Genetic Dissection of the Signaling Domain of a - NCBI - NIH

Molecular Biology of the Cell Vol. 3, 1245-1257, November 1992

Genetic Dissection of the Signaling Domain of a Mammalian Steroid Receptor in Yeast Michael J. Garabedian and Keith R. Yamamoto Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, California 94143-0448 Submitted July 10, 1992; Accepted August 24, 1992

The mechanism of signal transduction by steroid receptor proteins is complex and not yet understood. We describe here a facile genetic strategy for dissection of the rat glucocorticoid receptor "signaling domain," a region of the protein that binds and transduces the hormonal signal. We found that the characteristics of signal transduction by the receptor expressed in yeast were similar to those of endogenous receptors in mammalian cells. Interestingly, the rank order of particular ligands differed between species with respect to receptor binding and biological efficacy. This suggests that factors in addition to the receptor alone must determine or influence ligand efficacy in vivo. To obtain a collection of receptors with distinct defects in signal transduction, we screened in yeast an extensive series of random point mutations introduced in that region in vitro. Three phenotypic classes were obtained: one group failed to bind hormone, a second displayed altered ligand specificity, and a third bound hormone but lacked regulatory activity. Our results demonstrate that analysis of glucocorticoid receptor action in yeast provides a general approach for analyzing the mechanism of signaling by the nuclear receptor family and may facilitate identification of nonreceptor factors that participate in this process. INTRODUCTION Steroid receptors are direct transducers of hormonal signaling information. For example, interaction of a cognate ligand with the glucocorticoid receptor protein triggers nuclear localization of the hormone-receptor complex, its binding to specific DNA sequences termed glucocorticoid response elements (GREs), and regulation of transcriptional initiation from nearby promoters (Yamamoto, 1985; Evans, 1988; Beato, 1989; Godowski and Picard, 1989). Within the 795 amino acid (aa) rat glucocorticoid receptor, a 250 aa C-terminal segment is sufficient for signal transduction: it binds hormone (Godowski et al., 1987; Picard et al., 1987; Rusconi et al., 1987) and also specifies a reversible "protein inactivation" function that in the absence of hormone inhibits other receptor activities (Godowski et al., 1988; Picard et al., 1988). Truncated receptors that lack this "signaling domain" bind constitutively to GREs and modulate transcription (Godowski et al., 1987). Protein inactivation can be conferred by fusion of the signaling domain to heterologous proteins, and inactivation is reversed by hormone binding (Picard and Yamamoto, 1987; Picard et al., 1988; Godowski et al., 1988; © 1992 by The American Society for Cell Biology

Yamamoto et al., 1988; Eilers et al., 1989; Hope et al., 1991; Superti-Furga et al., 1991; Umek et al., 1991). The ability of the unliganded signaling domain to confer a dominant inhibitory effect that is independent of protein structure appears to exclude stereospecific or allosteric models of signal transduction, which invoke either masking and unmasking of functional regions or induced folding of functional domains transduced intramolecularly from the signaling region. The inactivation function requires a subregion of the domain that is also necessary for interaction of the receptor with the heat shock protein Hsp9O (Howard et al., 1988, 1990; Pratt et al., 1988; Dalman et al., 1991), which associates with the signaling domain of the unliganded receptor (aporeceptor) and dissociates on hormone binding. Genetic studies indicate that Hsp9O participates in the signal transduction pathway for steroid receptors (Picard et al., 1990a), and the Hsp9O-receptor interaction appears to be necessary for specific steroid binding (Bresnick et al., 1989; Dalman et al., 1989; Nemoto et al., 1990; Scherrer et al., 1990). One speculation is that Hsp9O actively alters the conformations of the aporeceptor and of fusion proteins bearing the signaling domain (Picard et al., 1988; Yamamoto et al., 1988). 1245

