complex. Thus the presence of elements of response from signaling pathways that regulate the expression of the gene in different directions may be detected in the promoter region. These elements recruit coactivators and corepressors, repression being one form of hormonal regulation of gene expression. It is obvious that the transcription machinery of such genes will couple or uncouple as a function of the relative influence of the cofactors that intervene in each moment.

The formation of a transcriptionally active complex requires the interaction of all transcription cofactors with their respective specific DNA sequences. Once they have bound to their specific sequences, it is on these that the remaining elements of the complex that do not interact directly with the DNA are assembled (Chin 1995; Filardo 2002). The elements of the complex that do not come into direct contact with DNA have their own specificity of interaction with the remaining proteins of the complex. Therefore, they include important restrictions so that a fully active transcriptional complex can be assembled with difficulty on a receptor dimer that has incorrectly recognized a HRE. Indeed, an incorrect interaction can imply a noticeable degree of transcription inhibition.

Many aspects relating to the specificity and intensity of gene transcription in response to hormones remain open. Nevertheless, a prudent conclusion permits establishing that two definite elements are intervening: the interaction of the receptor dimer with the palindrome and several protein-protein interactions that are produced between the dimer and the remaining components of the transcription machinery (Beato 1989).


Nuclear Hormone Receptors and Endocrine Disruptors

The chemical structure of the substances capable of interacting with a determinate nuclear receptor is tremendously varied. For now no pattern exists that permits one to assure that a particular substance is going to interact with the receptor to produce an agonist or antagonist effect. In recent years the concept of “endocrine disruptors” has been introduced to describe the substances that are capable of modifying the endocrine equilibrium. Some of them act by binding with nuclear hormone receptors, while others interfere with the processes of regulation of hormone secretion (Lathers 2002; Melnick et al. 2002; Nakata 2002; Powles 2002; Brown et al. 2002; Sonnenschein et al. 1998).

Endocrine disruptors apparently affect all nuclear receptors. Thus, a notable increment in impotence, alterations of the libido and of oligospermia in workers exposed to pesticides has been described. These alterations are due to the action of some compounds with estrogen-mimetic action and to their interaction with the androgen receptor. Additionally, alterations of thyroid function have been detected in rats exposed to dioxin and other toxic agents, though it is not sure if this effect is produced by direct interaction with the thyroid hormone receptor.

Among the endocrine disruptors that interact with the nuclear hormone receptors are the chemical substances that have an estrogen character. Thus, it has been discovered that substances as dissimilar as polychlorated biphenyls used as pesticides, numerous vegetable compounds (phytoestrogens), and components of plastic, paint, and detergent have weak estrogen activity. Given the abundance of these compounds in the diet or in western lifestyles, the substances are partly blamed for the growing incidence of breast cancer (Anderson 2002; Badger et al. 2002; Clemons et al. 2001; Colditz 1998; Jacobs et al. 2002; Safe 1998).


Regulation of Intensity of Hormone Response

The molecular details of the mechanism of hormonal action do not always clarify the numerous unknowns of the way in which the intensity of the hormone response is regulated from cell to cell and from minute to minute. There are numerous factors implicated in this process directed at achieving the greatest functional equilibrium of the organism. From the entry of the hormones into the cell to the conclusion of hormonal action, an ensemble of factors arises that intervenes in the process in a decisive way.


Membrane Receptors for Steroid Hormones

The entrance of steroid hormones into the cells has always been assumed to be a passive phenomenon, based on its solubility in the phospholipids of the cell membrane. Nevertheless, the existence ofspecific fixation ofsteroid hormones to cell membranes has opened the possibility of their entrance into the cells mediated by proteins of the membrane (Levin 2002). Nevertheless, it has not been possible to verify that they participate in some way in the transportation of steroids to the interior of the cell (Beato et al. 1996; Beato 1989). For them, other possible extragenomic actions have been postulated such as enzymes that participate in the metabolism of hormones or even membrane receptors (Beato et al. 1996; Chirino et al. 1991; Fernandez et al. 1994; Gruber et al. 2002; Revelli et al. 1998).

There is growing evidence that the membrane receptors for estrogens are very important in tissues as the vascular endothelium (Chambliss et al. 2002; Hodgin et al. 2002; Mendelsohn 2002; White 2002). In the endothelial cells ERs appear to be located in specific zones of the membranes called caveolas, but not in the greater part of the membrane. Such receptors mediate rapid responses to estrogens, such as the activation of NOS (nitric oxide synthase) in the vascular endothelium (Chambliss et al. 2002).

There are also numerous enzymes anchored in membranes of the microsomal cell fraction that participate in the metabolism of steroid hormones. Thus, those of the p450 family, which carry out molecular oxidation, or the sulfatases and sulfotransferases, more or less specific to several hormones (Pasqualini et al. 1995). The affinity of steroid hormones for proteins of the membrane (Kd between 10 and 100 nM) is frequently greater than that which some of these enzymes present for their substrates (Luzardo et al. 2000). Therefore, it is unlikely that a part of the proteins of the membrane that bind steroids is in reality enzymes metabolizing these hormones.

