Selective Estrogen Receptor Modulators. Antonio Cano

Chapter 3. Action of Selective Estrogen Receptor Modulators (SERMs) Through the Classical Mechanism of Estrogen Action

• Fernando Marin

• Ma Carmen Barbancho



The molecular mechanisms through which SERMs have an estrogen-agonist or antagonist effect, depending on the tissue in question, has been a topic of intense investigation during the last decade (Yang et al. 1996b; Shang et al. 2002; Meegan et al. 2003; Smith et al. 2004). Recent advances in the molecular biology of ERs have revealed the enormously complex nature of this process. As mentioned in Chapter XXX, estrogens regulate the activation of genes by means of a series of molecular events triggered after they bind to the estrogen receptor (ER), which is simply a transcription factor that can be activated by a ligand. The same course of events is set in motion by selective estrogen receptor modulators (SERMs). In short, the high-affinity binding to the ER, and to the ER alone, is a fundamental characteristic of SERMs. The absence of crossbinding with other members of the nuclear receptor superfamily (androgen, progesterone, glucocorticoids, mineralocorticoids, retinoic acid, vitamin D receptors, etc.) is a critical stage in the target molecule selection process. The binding of the ligand to the ER leads to a conformational change in the ligand-ER complex, causing disassociation of the stress proteins associated to the inactive receptor (heat-shock proteins). The inactive receptors, which are monomers in basal conditions, dimerize and get phosphorylated until they finally bind to a series of nuclear proteins called adaptor or coregulator proteins. This ligand-ER complex binds to one of the DNA response elements (EREs: estrogen response elements), generally located in the promoter region of estrogen target genes, and trigger the mRNA transcription and synthesis process. Depending on the cell type, coregulator protein load, ratio of coactivator and corepressor proteins, and different gene promoters, the ligand-ER dimer may activate or inhibit the gene transcription.

It is of great interest that new findings suggest that several effects of estrogen and SERMs also may be mediated by nonnuclear ER actions derived from fast activation (in a matter of minutes) of the protein-kinase cascades and other signaling processes associated to cell membrane ERs. These activation routes, described in detail in Chap. 1 by Escrich et al. are not mediated by transcription factors and nuclear ER activation. The functional effects associated with these pathways include some of the cardiovascular effects of estrogens, such as nitric oxide (NO)-dependent vasodilatation, inhibition of endothelial damage response, and reduced ischemic damage to the myocardium in experimental conditions (Ho et al. 2002; Salom et al. 2002). Raloxifene and its analog LY-117018 stimulate endothelial NO synthetase (eNOS) and lead to coronary artery relaxation and improved vasodilatation in hypertensive rats (Simoncini et al. 2002a; Wassman et al. 2002). The eNOS activation by means of protein- kinase activation (MAPKs and PI3K-Akt) has also been reported with EM-800 (Simoncini et al. 2002b). These nongenomic pathways of estrogen action have also been reported in neuronal cells (Qiu et al. 2003) and bone cells (Kousteni et al. 2003), but their contribution in the action of specific SERM compounds, and the role of different signaling pathways in tissue-specific actions, is undefined at this time.

In addition to these rapid nongenomic events, the pharmacodynamic profile of different SERMs may depend on the subtle mesh of the combined action of complex mechanisms that govern the transcription mediated by ERs. Basically, this can occur at four levels: (1) tissue amount and distribution of the different ER subtypes; (2) impact on binding capacity to promoters provoking the different 3D conformations of the SERM-ligand complex, which ultimately dictate specific coregulator interactions; (3) different content of nuclear coregulator proteins (coactivators and corepressors) in various cells; and (4) presence of different types of estrogen response elements (EREs) in the gene promoters, including the ability of ERs to affect gene expression without directly binding to target DNAs in a process known as transrepression.


Estrogen Receptor Subtypes

Until 1995, it was believed that only one gene-encoding ER existed. The description and characterization of a new ER subtype, known as ERβ, has modified the classical vision of the molecular pathways activated by estrogens (Kuiper et al. 1996). Furthermore, it is highly likely that each subtype has multiple isoforms. Although the majority of tissues express both ER subtypes (ERa and ERβ), there is a certain degree of tissue-specific distribution (Kuiper et al. 1997). For example, ERa predominates in the breast, liver, uterus, ovaries, and central nervous system. ERβ has a slightly different tissue distribution pattern, and a high expression has been reported in endothelial cells, bones, lungs, urogenital tract, ovaries, central nervous system, and prostate. While ERa acts predominantly as an activator, ERβ can down-regulate this response if it binds to ERa to form a heterodimer. Therefore, an attractive hypothesis to explain the pharmacology of SERMs is the relative content of both ER subtypes in a certain tissue, additionally considering the different relative affinities of the various SERMs to the two described ER subtypes. However, there is no clear proof to date that demonstrates that the selective action of a certain SERM is the result of its preferential binding to one of the two ER subtypes. In fact, tamoxifen and raloxifene, the two SERMs for which most data are available, show a similar degree of binding to both ER isoforms (Kuiper et al. 1997).