M.J. Garabedian and K.R. Yamamoto

The ligand binding properties of the receptor reveal other complexities in the signaling process; namely, the high-affinity and stereospecific hormone-receptor interactions can be accomplished with a series of ligands that is quite diverse in structure (Samuels and Tomkins, 1970; Rousseau et al., 1972; Rousseau and Schmit, 1977). Some ligands, such as dexamethasone and cortisol, are agonists and therefore activate receptor functions; others, such as RU486 and progesterone, are antagonists that bind competitively with agonists but evoke little or no receptor action. Biochemical studies imply that two closely spaced cysteines may be involved in steroid binding, and three residues (cysteine 656, methionine 662, and cysteine 754) have been crosslinked to bound ligands (Simons et al., 1987; CarlstedtDuke et al., 1988; Miller and Simons, 1988), but the nature of ligand specificity and the relationship of ligand structure to receptor activity are not understood. In principle, it might be possible to dissect the signaling process by "reverse genetics," in which sequence alterations are introduced in vitro in the domain, and their biological effects assessed in cultured cells lacking endogenous receptors. Although a few interesting mutants have been identified by this strategy (Chakraborti et al., 1991; Benhamou et al., 1992; Danielian et al., 1992), it is labor-intensive and time-consuming, and does not permit screening of large numbers of lesions. Indeed, insertion and deletion of clusters of amino acids throughout the signaling domain have yielded only derivatives that are fully defective for all receptor activities (Giguere et al., 1986; Rusconi and Yamamoto, 1987). Thus, the entire segment seems to be involved in configuring the functional domain, but precise structural determinants, and the roles of Hsp9O or other putative nonreceptor participants, have not emerged with the use of these methods. To pursue the functional complexity of steroid receptor signaling, we sought to develop a rapid and efficient genetic approach by capitalizing on the capacity of the rat glucocorticoid receptor to mediate hormone-dependent transcriptional enhancement when expressed in Saccharomyces cerevisiae (Schena and Yamamoto, 1988; Schena et al., 1989; Wright et al., 1990). This activity is representative of the emerging generality that many biological processes have been evolutionarily conserved to such an extent that molecular components from diverse species will interact faithfully to form functional complexes. For example, several factors that govern transcription initiation and its regulation, as well as enzymes that control the cell division cycle, have been shown to complement, in vivo or in vitro, counterparts from species as divergent as yeast and humans (Lee and Nurse, 1987; Buratowski et al., 1988; Becker et al., 1991; Cormack et al., 1991; Gill and Tjian, 1991; Leopold and O'Farrell, 1991; Lew et al., 1991). In principle, then, analysis of a complex process in mammals may be approached profitably in an experimentally simpler and 1246

genetically manipulable organism such as yeast or Drosophila. In the case of mammalian steroid receptor action, no corresponding endogenous machinery has been identified in yeast. Nevertheless, the receptors are functional in this heterologous setting, and the yeast Hsp9O homologues replace mammalian Hsp9O to potentiate receptor signaling (Picard et al., 1990a). Thus despite the absence of an endogenous pathway for glucocorticoid regulation in yeast, yeast components required for activity of the mammalian factor are present and competent. Whether nonreceptor factors in addition to Hsp9O are also required for signaling was unknown, and the extent to which receptor signaling in yeast mimicks precisely the mammalian process had not been examined. At the outset of our studies, then, we sought to assess whether analysis of steroid receptor action in yeast would provide information relevant to their activities in their normal settings. First, we compared the general properties of the signaling domain, in yeast and mammalian cells, by testing the effects of various deletion, substitution, and point mutations in the C-terminal segment. Second, we examined in yeast the binding affinities and functional effects of a series of steroid ligands. With this information in hand, we then screened a large set of random point mutations in the signaling domain, isolating and characterizing a subset of mutant receptors with distinct defects in steroid binding and response. MATERIALS AND METHODS

Plasmid Constructs The pGPD vector harboring the glucocorticoid receptor derivatives contains the glyceraldehyde-3-phosphate dehydrogenase (GPD) promoter, the TRP1 gene and 2,u plasmid replication origin from yeast and the Col El replication origin and ampicillin resistance gene of pUC 18 (Schena and Yamamoto, 1988; Schena et al., 1989). Reporter plasmids GRE-CYC1-LACZ (pAS26X) and LEX-CYCl-LACZ were constructed by inserting at position -178 of a truncated CYCl promoter three tandem 26-bp oligonucleotides derived from the tyrosine aminotransferase GRE or a single E. coli LEX operator, respectively (Godowski et al., 1988; Schena et al., 1989); these plasmids also contained the URA3 gene as a selectable marker, the Col El replication origin and ampicillin resistance gene of pUC 18 and the yeast 2,u plasmid replication origin. For integration into the yeast chromosome, plasmid pl-G26. 1 was constructed and contains the yeast URA3 gene, a GRE-linked CYC1-LAC Z fusion and a 760-bp fragment of the LEU 2 gene. Plasmid VARO (Picard and Yamamoto, 1987) was used for transient expression of wild-type and mutant receptor sequences in cultured mammalian cells. This plasmid contains the SV40 enhancer and aglobin promoter as well as the rabbit f,-globin splice and poly(A) addition sites, upstream and downstream, respectively, of the inserted receptor sequences. Reporter vector G46TCO contains the bacterial chloramphenicol acetyltransferase (CAT) gene driven by the thymidine kinase promoter (-109 to +54 relative to the start site), a 46-bp synthetic GRE derived from the murine mammary tumor virus long terminal repeat, and the Col El replication origin and ampicillin resistance gene of pUC 18 (Sakai et al., 1988; LaBaer, 1989). Molecular Biology of the Cell

