Molecular Mechanisms of Steroid Hormone Action

Home , Steroid hormone, Steroid hormone receptor

Endocrine-Related Cancer (1998) 5 1-14 Molecular mechanisms of steroid hormone action

R White and M G Parker

Molecular Endocrinology Laboratory, Imperial Cancer Research Fund, 44 Lincoln’s Inn Fields, London WC2A 3PX, UK (Requests for offprints should be addressed to M G Parker)

Introduction by hormonal signals but also by changes in other signalling pathways, which undoubtedly take place during The ability of oestrogens and androgens to stimulate the breast and prostate cancer progression. Receptors are also growth of a number of endocrine cancers is well capable of regulating the activity of a number of other established and several endocrine therapies are widely transcription factors, either directly or indirectly, and used to reduce the availability of the hormones or to block thereby modulate the ability of other signalling pathways their action. These include non-steroidal hormone to control gene transcription (Shemshedini et al. 1991, antagonists such as tamoxifen and flutamide, steroidal Philips et al. 1993, Stein & Yang 1995, Webb et al. 1995). compounds which include ICI 182780 and cyproterone Although the precise role of steroid hormones in cell acetate, and both steroidal and non-steroidal aromatase proliferation is still ill-defined, tremendous progress has inhibitors. Although we have learned a great deal about the been made in elucidating the role of hormones in receptor molecular mechanism of steroid hormone action, it is still activation and the mechanism by which hormone unclear how hormones stimulate the proliferation of antagonists block this activity. In particular two recent tumour cells and how hormone antagonists function. In advances have been made which provide new insights into part this is a result of the ability of oestradiol and steroid hormone action and the function of nuclear testosterone to regulate the expression of many proteins receptors in general. First, models for the three- implicated in the control of cell proliferation making it dimensional structure of the ligand binding domain of difficult to identify the crucial targets. Some of these several receptors have been determined (Bourguet et al. targets are growth factors and/or their receptors which 1995, Renaud et al. 1995, Wagner et al. 1995, Brzozowski suggests that the mitogenic effects of steroids may be et al. 1997) and secondly, novel proteins have been mediated by indirect autocrine or paracrine mechanisms identified (Cavaillès et al. 1994, Halachmi et al. 1994) (Clarke et al. 1991, Roberts & Sporn 1992). Alternatively which interact with receptors in a ligand dependent since steroids regulate the expression of certain cyclins or manner and may play a role in gene transcription. These kinase inhibitors (Musgrove & Sutherland 1994, Altucci developments are helping to provide a better et al. 1996) they may control cell cycle progression understanding of the mechanism of action of both directly. Recent work suggests that as well as the cyclin hormones agonists and antagonists and the links between D1 gene, cyclin D1 itself may be a crucial target (Zwijsen nuclear receptors and other signalling pathways. In this et al. 1997) but additional proteins could also be important review we will outline these advances, with reference to in different subsets of tumours. the oestrogen receptor (ER), and discuss their relevance to The identification of target genes for steroid hormones antioestrogen therapy and tamoxifen resistance. is further complicated by the observation that receptor signalling is cross-coupled with that of other signalling Steroid hormone receptors are members pathways. It used to be thought that signalling by receptors was relatively straightforward compared with most other of the nuclear receptor family signalling pathways, since the receptor itself is a Steroid hormone receptors are members of the nuclear transcription factor. It is now clear, however, that growth receptor family, a large group currently totalling factors, neurotransmitters and other hormones are able to approximately 150 different proteins, which function as modulate the activity of steroid hormone receptors (Power transcription factors in many different species including et al. 1991, Aronica & Katzenellenbogen 1993, Ignar- both invertebrates and vertebrates. In addition to steroid Trowbridge et al. 1993). This means that alterations in the hormone receptors, the nuclear receptor family consists of activity of receptors and the expression of individual target receptors for retinoids, thyroid hormone, fatty acids and genes involved in cell proliferation is determined not only prostaglandins and a number of so-called orphan

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

Figure 1 (A) Organisation of functional domains in the nuclear receptors. DBD and LBD refer to the positions of the DNA binding domain and the ligand binding domain respectively. The relative positions of the two transactivation domains AF1 and AF2 are also shown. (B) Schematic diagram of steroid hormone receptor binding to DNA. The receptors bind as homodimers to palindromic sequences. The specific sequences of simple response elements for the ER and other steroid hormone receptors are shown.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14 receptors, the ligands of which have yet to be identified transcription factors such as AP-1 (Philips et al. 1993, (Parker 1993, Mangelsdorf et al. 1995). They are highly Webb et al. 1995) and NF-κB (Stein & Yang 1995). To related in both primary amino acid sequence and the date most of the analysis which has been carried out has organisation of functional domains suggesting that many been to determine the mechanism by which receptors aspects of their mechanism of action are conserved. Thus, function as DNA dependent transcription factors but it is progress in our understanding of steroid hormone action likely that the transcription of many genes is regulated has been facilitated by studies of many nuclear receptor indirectly by non-classical mechanisms. family members. The sex steroid receptors are not only expressed in sex Nuclear receptors function as ligand accessory tissues but in many other types of cells including liver, bone, pituitary and cardiovascular cells. dependent transcription factors Androgen receptors are encoded by a single gene whereas In the absence of hormone, steroid hormone receptors there are two genes for ERs, ERα and ERβ (Kuiper et al. exist as inactive oligomeric complexes with a number of 1996, Mosselman et al. 1996). Our current knowledge of other proteins including chaperon proteins, namely the the distribution of ERβ (Kuiper et al. 1997) is based on heat shock proteins Hsp90 and Hsp70 and cyclophilin-40 mRNA analysis either by RT-PCR or by in situ and p23 (Smith & Toft 1993, Pratt & Toft 1997). The role hybridisation and the relative levels of ERα and ERβ of Hsp90 and other chaperons may be to maintain the protein levels have yet to be determined. Analysis of mice receptors folded in an appropriate conformation to made devoid of ERα by targeted gene disruption respond rapidly to hormonal signals. Whether different (Bocchinfuso & Korach 1977) indicates that this receptor types of receptor complexes occur in cells, each is essential for uterine growth and mammary gland containing a subset of chaperons and able to respond to development but not for mediating the inhibitory effects of different types of signal, or whether receptors are folded oestrogens in vascular injury (Jafrati et al. 1997). Thus, it into a mature complex containing chaperons in an is possible that ERα and ERβ have distinct functions in assembly line type of process is as yet unclear (Bohen et some tissues but not in others. Undoubtedly the role of al. 1995, Pennisi 1996). Following hormone binding, the these individual receptors will be confirmed in the future oligomeric complex dissociates allowing the receptors to by gene knock-out experiments to generate mice lacking function either directly as transcription factors by binding either both receptors or one isoform only and analysing the to DNA in the vicinity of target genes or indirectly by resulting phenotypes. modulating the activity of other transcription factors. A simplified diagram of the organisation of functional Nuclear receptors may be classified into four types domains in nuclear receptors is shown in Fig. 1A. The based on their dimerisation and DNA binding properties. receptors are characterised by a highly conserved DNA Steroid hormone receptors bind to DNA as homodimers binding domain and a moderately conserved ligand either to simple response elements comprising short binding domain which also functions in dimer formation palindromic sequences or to composite response elements and transcriptional activation (Parker 1993, Mangelsdorf with other transcription factors. Oestrogen response et al. 1995). A second dimerisation domain is located elements consist of inverted repeats of the sequence A/ within the DNA binding domain itself and the receptors GGGTCA separated by three nucleotides, whereas the contain two transcriptional activation functions by which receptors for androgens, glucocorticoids and progestins they are able to regulate the expression of target genes; bind to inverted repeats of the sequence AGA/GACA, also AF1, which is located in the N-terminal or A/B domain, separated by three nucleotides. In contrast, a second group and AF2 in the ligand binding domain. When activated by of nuclear receptors, including the retinoic acid receptor hormone binding, the receptors may act directly as (RAR), thyroid hormone receptor (TR) and vitamin D transcription factors by binding to specific DNA receptor, bind to DNA as heterodimers in combination sequences, termed hormone response elements, found in with the retinoid X receptor (RXR) (Mangelsdorf et al. the vicinity of target genes (Evans 1988, Beato 1989). The 1995). The binding sites for these receptors consist of sequences of simple binding sites for the steroid receptors direct repeats of the sequence A/GGTCA separated by up are shown in Fig. 1B. Composite response elements have to five nucleotides with the DNA binding specificity of also been identified which bind receptors in addition to different receptors being determined to some extent by the other transcription factors such that the binding of one spacing of the repeats. In addition RXR itself and some influences, either positively or negatively, the activity of orphan receptors bind to direct repeats as homodimers the other (Diamond et al. 1990). In addition, it is now while a final group of orphan receptors are also able to evident that nuclear receptors are also capable of bind as monomers to single copies of the extended binding regulating the transcription of genes that lack hormone site sequence TCAAGGTCA. The major part of the response elements by modulating the activity of other dimerisation interface is formed from sequences within