In the case of vitamin D3, there is a membrane receptor that, after being bound to this compound, and by the mediation of a G-protein, activates the opening of channels for the entrance of calcium into the cell (Bouillon et al. 1995). There are also membrane receptors for progesterone that mediate, among other processes, the reaction of acrosomes in spermatozoa. Finally, evidence of extragenomic participation of estrogens in exocytosis does exist (Machado et al. 2002).


Regulation of Concentration of Receptors per Cell

Abundance of receptors is one of the most important factors in the regulation of the intensity of hormone response. This depends on the degree of expression of their respective genes and on the speed with which the receptors are eliminated.

During embryonic development, profound changes are produced in the expression of the genes for receptors and in the corporal distribution of the cells capable of expressing them. The thyroid hormone receptors are among the more ubiquitous of this family of receptors and are present in all cells. Nevertheless, both their abundance and the type of receptor that is expressed vary with age and from one tissue to another. The other receptors vary extensively among the different tissues. This unequal cell distribution conditions the response to the hormone, which has given rise to the concept of the target cell.

The gene that encodes for a receptor can be subjected to regulation by signals of diverse origin, as occurs with any other gene. The regulation of gene expression of the nuclear hormone receptors does not follow a single pattern. The hormone itself acts to negatively regulate the gene transcription of the receptor, particularly when the hormone is in excess. This diminishes the protein excess in the interior of the cell. There are, however, some exceptions since physiological doses of estrogens or androgens induce the synthesis of their own receptors.

There are numerous examples of how hormones regulate other receptors. Of recognized physiological importance, the synthesis of progesterone receptors is induced by estrogens in the endometrium, a process that regulates the transition of the proliferative phase to the secretory phase in the menstrual cycle. Additionally, androgen receptors are induced by FSH in Leydig cells in a process that is decisive in the regulation of testicular steroidogenesis (Mester et al. 1995).

An example of the complexity involved in the regulation of nuclear hormone receptors is shown in the case of the ER in the liver. Its synthesis is induced by estradiol, growth hormone, thyroid hormones, and glucocorticoids.


Receptor Destiny After Activation

The regulation of transcription by hormones requires a controlled limitation in order to guarantee that the protein is produced in adequate amounts. This process, nevertheless, is very poorly understood for members of the family of nuclear receptors.

The receptors are continuously subjected to a process of synthesis and destruction, which achieves a steady state. The concentration of receptors in the cell only reflects the situation of the steady state at that moment. As a consequence of hormone action, the number of receptors per cell drastically diminishes in the hours that follow. This observation has led to the postulation that the receptors undergo a process of destruction, or “processing”, induced by the hormone. Despite all efforts, receptor processing has not been deciphered. The destruction of receptors implies the existence of a proteolysis process. Nevertheless, signs of proteolysis, in the form of small peptides of degradation originating in the receptor, have not been detected in the cell. Therefore, if there is a process of receptor proteolysis, it has to be very fast and complete (Beato 1989; Edwards et al. 2002; Kassis et al. 1983).

The possibility of receptor reutilization, once its function has been performed, has been advanced for a long time, but it has been verified only in the case of glucocorticoid receptors (Munck et al. 1995). Very little is known of the details of that process.


Cross-Talk Signaling

Numerous intracellular signaling pathways initiated in the membrane receptors include processes of phosphorylation (Aaronica et al. 1993; Munck et al. 1995). Eventually, nuclear receptors can act as substrata of phosphorylation- dephosphorylationin response to signals originating in other pathways. The state of phosphorylation of the nuclear receptors integrates them within the system of cell membrane signaling.

In this way, the gene transcription activity induced by the nuclear receptors modifies the abundance and activity of the proteins that participate in the pathways of membrane signaling. There also is evidence that the steroid hormone receptors activated by their ligands can interact with elements of the membrane signaling system by activating the pathway in the absence of a corresponding extracellular signal. In this way, a real crisscross of signals from membrane and nuclear receptors is produced, originating in membrane and in nuclear receptors, that maintains the cellular activity integrated into the individual whole.

Cross-talk signaling is an area of very recent investigation that is acquiring greater importance each day. As an example, cyclic AMP, under very specific conditions, enlarges the transcription capacity of ERs (Aaronica et al. 1993). Indeed, the members of the hormone receptor family are the object of phosphorylation. It has been described that this occurs in serine residues for all of them, although it has been described in serine and in tyrosine for ERs (Gruber et al. 2002; Powles 2002).

Regulation of the proteic activity by means of phosphorylation and dephosphorylation is well known. In the case of nuclear receptors, it has been described that the state of phosphorylation affects not only their affinity for the hormone, but also their transcription activity. The process of phosphorylation seems to occur after the receptor binds with the hormone and frees the hsp90, which is a phosphoprotein (Mester et al. 1995).

The enzymes that carry out the processes of receptor phosphorylation are kinases belonging to the signaling pathways of membrane receptors. The pathway of the MAP kinases, activated by different growth factors (EGF, Heregulin, IGF-1, TGF-ALPHA), phosphorylate specifically the ER in the serines S118 and S167 (Gruber et al. 2002; Nilsson et al. 2001; Powles 2002). These serines form part of the TAF1 region of the receptor and are activated by this procedure independently of the binding of the receptor to estrogen. Once phosphorylated, the receptor is capable of dimerizing and activating the expression of some genes in the absence of estrogen.