Genetics of Steroid Receptor Signaling

Yeast Strains and Growth Conditions The protease deficient strain BJ2168 (a, pep4-3, prcl-407, prb-1122, ura3-52, trpl, leu2) (ones, 1977) was used in all experiments. Strain BJG26.1 was constructed by integrating plasmid pl-G26.1 at leu2 of BJ2168 (Schena et al., 1989). Cultures were propagated at 30°C in minimal yeast medium (yeast nitrogen base without amino acids; Difco, Detroit, MI) supplemented with amino acids and 2% glucose.

Steroids Dexamethasone (DEX), triamcinolone acetonide (TA), hydrocortisone (HC), deoxycorticosterone (DOC), progesterone (PROG), ,B-estradiol (E2), aldosterone (ALDO), and testosterone (TEST) were purchased from Sigma (St. Louis, MO); RU28362 was obtained from New England Nuclear (Boston, MA); deacylcortivazol (DAC) and RU 486 were generous gifts of S. Simons (NIH) and Roussel-Uclaf, respectively. Radiolabeled ligands 3H-dexamethasone (45 Ci/mmol) and 3Htriamcinolone acetonide (51.4 Ci/mmol) were purchased from New England Nuclear. Stock solutions of steroids were prepared in 95% ethanol and stored in the dark at -20°C; under these conditions the steroids are stable for many months. Steroids were added to media as 103-fold stocks for all concentrations; vehicle alone was added to control cultures.

Mammalian Cell Culture and Transfections CV-1, COS 7, HeLa, and the rat hepatoma HTC cell derivative GrH2, which expresses receptor at approximately eight-fold higher than normal levels (Vanderbilt et al., 1987; Howard et al., 1990) were propagated in Dulbecco's modified Eagle's medium H-16 (Cell Culture Facility, UCSF) supplemented with 5% fetal calf serum (Hyclone, Logan, UT). Cells were grown at 37°C in a humidified incubator (Rh 90%) maintained at 8% CO2. For CAT assays, subconfluent cultures (2 X 105 cells/60-mm dish) were cotransfected with the calcium phosphate procedure (Godowski et al., 1988) with 0.2 Mug VARO receptor expression vector and 1 ,ag G46TCO reporter vector. For in vitro hormone binding and analysis of receptor protein, COS 7 cells (6 X 105 cells/100-mm dish), which lack endogenous receptor, were transfected with 25 ,ug of VARO DNA as described above. In these cells the transfected DNA is competent to replicate, yielding higher levels of receptor expression and thus facilitating the assay. -

CAT and A-Galactosidase Assays For CAT assay, cell cultures were incubated for 14 h with transfecting DNA precipitates, after which they were washed twice with phosphate buffered saline (PBS), and 4 ml of fresh medium was added together with the indicated ligand. After an additional 24 h, cells were rinsed twice with 5 ml of cold PBS, and with the use of a rubber policeman, they were scraped from the culture dish in 1 ml of PBS and transferred to a 1.5 ml microcentrifuge tube on ice. The cells were pelleted at 5000 X g for 1 min in a microcentrifuge. The cell pellets were resuspended in 125 Iul of 0.25 M tris(hydroxymethyl)aminomethane(Tris) pH 7.5. Extracts were prepared by four freeze-thaw cycles (-70°C and 65°C) followed by centrifugation for 5 min at 12 000 X g. Heattreated extracts (5 min at 65°C) were normalized to protein content, and a constant amount of protein (50 Mg) was subjected to the thinlayer chromatographic CAT assay (Godowski et al., 1988). CAT activity was quantitated by Beta Scope fluorometry. With yeast liquid cultures, quantitative fl-galactosidase measurements were carried out as described by Schena et al. (1989). Cultures were grown at 30°C in the appropriate selective media containing 2% glucose, subcultured 1:10, treated with hormone or with vehicle and grown for an additional 8 h. Typically, 0.5 ml of cells were transferred to a microcentrifuge tube, pelleted, washed once with 0.5 ml of LacZ buffer (10 mM KCl, 1 mM MgSO4, 50 mM f-mercaptoethanol, and 100 mM NaPO4, pH 7.0), pelleted again and resuspended in 50 MAl of LacZ buffer. To permeabilize the yeast cells, 50 Ml of CHCl3 and Vol. 3, November 1992