Conformation of Ligand-ER Complex

One of the most significant findings in this field is the demonstration that the tertiary structure of the ligand-ER complex depends on the molecular characteristics of the ligand. The ligand-binding domain (LBD) of the ER consists of a hydrophobic open pocket formed by 12 short helices. Near its carboxy-terminal end, it includes an amino acid sequence called Activating Function-2 (AF-2), which is considered essential in the activation of genes that mediate estrogenic activity in reproductive tissues, such as the uterus and breast. AF-2 is dependent on the binding of the ligand. Therefore, if AF-2 gets blocked, the transcription of these genes would be compromised. When the ER-LBD binds to 4-hydroxy-tamoxifen, the active metabolite of tamoxifen, the change in the alignment of helix 12 prevents the interaction between the ligand- ER complex and a coactivator protein. In contrast, if the ligand is estradiol, helix 12 goes over the ligand, snuggling it in a tight fitting pocket that permits the coactivator proteins to bind to this region (Shiau et al. 1998). Similar findings have been reported with X-ray protein crystallization techniques that have analyzed the 3D structure of the ERa and raloxifene complex. The basic side chain of the raloxifene molecule is very rigid and does not fit in the pocket formed by the ligand-binding domain of the ERa. Therefore, helix 12 cannot fit properly over the ligand and remains outside the “pocket”, preventing AF-2 from interacting with the coregulator protein in order for the transcription to take place (Brzozowski et al. 1997). The different SERMs give rise to different 3D conformations of the SERM-ligand complex, forming a continuum of different intermediary forms, from the binding of estradiol at one end to the binding of a pure antiestrogen at the other (McDonnell et al. 2002).


Coregulator Protein Cell Content and Coactivators/Corepressors Ratio

In order for estrogen-mediated genomic activation to occur, the ligand-ER complex must bind to other nuclear proteins (coregulator proteins) that can either act as coactivators (stimulators of gene transcription) or corepressors (inhibitors of gene transcription) for access of the complex to the ER Es (for a review see Rosenfeld et al. 2001) (Fig. 3.1). The discovery and cloning of these coregulator proteins has been another key milestone in our understanding of gene expression pathways mediated by transcription factors (Smith et al. 2004). Several coregulator proteins are known to be capable of binding to ERs and modulating their function. Major ER coactivators fall into three groups of proteins: steroid receptor coactivator-1 (SRC-1), SRC-2, and SRC-3. In addition, proteins such as cyclic AMP response element binding protein (CBP/p300) act as coactivators for multiple transcription factors. Some of the corepressors associated with raloxifene and 4-hydroxy-tamoxifen antiestrogenic actions are the silencing mediator of retinoic acid and thyroid hormone receptor (SMRT) and the nuclear receptor corepressor (NcoR). Whether the coregulator protein that binds to the ligand-ER complex is a coactivator or corepressor protein depends on the conformational alteration that the ER undergoes, which in turn depends on the type of ligand to which it is bound. Furthermore, the amount of coregulator proteins, in both absolute and relative terms, varies according to the different cells that respond to estrogens. An extraordinary example of this fact was described recently by Shang et al. (2002). These authors demonstrated that the estrogen-antagonist action of tamoxifen and raloxifene on breast cells is mediated by the action of corepressor proteins present in these cells. However, the opposite action of these two SERMs on the endometrium is explained by the capacity that tamoxifen shows in facilitating the recruitment of coactivator proteins of a group of genes in endometrial cells. In short, the agonist effect of tamoxifen on the endometrium is based on the induction in this tissue of a high expression of coactivator SRC-1. This effect is not observed with raloxifene. Therefore, differences in cell type and ratio of different coregulator proteins in such cells may determine the response to different types of SERMs (Shang et al. 2002).

Fig. 3.1. Coregulator proteins and gene transcription. The DNA-bound receptor can either positively or negatively regulate target gene transcription. Agonist-bound ERs can recruit transcriptional adaptors, proteins that permit the receptor to transmit its regulatory information to the cellular transcriptional apparatus. Among those adaptors are coactivators (stimulators of gene transcription) and correpresors (inhibitors of gene transcription)


Transrepression: Regulation of Gene Expression by an ERE-Independent Mechanism

One last mechanism that has been put forward to explain the tissue selectivity of SERMs, or even gene-specific regulation within the same cell, derives from the existence of estrogen-dependent genes containing not the classic ERE sequence in the gene promoter region but other alternative response sequences, in such a way that only the genes that contain these non-ERE sequences are transcribed when the tamoxifen-ERa or raloxifene-ERa complexes interact with the gene promoter (Paech et al. 1997; Yang et al. 1996b). The gene expression in this ERE-independent mechanism involves other DNA-bound transcription factors (Kushner et al. 2000; Abdelrahim et al. 2002). It should be noted that in many estrogen-responsive genes a direct DNA binding of ER is not required for ER-mediated activation of transcription. ER, through protein-to-protein contacts with other transcription factors, such as AP-1 and Sp-1, allows increased efficiency of transcription mediated by these factors (Kushner et al. 2000). This is the case of the Transforming Growth Factor β3 (TGF-β3) gene, a highly abundant growth factor in the bone matrix with a highly potent antiresorptive activity whose expression has been reported with raloxifene and other SERMs but not with 17-β-estradiol through a nonclassical ERE-independent mechanism (Yang et al. 1996a,b).