The cycline-dependent kinases (complex cyclines A/E and CDK2) are able to phosphorylate the ER in serine, particularly in the S104 and S106 belonging to TAF1, and with consequences similar to those of the pathway of the MAP kinases (Osborne et al. 2001; Powles 2002).

Some signal pathways that activate the adenylate cyclase phosphorylate tyrosine residues (Y535) of the TAF2 domain, which, as previously mentioned, is dependent on hormone binding. In this case, modulation of the transcription induction activity is produced in the presence of the hormone (Lee et al. 2002; Osborne et al. 2001).

Inversely, the ER bound to estradiol may be capable of binding with src proteins that form part of the signals transmission complex from the EFG. The consequence is that activation of the EGFR-dependent pathway is produced in the absence of EGF, with the consequent cascade of reactions due to such active agents. It is very probable that this type of reaction mediates the induction of cell proliferation in the endometrium, where estrogen and EGF play dominant, mutually dependent roles that are hardly separable (Lee et al. 2002; Osborne et al. 2001; Powles 2002). Moreover, cross-talk between proteins of the src family and the ER has been described. This signaling includes the phosphorylation of the receptor, in a process initiated by the interaction of src-1 with progesterone (Lee et al. 2002; Miglaccio et al. 1998; Powles 2002).

Experimental evidence exists that some mutated receptors that cannot bind the hormone are constitutionally active from a transcriptional point of view. In these cases, the phosphorylation of the receptor can play an important role in the transcription of some genes in the absence of the hormone.

These findings complete the panorama relative to the mechanisms of hormonal action mediated by nuclear receptors. Thus, gene activation mediated by nuclear receptors can respond to three clearly differentiated modalities: (1) receptor bound to hormone and not phosphorylated, (2) receptor bound to hormone and phosphorylated, and (3) receptor not bound to hormone and phosphorylated (Filardo 2002; Lee et al. 2002; Powles 2002).

Although it is difficult to establish firm conclusions in this area, a prudent formulation of the concept of cross-talk should stress that phosphorylation is a prominent process in the regulation of the activity of the different members of the family of nuclear receptors. This knowledge opens new perspectives in the global comprehension of the processes of cell regulation and illustrates the points of contact among the pathways of intracellular signaling of steroid hormones on the one hand and of the growth factors and peptide hormones on the other. Both pathways were, until recently, considered separate and relatively independent.


Silencing of Genes for Nuclear Hormone Receptors

The technology for the production of animals completely lacking the gene of one of the receptors (knockout mice) has erupted with extraordinary force in the generation of knowledge on multiple facets of hormonal action (Korach 1994). The coincidental discovery of human subjects with a deficit of some of these genes has brought, moreover, the possibility of verifying up to what point the conclusions obtained in mice are applicable to the human species.

Table 1.2. Principal effects of knocking out genes for estrogen receptors a and в in the mouse


Not lethal

Both sexes infertile

Normal RE-β expression


Not lethal

Male fertile; female subfertile

Normal prenatal development of reproductive tract, insensitive to estrogens and antiestrogens

Normal uterus

Normal prenatal and postnatal ovarian development, with multiple nonovulatory hemorrhagic follicles as an adult, 30-40% incidence of ovarian cancer in 18 months

Ovary apparently normal in its development, but does not present normal frequency of spontaneous ovulations.

Normal prenatal development, but insensitive to the development promoted by estrogens during puberty. Sensitive to progesterone and prolactin.

Breast indistinguishable from normal type in virgin mice.

Normal differentiation during pregnancy and lactation.

Normal pre- and postnatal male reproductive tract development. Progressive atrophy with age of rete testis and seminal tubules. Diminution of fertilizing capacity of sperm.

Normal development of masculine tract.

No evidence of problems related to sperm or to fertility.

Females: Neuroendocrine system apparently normal, except for an excess of transcription of gonadotropin genes.

Elevated levels of estradiol, and testosterone and LH, but normal for FSH and progesterone.

Rapid hypocampal response to estradiol maintained.

Normal level of circulating estradiol.

Males: Neuroendocrine system apparently normal, exceptfor an excess of transcription of the gonadotropin genes.

Elevated levels of estradiol, and testosterone and LH, but normal for FSH and progesterone.

Normal level of circulating estradiol.

Females: Mating behavior response lacking under influence of estradiol. Greater aggressiveness and infanticide.

No apparent defects in sexual behavior.

Males: Normal mounting, but without penetration or ejaculation.

No apparent defects in sexual behavior.

Data from Couse JF, Korack SK, Endocr Rev 20:358 (1999)

Particularly exciting are the advances in techniques of molecular biology applied to human endocrinology. The case of a man homozygous for a type of ER truncated at the beginning of the molecule is a good example. The receptor lacked all the functionally relevant domains (Smith et al. 1994). The characteristics of this subject, and the subject’s data as compared with those of “knockout” mice, have revealed the determinant role of estrogens on aspects such as inhibiting growth (the individual continued growing at the age of 27), spermatogenesis (he had oligospermia), or hypothalamic feedback (he secreted an excess of LH despite a normal level of androgens).