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

Figure 2 Diagrams of models of the structure of the ligand binding domains of nuclear receptors to demonstrate the effect of agonist binding. Panel A shows the arrangement of sequence elements for three different receptors, human RXRα in the absence of ligand and human RARγ and rat TRα in the presence of their respective ligands. The numbers 1 to 12 correspond to separate α-helices and s1 to s4 to β-strands. The position of amino acids which are in close contact with ligand are marked and are found in helices 3, 5, s1 and s3, the loop between helices 6 and 7 and helices 11 and 12. Panels B and C show diagrams of the folded structure in the absence and presence of ligand and the relative positions of helices 1 to 12 (H1-H12). This shows the realignment of helices 10 and 11 and the alteration in position of helix 12 to form a lid over the ligand binding pocket. This repositioning results in the formation of a new surface on the ligand binding domain, indicated by the white circles, which is believed to be required for the interaction with coactivator proteins. the ligand binding domain of the receptors while a second appear to interact with each other in the context of the dimerisation function is located within the DNA binding intact receptor in a way which has yet to be determined. domain itself (Kumar & Chambon 1988, Fawell et al. The N-terminal activation function, AF1, varies 1990a). considerably in both size and primary amino acid Transcriptional activation by steroid receptors is sequence in different steroid hormone and nuclear mediated by at least two distinct activation functions; one, receptors. A number of studies have shown that this referred to as AF1, is located in the N-terminal domain and domain is a target for phosphorylation by other signalling a second, AF2, is in the hormone binding domain (Lees et pathways (Ali et al. 1993, Kato et al. 1995, Bunone et al. al. 1989, Tora et al. 1989). In the ER, and probably in other 1996). The second activation function, AF2, is induced by receptors as well, the absolute and relative activities of hormone binding and a short sequence encoding an these two domains vary depending on the target promoter amphipathic α-helix in the C-terminal part of the ligand and the cell type. However, while the two activation binding domain, which is conserved in all trans- domains have the potential to act independently, they criptionally active nuclear receptors, has been shown to be

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14

essential in receptor function (Danielian et al. 1992, with one major exception. The C-terminal helix, helix 12, Saatcioglu et al. 1993, Barettino et al. 1994, Durand et al. which protrudes beyond the core of the ligand binding 1994). However, it is doubtful whether this short sequence domain in the unliganded RXRα, is folded back across the functions autonomously since mutation of conserved core of the domain in both RARγ and TRα. It has therefore residues in other parts of the ligand binding domain, which been proposed that in the case of RARγ, ligand binding do not affect ligand binding or dimerisation, also generate results in a realignment of helices 10 and 11 which re- transcriptionally defective receptors suggesting that AF2 arrange to form a continuous helix. This creates a is derived from a number of different elements (O’Donnell shortened helix 12, corresponding to the amphipathic α- & Koenig 1990, Henttu et al. 1997). helix described earlier, which is released from the core of the domain to fold back over the top of the ligand binding pocket where it forms both contacts with the ligand and a Structure of the ligand binding domain of salt bridge in helix 4. Helix 12 is similarly aligned in the nuclear receptors case of the TRα although a salt bridge is not formed. A major advance in our understanding of how nuclear Protein sequence comparisons between different receptors receptors function has been the determination of models suggest that their overall helical structure is conserved for the ligand binding domain based on crystal structures (Wurtz et al. 1996) and unpublished results indicate that of three different receptors either in the absence or the structures of peroxisome proliferator activated presence of their cognate ligands. The secondary structure receptor (PPAR) and ERα are consistent with this of all three nuclear receptors is shown in Fig. 2A and prediction. Thus it appears that a major role for ligand schematic diagrams of models of the structures obtained binding is an alteration in the conformation of the ligand in the absence and presence of ligand in Fig. 2B and C binding domain to form a novel surface. This may then respectively. The structure of the ligand binding domain of allow the binding of coactivators and other regulatory RXRα crystallised in the absence of ligand, consists of 12 proteins. α-helices and two β-strands arranged in three layers to α form an antiparallel -helical sandwich (Bourguet et al. Recruitment of receptor interacting 1995). This receptor was crystallised in the form of a dimer with the dimer interface formed mainly from helices proteins 9 and 10. Subsequently, crystals of the ligand binding Gene transcription depends on the formation of a pre- domains of RARγ (Renaud et al. 1995) and TRα (Wagner initiation transcription complex that includes basal et al. 1995) generated in the presence of their respective transcription factors and one of the roles of transcriptional ligands were shown to have a similar overall structure, but activators, including steroid hormone receptors, is thought

Table 1 Receptor interacting proteins.