Fig. 3.2. Potential genomic and nongenomic activation routes that may be induced by natural or synthetic ER ligands. The hypotheses that explain the tissue-selective action of SERMs include (a) selective activation of ER subtypes, (b) different changes in the 3D configuration of the ligand-ER complex, (c) recruitment of coactivator and corepressor proteins, and (d) activation of alternative response elements in certain inducible genes

Figure 3.2 provides a summary of the different mechanisms that have been hypothesized to explain the selective tissue action of SERMs.


1. Abdelrahim M, Samudio I, Smith R, Burghart R, Safe S (2002) Small inhibitory RNA duplexes for Sp! MRNA block basal and estrogen-induced gene expression and cell cycle progression in MCF-7 breast cancer cells. J Biol Chem 77:28815-28822

2. Brzozowski AM, Pike ACW, Dauter Z, Hubbard RE, Bonn T, Engstrom O, et al. (1997) Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 89:753758

3. Ho KJ, Liao JK (2002) Non-nuclear actions of estrogen. Arterioscler Throm Vas Biol 2:1952-1961

4. Kousteni S, Han L, Chen JR, Almeida M, Plotkin LI, Bellido T, Manolagas SC (2003) Kinase-mediated regulation of common transcription factors accounts for the bone- protective effects of sex steroids. J Clin Invest 11:1651-1664

5. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA (1997) Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 38:863-870

6. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA (1996) Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 3:5925-5930

7. Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uth R, et al. (2000) Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311-317

8. McDonnell DP, Wijayarate A, Chang C, Norris JD (2002) Elucidation of the molecular mechanism of action of selective estrogen receptor modulators. Am J Cardiol 0(suppl):35F-43F

9. Meegan MJ, Lloyd DG (2003) Advances in the science of estrogen receptor modulation. Curr Med Chem 0:181-210

10. Paech K, Webb P, Kuiper GGJM, Nilsson S, Gustafsson JA, Kushner PJ, et al. (1997) Differential ligand activation of estrogen receptors ERa and ERβ at AP1 sites. Science 77:1508-1510

11. Qui J, Bosch MA, Tobias SC, Grandy DK, Scanlan TS, Ronnekleiv OK, et al. (2003) Rapid signaling of estrogen in hypothalamic neurons involves a novel G-protein-coupled estrogen receptor that activates protein kinase C. J Neurosci 3:9529-9540

12. Rosenfeld MG, Glass CK (2001) Coregulator codes of transcriptional regulation by nuclear receptors. J Biol Chem 276:36865-36868

13. Salom JB, Burguete MC, Pérez-Asensio FJ, Centeno JM, Torregrosa G, Alborch E (2002) Acure relaxant effects of 17-β-estradiol through non-genomic mechanisms in rabbit carotid artery. Steroids 7:339-346

14. Shang Y, Brown M (2002) Molecular determinants for the tissue specificity of SERMs. Science 295:2465-2468

15. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard A, Greene GL (1998) The structural basis of estrogen receptor/co-activator recognition and the antagonism of this interaction with tamoxifen. Cell 5:927-937

16. Simoncini T, Genazzani AR, Liao JK (2002a) Nongenomic mechanisms of endothelial nitric oxide synthetase activation by the selective estrogen receptor modulator raloxifene. Circulation 105:1368-1373

17. Simoncini T, Varone G, Fornari L, Mannella P, Luisi M, Labrie F, et al (2002b) Genomic and nongenomic mechanisms of nitric oxide synthesis induction in human endothe- lia! cells by a fourth-generation selective estrogen receptor modulator. Endocrinology 43:2052-2061

18. Smith CL, O’Malley BW (2004) Co-regulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr Rev 25:45-71

19. Wassmann S, Laufs U, Stamenkovic D, Linz W, Stasch JP, Ahlbory K, Rosen R, Bohm M, Nickening G (2002) Raloxifene improves endothelial dysfunction in hypertension by reduced oxidative stress and enhanced nitric oxide production. Circulation 105:20832091

20. Yang NN, Bryant HU, Hardikar S, Sato M, Galvin RJS, Glasebrook AL, et al. (1996a) Estrogen and raloxifene stimulate transforming growth factor-3β expression in rat bone: a potent mechanism for estrogen- or raloxifene-mediated bone maintenance. Endocrinology 37:2075-2084

21. Yang NN, Venugopalan M, Hardikar S, Glasebrook AL (1996b) Identification of an estrogen response element activated by metabolites of 17-β-estradiol and raloxifene. Science 73:1222-1225. [Letter. Published errata: Science 275:1249, 1997 Feb 28]

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