The majority of descriptions on the effects of ER gene suppression are anatomical (Table 1.2), although the functional studies in mice are already erupting with force (Couse et al. 1999a; Couse et al. 1999b; Krege et al. 1998). An important surprise from these experiments includes fewer incidents than expected when one receptor is absent, for example, the viability of gametes lacking ERa or the scarcity of defects produced by the absence of a thyroid hormone receptor (Kastner et al. 1995; Korach et al. 1996; Mangelsdorf et al. 1995).

Globally, those experiments with knockout mice suggest that the implication of more than one member of the nuclear receptor family may have prominent effects in organogenesis, even if their expression is for a brief period during embryonic life. The mice with double or triple knock-outs, lacking two or three receptors, will surely contribute to finally clarifying the roles of each hormone and each receptor.



Steroid hormones regulate a very extensive assembly of functions in numerous corporal tissues. Estrogens, the steroid hormones to which the majority of this chapter is dedicated, regulate from basic functions related to reproduction, the development of the skeleton, the maintenance of arterial tension, or diverse nervous functions. The molecular studies on the mechanism of action of estrogens have set the foundations that will permit us to understand how they carry out such diverse functions in such dissimilar tissues as well as how some substances that act through the estrogen signaling pathway can exercise opposite functions in different tissues. In this respect, there are five facts of particular importance that constitute the central nucleus of this revision:

1. There are two types of intracellular ER (ERa and ERβ) that are the product of different genes, have different patterns of tissue expression, have different pharmacological properties, activate different groups of genes, and can even carry out opposite actions when they are simultaneously present in the same cell.

2. The vast majority of the actions for which the estrogens in tissues are known are mediated by one of their intracellular receptors and imply the modification of the expression of extensive groups of genes that vary from one tissue to another. The estrogen-receptor complexes recognize the genes regulated by estrogens through short sequences of nucleotides (estrogen response elements) in their promoter region that specifically anchor the receptor. Once the receptor is bound to DNA, cofactors of transcription (coactivators or corepressors) capable of influencing the efficiency of the gene transcription machinery are recruited.

3. Target cells have different collections of coactivators and corepressors that

a) are not functionally equivalent,

b) are not recruited with the same efficiency by the hormone-receptor complex,

c) do not influence the transcription machinery of each cell with the same efficacy, and

d) do not behave identically in each gene promoter regulated by estrogens.

4. There is growing evidence on the existence of ERs anchored to specific regions of the plasmatic membrane of target cells. These receptors mediate fast actions of estrogens that are executed by their own signaling mechanisms and that are different from the actions used at the genome level by the intracellular receptors.

5. Evidence is also accumulating on the existence of a system of interconnected signals among ERs and signaling systems originating in membrane receptors for growth factors. The use of ER free of ligand as one of the steps in the signaling pathways of membrane receptors for growth factors has also been observed.

In this way, although ERs participate in all cases and in all cells capable of responding to estrogens, the nature and intensity of the response is conditioned by the receptor interaction with three different types of molecules: estrogen (steroid or not), DNA (through the HRE sequences), and the protein-protein interactions, including cofactors of transcription as well as elements of the signaling pathway from membrane receptors.

In the West, where demographic trends suggest that women will live on average 30 years after menopause, the need to replace the ovarian source of estrogens has become evident. The Gordian knot (Diamanti-Kandarakis et al. 2003) rests in finding drugs (SERMs) that replace the functions of estrogens without producing estrogen-dependent tumors and other adverse consequences.

Only in-depth knowledge of the mechanisms of the action of estrogens and of other ligands for their receptors will permit a deeper understanding of the foundations on which the specificity of action on tissue for each SERM are based. This is perhaps among those challenging frontiers of knowledge that carry with it the potential to impact society in a profound way.


1. Aaronica SM, Katznellenbogen BS (1993) Stimulation of estrogen receptor-mediated transcription by estrogen, cyclic-AMP and IGF-1. Mol Endocrinol 7:743

2. Akner G, Wikstrom AC, Gustafsson JA (1995) Subcellular distribution of the glucocorticoid receptor and evidence for its association with microtubules J Steroid Biochem Mol Biol 52:1

3. Ali S, Metzger D, Bornert JM, Chambon P (1993) Modulation of transcriptional activation by ligand-dependent phosphorylation of the human oestrogen receptor A/B region. EMBO J 12:1153

4. Anderson E (2002) The role of oestrogen and progesterone receptors in human mammary development and tumorigenesis. Breast Cancer Res 4:197

5. Badger TM, Ronis MJ, Hakkak R, Rowlands JC, Korourian S (2002) The health consequences of early soy consumption. J Nutr 132:559S

6. Beato M, Sanchez-Pacheco A (1996) Interaction of steroid hormone receptors with the transcription initiation complex. Endocr Rev 1:587