Predicted size Group Name(s) Protein (kDa) Receptor binding Function(s)

I CBP/p300 CBP 265 ER, RAR, RXR, TR Histone acetylase p300 300 ER, RAR, RXR, TR Bind transcription factors

p160 SRC-1a 156.7 ER, RAR, RXR, TR, PR RIP160 SRC-1e/NCoA-1 152.3 ERAP160 TIF2 159.6 ER, RAR, RXR, TR, PR Form complexes with CBP/p300 GRIP1 NCoA-2 pCIP 152 ER, RAR, TR, PR

II ARA70 ARA70 70 AR Androgen receptor coactivator RIP140 RIP140 128 ER, RAR, RXR, TR, PR Multiple nuclear receptor binding sites TIF1 TIF1α 112 ER, RAR, TR, PR Chromatin remodelling? SUG1 SUG1 46 RAR, RXR, ER, TR, PR, VDR Associated with proteosome TRIP1

AR, androgen receptor; PR, progesterone receptor; VDR, vitamin D receptor.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action to be the stabilisation of this complex. Although receptors of more copies of a short sequence motif LXXLL which is have been shown to bind directly to basal transcription necessary and sufficient to mediate the binding of ligand factors in vitro the significance of these interactions is bound nuclear receptors. This motif has been described as unclear since they are unaffected by mutations in the a signature sequence for the interaction of proteins with receptor that destroy transcriptional activity (Sadovsky et nuclear receptors (Heery et al. 1997, Torchia et al. 1997). al. 1995). The involvement of additional targets is sug- Other proteins have also been isolated which interact gested by the observation that AF2 activity is inhibited with other regions of the ligand binding domain. For when receptors are overexpressed suggesting that limiting example two related proteins, N-CoR and SMRT, which downstream target proteins, required for gene trans- function as repressors for RXR and TR in the absence of cription, are being sequestered (Tasset et al. 1990). A ligand (Horlein et al. 1995, Heinzel et al. 1997, Nagy et number of candidate target proteins have been detected in al. 1997), have been reported to bind to oestradiol and extracts from mammalian cells which bind to the ligand progesterone receptors in the presence of antagonists binding domain of receptors in a ligand dependent manner (Jackson et al. 1997, Smith et al. 1997). A detailed (Cavaillès et al. 1994, Halachmi et al. 1994). Since these description of all these proteins, both activators and proteins fail to interact with mutant receptors which are repressors, is beyond the scope of this article; they are transcriptionally inactive it has been proposed that they reviewed in detail elsewhere (Horwitz et al. 1996, Heery may have a role in ligand dependent transcriptional & Parker 1997) and therefore we will focus on the activation by AF2. importance of coactivators in steroid hormone action Receptor interacting proteins may be classified into together with potential interactions between receptors and two groups (Table 1). The first group consists of two other transcriptional regulators. families of proteins which appear to stimulate transcription and represent bona-fide coactivators. One family, originally identified as proteins of 160 kDa, consists of a Structure and function of coactivators for number of related proteins; the steroid receptor coactivator nuclear receptors proteins (SRC-1a and SRC-1e) (Onate et al. 1995, Kamei To date, it appears that the p160 proteins and CBP/p300 et al. 1996), transcription intermediary factor TIF2 are the most important of the receptor interacting proteins (Voegel et al. 1996) (independently identified as GRIP-1 in terms of their ability to potentiate the transcriptional (Hong et al. 1996)) and CREB binding protein (CBP) activity of receptors. The p160 proteins SRC1, TIF2 and interacting protein (p/CIP) (Torchia et al. 1997). The p/CIP share significant sequence homology, most notably second family comprises CBP and p300, which were in a central region found to be required for interaction with originally shown to function as coactivators for cAMP the ligand binding domain of nuclear receptors. They response element binding protein (CREB), the trans- show ligand dependent binding to nuclear receptors in cription factor that mediates responses to protein kinase A vitro but fail to bind to transcriptionally inactive receptors stimulation. Subsequently, however, CBP/p300 were in which mutations have been introduced in the conserved shown to function as coactivators for many other hydrophobic residues in helix 12 (Kamei et al. 1996). transcription factors and may play a central role in many Similarly mutations in conserved residues in helix 3 which signalling pathways (Janknecht & Hunter 1996, Shikama also impair AF2 also prevent binding of these coactivators et al. 1997). (Henttu et al. 1997). This suggests that, in the presence of The second group contains a number of unrelated ligand, helices 3 and 12 form a composite surface proteins which also interact with the ligand binding recognised by the p160 proteins, a hypothesis which is domain but their role in receptor function is still unclear. consistent with the models of the holo ligand binding This group includes a 140 kDa protein termed RIP 140 domains. Similarly, CBP/p300 have been reported to bind (Cavaillès et al. 1995), TIF1 (Le Douarin et al. 1995), to the ligand binding domains of nuclear receptors in a ARA70 (Yeh & Chang 1996), which has been shown to ligand dependent manner and increase AF2 activity in stimulate transactivation mediated by the androgen transfected cells (Chakravarti et al. 1996, Hanstein et al. receptor, and TRIP1/SUG1 (Lee et al. 1995, vom Baur et 1996, Kamei et al. 1996, Yao et al. 1996). This binding is al. 1996). RIP 140 stimulates receptor mediated trans- also dependent on the integrity of helix 12, implying that cription in yeast (Joyeux et al. 1996) and contains an these proteins recognise a similar if not identical binding ‘activation domain’ which is functional in mammalian site (Heery et al. 1997). The p160 and CBP/p300 family cells (L’Horset et al. 1996) but it is still unclear whether it members also interact directly with one another (Hanstein is a bona-fide coactivator. TIF1 has been shown to interact et al. 1996, Kamei et al. 1996, Yao et al. 1996) but we have with proteins associated with heterochromatin and may little information about the stoichiometry of binding of play a role in chromatin remodelling (Le Douarin et al. any of these proteins to receptors. For example, it is 1996). All these proteins have been shown to contain one unclear whether the proteins bind to receptors

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14 independently or as complexes, whether they bind level of sequence conservation in the nuclear receptor sequentially or simultaneously, which may pose steric family, allows a model to be proposed for the mechanism problems given their relative sizes, and whether there is of action of hormone agonists and antagonists. The redundancy or receptor selectivity. These questions are realignment of helix 12 as a lid over the ligand binding being tackled by many laboratories at the present time. pocket provides a potential molecular explanation for the When coactivators are recruited to receptors they may disparate activities of different ligands. Ligands which serve at least two distinct roles. First, they may stabilise function as agonists would be expected to allow the the formation of a pre-initiation transcription complex of formation of an interacting surface made up from residues DNA polymerase and basal transcription factors, and in helices 3 and 12. In contrast, ligands that fail to allow secondly they may remodel chromatin by destabilising the the correct realignment of helix 12 are likely to function binding of nucleosomes. Regions required for trans- as antagonists. Interestingly, although the structure of the criptional activation have been found in both p160 and ligand appears to determine its affinity for binding to the CBP/p300 family members. These have been identified by receptor it does not seem to be as important in determining fusing fragments of these proteins to a DNA binding AF2 activity, implying that the role of agonists may be domain and testing their ability to stimulate transcription simply to release helix 12 from an inactive conformation of reporter genes in transfected cells. In SRC1, one rather than to promote its correct realignment. This may corresponds to the region involved in CBP binding and be true in the case of the ER since many environmental may reflect the recruitment of CBP and/or p300 to the oestrogens that bind relatively poorly still function as promoter. In CBP, an activation domain corresponds to a agonists (Soto et al. 1991, White et al. 1994, Jobling et al. region that binds to TATA-box binding protein (TBP) and 1995). so may directly interact with the basal transcription Hormone antagonists may be classified into two basic machinery. Both families of proteins also encode histone types, either mixed agonists/antagonists or pure ant- acetyl transferases that acetylate lysine residues on agonists, exemplified by the antioestrogens tamoxifen and specific histones which may destabilise nucleosomes to ICI 182780 respectively. The binding site in the receptor promote gene transcription (Ogryzko et al. 1996). In for both types of compound overlaps with that of oestra- addition, they are also reported to be capable of recruiting diol and so they act as competitive inhibitors of oestrogen P/CAF, another histone acetyl transferase (X-J Yang et al. action (Dauvois & Parker 1993). In vitro assays and 1996). What emerges from these studies is that co- transient transfection studies suggest that tamoxifen activators are multidomain proteins that presumably binding allows the receptor to dimerise and bind to DNA function either independently or as complexes and the with high affinity but blocks transcriptional activity goal is now to determine the precise roles and relative mediated by AF2. The agonist activity is thought to be importance of each of these domains when they are derived primarily from AF1, which is active even when recruited by different receptors to target genes. A model tamoxifen is bound to the receptor (Berry et al. 1990, showing some of the functions of coactivator proteins is Danielian et al. 1993, Tzukerman et al. 1994). Therefore shown in Fig. 3. tamoxifen has the potential to act as an antagonist when the transcriptional activity of the receptor is mediated by Role of hormonal agonists and AF2 but as an agonist on promoters or in cell types where antagonists in coactivator recruitment AF1 is active. The pure antioestrogens, however, seem to block the activity of both AF1 and AF2. Evidence is Although crystal structures have yet to be solved for the emerging that part of the mechanism of antagonism may ligand binding domain of any one receptor in the liganded be a result of the recruitment of corepressors to the and unliganded state, the overall similarity of the receptor (Horwitz et al. 1996, Jackson et al. 1997, Smith structures for apo-RXRα and holo-RARγ and the high et al. 1997). The lack of AF2 activity in the presence of

