sign in sign up
<<
Home , Steroid hormone receptor

Cell, Vol. 83, 851-857, December 15, 1995, Copyright 0 1995 by Cell Press Steroid Receptors: Review Many Actors in Search of a Plot

Miguel Beato,’ Peter Herrlich,t and Giinther Schlitz* ization (Figure 1; reviewed by Glass, 1994). After binding *Institut ftir Molekularbiologie und Tumorforschung to DNA, the is thought to interact with compo- Philipps-Universitat Marburg nents of the basal transcriptional machinery and with se- Emil-Mannkopff-Strasse 2 quence-specific factors. Although a number D-35037 Marburg of such interactions have been described, the actual Federal Republic of Germany mechanism of action is still far from being tForschungszentrum Karlsruhe understood. We know many actors, but we do not know lnstitut fiir Genetik the plot. The only certainty is that there are many more D-76021 Karlsruhe actors than expected and that the plot they are involved Federal Republic of Germany in is neither simple nor unique. In reviewing the wealth of *Deutsches Krebsforschungszentrum recent reports on SHRs, we will describe various levels Im Neuenheimer Feld 280 of regulation, focusing on a few well-characterized exam- D-691 20 Heidelberg ples of hormonal induction and repression and on the in- Federal Republic of Germany sights gained by targeted disruption of the for SHRs.

The Curtain Rises: The Unliganded SHR Complex It tookalmost aquarterof acenturyfrom the earliest indica- In contrast with other nuclear receptors, all unliganded tion that steroid play a role in transcriptional SHRs are associated with a large multiprotein complex of control, triggered by the observation by Ulrich Clever of chaperones, including and the immunophilin ecdysoneinduced giant chromosome puffs, and from the Hsp56, which maintains the receptors in an inactive but earliest detection of steroid hormone receptors (SHRs) ligand-friendly conformation (reviewed by Pratt, 1993). to the cloning of their genes (reviewed by Evans, 1988). SHRs introduced into yeast can be activated upon ligand Although availability of the first SHR cDNA clones 10 years addition. Data obtained in mutant yeast strains suggest ago triggered the isolation of the now huge superfamily that the chaperoning proteins play an active role in keeping of nuclear receptors by homology screening with the DNA- SHRs functional. In yeast strains expressing the glucocor- binding domain (DBD) (Mangelsdorf and Evans, 1995 [this ticoid receptor, disruption of the Hsp90 homologs does issue of Cell]; Thummel, 1995 [this issue of CeW]), the not lead to constitutive activation of the receptor but rather vertebrate SHRs have remained a distinct class that are to a significant impairment of hormone induction (Bohen different in several respects from all other nuclear re- and Yamamoto, 1993). Chaperones in addition to Hsp90 ceptors. are required for SHR function, as suggested by mutants of the yeast dnaJ homolog YDJl , which also associates Prologue: The Main Actors with the unliganded SHR complex. In contrast with mu- SHRs exert their influence in embryonic development and tants found in hsp90, one ydil allele generates constitu- adult homeostasis as hormone-activated transcriptional tively active and glucocorticoid receptors (Cap- regulators. Their modular structure, consisting of a DBD, Ian et al., 1995; Kimura et al., 1995). nuclear localization signals, a ligand-binding domain (LBD), and several transcriptional activation functions Other Members of the Cast: Basal Transcription (AFs) (Figure l), is conserved with other members of the Factors and Coactivators family. Unique to the SHRs is their ability To regulate transcription, liganded SHRs must talk to the upon activation to bind to palindromic DNA sequences, transcription initiation complex. It is currently debated called hormone response elements (HREs) (Figure l), ex- whether transcription initiation complexes assemble at the clusively as homodimers, at least in vivo. The receptors TATA box in an ordered stepwise fashion (TFIID > TFIIB for glucocorticoids, mineralocorticoids, , and > RNA polymerase II + TFIIF > TFIIE > TFIIH) or are androgens recognize the same DNA sequence (AGAACA recruited as preformed complexes. Such preformed holo- as half-site) that creates a specificity problem to be dis- enzyme complexes, containing RNA polymerase II and all cussed later, while the recognizes AGG- relevant general transcription factors along with several TCA, identical with the half-site used by the nonsteroid additional polypeptides, exist in yeast (Kim et al., 1994; nuclear receptors. Mutant data, nuclear magnetic reso- Koleske and Young, 1994) and in higher eukaryotes (Ossi- nance studies, and X-ray analyses of DBDlHRE cocrystals pow et al., 1995). of glucocorticoid and estrogen receptors have shown that SHRs have been shown to interact in vitro directly with half-sites are distinguished by several amino acids (origi- components of the transcription initiation complex (re- nally named the P box by Umesono and Evans, 1989) of viewed by Tsai and O’Malley, 1994), but the physiological a recognition helix that is coordinated by a zinc-binding significance of these interactions remains unclear. There motif and makes base-specific contacts within the major have also been indications for the existence of coactiva- groove. A second zinc atom organizes both an a helix, tors that would act as bridging factors between SHRs and which is oriented alongside the axis of the DNA, and the the transcription initiation complex. A number of such in- D box, responsible, at least in part, for specific homodimer- termediary factors that interact with AF2 at the C-terminus Cell 852

-‘-iso .-.AFPj -’ -- -7 oeo / Figure 1. Domain Structure of SHRs ------.-..,‘i ___...____1... ..~- .. ..__.... ._ The is taken as a model tovisualize general features: thecore transacti- “xET icOiC;iVatorsl 1 vation domains AF1 (also called ~1) (which is RIP 140 hormone independent) and AF2 (which is hor- TIF-I / SUG-1 : mone dependent). AF2, to which the putative ..-__-- coactivators introduced in the text bind, is lo- cated in the C-terminal portion of helix 11 of the LBD. The backbone drawing is derived from the crystal structure of the RXRa LBD (Bourguet et al., 1995), possibly resembling v those of SHRs. The DBD of the glucocorticoid receptor (schematic derived from Glass, 1994) is assembled as a dimer on the palindromic HRE. The organization by the zinc-binding mo- tifs is made visible by the expanded drawing. @stands for hydrophobic in the pu- tative AF2 consensus sequence.