7. Beato M (1989) Gene regulation by steroid hormones. Cell 56:335

8. Bouillon R, Okamura HO, Norman AW (1995) Structure-function relationship in the vitamin D endocrine system. Endocr Rev 16:200

9. Brann DW, Hendry LH, Mahesh V (1995) Emerging diversities in the mechanism of action of steroid hormones. J Steroid Biochem Mol Biol 52:2

10. Brzozowski AM, Pike AC, Dauter Z (1997) Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753

11. Brzozowski AM, Pike ACW, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustaffson JA, Carlquist M (1997) Molecular basis of agonism and antagonism in the estrogen receptor. Nature 389:753

12. Castellano Diaz E, Gonzalez Quijano MI, Liminana JM, Diaz Chico BN (1989) Tamoxifen decreases the estradiol induced progesterone receptors by interfering with nuclear estrogen receptor accumulation. J Steroid Biochem 33:133

13. Chambliss KL, Shaul PW (2002) Estrogen modulation of endothelial nitric oxide synthase. Endocr Rev 23:665

14. Chan S (2002) A review of selective estrogen receptor modulators in the treatment of breast and endometrial cancer. Semin Oncol 29:129

15. Chin WL (1995) Hormonal regulation of gene expression. In: DeGroot L (ed) Endocrinology. Saunders, Philadelphia, p 6

16. Chirino R, Fernandez L, Lopez A, Navarro D, Rivero JF, Dfaz-Chico JC, Dfaz-Chico BN (1991) Thyroid hormones and glucocorticoids act synergistically in the regulation of the low affinity glucocorticoid binding sites in the male rat liver. Endocrinology 129:3118

17. Clemons M, Goss P (2001) Estrogen and the risk of breast cancer. N Engl J Med 344:276

18. Colditz GA (1998) Relationship between estrogen levels, use of replacement theraphy and breast cancer. J Natl Cancer Inst 90:814

19. Cosman F, Lindsay R (1999) Selective estrogen receptor modulators: clinical spectrum. Endocr Rev 20:418

20. Couse JF, Hewitt SC, Bunch DO, Sar M, Walker VR, Davies BJ, Korack K (1999a) Postnatal sex reversal of the ovaries in mice lacking estrogen receptors alpha and beta. Science 286:2328

21. Couse JF, Korack KS (1999b) Estrogen receptor null mice: what have we learned and where will they lead us? Endocr Rev 20:358

22. Diamanti-Kandarakis E, Sykiotis GP, Papavassiliou AG (2003) Selective modulation of postmenopausal women: cutting the Gordian knot of hormone replacement therapy with breast carcinoma. Cancer 97:12

23. Dfaz-Chico BN, Ogasawara Y, Chamness GC, Salman M, McGuire WL (1988) A 46 kDa antigen associated with estrogen receptors in human breast cancer. J Steroid Biochem 30:315

24. Edwards DP, Altaman M, DeMarzo A, Zhang Y, Weigel N, Beck CA (1995) Progesterone receptor and the mechanism of action of progesterone antagonists. J Steroid Biochem Mol Biol 53:449

25. Edwards DP, Wardell SE, Boonyaratanakornkit V (2002) Progesterone receptor interacting coregulatory proteins and cross talk with cell signaling pathways. J Steroid Biochem Mol Biol 83:173

26. Evans RM, Hollenberg S (1988) Zinc fingers: gilt by association. Cell 52:1

27. Evans RM (1989) Molecular characterization of the glucocorticoid receptor. Rec Prog Horm Res 45:1

28. Evans RM (1988) The steroid and thyroid receptor family. Science 240:889

29. Fawell SE, Lees JA, White R, Parker MG (1990) Characterization and colocalization of steroid binding and dimerization activities in the mouse estrogen receptor. Cell 60:953

30. Fernandez L, Chirino R, Boada LD, Navarro D, Cabrera N, Del Rio I, Dfaz-Chico BN (1994) Stanozolol and danazol, unlike natural androgens, interact with the low affinity glucocorticoid binding sites from male rat liver microsomes. Endocrinology 134:1401

31. Filardo EJ (2002) Epidermal growth factor receptor (EGFR) transactivation by estrogen via the G-protein-coupled receptor, GPR30: a novel signaling pathway with potential significance for breast cancer. J Steroid Biochem Mol Biol 80:231

32. Freedman LP, Luisi BF, Korszum ZR, Basavappa R, Sigler PB, Yamamoto KR (1988) The function and structure of the metal coordination sites within the glucorticoid receptor DNA binding domain. Nature 334:543

33. Fuqua SA, Fitzgerald SD, Allred DC, Elledge RM, Nawaz Z, McDonnell DP, O’Malley BW, Greene G, McGuire WL (1992) Inhibition of estrogen receptor action by a naturally occurring variant in human breast tumors. Cancer Res 52:483

34. Fuqua SA (2001) The role of estrogen receptors in breast cancer metastasis. J Mammary Gland Biol Neoplasia 6:407

35. Garcia T, Lehler S, Bloomer WD, Schachter B (1988) A variant estrogen receptor messenger ribonucleic acid is associated with reduced levels of estrogen binding in human mammary tumors. Mol Endocrinol 2:785