20,Ml of 0.1% SDS was added to the suspension and vortexed on the maximum setting for 10 s. The cells were then placed at 30°C for 5 min to equilibrate temperature before the addition of 3-galactosidase substrate o-nitrophenyl-f-galactoside (0.7 ml of 2 mg/ml ONPG [Sigma] in LacZ buffer). After 1-10 min at 30°C, the reaction was stopped by the addition of 0.5 ml of 1 M Na2CO3. ,B-galactosidase units were defined as 103 times the change in optical density (OD) at 420 nm (due to the hydrolysis of o-nitrophenyl-f-galactoside) divided by the product of the assay duration (min) times the culture volume (ml) times the OD at 600 nm of the culture.

Preparation of Extracts for Hormone Binding Assay Yeast extracts were prepared from 50-ml cultures at -0.5-0.9 OD600 (strain BJ2168 containing pGPD expressing full-length glucocorticoid receptor [N795]) at 30°C in minimal medium containing 2% glucose. Cells were pelleted by centrifugation (5000 X g for 5 min), washed once with ice cold phosphate-buffered saline (PBS), and pelleted again. PBS was removed and cells were suspended in one volume (typically 200-500 Ml) of receptor buffer (50 mM NaCl, 10 mM fris pH 7.5, 1 mM EDTA, 1 mM DTT, 15 mM MgCl2, 20 mM sodium molybdate, 20% glycerol, 1 mg/ml each of protease inhibitors leupeptin [Boehringer-Mannheim, Indianapolis, IN], pepstatin A [Sigma], aprotinin [Boehringer-Mannheim], 1 mM phenylmethylsulfonyl fluoride [Sigma]). Cells were transferred to a 1.5-ml microcentrifuge tube on ice, filled with glass beads to near the top, leaving a slight liquid meniscus, and agitated on an Eppendorf (Freemont, CA) horizontal mixer for 10 min at 4°C to effect cell lysis. After homogenization, a hole was punched through the bottom of the tube with a 20-gauge needle and the cell lysate was harvested in a 1.9-ml microcentrifuge tube by brief (2-5 s) centrifugation. The lysate was then centrifuged in a microcentrifuge at 12 000 X g for 10 min at 4°C; the supernatant was transferred to a fresh tube, carefully avoiding the floating lipids, and the protein concentration (typically 5-20 mg/ml) determined (Bio Rad, Richmond, CA; assay) with bovine serum albumin as a standard. For mammalian cell extracts, cells were scraped from a 100-mm tissue culture dish into 5 ml of cold PBS, centrifuged, and lysed by sonication in 200-500 Ml receptor buffer. The lysate was centrifuged at 12 000 X g for 10 min at 4°C; protein concentration of the extract supematant was typically 5-10 mg/ml.

Hormone Binding Assay Cell extract (25-50,Ml) was incubated with various concentrations of

a radiolabeled ligand in the presence or absence of a 100- to 200-fold excess of unlabeled ligand, in a final volume of 75 Ml of receptor buffer. After an 18-h incubation on ice to reach equilibrium, the extract was mixed with an equal volume of a 10 mg/ml acid washed charcoal suspended in receptor buffer, incubated for 10 min at 0°C, and centrifuged at 12 000 X g for 2 min. The supematant (135 Ml) was spotted onto a Whatman (Clifton, NJ) 2.4 cm GF/C filter, which was washed in batch for 10 min in 1 1 of ice cold 50 mM NaCl, 10 mM Tris pH 7.5, dried under a heat lamp, and counted in Safety Solve (Research Products International, Mt Prospect, IL). Specific binding was computed as counts bound in the absence of excess unlabeled ligand minus counts bound in its presence.