of SHR (Figure 1) in an agonist-dependent fashion have interference of nuclear receptors with two groups of tran- been identified (Halachmi et al., 1994). One of them, scription factorsof particular physiologic importance, AP-1 RIP140, binds only to transcriptionally active variants of and NF-~6. estrogen receptor and appears to interact with estrogen Inhibition of AP-l-dependent genes by nuclear recap- receptor in vivo (CavaiWs et al., 1995). Another protein, tors is transcriptional and rapid, does not require protein TlFl , interacts with y (RX@) and estro- synthesis, and can be traced to an interaction of AP-1 with gen and progesterone receptors and belongs to a group nuclear receptors (reviewed by Schijle and Evans, 1991; of so-called RING proteins (Le Douarin et al., 1995). The Saatcioglu et al., 1994; Herrlich and Ponta, 1994). The RING family includes PML, the to relationship is mutual in that elevated expression of AP-1 which a (RARa) is fused in acute subunits or their activation in response to growth factors promyelocytic leukemia, which itself enhances transacti- or phorbol ester inhibits HRE promoters. As an important vation by (Guiochon-Mantel et al., feature of this mutual inhibition, the interfering factor does 1995), and the estrogen-responsive finger protein Efp (ln- not seem to contact DNA, and the repressed factor re- oue et al., 1993). As also induce expression of mains DNA bound. No major change in genomic dimethyl the progesterone receptor, complex cascades of hormone sulfate footprint has been detected at the AP-1 site of the regulation as in insects (Thummel, 1995) may also exist glucocorticoid-repressed endogenous collagenase pro- in mammals. moter (Kiinig et al., 1992) nor over the HRE of the glucocor- Another AF2-binding protein, SUGl, interacts with sev- ticoid-induced and phorbol ester-repressed (and thus eral nuclear receptors, including SHRs, and is a compo- probably AP-1 -repressed) tyrosine aminotransferase nent of the RNA polymerase II holoenzyme (see (Reik et al., 1994). The mutual interference and the stoichi- Mangelsdorf and Evans, 1995). Therefore, SUGl could ometry suggested from cotransfection experiments argue be contacted by SHR for recruiting the holoenzyme. Addi- for direct interaction between SHR and AP-1. Direct inter- tional SHR-interacting proteins have been identified action occurs between in vitro translated glucocorticoid (Oiiate et al., 1995), and it is also possible that the action receptor and Jun (reviewed by Yamamoto et al., 1993; of coactivators presupposses the ligand-dependent dis- Saatcioglu et al., 1994; Herrlich and Ponta, 1994), but placement of corepressors, as described for other nuclear there is yet no convincing proof of its in vivo significance. receptors (reviewed by Mangelsdorf and Evans, 1995). The AP-1 inhibitory property of SHRs is clearly distinct The emerging picture outlines several possible interac- from their transactivating function. It is ligand dependent tions of SHRs with components of the initiation complex but appears to occur at lower ligand concentration than but also with a number of intermediary factors. The latter transactivation (Jonat et al., 1990; see references in Saat- probably form a large family with differential affinities for cioglu et al., 1994). Several antiglucocorticoids, antipro- various SHRs. Sorting out the meaning of these interac- gestins, and antiandrogens interfere with DNA binding tions is a challenge for the near future. (e.g., Becker et al., 1986; Truss et al., 1994; Heck et al., 1994), but induce transrepression (Heck et al., 1994), sug- Cross-Talk: To Be or Not to Be-Active gesting that different ligands can bring about substantially SHRs are not only capable of stimulating gene activity, different SHR conformations. Repressing and activating but are also competent transcriptional repressors. Theo- properties of SHRs are further discriminated by receptor retically, transcriptional repression occurs by competition mutations (Heck et al., 1994; Helmberg et al., 1995). In for the DNA-binding site (see examples in Thummel, particular, mutants that cannot bind to DNA, mutants with 1995), by competition for common mediators to the tran- defective AFs, and D box mutants that (in case of the scription initiation complex, or by sequestration of the tran- glucocorticoid receptor, with no known other dimerization scription factors into inactive forms. The last of these pos- interphase) cannot dimerize repress AP-1 -dependent pro- sible mechanisms is exemplified in the transcrip0onal moters perfectly well, presumably as monomers. Mutual Review: Steroid Hormone Receptors 853