36. Green S, Kumar V, Theulaz I, Wahli W, Chambon P (1988) The N-terminal DNA- binding “zinc finger” of the oestrogen and glucocorticoid receptors determines target gene specificity. EMBO J 7:3037

37. Green S, Walter P, Kumar V, Krust A, Bornert JM, Argos P, Chambon P (1986a) Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 320:134

38. Greene GL, Gilna P, Waterfield M, Baker A, Hort Y, Shine J (1986b) Sequence and expression ofhuman estrogen receptor complementary DNA. Science 231:1150

39. Gruber CJ, Tschugguel W, Schneeberger C, Huber JC (2002) Production and actions of estrogens. N Engl J Med 346:340

40. Guiochon M, Loosfelt H, Lescop P, Sar S, Atger M, Perrot A, Milgrom E (1989) Mechanisms of nuclear localization of the progesterone receptor: evidence for interaction between monomers. Cell 57:1147

41. Hall JM, McDonnell DP (1999) The estrogen receptor beta-isoform of the human estrogen receptor modulates ER-alpha transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 140:5566

42. Hodgin JB, Maeda N (2002) Minireview: estrogen and mouse models of atherosclerosis. Endocrinology 143:4495

43. Jacobs MN, Lewis DF (2002) Steroid hormone receptors and dietary ligands: a selected review. Proc Nutr Soc 61:105

44. Jensen EV (1991) Overview of the nuclear receptor family. In: Parker M (ed) Nuclear hormone receptor. Academic, London, p 1

45. Jordan VC, Murphy CS (1990) Endocrine pharmacology of antiestrogens as antitumor agents. Endocr Rev 11:578

46. Jordan VC (2001) Selective estrogen receptor modulation: a personal perspective. Cancer Res 61:5683

47. Jordan VC (2002) The secrets of selective estrogen receptor modulation: cell-specific coregulation. Cancer Cell 1:215

48. Kassis JA, Gorski J (1983) On the mechanism of estrogen receptor replenishment: recycling, resynthesis and/or processing. Mol Cell Biochem 52:27

49. Kastner P, MarkM, Chambon P (1995) Nonsteroid nuclear receptors: what are genetic studies telling us about their role in real life? Cell 83:859

50. Kelly MJ, Wagner EJ (1999) Estrogen modulation of G-protein-coupled receptors. Trends Endocrinol Metab 10:369

51. King WJ, Greene GL (1984) Monoclonal antibodies localize oestrogen receptor in the nuclei of target cells. Nature 307:745

52. Klug A, Rhodes D (1987) “Zinc finger”: a novel protein motif for nucleic acid recognition. Trends Biochem Sci 12:464

53. Korach KS (1994) Insights from the study of animals lacking functional estrogen receptor. Science 266:1524

54. Korach SK, Couse JF, Curtis SW, Wahsburn TF, Linzdey J, Kimbro KS, Eddy EM, Miglacchio S, Snedeker SM, Lubhan DB, Schomberg DW, Smith EP (1996) Estrogen receptor gene disruption: molecular characterization and experimental and clinical phenotypes. Rec Prog Horm Res 51:159

55. Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korack K, Gustafsson JA (1998) Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc Natl Acad Sci USA 95:15677

56. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA (1996) Comparison of the ligand binding specificity ans transcript tissue distribution of estrogen receptor alpha and beta. Endocrinology 138:863

57. Kuiper GG, Enmark E, Haggblad J, Pelto-Huikko M, Nilsson S, Gustafsson JA (1996) Cloning of a novel estrogen receptor in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925

58. Kumar V, Chambon P (1988) The estrogen receptor binds tightly to its responsive elements as a ligand-induced homodimer. Cell 55:145

59. Kumar V, Green S, Stack G, Berry M, Jin JR, Chambon P (1987) Functional domains of the human estrogen receptor. Cell 51:941

60. Lathers CM (2002) Endocrine disruptors: a new scientific role for clinical pharmacologists? Impact on human health, wildlife, and the environment. J Clin Pharmacol 2:7

61. Lee AV, Cui X, Oesterreich S (2002) Cross-talk among estrogen receptor, epidermal growth factor, and insulin-like growth factor signaling in breast cancer. Clin Cancer Res 7(12 Suppl):4429s

62. Lee MS, Gippert, GP, Soman KV, Case DA, Wright PE (1989) Three-dimensional solution structure of a single zinc finger DNA-binding domain. Science 245:635

63. Levin ER (2002) Cellular functions of plasma membrane estrogen receptors. Steroids 67:471

64. Lones D,MarkM,Mendehlson C, Dollé P, Decimo D, Chambon P (1995) Developmental roles of the retinoic acid receptors. J Steroid Biochem Mol Biol 53:475

65. Luzardo OP, Machin RP, Dfaz-Chico BN, Fernandez L (2000) Photoaffinity labeling identification of a specific binding protein for the anabolic steroids stanozolol and danazol: an oligomeric protein regulated by age, pituitary hormones, and ethinyl estradiol. Endocrinology 141:3377