Figure 3 Interactions between receptors and coactivator proteins. Activated receptors bind p160 proteins and proteins of the CBP/p300 family in a ligand dependent manner. This DNA bound complex may then stimulate the expression of hormone responsive genes by a number of possible mechanisms including the remodelling of chromatin, interaction with other transcription factors and the stabilisation of the pre- initiation complex which consists of DNA polymerase and basal transcription factors.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

Figure 4 A model for the structure of the ligand binding domain in the presence of hormone agonists and antagonists. Although antagonists may bind with high affinity the structure of the domain may be unable to fold to generate the surface required for the binding of other proteins and realignment of helix 12 with helices 3 and 5 may be prevented. The helix numbers are based on those described by Brzozowski et al. (1997). These differ from those described by Wurtz et al. (1996) and presented in Fig. 2 due to the presence of an additional helix in ERα, helix 4. either type of antioestrogen is likely to reflect the incorrect action of ERs in other ways. They clearly reduce the realignment of helix 12. This has recently been confirmed intracellular concentration of ERs by reducing their half by the determination of models for the structure of the life (Gibson et al. 1991, Dauvois et al. 1992). In the ligand binding domain of the ER both in the presence of presence of ICI 182780, receptors accumulate in the oestradiol and in the presence of the antagonist raloxifene cytoplasm as a consequence of a block in nuclear uptake (Fig. 4). The major difference in the two structural models of the protein (Dauvois et al. 1993) and it is conceivable is an alteration in the positioning of helix 12 which is a that this aberrant accumulation and/or conformation consequence of the need to accommodate the long, bulky triggers degradation. It has also been proposed that ICI side chain of the antioestrogen in the ligand binding cavity 182780 blocks receptor dimerisation and, as a (Brzozowski et al. 1997). The failure to correctly re- consequence, inhibits DNA binding (Fawell et al. 1990b). position helix 12 and the inability to form an appropriate These observations, however, were based on in vitro surface for the binding of coactivator proteins in the experiments and have been questioned by a number of presence of hormone antagonists may be analysed in vitro. workers analysing the effects of antioestrogens in intact Addition of either tamoxifen or ICI 182780 prevents cells (Reese & Katzenellenbogen 1991, Metzger et al. binding in assay systems in which coactivator proteins are 1995). analysed for their ability to associate with the ligand Although it may be possible to assign all anti- binding domain of the ER (Henttu et al. 1997). Similarly oestrogens as either mixed agonists/antagonists or pure these antagonists inhibit the stimulatory effects of antagonists it is unlikely that their mechanism of action coactivators such as SRC1 on receptor mediated trans- will be precisely the same as either tamoxifen or ICI criptional activation in mammalian cells. Finally it may be 182780. Partial agonists are particularly interesting significant that many steroid antagonists, e.g. tamoxifen, because their agonist activity is cell specific. Thus, both ICI 182780 and RU486, are of greater size than the natural raloxifene and tamoxifen act as antagonists in breast ligand suggesting that interference with the realignment of cancer and agonists in bone while their action differs in the helix 12 may be a common mechanism of hormone endometrium where only raloxifene is an antagonist antagonism. (Draper et al. 1996, N N Yang et al. 1996). This difference In addition to their effects on the realignment of helix must reflect distinct conformations induced in the receptor 12 pure antioestrogens have been found to suppress the by the two antioestrogens which might, for example,

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14 differentially affect the sensitivity of the N-terminal lines (Ignar-Trowbridge et al. 1993, 1996) and shown to activation domain to phosphorylation by other signalling be mediated by the MAP kinase pathway which pathways or alter interactions between AF1 and AF2. phosphorylates a serine residue at position 118 in AF1 (Ali et al. 1993, Kato et al. 1995, Bunone et al. 1996). Phosphorylation of this residue seems to be required for Ligand independent stimulation of ERs AF1 activity but little is known about its role in The transcriptional activity of the ER has been reported to transcriptional activation. Since AF1 is functional in the be modulated by a number of other signalling pathways presence of tamoxifen it is likely to be important in the involving different protein kinase cascades. The best agonist effects of this antioestrogen. Cell specific characterised of these is the ability of epidermal growth variations in AF1 activity probably reflect differences in factor (EGF) to modulate the activity of the ER. It was the stimulation of the MAP kinase pathway in response to initially observed that the effects of oestrogen on the growth factors and this may account for the cell-specific mouse uterus can be mimicked by the administration of differences observed for the agonist activity of tamoxifen. EGF to animals and that this is a result of the activation of Agents that increase cyclic AMP levels and stimulate the ER in an oestrogen independent manner (Ignar- protein kinase A activity also increase transactivation by Trowbridge et al. 1992). These results have been the ER, an effect which was first demonstrated by treating confirmed by analysing the uterotrophic response in ER cells with the neurotransmitter dopamine (Power et al. knock-out mice, where the oestrogen-like effects of EGF 1991). However, it is not known whether this is achieved are absent, demonstrating that these responses are by phosphorylation of the receptor itself or other proteins dependent on the expression of the ER (Curtis et al. 1996). involved in transcriptional activation. The activation of ERs by EGF has been analysed in cell

Figure 5 Activation of the oestrogen receptor. The receptor may be activated by ligand binding and in addition by stimulation with growth factors acting via MAP kinase (MAPK) to phosphorylate AF1. The receptor may also be a target for tyrosine kinase(s) resulting in activation of AF2. Partial antagonists such as tamoxifen block ligand dependent activity but not growth factor stimulation of AF1. Pure antagonists prevent ligand dependent activation and may inhibit other pathways by completely blocking the action of the receptor protein.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