repression requires N-terminal sequences of SHRs and the basic- region of Jun (see also discussion in Saatcioglu et al., 1994). Many of these conclusions rely on cotransfection experiments and therefore need to be taken with caution. How can bona fide transcription factors be converted into repressors? An interesting hint has come from the observation that glucocorticoid receptor and Jun homodi- mers synergize, while glucocorticoid receptor represses B glucocorticoid Fos-Jun heterodimers (Yamamoto et al., 1993; Teurich receptor and Angel, 1995), which suggests conformational changes by protein-protein interaction as basis for altered activity. An influence of DNA on SHR conformation has been pos- tulated (reviewed by Yamamoto et al., 1993; Lefstin et al., 1994). While well known for RXR heterodimers (Mangels- dorf et al., 1991; Saatcioglu et al., 1994), data for glucocor- ticoid receptor modulation by so-called negative HREsand composite elements are not persuasive. Binding of gluco- C glucocorticoid corticoid receptor to such elements has yet only been CXf2pt”r shown in vitro (but not in vivo, e.g., by genomic foot- printing) and often requires high concentration of recombi- nant receptor. It therefore appears that, as a common principle, SHRs can exist in a transactivating or a repressing conformation Figure 2. Synergizing and Repressing Interactionsof SHRs with Other in which the activation domains are disguised (Figure 2). Transcription Factors Interacting proteins and ligands convert one form into the In the well-known assembly with individual DNA elements in the same other. Protein-protein interaction can be mutual, and not promoter (A), transcription factors in their transactivating (circle) con- only the synergy but also the inhibition occur with only formation interact with coactivators and, in an unknown fashion, bun- one factor bound to DNA. It is not yet clear whether these dle their stimuli to the transcriptional initiation complex. Two different transcription factors can also act from one promoter element, envis- rules of mutual interactions as described for SHRs and aged here as synergy (6) of, e.g., a Jun homodimer at an AP-I-binding AP-1 can be applied to other transcription factors such as site with the glucocorticoid receptor and as repression (C). The latter GATAl and Spil and to putative cell type-specific factors interaction needs to alter the conformation of the partners (shown modulating SHR transcription factor cross-talk, all of which as rectangles, concealing the activation domains). The glucocorticoid we have not covered here. receptor may repress as a monomer (see text) and engage corepres- sion. Direct protein-protein interaction of the transcription factors, per- The interference of SHRs with the other important factor haps with participation of tissue-specific additional factors, determines of the inflammatory response, NF-KB, has only recently the regulatory properties. been studied. Overexpression of ~65, one of the transcrip- tionally active subunits of NF-KB, and of glucocorlicoid al., 1995; Berko-Flint et al., 1994), the relevance of nega- receptor, as well as in vitro binding between p65 and re- tive regulation by SHRs in the intact organism awaits con- ceptors for either glucocorticoid, progesterone, or estrogen vincing demonstration, e.g., by appropriate rodent or hu- suggest a mutual interference mechanism as for AP-1 man mutants. (Stein et al., 1993; Ray and Prefontaine, 1994; Stein and Yang, 1995; Caldenhoven et al., 1995; Scheinman et al., Do the Actors Need a Revolving Stage?: 1995a). In keeping with this notion, interleukin-2 expres- Role for Chromatin sion by a leukemic cell line selected for glucocorticoid- The interaction between proteins and DNA and among inducible apoptosis is inhibited by a transactivation- SHRs, transcription factors, and the initiation complex has defective glucocorticoid receptor mutant (Helmberg et al., to cope with the structural organization of DNA in the nu- 1995). The glucocorticoid receptor seems to block NF-KB cleus. Genetic analysis has revealed a widespread DNA binding as measured by bandshifts in vitro. Inhibition involvement of chromatin structure in gene regulation. by glucocorticoid of promoters containing NF-KB sites Transactivation by glucocorticoid receptor in yeast re- could, on the other hand, be explained by the recent find- quires components of the SWllSNF complex (Yoshinaga ing of rapid induction of lKBa synthesis in response to et al., 1992), a set of pleiotropic transactivators that coun- hormone (Scheinman et al., 1995b; Auphan et al., 1995). teract repressing functions of chromatin and are therefore IKB traps NF-KB in the cytoplasm, and its increased syn- important for transcription of inducible genes (Winston and thesis may revert NF-KB binding to promoters. These inter- Carlson, 1992). In human cells lacking a homolog of SWl2, esting findings reopen the debate on the antiinflammatory human Brm (hBrm), transactivation by glucocorticoid re- action and induction of apoptosis by SHRs. Is the balance ceptor is weak and can be selectively enhanced by expres- between apoptosis and survival regulators disturbed by sion of hBrm (Muchardt and Yaniv, 1993). Like SW12 in the interference with a survival pathway or by induction yeast, hBrm is part of a large multiprotein complex that of a suicide gene? Although indications exist (Auphan et mediates ATP-dependent disruption of a nucleosome and Cell 854

Nucleosome enables binding of GAL-linked transactivators to GALC Free DNA binding sites in nucleosomes (Kwon et al., 1994). Asecond Orientation 1 Orientation 2 human homolog, BRGl, is a nuclear protein that can re- store glucocorticoid receptor-dependent transcription in yeast strains lacking SW12 (Khavari et al., 1993). BRGl A binds specifically the retinoblastoma gene product Rb, and Rb up-regulates glucocorticoid receptor-mediated transactivation only in the presence of hBrm (Singh et al., 1995). These results document the link between SHR and the complex cellular machinery involved in chromatin dy- namics and cell cycle control.