Steroid Chromatography Yeast extract (100 ,l, 10 mg/ml) was brought to 500 nM 3H-dexa-

methasone and incubated for 2 or 24 h at 30°C. The steroid was extracted into four volumes of ethyl acetate by vortexing for 30 s and centrifuging 12 000 X g for 1 min to separate organic and aqueous phases. The upper organic phase, which contained >98% of the radioactivity, was transfered to a fresh tube, evaporated to dryness in a Speedvac (Savant, Farmingdale, NY) and redissolved in 30 Ml of ethyl acetate. The samples were spotted onto 20 X 20 cm silica gel thin-layer sheets and chromatographed in an ethyl acetate-chloroform (50:50 vol/vol) solvent system at room temperature (Samuels and

1247


Tomkins, 1970) until the solvent front reached the top of the plate. The plates were air dried, sprayed with Enhance (New England Nuclear), air dried again, and exposed to X-ray film for 2 d at -70°C.

Chemical Mutagenesis An 820-bp Sph I-Eco RI fragment was excised from rat glucocorticoid receptor cDNA; the fragment, which encodes amino acids 494-768 of the 795 aa receptor and includes 90% of the signaling domain, was transferred to a single-stranded vector (M13mpl8) for mutagenesis. Single-stranded DNA (10 qg) representing the noncoding strand was treated for 6 min with 2 M sodium nitrite (Myers et al., 1985; Schena et al., 1989), and reverse transcriptase (Boehringer-Mannheim) was used to extend through the mutagenized region from a primer hybridizing just upstream of the M13 polylinker (primer 1211; New England Biolabs, Beverley, MA). Mutagenized double stranded fragments were excised with Sph I and EcoRI, purified on low-melt agarose (SeaPlaque 50122; FMC, Rockland, ME), and reinserted into the wildtype receptor backbone in the pGPD yeast expression plasmid. The ligation mixture was transformed into E. coli strain DH 5a, and plasmid DNA was prepared from 2 X 105 bacterial transformants, yielding a mutagenized receptor pool. These species were transformed by the lithium acetate method into yeast strain BJG26.1, which bears an integrated GRE-linked reporter gene (Schena et al., 1989). Transformants were selected on minimal plates deficient in uracil and tryptophan and containing 10 ,M DOC. Colonies (-200-300/plate) were transferred to nitrocellulose filters, lysed in liquid nitrogen, and scored for (3-galactosidase activity with 0.2 mg/ml of 5-bromo-4-chloro-indolyl13-D-galactopyranoside (X-Gal) for 10 min at 30'C (Schena et al., 1989). Yeast transformants expressing low levels of f-galactosidase were detected as white colonies in the presence of hormone (frequency -0.3%). Transformants expressing ,B-galactosidase in the absence of hormone were scored as blue colonies in the presence of X-Gal (frequency -0.2%). Yeast plasmid isolation and DNA sequencing were as described in Schena et al. (1989). Briefly, cells from 1.5 ml of stationary yeast cultures were collected by centrifugation, washed in 1 ml of 1 M sorbitol, and incubated in 250 ,ul of 1 M sorbitol, 50 mM Tris pH 7.5, 20 mM ,B-mercaptoethanol, and 2 mg/ml zymolyase 1OOT (ICN, Irvine, CA) at 37°C for 30 min. Spheroplasts were pelleted, resuspended in 125 ul of 50 mM EDTA and 0.3% SDS, and incubated at 68°C for 20 min. Cell debris was removed by ammonium acetate precipitation and the plasmid DNA was phenol extracted, ethanol precipitated, and dissolved in 25 yl of 10 mM Tris pH 8.0, 1 mM EDTA, containing 100 ,g/ml of RNAse A; 1 ,gl was used to transform E. coli strain DH 5a. Double stranded DNA was sequenced with oligonucleotide primers (Biomolecular Resource Center, UCSF) complementary to receptor sequences flanking the mutagenized region.