66. MacGregor JI, Jordan VC (1998) Basic guide to the mechanisms of antiestrogen action. Pharmacol Rev 50:151

67. Machado JD, Alonso C, Morales A, Borges R (2002) A novel nongenomic action of estrogens: the regulation of exocytotic kinetics. Ann NY Acad Sci 971:284

68. Mangelsdorf DJ, Thrummel C, Beato M, Herrlich P, Schütz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM (1995) The nuclear receptor superfamily: the second decade. Cell 83:835

69. McDonnell DP, Clemm DL, Imhof MO (1994) RU486 exerts antiestrogenic activities through a novel progesterone receptor A form-mediated mechanism. J Biol Chem 269:11945

70. McDonnell DP, Norris JD (2002) Connections and regulation of the human estrogen receptor. Science 296:1642

71. McDonnell DP (1999) The molecular pharmacology of SERMs. Trends Endocrinol Metab 10:301

72. McEwen BS, Alves SE (1999) Estrogen actions in the central nervous system. Endocr Rev 20:279

73. McGuire WL, Chamness GC, Fuqua SAW (1991) Estrogen receptor variants and clinical breast cancer. Mol Endocrinol 5:1571

74. McKenna NJ, Lanz RB, O’Malley BW (1999) Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321

75. Melnick R, Lucier G, Wolfe M, Hall R, Stancel G, Prins G, Gallo M, Reuhl K, Ho SM, Brown T, Moore J, Leakey J, Haseman J, Kohn M (2002) Summary of the National Toxicology Program’s report of the endocrine disruptors low-dose peer review. Environ Health Perspect 110:427

76. Mendelsohn ME (2002) Genomic and nongenomic effects of estrogen in the vasculature. Am J Cardiol 90:3F

77. Mester J, Baulieu EE (1995) Nuclear receptor superfamily. In: DeGroot L (ed) Endocrinology. Saunders, Philadelphia, p 93

78. Miglaccio A, Piccolo D, Castoria G, DiDomenico M, Bilancio A, Lombardi M, Gong W, Beato M, Auricchio F (1998) Activation of the Src/p21(ras)/Erk pathway by progesterone receptor via cross-talk with estrogen receptor. EMBO J 17: 2008

79. Munck A, Toth ANF (1995) Glucocorticoid action. In: DeGroot L (ed) Endocrinology. Saunders, Philadelphia, p 1642

80. Nakata K (2002) Theoretical approach to endocrine disruptors. Front Biosci 7:c68

81. Navarro D, Leon L, Chirino R, Fernandez L, Pestano J, Dfaz-Chico BN (1998) The two native estrogen receptor forms of 8S and 4S present in cytosol from human uterine tissues display opposite reactivity with tamoxifen aziridine and the estrogen response element. J Steroid Biochem Mol Biol 64:49

82. Navarro D, Luzardo OP, Fernandez L, Chesa N, Dfaz-Chico BN (2002) Transition to androgen independence in prostate cancer. J Steroid Biochem Mol Biol 171:1

83. Nilsson S, Makela S, Treuter E, Tujangue M, Thomsen J, Anderson G, Enmark E, Pettersson K, Warner M, Gustafsson JA (2001) Mechanisms of estrogen action. Physiol Rev 81:1535

84. O’Malley B, Means AR (1974) Female steroid hormones and target cell nuclei. Science 183:610

85. O’Malley BW (1990) The steroid receptor superfamily: more excitement predicted for the future. Mol Endocrinol 4:363

86. Osborne CK, Schiff R, Fuqua SA, Shou J (2001) Estrogen receptor: current understanding of its activation and modulation. Clin Cancer Res 7(12 Suppl): 4338s

87. Palmieri C, Cheng GJ, Saji S, Zelada-Hedman M, Warri A, Weihua Z, Van Noorden S, Wahlstrom T, Coombes RC, Warner M, Gustafsson JA (2002) Estrogen receptor beta in breast cancer. Endocr Relat Cancer 9:1

88. Pasqualini JR, Chetrite G, Nguyen BL, Maoloche C, Delalonde M, Talbi M, Feinstein MC, Blacker C, Botella J and Paris J (1995) Estrone sulfate-sulfatase and 17β-hidroxisteroid dehydrogenase activities: a hypothesis for their role in the evolution of human breast cancer from hormone-dependence to hormone-independence. J Steroid Biochem Mol Biol 53:407

89. Pettersson K, Delaunay F, Gustafsson J-Â (2000) Estrogen receptor beta acts as a dominant regulator of estrogen signaling. Oncogene 19:4970

90. Picard D, Yamamoto KR (1987) Two signals mediate hormone dependent nuclear localization of the glucocorticoid receptor. EMBO J 6:3333

91. Ponglikitmongkol M, Green S, Chambon P (1988) Genomic organization of the human oestrogen receptor gene. EMBO J 7:3385

92. Powles TJ (2002) Anti-oestrogenic prevention of breast cancer - the make or break point. Nat Rev Cancer 2:787

93. Quingley CA, DeBellis A, Marschke KB, El-Awady MK, Wilson EM, French FS (1995) Androgen receptor defects: historical, clinical and molecular perspectives. Endocr Rev 16:271