One of the Scenes: The Mouse Mammary Tumor Virus Promoter The mouse mammary tumor virus (MMTV) promoter is a well-documented example of transcriptional control by steroid hormones. The SHRs bind to several HREs and facilitate the interaction of other transcription factors, in- cluding nuclear factor 1 (NFl) and the octamer transcrip- tion factor OTFl , with the MMTV promoter (reviewed by Truss and Beato, 1993). Nucleosomes are nonrandomly distributed on the MMTV promoter (Richard-Foy and Hager, 1987), though a more heterogeneous distribution Figure 3. Influence of Nucleosomal Phase on Protein-DNA Interac- of nucleosome positions is found by formaldehyde fixation tions (Fragoso et al., 1995). One dominant nucleosome phase (A) Proteins, such as NFI or OTFl, that interact with over half the helix circumference (stippled) cannot bind to their nucleosomally organized found both in mammalian cells and in yeast carrying an cognate sites, irrespective of the rotational orientation of the major MMTV promoter permits SHR binding to HREs while pre- groove. cluding binding of NFI (Truss et al., 1995; Chavez et al., (8) Proteins, like SHRs, which contact only a narrow sector of the 1995). This difference probably reflects the different ways helix (around 100°), would bind if the major groove were exposed in which various proteins recognize their cognate DNA (orientation l), but not if it pointed to the histone octamer (orientation 2) (Li and Wrange, 1995). sites (Figure 3). Such data imply that DNA contains confor- mational or topological information that is implemented in chromatin and modulates the accessibility to &-acting the recent description of a male with a mutation in the elements. estrogen receptor gene (Smith et al., 1994), no complete Hormone induction was believed to cause a displace- loss-of-function mutation in other human or murine SHR ment of the nucleosome over the HREs, thus allowing genes has been described that suggested that complete free access of NFl to its binding site and transcriptional loss of any one of these receptors might lead to embryonic activation (Richard-Foyand Hager, 1987). However, geno- lethality. This suspicion has been substantiated for null mic footprinting of the chromosomal MMTV promoter mutations in mice of the glucocorticoid receptor, but not shows that hormone induction does not lead to displace- for the sex steroid receptors. ment, but rather to a rearrangement of the nucleosome Glucocorticoid Receptor that enables simultaneous binding of receptors, NFl and Disruption of the glucocorticoid receptor gene is expected OTFl (Truss et al., 1995). Since these factors cannot bind to interfere with many physiological processes, such as simultaneously to the MMTV promoter on free DNA, the regulation of carbohydrate, protein and metabolism, organization in chromatin may beaprerequisiteforoptimal and modulation of immune and central nervous system induction of the MMTV promoter. One attractive possibility (CNS) responses. Unexpectedly, the analysis of the gluco- is that the hormone-induced nucleosomal change may be corticoid receptor-negative mice revealed that the recep- related to the recently observed receptor-mediated recruit- tor is also required for maturation of several organ sys- ment of the SWllSNF complex or of other chromatin re- tems, e.g., lung and adrenal gland (Cole et al., 1995), modeling factors (Figure 4). perhaps also explaining why so far only partial loss-of- function mutations have been observed in humans. Most Knockout of the Players Provides New Insights of the glucocorticoid receptor-deficient mice die shortly into Old Problems after birth owing to respiratory failure caused by lack of Even though a wealth of information on steroid action is inflation of the lungs, likely resulting from lowered produc- available, the generation of mice with mutations in the tion of surfactants and from deficiency of a glucocorticoid- major vertebrate SHRs by homologous recombination in inducible sodium channel. The adrenals of mutant mice embryonic stem cells has generated new and often unex- lack adrenergic chromaffine cells from day 13 of embry- pected insights. With the exception of the androgen insen- onic development. These cells are derived from a bipoten- sitivity syndrome (reviewed by McPhaul et al., 1993) and tial neural crest cell population that, depending on environ- Review: Steroid Hormone Receptors 855

an altered have been frequently ob- served in several species, including humans, rats, and mice. The wide phenotypic spectrum observed in patients with the androgen insensitivity syndrome will be of great value for understanding the detailed structure/function re- lationship of this receptor in vivo (McPhaul et al., 1993). Recently, amplificatiotl of the androgen receptor locus has SHR been observed in hormone-insensitive prostate cancers (Visakorpi et al., 1995), which is likely to be of importance + in prostate cancer development from a hormone-sensitive SWVSNF to a hormone-refractory state.