The ER is also modified by phosphorylation on Webb et al. 1995). A number of alternative mechanisms tyrosine residues, in particular at Y537 in the human have been described to explain how nuclear receptors may protein (Arnold et al. 1995), which is located at the N modulate the activity of other transcription factors. These terminus of helix 12 in the hormone binding domain. include the binding of receptors to DNA at composite Phosphorylation of this tyrosine residue is not absolutely response elements in association with other factors, e.g. required for AF2 activity since it can be replaced with the control of proliferin gene expression by the gluco- phenylalanine without affecting oestrogen dependent corticoid receptor and AP-1 (Diamond et al. 1990), and activation (White et al. 1997). However, analysis of other the oestrogen regulation of ovalbumin gene expression mutations introduced at this position suggests that (Gaub et al. 1990). The recent discovery of the ability of phosphorylation may represent an alternative mechanism receptors to associate with proteins such as CBP/p300, for ligand independent activation of the ER for the which themselves have the potential to make multiple con- following reason. Replacement of the tyrosine residue tacts with other transcription factors, suggests a possible with charged residues, either aspartic acid or glutamic mechanism for the integration of different signalling acid, which might mimic tyrosine phosphorylation, pathways at target gene promoters. For example, it has generates constitutively active receptors (Weis et al. 1996, been reported that the ability of receptors to repress AP-1 White et al. 1997). We think that this is likely to occur as activity results from the sequestering of CBP (Kamei et al. a consequence of the realignment of helix 12 to form the 1996) although other factors, as yet unidentified, have also interacting surface required to recruit coactivators. been proposed (Saatcioglu et al. 1994, 1997). Surprisingly an alanine replacement or a serine at this position has a similar effect. One interpretation of these results is that the tyrosine residue takes part in The problem of tamoxifen insensitivity and hydrophobic interactions which maintain the receptor in tamoxifen resistance an inactive state and that activation is achieved by There is no a priori reason why ER positive breast disruption of these interactions allowing the release of tumours should all be sensitive to tamoxifen treatment helix 12, analogous to the conformational change that since their proliferation may be completely independent of occurs as a result of ligand binding. The importance of this the activity of the ER. Similarly, tumours may become residue, and the possibility that phosphorylation may tamoxifen resistant as a consequence of further genetic represent a means for activating the receptor, is suggested changes that allow them to proliferate in the absence of from the sequence conservation between both ERα and functional ER. However, a relatively high proportion of ERβ, which have been analysed from a number of tamoxifen resistant tumours are still sensitive to alternate different species all of which have a tyrosine residue in endocrine therapies suggesting that the proliferation of a this position. However, the identification of a growth subset of tumours is receptor dependent. factor/kinase pathway which both phosphorylates and The molecular basis for the tamoxifen resistance of activates the receptor at this position remains elusive. A this type of tumour is unknown but several possibilities summary of the various pathways which may be involved have been proposed. A point mutation has been identified in the activation of the receptor is shown in Fig. 5. at amino acid 537 in the human ER from a breast tumour sample in which a tyrosine residue, previously shown to Interactions between nuclear receptors be a site of protein modification by phosphorylation, as and other transcription factors described earlier, is altered to asparagine. This generates a constitutively active protein which appears to be Interactions may also occur within the cell nucleus insensitive to the affects of antioestrogens (Zhang et al. between receptors and other transcription factors, for 1997). However, mutation of this residue to a number of example AP-1. This transcription factor is implicated as a other amino acids, including alanine and serine, results in critical target for many signalling pathways that regulate receptor proteins which although constitutively active are cell differentiation, proliferation and transformation inhibited by treatment with either tamoxifen or ICI (Angel & Karin 1991). It is important as a target for 182780 (Weis et al. 1996, White et al. 1997). Receptor growth factors which stimulate the phosphorylation of mutations (Jordan et al. 1995) and variants (Fuqua 1994, Fos/Jun family members and may also play a role in Fuqua et al. 1995) have also been proposed to account for growth inhibition by retinoids which have been shown to tamoxifen insensitivity. In particular, a variant lacking the block AP-1 activity (Pfahl 1993, Chen et al. 1995). oestrogen binding domain has been reported to exhibit Oestrogens, which stimulate the proliferation of MCF-7 constitutive activity and stimulate the proliferation of breast cancer cells, increase growth factor induced AP-1 breast cancer cells in vitro in the presence of tamoxifen. activity whereas antioestrogens, which inhibit growth of We have been unable to confirm this observation when the these cells, inhibit the activity of AP-1 (Philips et al. 1993, expression of this variant was induced in cell lines and we

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14 conclude that changes in other signalling pathways must Arnold SF, Obourn JD, Jaffe H & Notides AC 1995 Phosphory- also be necessary (Rea et al. 1996). One candidate is the lation of the human estrogen receptor on tyrosine-537 in vivo MAP kinase pathway, which potentiates the activity of the and by Src family tyrosine kinases in vitro. Molecular N-terminal activation domain, AF1, even in the presence Endocrinology 9 24-33. of tamoxifen. Tamoxifen, like oestrogen, seems to Aronica SM & Katzenellenbogen B 1993 Stimulation of estrogen promote the binding of the receptor to stimulate the receptor-mediated transcription and alteration in the expression of target genes and it is conceivable that AF1 phosphorylation state of the rat uterine estrogen receptor by estrogen, cyclic adenosine monophosphate, and insulin-like mediated transcription may therefore be increased in cells growth factor-I. Molecular Endocrinology 7 743-752. with enhanced MAP kinase activity. These target genes, Barettino D, Ruiz MDMV & Stunnenberg HG 1994 however, may remain sensitive to alternative endocrine Characterization of the ligand-dependent transactivation treatments. For example, ER dependent transcription domain of thyroid-hormone receptor. EMBO Journal 13 would remain sensitive to aromatase inhibitors because 3039-3049. there would be insufficient available oestrogen to activate vom Baur E, Zechel C, Heery D, Heine MJS, Garnier JM, Vivat the receptor. Similarly, ICI 182780 treatment would not V, Le Douarin B, Gronemeyer H, Chambon P & Losson R only reduce ER levels but may inhibit both activation 1996 Differential ligand-dependent interactions between the domains. Thus, the ability of alternative endocrine AF-2 activating domain of nuclear receptors and the putative therapies to block tumour growth in tamoxifen resistant transcriptional intermediary factors mSUG1 and TIF1. EMBO patients might reflect the failure of the receptor to activate Journal 15 110-124. a subset of target genes even when the N-terminal domain Beato M 1989 Gene regulation by steroid hormones. Cell 56 335- AF1 is stimulated by other signalling pathways. 344. Berry M, Metzger D & Chambon P 1990 Role of the two activating domains of the oestrogen receptor in the cell-type Future prospects and promoter-context dependent agonistic activity of the anti- An understanding of the molecular mechanisms of action oestrogen 4-hydroxytamoxifen. EMBO Journal 9 2811-2818. of the ER is not only beginning to provide an explanation Bocchinfuso WP & Korach KS 1997 Biological impact of a for the function of clinically useful antioestrogens but is disrupted estrogen receptor gene on estrogen-related cancer. Endocrine-Related Cancer 4 387-406. also suggesting novel therapeutic targets. These include, Bohen SP, Kralli A & Yamamoto KR 1995 Hold ‘em and fold for example, the interactions between coactivators and ‘em: chaperons and signal transduction. Science 268 1303- receptors which can be disrupted by specific peptides in 1304. vitro (Heery et al. 1997, Torchia et al. 1997). It may Bourguet W, Ruff M, Chambon P, Gronemeyer H & Moras D therefore be possible to prevent these interactions in vivo 1995 Crystal-structure of the ligand-binding domain of the and block the agonistic effects of oestrogens. The two human nuclear receptor RXR-alpha. Nature 375 377-382. areas in which progress has been relatively slow in the Brzozowski AM, Pike ACW, Dauter Z, Hubbard RE, Bonn T, past, namely the identification of the crucial oestrogen Engstrom O, Ohman L, Greene GL, Gustafsson J-Å & regulated target genes involved in cell proliferation and Carlquist M 1997 Molecular basis of agonism and antagonism the mechanism of tamoxifen resistance, remain very in the oestrogen receptor. Nature 389 753-758. active areas of research. Hopefully, with a better Bunone G, Briand P-A, Miksicek RJ & Picard D 1996 Activation understanding of both the mammalian cell cycle and of the unliganded estrogen receptor by EGF involves the MAP alternate intracellular signalling mechanisms, more kinase pathway and direct phosphorylation. EMBO Journal progress will continue to be made in the future. 15 2174-2183. Cavaillès V, Dauvois S, Danielian PS & Parker MG 1994 Inter- action of proteins with transcriptionally active estrogen References receptors. Proceedings of the National Academy of Sciences Ali S, Metzger D, Bornert J-M & Chambon P 1993 Modulation of the USA 91 10009-10013. of transcriptional activation by ligand-dependent phosphory- Cavaillès V, Dauvois S, L’Horset F, Lopez G, Hoare S, Kushner lation of the human oestrogen receptor A/B region. EMBO PJ & Parker MG 1995 Nuclear factor RIP140 modulates Journal 12 1153-1160. transcriptional activation by the estrogen receptor. EMBO Altucci L, Addeo R, Cicatiello L, Dauvois S, Parker MG, Truss Journal 14 3741-3751. M, Beato M, Sica V, Bresciani F & Weisz A 1996 17-beta- Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman estradiol induces cyclin d-1 gene-transcription, p36(d1)- IG, Juguilon H, Montminy M & Evans RM 1996 Role of CBP/ p34(cdk4) complex activation and p105(rb) phosphorylation P300 in nuclear receptor signalling. Nature 383 99-103. during mitogenic stimulation of g(1)-arrested human breast- Chen JY, Penco S, Ostrowski J, Balaguer P, Pons M, Starrett JE, cancer. Oncogene 12 2315-2324. Reczek P, Chambon P & Gronemeyer H 1995 Rar-specific Angel P & Karin M 1991 The role of Jun, Fos and the AP-1 agonist/antagonists which dissociate transactivation and AP-1 complex in cell-proliferation and transformation. Biochimica transrepression inhibit anchorage-independent cell- et Biophysica Acta 1072 129-157. proliferation. EMBO Journal 14 1187-1197.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