94. Ratajczak T, Carrello A, Mark PJ, Warner BJ, Simpson B, Moritz RL, House AK (1993) The cyclophilin component of the unactivated estrogen receptor contains a tetratri- copeptide repeat domain and shares identity with p59 (FKBP59). J Biol Chem 268:13187

95. Redeuilh G, Moncharmont B, Secco C, Baulieu EE (1987) Subunit composition of the molybdate stabilized 8-9S nontransformed estradiol receptor purified from calf uterus. J Biol Chem 262:6969

96. Revelli A, Massobrio M, Tesarik J (1998) Nongenomic actions of steroid hormones in reproductive tissues. Endocr Rev 19:3

97. Riggs BL, Hartmann LC (2003) Selective estrogen-receptor modulators - mechanisms of action and application to clinical practice. N Engl J Med 348:618

98. Rubanyi GM, Kauser K, Johns A (2002) Role of estrogen receptors in the vascular system. Vascul Pharmacol 38:81

99. Safe S (1998) Interactions between hormones and chemicals in breast cancer. Annu Rev Pharmacol Toxicol 38:121

100. Scott GK, Kushner P, Vigne JL, Benz CC (1991) Truncated forms of DNA-binding estrogen receptors in human breast cancer. J Clin Invest 88:700

101. Segars JH, Driggers PH (2002) Estrogen action and cytoplasmic signaling cascades. Part I: Membrane-associated signaling complexes. Trends Endocrinol Metab 13:349

102. Segnitz B, Ghering U (1995) Native structure of the estrogen receptor of MCF-7 cells “in vivo”. Proc Natl Acad Sci USA 92:2179

103. Seiler-Tuyns A, Walker P, Martinez E, Merillat AM, Givel F, Wahli W (1986) Identification of estrogen-responsive DNA sequences by transient expression experiments in a human breast cancer cell line. Nucleic Acids Res 14:8755

104. Shang Y, Brown M (2002) Molecular determinants for the tissue specificity of SERMs. Science 295:2465

105. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL (1998) The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927

106. Brown SMT, Moore J, Leakey J, Haseman J, Kohn M (2002) Summary of the National Toxicology Program’s report of the endocrine disruptors low-dose peer review. Environ Health Perspect 110:427

107. Smith CL, Nawaz Z, O’Malley BW (1997) Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed antiestrogen, 4-hydroxytamoxifen. Mol Endocrinol 11:657

108. Smith EP, Boyd J, Frank GP, Lubhan DB, Korack KS (1994) Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. N Engl J Med 331:1056

109. Smith FS, Toft DO (1993) Steroid receptors and their associated proteins. Mol Endocrinol 7:4

110. Sonnenschein C, Soto AM (1998) An updated review of environmental estrogen and androgen mimics and antagonists. J Steroid Biochem Mol Biol 65:143

111. Toft DO, Gorski J (1966) A receptor molecule for estrogens: isolation from rat uterus and preliminary characterization. Proc Natl Acad Sci USA 5:1574

112. Tora L, Gaub MP, Mader S, Dierich A, Bellard M, Chambon P (1988) Cell-specific activity of a GGTCA half-palindromic oestrogen responsive element in the chicken ovoalbumin gene promoter. EMBO J 7:3771

113. Tora L, White J, Brou C, Tasset D, Webster N, Scheer E, Chambon P (1989) The human estrogen receptor has two independent nonacidis transcriptional activation functions. Cell 59:477

114. Tzukerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP (1994) Human estrogen receptor transactivational capacity is deter- mined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21

115. Vasudevan N, Ogawa S, Pfaff D (2002) Estrogen and thyroid hormone receptor interactions: physiological flexibility by molecular specificity. Physiol Rev 82:923

116. Vigushin DM, Coombes RC (2002) Histone deacetylase inhibitors in cancer treatment. Anticancer Drugs 13:1

117. Wakeling AE (1993) The future of new pure antiestrogens in clinical breast cancer. Breast Cancer Res Treat 25:1

118. Welshons WV, Lieberman ME, Gorski J (1984) Nuclear localization of unoccupied oestrogen receptors. Nature 307:747

119. White RE (2002) Estrogen and vascular function. Vascul Pharmacol 38:73

120. Cowley SM, Parker MG (1999) A comparison of transcriptional activation by ER alpha and ER beta. J Steroid Biochem Mol Biol 69:165

121. McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW (1995) Analysis of estrogen receptor function in vitro reveals three distinct classes of antiestrogens. Mol Endocrinol 9:659

122. McInerney EM, Weis KE, Sun J, Mosselman S, Katzenellenbogen BS (1998) Transcription activation by the human estrogen receptor subtype beta (ERb) studied with ERa and ERb receptor chimeras. Endocrinology 139:4513

123. Barkhem T, Carlsson B, Nilsson Y, Enmark E, Gustafsson J-Â, Nilsson S (1998) Differential response of estrogen receptor alpha and estrogen receptor beta to partial estrogen agonists/antagonists. Mol Pharmacol 54:105



If you find an error or have any questions, please email us at Thank you!