Perspectives: The Curtain Does Not Fall The availability of SHR knockouts and of SHR mutant mice to come will help in numerous open questions touched upon in this review, e.g., the role of glucocorticoid receptor in T cell formation and apoptosis and in the control of the acute phase response, the in vivo significance of activating and repressing SHR functions, the functional significance of SHR variants produced from nested primers or by alter- Figure 4. Model for Remodeling of Chromatin by SHR native splicing (e.g., the 0 isoform of the glucocorticoid SHR could remodel chromatin by recruiting the SWILSNF complex or other factors that, e.g., catalyze the displacement of histone H2A/ receptor and the A form of progesterone receptor; see histone H2B dimers (Cbt6 et al., 1994). also similar variants of nonsteroid receptors in Thummel, 1995), SHR function in the CNS, and the role of estrogens in bone formation and osteoporosis and in the develop- mental cues, gives rise to chromaffine cells in the ment of steroid hormone-dependent cancers. Receptor- presence of glucocorticoid or to sympathetic neurons in deficient mice will also be a prerequisite for a critical analy- the presence of nerve growth factor. sis of putative nonsteroid receptor-mediated effects of Sex Steroid Receptors gluco- and mineralocorticoids as well as of (Mani Two general conclusions can be drawn from the analyses et al., 1994; Aronica et al., 1994). of mice lacking a functional receptor for either estrogens, SHRs utilize the same or highly related DNA-binding progesterone, or : lack of these receptors is sites. This prompts the following question: how does an not lethal, and the balanced sex ratio observed suggests ubiquitous hormonal signal become interpreted in a tem- that their absence does not affect the processes leading porally and spatially restricted manner? Recent investiga- to sex determination. Ablation implicates the progesterone tions have revealed several mechanisms through which receptor and its ligand in many functions other than preg- selectivity might be achieved. One obvious mechanism nancy (Lydon et al., 1995). Mice without progesterone re- to achieve steroid-specific gene activation is differential ceptor develop normally, but female homozygous mice expression of the receptor itself. Indeed, expression of are infertile owing to abnormalities in the reproduction sys- the progesterone receptor in hepatoma cells, where it is tem. In addition to pregnancy, ovulation, luteinization, and normally not expressed, has led to activation of glucocorti- mammary gland development are impaired. coid-dependent genes. Selective steroid transport has In contrast with the plethora of spontaneous human and been suggested from studies in yeast (Kralli et al., 1995). rodent androgen receptor mutations, the lack of a mutation Selective inactivation of hormone in tissues, as seen for of the estrogen receptor suggested its involvement in glucocorticoids in the collecting ducts of the kidneys, is some vital function. Therefore, it was surprising that mice another important mechanism to guarantee selectivity without estrogen receptor are viable, even though they (Funder, 1993). are severely compromised in reproductive functions (Lu- A clinically very important open question concerns the bahn et al., 1993). Mice of both sexes are infertile, an mechanism of mitogenic effects of steroid hormones, in unexpected finding for males devoid of estrogen receptor. particular the proliferative actions of estrogens and pro- The testes are smaller, and the seminiferous tubules are gesterone in uterine and mammary tissues. Ovarian hor- structurally altered. Apparently, the estrogen receptor has mones up-regulate the transcription of immediate-early a direct role in the spermatogenic process. In the female, genes such as c-fos, c-jun, and cyclin Dl . Targeted disrup- follicular development is impaired and hemorrhagic cystic tion of the cyclin Dl gene prevents proliferation of the ovaries develop, possibly owing to excessive gonadotro- mammary gland during pregnancy (Sicinski et al., 1995) pin stimulation. Since one of the well-established targets and yields a phenotype similar to that observed in mice of estrogen action is the progesterone receptor gene, the deprived of progesterone receptor (Lydon et al., 1995). relative contributions of either receptor remain to be de- The pathway leading to cyclin Dl activation is under inves- fined. tigation and may be a target area for cancer therapy. Abnormalities of male phenotypic development due to This summary of recent developments shows that con- Cell 858