Clarke R, Dickson RB & Lippman ME 1991 The role of steroid Gaub M-P, Bellard M, Scheuer I, Chambon P & Sassone-Corsi P hormones and growth factors in the control of normal and 1990 Activation of the ovalbumin gene by the estrogen malignant breast. In Nuclear Hormone Receptors, pp 297- receptor involves the Fos-Jun complex. Cell 63 1267-1276. 319. Ed. M Parker. London: Academic Press. Gibson MK, Nemmers LA, Beckman Jr WC, Davis VL, Curtis Curtis SW, Washburn T, Sewall C, Diaugustine R, Lindzey J, SW & Korach KS 1991 The mechanism of ICI 164,384 Couse JF & Korach KS 1996 Physiological coupling of antiestrogenicity involves rapid loss of estrogen receptor in growth-factor and steroid-receptor signaling pathways - uterine tissue. Endocrinology 129 2000-2010. estrogen-receptor knockout mice lack estrogen-like response Halachmi S, Marden E, Martin G, MacKay H, Abbondanza C & to epidermal growth-factor. Proceedings of the National Brown M 1994 Estrogen receptor-associated proteins - Academy of Sciences of the USA 93 12626-12630. possible mediators of hormone-induced transcription. Science Danielian PS, White R, Lees JA & Parker MG 1992 264 1455-1458. Identification of a conserved region required for hormone Hanstein B, Eckner R, DiRenzo J, Halachmi S, Liu H, Searcy B, dependent transcriptional activation by steroid hormone Kurokawa R & Brown M 1996 p300 is a component of an receptors. EMBO Journal 11 1025-1033. estrogen receptor coactivator complex. Proceedings of the Danielian PS, White R, Hoare SA, Fawell SE & Parker MG 1993 National Academy of Sciences of the USA 93 11540-11545. Identification of residues in the estrogen receptor which Heery DM & Parker MG 1997 Ligand induced transcription by confer differential sensitivity to estrogen and hydroxy- nuclear receptors. Retinoids 13 26-30. tamoxifen. Molecular Endocrinology 7 232-240. Heery DM, Kalkhoven E, Hoare S & Parker MG 1997 A Dauvois S & Parker MG 1993 Mechanism of action of hormone signature motif in transcriptional co-activators mediates antagonists. In Steroid Hormone Action, pp 161-185. Ed. M binding to nuclear receptors. Nature 387 733-736. Parker. Oxford: IRL Press. Heinzel T, Lavinsky RM, Mullen T-M, Soderstrom M, Laherty Dauvois S, Danielian PS, White R & Parker MG 1992 Anti- CD, Torchia J, Yang W-M, Brard G, Ngo SD, Davie JR, Seto estrogen ICI 164,384 reduces cellular estrogen receptor E, Eisenman RN, Rose DW, Glass CK & Rosenfeld MG 1997 content by increasing its turnover. Proceedings of the National A complex containing N-Cor, mSin3 and histone deacetylase Academy of Sciences of the USA 89 4037-4041. mediates transcriptional repression. Nature 387 43-48. Dauvois S, White R & Parker MG 1993 The antiestrogen ICI Henttu PMA, Kalkhoven E & Parker MG 1997 AF-2 activity and 182780 disrupts estrogen receptor nucleocytoplasmic recruitment of steroid receptor coactivator 1 to the estrogen shuttling. Journal of Cell Science 106 1377-1388. receptor depend on a lysine residue conserved in nuclear Diamond MI, Miner JN, Yoshinaga SK & Yamamoto KR 1990 receptors. Molecular and Cellular Biology 17 1832-1839. Transcription factor interactions: selectors of positive or Hong H, Kohli K, Trivedi A, Johnson DL & Stallcup MR 1996 negative regulation from a single DNA element. Science 249 Grip1, a novel mouse protein that serves as a transcriptional 1266-1272. coactivator in yeast for the hormone-binding domains of Draper MW, Flowers DE, Huster WJ, Neild JA, Harper KD & steroid-receptors. Proceedings of the National Academy of Arnaud C 1996 A controlled trial of raloxifene (LY139481) Sciences of the USA 93 4948-4952. HCl - impact on bone turnover and serum-lipid profile in Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B, Kurokawa healthy postmenopausal women. Journal of Bone and Mineral R, Ryan A, Kamei Y, Soderstrom M, Glass CK & Rosenfeld Research 11 835-842. MG 1995 Ligand-independent repression by the thyroid Durand B, Saunder M, Gaudon C, Roy B, Losson R & Chambon hormone receptor mediated by a nuclear receptor co- P 1994 Activation function 2 (AF-2) of retinoic acid receptor repressor. Nature 377 397-404. and 9-cis retinoic acid receptor: presence of a conserved Horwitz KB, Jackson TA, Bain DL, Richer JK, Takimaoto GS & autonomous constitutive activating domain and influence of Tung L 1996 Nuclear receptor coactivators and corepressors. the nature of the response element on AF-2 activity. EMBO Molecular Endocrinology 10 1167-1177. Journal 13 5370-5382. Ignar-Trowbridge DM, Nelson KG, Bidwell MC, Curtis SW, Evans RM 1988 The steroid and thyroid hormone receptor Washburn TF, McLachlan JA & Korach KS 1992 Coupling of superfamily. Science 240 889-895. dual signaling pathways: epidermal growth factor action Fawell SE, Lees JA, White R & Parker MG 1990a involves the estrogen receptor. Proceedings of the National Characterization and colocalization of steroid binding and Academy of Sciences of the USA 89 4658-4662. dimerization activities in the mouse estrogen receptor. Cell 60 Ignar-Trowbridge DM, Teng CT, Ross KA, Parker MG, Korach 953-962. KS & McLachlan JA 1993 Peptide growth factors elicit Fawell SE, White R, Hoare S, Sydenham M, Page M & Parker estrogen receptor-dependent transcriptional activation of an MG 1990b Inhibition of estrogen receptor-DNA binding by estrogen-responsive element. Molecular Endocrinology 7 the ‘pure’ antiestrogen ICI 164,384 appears to be mediated by 992-998. impaired receptor dimerization. Proceedings of the National Ignar-Trowbridge DM, Pimentel M, Parker MG, McLachlan JA Academy of Sciences of the USA 87 6883-6887. & Korach KS 1996 Peptide growth factor cross-talk with the Fuqua SAW 1994 Estrogen receptor mutagenesis and hormone estrogen receptor requires the A/B domain and occurs resistance. Cancer 74 1026-1029. independently of protein kinase C or estradiol. Endocrinology Fuqua SAW, Wiltschke C, Castles C, Wolf D & Allred DC 1995 137 1735-1744. A role for estrogen-receptor variants in endocrine resistance. Jackson TA, Richer JK, Bain DL, Takimoto GS, Tung L & Endocrine Related Cancer 2 19-25. Horwitz KB 1997 The partial agonist activity of antagonist-