trol of gene expression by steroid hormones is far more Funder, J.W. (1993). Mineralocorticoids, glucocorticoids, receptors complex than was apparent at the time when the genes for and response elements. Science 259, 1132-I 133. SHRs were isolated. With more and more players getting Glass, C.K. (1994). Differential recognition of target genes by nuclear on stage, we realize not only this complexity but also the receptor monomers, dimers, and heterodimers. Endocrine Rev. 75, 391-407. persuasive role steroid hormones play in a vast number Guiochon-Mantel, A., Savouret, J.F., Qignon, F., Delabre, K., Milgrom, of physiologic and pathologic processes. E., and de The, H. (1995). Effect of PML and PML-RAR on the tran- scription properties and subcellular distribution of steroid hormone Acknowledgments receptors. Mol. Endocrinol., in press. Halachmi, S., Marden, E., Martin, G., MacKay, I.,Abbondanza, C., and We are grateful to numerous colleagues who sent their published and Brown, M. (1994). Estrogen receptor-associated proteins: possible unpublished data. We apologize that because of limited space not mediators of hormone-induced transcription. Science 264, 1455- all work could be included, and many are referred to through review 1458. articles. We thank our colleagues who contributed criticisms and infor- mation: Peter Angel, Andrew Cato, Martin GBttlicher, Jijrg Klug, Rein- Heck, S., Kullmann, M.,Gast, A., Ponta, H., Rahmsdorf, H.J., Herrlich, hard Ltihrmann, Paula Monaghan, Rolf MOller, Mark Nichols, Francis P., and Cato, A.C.B. (1994). Adistinct modulating domain in glucocorti- Stewart, Guntram Suske, and MathiasTruss. This workwas supported coid receptor monomers in the repression of activity of the transcription by the Deutsche Forschungsgemeinschaft. factor AP-1. EMBO J. 73, 4087-4095. Helmberg, A., Auphan, N., Caelles, C., and Karin, M. (1995). Glucocor- References ticoid-induced apoptosis of human leukemic cells is caused by the repressive function of the glucocorticoid receptor. EMBO J. 14, 452- Aronica, M.S., Kraus, W.L., and Katzenellenbogen, B.S. (1994). Estro- 460. gen action via the CAMP signaling pathway: stimulation of adenylate Herrlich, P., and Ponta, H. (1994). Mutual cross-modulation of steroid- cyclase and CAMP-regulated gene transcription. Proc. Natl. Acad. Sci. retinoic acid receptor and AP-1 transcription factor activities: a novel USA 91, 8517-8521. propertywith practical implications. Trends Endocrinol. Metab. 5,341- Auphan, N., Didonato, J.A., Rosette, C., Hemberg, A., and Karin, M. 346. (1995). lmmunosuppression by glucocorticoids: inhibition of NF-KB Inoue, S., Orimo, A., HosQi, T., Kondo, S., Toyoshima, H., Kondo, T., activity through induction of IKB synthesis. Science 270, 286-290. Ikegami, A., Ouchi, Y., Orimo, H., and Muramatsu, M. (1993). Genomic Becker, P.B., Gloss, B., Schmid, W., Strshle, U., and Schlitz, G. binding-sitecloning revealsan estrogen-responsive gene that encodes (1986). In viva protein-DNA interactions in a glucocorticoid response a RING finger protein. Proc. Natl. Acad. Sci. USA 90, 11117-l 1121. element require the presence of the hormone. Nature 324, 686-688. Jonat, C., Rahmsdorf, H.J., Park, K.-K., Cato,A.C.B., Gebel, S., Ponta, Berko-Flint, Y., Levkowitz, G., and Vardimon, L. (1994). Involvement H., and Herrlich, P. (1990). Anti-tumor promotion and antiinflammation: of c-Jun in the control of glucocorticoid receptor transcriptional activity down-modulation of AP-1 (Fos/Jun) activitybyglucocorticoid hormone. during development of chicken retinal tissue. EMBO J. 73, 646-654. Cell 62, 1189-l 204. Bohen, S.P., and Yamamoto, K.R. (1993). Isolation of Hsp90 mutants Khavari, P.A., Peterson, C.L., Tamkun, J.W., Mendel, D.B., and by screening for decreased steroid receptor function. Proc. Natl. Acad. Crabtree, G.R. (1993). BRGl contains a conserved domain of the Sci. USA 90, 11424-11428. SW12/SNF2 family necessary for normal mitotic growth and transcrip Bourguet, W., Ruff, M., Chambon, P., Gronemeyer, H., and Moras, tion. Nature 366, 170-I 74. D. (1995). Crystal structure of the ligand-binding domain of the human Kim,Y.J., Bjorklund, S., Li,Y., Sayre, M.H., and Kornberg, R.D.(1994). nuclear receptor RXA-a. Nature 375, 377-382. A multiprotein mediator of transcriptional activation and its interaction Caldenhoven, E., Liden, J., Wissnik, S., Van de Stolpe, A., Raaijmak- with the C-terminal repeat domain of RNApolymerase II. Cell 77,599- ers, J., Koenderman, L., Okret, S., Gustafsson, J.-A., and Van der 608. Saag, P.T. (1995). Negative cross-talk between RelAand the glucocor- Kimura, Y., Yahara, I., and Lindquist, S. (1995). Role of the protein ticoid receptor: a possible mechanism for the antiinflammatory action YDJI in establishing hsp90-mediated of glucocorticoids. Mol. Endocrinol. 