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access Endocrine-Related Cancer (1998) 5 1-14

occupied steroid receptors is controlled by a novel hinge means of two distinct sites. Molecular and Cellular Biology domain-binding coactivator L7/SPA and the corepressors N- 16 6029-6036. Cor or SMRT. Molecular Endocrinology 11 693-705. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Jafrati MD, Karas RH, Aronovitz M, Kim S, Sullivan Jr TR, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P & Lubahn DB, O’Donnell Jr TF, Korach KS & Mendelsohn ME Evans RE 1995 The nuclear receptor superfamily: the second 1997 Estrogen inhibits the vascular injury response in decade. Cell 83 835-839. estrogen receptor α-deficient mice. Nature Medicine 3 545- Metzger D, Berry M, Ali S & Chambon P 1995 Effect of 548. antagonists on DNA-binding properties of the human Janknecht R & Hunter T 1996 A growing coactivator network. estrogen-receptor in vitro and in vivo. Molecular Nature 383 22-23. Endocrinology 9 579-591. Jobling S, Reynolds T, White R, Parker MG & Sumpter JP 1995 Mosselman S, Polman J & Dijkema R 1996 ERβ: identification A variety of environmentally persistent chemicals, including and characterization of a novel human estrogen receptor. some phthalate plasticizers, are weakly estrogenic. FEBS Letters 392 49-53. Environmental Health Perspectives 103 582-587. Musgrove EA & Sutherland RL 1994 Cell cycle control by Jordan VC, Catherino WH & Wolf DM 1995 A mutant receptor steroid hormones. In: Seminars in Cancer Biology vol 5, as a mechanism of drug resistance to tamoxifen treatment. pp 381-389. Ed. M Parker. London: Academic Press. Annals of the New York Academy of Sciences 761 138-147. Nagy L, Kao H-Y, Chakravarti D, Lin RJ, Hassig CA, Ayer DE, Joyeux A, Cavaillés V, Balaguer P & Nicolas JC 1996 RIP 140 Schreiber SL & Evans RM 1997 Nuclear receptor repression enhances nuclear receptor-dependent transcription in vivo, in mediated by a complex containing SMRT, mSin3A, and yeast. Molecular Endocrinology 11 193-202. histone deacetylase. Cell 89 373-380. Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin O’Donnell AL & Koenig RJ 1990 Mutational analysis identifies S-C, Heyman RA, Rose DW, Glass CK & Rosenfeld MG 1996 a new functional domain of the thyroid hormone receptor. A CBP integrator complex mediates transcriptional activation Molecular Endocrinology 4 715-720. and AP-1 inhibition by nuclear receptors. Cell 85 403-414. Ogryzko VV, Schiltz RL, Russanova V, Howard BH & Nakatani Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S, Sasaki Y 1996 The transcriptional coactivators p300 and CBP are H, Masushige S, Gotoh Y, Nishida E, Kawashima H, Metzger histone acetyltransferases. Cell 87 953-959. D & Chambon P 1995 Activation of the estrogen-receptor Onate SA, Tsai SY, Tsai M-J & O’Malley BW 1995 Sequence through phosphorylation by mitogen-activated protein-kinase. and characterization of a coactivator for the steroid hormone Science 270 1491-1494. receptor superfamily. Science 270 1354-1357. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S & Parker MG 1993 Steroid and related receptors. Current Opinion Gustafsson J-A 1996 Cloning of a novel estrogen receptor in Cell Biology 5 499-504. expressed in rat prostate and ovary. Proceedings of the Pennisi E 1996 Expanding the eukaryote’s cast of chaperons. National Academy of Sciences of the USA 93 5925-5930. Science 274 1613-1614. Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Pfahl M 1993 Nuclear receptor AP-1 interaction. Endocrine Nilsson S & Gustafsson J-A 1997 Comparison of the ligand Reviews 14 651-658. binding specificity and transcript tissue distribution of Philips A, Chalbos D & Rochefort H 1993 Estradiol increases estrogen receptors α and β. Endocrinology 183 863-870. and anti-estrogens antagonize the growth factor-induced Kumar V & Chambon P 1988 The estrogen receptor binds tightly activator protein-1 activity in MCF7 breast cancer cells to its responsive element as a ligand-induced homodimer. Cell without affecting c-fos and c-jun synthesis. Journal of 55 145-156. Biological Chemistry 268 14103-14108. Le Douarin B, Zechel C, Garnier JM, Lutz Y, Tora L, Pierrat B, Power RF, Mani SK, Codina J, Conneely OM & O’Malley BW Heery D, Gronemeyer H, Chambon P & Losson R 1995 The 1991 Dopaminergic and ligand-independent activation of N-terminal part of TIF1, a putative mediator of the ligand- steroid hormone receptors. Science 254 1636-1639. dependent activation function (AF-2) of nuclear receptors, is Pratt WB & Toft DO 1997 Steroid receptor interactions with heat fused to B-raf in the oncogenic protein T18. EMBO Journal shock proteins and immunophilin chaperons. Endocrine 14 2020-2033. Reviews 18 306-360. Le Douarin B, Nielsen AL, Garnier J-M, Ichinose H, Jeanmougin Rea D & Parker MG 1996 Effects of an exon-5 variant of the F, Losson R & Chambon P 1996 A possible involvement of estrogen-receptor in MCF-7 breast-cancer cells. Cancer TIF1α and TIF1β in the epigenetic control of transcription by Research 56 1556-1563. nuclear receptors. EMBO Journal 15 6701-6715. Reese JC & Katzenellenbogen BS 1991 Differential DNA- Lee JW, Ryan F, Swaffield JC, Johnston SA & Moore DD 1995 binding abilities of estrogen receptor occupied with two Interaction of thyroid-hormone receptor with a conserved classes of antiestrogens: studies using human estrogen transcriptional mediator. Nature 374 91-94. receptor overexpressed in mammalian cells. Nucleic Acids Lees JA, Fawell SE & Parker MG 1989 Identification of two Research 19 6595-6602. transactivation domains in the mouse oestrogen receptor. Renaud J-P, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer Nucleic Acids Research 17 5477-5488. H & Moras D 1995 Crystal structure of the RAR-γ ligand- L’Horset F, Dauvois S, Heery DM, Cavaillés V & Parker MG binding domain bound to all-trans retinoic acid. Nature 378 1996 RIP-140 interacts with multiple nuclear receptors by 681-689.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access White and Parker: Molecular mechanisms of steroid hormone action