9, 401-412. pathways. Science 268, 1382-1365. Caplan, A.J., Langley, E., Wilson, E.M., and Vidal, J. (1995). Hormone- Koleske, A.J., and Young, R.A. (1994). An RNA polymerase-II holoen- dependent transactivation by the human androgen receptor is regu- zyme responsive to activators. Nature 368, 466-469. lated by a DNAJ protein. J. Biol. Chem. 270, 5251-5257. KBnig, H., Ponta, H., Rahmsdorf, H.J., and Herrlich, P. (1992). Interfer- Cavaill&s, V., Dauvios, S., L’lllorset, F., Lopez, G., Hoarc, S., Kushner, ence between pathway-specific transcription factors: glucocorticoids P.J., and Parker, M.G. (1995). Nuclear factor RIP140 modulates tran- antagonize phorbol ester-induced AP-1 activity without altering AP-I scriptional activation by the estrogen receptor. EMBO J. 14, 3741- site occupation in viva. EMBO J. 77, 2241-2246. 3751. Kralli, A., Bohnen, S.P., and Yamamoto, K.R. (1995). LEMl, an ATP- Chavez, S., Candau, R., Truss, M., and Beato, M. (1995). Constitutive binding-cassette transporter, selectively modulates the biological po- repression and -dependent hormone activation of the tencyof steroid hormones. Proc. Natl. Acad. Sci. USA92,4701-4705. mouse mammary tumor virus promoter in yeast. Mol. Cell. Biol. 75, Kwon, H., Imbalzano, A.N., Khavari, P.A., Kingston, R.E., and Green, 6987-6998. M.R. (1994). Nucleosome disruption and enhancement of activator Cole, T.J., Blendy, J.A., Monaghan, A.P., Krieglstein, K., Schmid, W., binding by a human SWlISNF complex. Nature 370, 477-481. Aguzzi, A., Fantuzzi, G., Hummler, E., Unsicker, K., and Schijtz, G. Le Douarin, B., Zechel, C., Garnier, J.M., Lutz, Y., Tora, L., Pierrat, (1995). Targeted disruption of the glucocorticoid receptor gene blocks B., Heery, D., Gronemeyer, H., Chambon, P., and Losson, R. (1995). adrenergicchromaffin cell development and severely retards lung mat- The N-terminal part of TIFI, a putative mediator of the ligand- uration. Genes Dev. 9, 1608-1621, dependent activation function (AF-2) of nuclear receptors, is fused to C6t& J., Quinn, J., Workmann, J.L., and Peterson, C.L. (1994). Stimu- B-raf in the oncogenic protein T18. EMBO J. 74, 2020-2033. lation of GAL4 derivative binding to nucleosomal DNA by the yeast Lefstin, J.A., Thomas, J.R., and Yamamoto, K.R. (1994). Influence of SWI/SNF complex. Science 265, 53-59. a steroid receptor DNA-binding domain on transcriptional regulatory Evans, R.M. (1988). The steroid and thyroid super- functions. Genes Dev. 8, 2842-2856. family. Science 240, 889-895. Li, Q., and Wrange, 0. (1995). Accessibility of a glucocorticoid re- Fragoso, G., John, S., Roberts, M.S., and Hager, G.L. (1995). Nucleo- sponse element in a nucleosome depends on its rotational positioning. some positioning on the MMTV LTR results from the frequency-biased Mol. Cell. Biol. 75, 4375-4384. occupancy of multiple frames. Genes Dev. 9, 1933-1947. Lubahn, D.B., Moyer, J.S., Golding, T.S., Couse, J.F., Korach, K.S., Review: Steroid Hormone Receptors 857

and Smithies, 0. (1993). Alteration of reproductive function but not gene in a man. N. Engl. J. Med. 337, 1056-1061. prenatal sexual development after insertional disruption of the mouse Stein, B., and Yang, M.X. (1995). Repression of the interleukin-8 pro- estrogen receptor gene. Proc. Natl. Acad. Sci. USA 90, 11162-I 1166. moter by estrogen receptor is mediated by NF-KB and CIEBP8. Mol. Lydon, J.P., DeMayo, F.J., Funk, C.R., Mani, SK., Hughes, AR., Cell. Biol. 15, 4971-4979. Montgomery, C.J.J., Shyamala, G., Conneely, O.M., and O’Malley, Stein, B., Baldwin, AS., Ballard, D.W., Greene, W.C., Angel, P., and B.W. (1995). Mice lacking progesterone receptor exhibit pleiotropic Herrlich, P. (1993). Cross-coupling of the NF-KB ~65 and FoslJun reproductive abnormalities. Genes Dev. 9, 2268-2278. transcription factors produces potentiated biological function. EMBO Mangelsdorf, D.J., and Evans, R.M. (1995). The RXR heterodimers J. 12, 3879-3891. and orphan receptors. Cell 83, this issue. Teurich, S., and Angel, P. (1995). The glucocorticoid receptor syner- Mangelsdorf, D.J., Umesono, K., Kliewer, S.A., Borgmeyer, U., Ong, gizes with Jun homodimers to activate AP-l-regulated promoters lack- ES., and Evans, R.M. (1991). A direct repeat in the cellular retinol- ing GR binding sites. Chem. Senses 20, 251-255. binding protein type II gene confers differential regulation by RXR and Thummel, C.S. (1995). From embryogenesis to metamorphosis: the RAR. Cell 66, 555-561. regulation and function of Drosophila nuclear receptor superfamily Mani, S.K., Allen, J.M., Clark, J.H., Blaustein, J.D., and O’Malley, B.W. members. Cell 83, this issue. (1994). Convergent pathways for steroid hormone- and neurotransmit- Truss, M., and Beato, M. (1993). Steroid hormone receptors: interac- ter-induced rat sexual behavior. Science 265, 1246-1249. tion with DNA and transcription factors. Endocrine Rev. 74, 459-479. McPhaul, M.J., Marcelli, M., Zoppi, S., Griffin, J.E., and Wilson, J.D. Truss, M., Bartsch, J., and Beato, M. (1994). Antiprogestins prevent (1993). Genetic basis of endocrine disease. IV. The spectrum of muta- progesterone receptor binding to hormone responsive elements in tions in the androgen receptor gene that causes androgen resistance. viva. Proc. Natl. Acad. Sci. USA 91, 11333-I 1337. J. Clin. Endocrinol. Metab. 78, 17-23. Truss, M., Bartsch, J., Schelbert, A., Hache, R.J.G., and Beato, M. Muchardt, C., and Yaniv, M. (1993). A human homologueof Saccharo- (1995). Hormone induces binding of receptorsand transcription factors myces cefevisiae SNFPlSWl2 and Drosophilia brm genes potentiates to a rearranged nucleosome on the MMTV promoter in viva. EMBO transcriptional activation by the glucocorticoid receptor. EMBO J. 72, J. 14, 1737-1751. 4279-4290. Tsai, M.J., and O’Malley, B.W. (1994). Molecular mechanisms of action Oftate, S.A., Tsai, M.J., and O’Malley, B.W. (1995). Sequence and of steroid/thyroid receptorsuperfamily members.Annu. Rev. Biochem. characterization of a for the steroid hormone receptor su- 63, 451-486. perfamily. Science, 270, 1354-1357. Umesono, K., and Evans, R.M. (1989). Determinants of target gene Ossipow, V., Tassan, J.-P., Nigg, E.A., and Schibler, U. (1995). A specificity for steroid/thyroid hormone receptors, Cell 57, 1139-I 146. mammalian RNA polymerase II holoenzyme containing all compo- Visakorpi, T., Hyytinen, E., Koivisto, P., Tanner, M., Keinanen, R., nents required for promoter-specific transcription initiation. Cell 83, Palmberg, C., Palotie, A., Tammela, T., Isola, J., and Kallioniemi, O.-P. 137-I 46. (1995). In viva amplification of the androgen receptor gene and pro- Pratt, W.B. (1993). The role of heat shock proteins in regulating the gression of human prostate cancer. Nature Genet. 9, 401-406. function, folding, and trafficking of the glucocorticoid receptor. J. Biol. Winston, F., and Carlson, M. (1992). Yeast SNFlSWl transcriptional Chem. 268, 21455-21458. activators and the SPTlSlN chromatin connection. Trends Genet. 8, Ray, A., and Prefontaine, K.E. (1994). Physical association and func- 387-391. tional antagonism between the ~65 subunit of transcription factor Yamamoto, K.R., Pearce, D.,Thomas, J., andMiner, J.N. (1993).Com- NF-KB and the glucocorticoid receptor. Proc. Natl. Acad. Sci. USA binatorial regulation at a mammalian composite element. In Transcrip- 91, 752-758. tional Regulation, S.L. McKnight and K.R. Yamamoto, eds. (Cold Reik, A., Stewart, A.F., and Schiitz, G. (1994). Cross-talk modulation Spring Harbor, New York: Cold Spring Harbor Laboratory Press), pp. OS signal transduction pathways: two mechanisms are involved in the 3-32. control of tyrosine aminotransferase gene expression by phorbol es- Yoshinaga, SK., Peterson, C.L., Herskowitz, I,, and Yamamoto, K.R. ters. Mol. Endocrinol. 8, 490-497. (I 992). Roles of SWII, SWl2, and SW13 proteins for transcriptional Richard-Foy, H., and Hager, G.L. (1987). Sequence-specific position- enhancement by steroid receptors. Science 258, 1598-1604. ing of nucleosomes over the steroid-inducible MMTV promoter. EMBO J. 6, 2321-2328. Saatcioglu, F., Claret, F.-X., and Karin, M. (1994). Negative transcrip- tional regulation by nuclear receptors. Semin. Cancer Viol. 5, 347- 359. Scheinman, R.I., Gualberto, A., Jewell, C.M., Cidlowski, J.A., and Baldwin, AS., Jr. (1995a). Characterization of mechanisms involved in transrepression of NF-KB activated glucocorticoid receptors. Mol. Cell. Biol. 75, 943-953. Scheinman, RI., Cogswell, PC., Lofquist, A.K., and Baldwin, AS., Jr. (1995b). Role of transcriptional activation of IrcBa in mediation of immunosuppression by glucocorticoids. Science 270, 283-288. Schule, R., and Evans, R.M. (1991). Cross-coupling of signal transduc- tion pathways: meets leucine zipper. Trends Genet. 7,377- 381. Sicinski, P., Donaher, J.L., Parker, S.B., Li, T.S., Fazeli, A., Gardner, H., Haslam, S.Z., Bronson, R.T., Elledge, S.J., and Weinberg, R.A. (1995). Cyclin Dl provides a link between development and oncogene- sis in the retina and breast. Cell 82, 621-630. Singh, P., Coe, J., and Hong, W.J. (1995). A role for in potentiating transcriptional activation by the glucocorticoid receptor. Nature 374, 562-585. Smith, E.P., Boyd, J., Frank, G.R., Takahashi, H., Cohen, R.M., Specker, B., Williams, T.C., Lubahn, D.B., and Korach, K.S. (1994). Estrogen resistance caused by a mutation in the estrogen-receptor