Roberts AB & Sporn MB 1992 Mechanistic interrelationships transactivational capacity is determined by both cellular and between two superfamilies: the steroid/retinoid receptors and promoter context and mediated by 2 functionally distinct transforming growth factor-β. In: Cancer Surveys, vol 14, pp intramolecular regions. Molecular Endocrinology 8 21-30. 205-220. Ed. MG Parker. Cold Spring Harbor: Cold Spring Voegel JJ, Heine MJS, Zechel C, Chambon P & Gronemeyer H Harbor Laboratory Press. 1996 TIF2, a 160 kDa transcriptional mediator for the ligand- Saatcioglu F, Bartunek P, Deng T, Zenke M & Karin M 1993 A dependent activation function AF-2 of nuclear receptors. conserved C-terminal sequence that is deleted in v-ErbA is EMBO Journal 15 101-108. essential for the biological activities of c-ErbA (the thyroid hormone receptor). Molecular and Cellular Biology 13 3675- Wagner RL, Apriletti JW, McGrath ME, West BL, Baxter JD & 3685. Fletterick RJ 1995 A structural role for hormone in the thyroid Saatcioglu F, Claret F-X & Karin M 1994 Negative trans- hormone receptor. Nature 378 690-697. criptional regulation by nuclear receptors. In: Seminars in Webb P, Lopez GN, Uht RM & Kushner PJ 1995 Tamoxifen Cancer Biology, vol 5, pp 347-359. Ed. M Parker. London: activation of the estrogen receptor/AP-1 pathway: potential Academic Press. origin for the cell-specific estrogen-like effects of Saatcioglu F, Lopez G, West BL, Zandi E, Feng W, Lu H, Esmaili antiestrogens. Molecular Endocrinology 9 443-456. A, Apriletti JW, Kushner PJ, Baxter JD & Karin M 1997 Weis KE, Ekena K, Thomas JA, Lazennec G & Mutations in the conserved C-terminal sequence in the thyroid Katzenellenbogen BS 1996 Constitutively active human hormone receptor dissociate hormone-dependent activation estrogen-receptors containing amino-acid substitutions for from interference with AP-1 activity. Molecular and Cellular tyrosine-537 in the receptor protein. Molecular Biology 17 4687-4695. Endocrinology 10 1388-1398. Sadovsky Y, Webb P, Lopez G, Baxter JD, Cavailles V, Parker MG & Kushner PJ 1995 Transcriptional activators differ in White R, Jobling S, Hoare SA, Sumpter JP & Parker MG 1994 their responses to overexpression of TATA-box-binding Environmentally persistent alkylphenolic compounds are protein. Molecular and Cellular Biology 15 1554-1563. estrogenic. Endocrinology 135 175-182. Shemshedini L, Knauthe R, Sassone-Corsi P, Pornon A & White R, Sjöberg M, Kalkhoven E & Parker MG 1997 Ligand- Gronemeyer H 1991 Cell-specific inhibitory and stimulatory independent activation of the oestrogen receptor by mutation effects of Fos and Jun on transcription activation by nuclear of a conserved tyrosine. EMBO Journal 16 1427-1435. receptors. EMBO Journal 10 3839-3849. Wurtz J-M, Bourguet W, Renaud J-P, Vivat V, Chambon P, Moras Shikama N, Lyon J & La Thangue NB 1997 The p300/CBP D & Gronemeyer H 1996 A canonical structure for the ligand- family: integrating signals with transcription factors and binding domain of nuclear receptors. Nature Structural chromatin. Trends in Cell Biology 7 230-236. Biology 3 87-94. Smith CL, Nawaz Z & O’Malley BW 1997 Coactivator and Yang NN, Venugopalan M, Hardikar S & Glasebrook A 1996 corepressor regulation of the agonist/antagonist activity of the Identification of an estrogen response element activated by mixed antiestrogen, 4-hydroxytamoxifen. Molecular metabolites of 17β-estradiol and raloxifene. Science 273 Endocrinology 11 657-666. 1222-1224. Smith DF & Toft DO 1993 Steroid-receptors and their associated proteins. Molecular Endocrinology 7 4-11. Yang X-J, Ogryzko VV, Nishikawa J-I, Howard BH & Nakatani Soto AM, Justicia H, Wray JW & Sonnenschein C 1991 p-Nonyl- Y 1996 A p300/CBP-associated factor that competes with the phenol: an estrogenic xenobiotic released from ‘modified’ adenoviral oncoprotein E1A. Nature 382 319-324. polystyrene. Environmental Health Perspectives 92 167-173. Yao T-P, Ku G, Zhou N, Scully R & Livingston DM 1996 The Stein B & Yang MX 1995 Repression of the interleukin-6 nuclear hormone receptor coactivator SRC-1 is a specific promoter by estrogen receptor is mediated by NF-κ and C/ target of p300. Proceedings of the National Academy of EBPβ. Molecular and Cellular Biology 15 4971-4979. Sciences of the USA 93 10626-10631. Tasset D, Tora L, Fromental C, Scheer E & Chambon P 1990 Yeh S & Chang C 1996 Cloning and characterization of a specific Distinct classes of transcriptional activating domains function coactivator, ARA70, for the androgen receptor in prostate by different mechanisms. Cell 62 1177-1187. cells. Proceedings of the National Academy of Sciences of the Tora L, White J, Brou C, Tasset D, Webster N, Scheer E & USA 93 5517-5521. Chambon P 1989 The human estrogen receptor has two independent nonacidic transcriptional activation functions. Zhang QX, Borg A, Wolf DM, Oesterreich S & Fuqua SA 1997 Cell 59 477-487. An estrogen receptor mutant with strong hormone- Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, Glass CK independent activity from a metastatic breast cancer. Cancer & Rosenfeld MG 1997 The transcriptional co-activator p/CIP Research 57 1244-1249. binds CBP and mediates nuclear-receptor function. Nature Zwijsen RML, Wientjens E, Klompmaker R, van der 387 677-684. Sman J, Bernards R & Michalides RJAM 1997 CDK- Tzukerman MT, Esty A, Santisomere D, Danielian P, Parker MG, independent activation of estrogen receptor by cyclin D1. Cell Stein RPJW & McDonnell DP 1994 Human estrogen-receptor 88 405-415.

Downloaded from Bioscientifica.com at 10/01/2021 12:23:40PM via free access