RESULTS Receptor Derivatives in Yeast Mammalian steroid receptors have been shown to mediate hormone-dependent transcriptional regulation when expressed in yeast (Metzer et al., 1988; Schena and Yamamoto, 1988; Mak et al., 1989; Wright et al., 1990). In principle, this heterologous host might facilitate genetic analysis of the signal transduction mechanism, assuming that the mechanism in yeast reflects accurately that used in mammalian cells. To examine this point, we tested in yeast the activities of various receptor derivatives whose signaling properties had been well characterized in mammalian cells. Rat glucocorticoid receptor cDNA sequences expressed from the yeast glycerol 3-phosphate dehydrogenase (GPD) 1248

promoter were stably introduced into a yeast strain, BJ2168, which also contains a reporter gene with tandem GREs upstream of the yeast cytochrome cl (CYC1) promoter fused to the E. coli 3-galactosidase (LACZ) coding sequences (Schena and Yamamoto, 1988; Picard et al., 1990b; Schena et al., 1991). As in mammalian cells, mutant receptors lacking 27 or 150 C-terminal amino acids (N768 and N645) failed to activate GREs in yeast either in the presence or absence of hormone (Godowski et al., 1987) (Figure 1, ac). Similarly, deletion mutants encoding receptor derivatives N508 and N464, which delete portions of the DNA binding domain and lack 287 and 331 C-terminal amino acids, respectively, failed to activate GREs both in yeast (Figure 1, f and g) and in mammalian cells (Godowski et al., 1987). Immunoblotting established that all receptor species were produced at equivalent levels. Receptor derivatives with deletions of 200 and 239 C-terminal amino acids (N595 and N556) displayed constitutive activity in yeast (Figure 1, d and e), again recapitulating their phenotypes in mammalian cells (Godowski et al., 1987). Likewise, the chimeric receptor NLxC (Godowski et al., 1988), in which the receptor DNA binding domain is replaced with that of bacterial repressor LexA, enhanced transcription in yeast from a promoter linked to the LEX operator. In this case, enhancement was hormone dependent; as expected, deletion of the signaling domain produced a species (NLx) that enhanced transcription constitutively (Figure 1, h and i). Finally, a receptor point mutant C656G (Chakraborti et al., 1991), which displays increased affinity for glucocorticoids in mammalian cells, behaved in a similar manner in yeast (Figure 1, bottom graph). Thus, every receptor derivative tested, including various deletion, substitution, and point mutants, displayed parallel phenotypes in yeast and mammalian cells, suggesting strongly that the same mechanism of signal transduction is employed in the two species.

Ligand Efficacies in Yeast When we expressed the full-length receptor in yeast, we found that the pattern of ligand specificity for glu-

cocorticoid receptor function was generally similar to that in mammalian cells (Figure 2). In both yeast and mammalian cells, deacylcortivazol (Simons et al., 1979) activated most strongly, with half-maximal induction in yeast achieved at 10 nM. RU28362 (Moguilewski and Raynaud, 1980) also evoked strong activation in both cases, albeit at a fivefold higher concentration than deacylcortivazol. In general, higher ligand concentrations were required for induction in yeast than in mammalian cells. Studies in a mutant yeast strain (Nitsis and Wang, 1988) that displays a general increase in permeability to small molecules suggested that inefficient ligand entry in yeast accounts at least in part for this difference; 10fold lower hormone concentrations are needed for half-

Molecular Biology of the Cell


3-Galactosidase Units

A

Yeast

o,

a

1.5

c

N 795 1xS103-fold lower affinity to the receptor when it was produced in yeast. Remarkably, when receptor is expressed in Schizosaccharomyces pombe Molecular Biology of the Cell


or Drosophila melanogaster cells, yet other species-specific ligand response patterns appear to emerge (Picard et al., 1990b; Yoshinaga and Yamamoto, 1991). Interestingly, the oncoprotein v-erbA, a derivative of the thyroid hormone receptor, fails to bind agonists or to activate reporter gene expression in mammalian cells, whereas it displays both activities in yeast (Privalsky et al., 1990). Together, these findings imply that one or more nonreceptor factors may interact similarly but nonidentically with nuclear receptors in each species, apparently affecting ligand binding and efficacy. By this view, subtle differences between the yeast and mammalian cell factors would produce the distinct receptor behaviors observed in the two species. An obvious candidate for such a putative nonreceptor factor is Hsp9O, which has been shown in other studies to affect receptor signaling (Picard et al., 1990a). However, expression of mammalian Hsp9O in yeast in place of the homologous yeast gene product is not sufficient to produce the mammalian pattern of ligand efficacy (Picard et al., 1990a). Although we are intrigued by the notion that a common homologous factor acts on the receptor in all four species tested, it is conceivable that the novel receptor behavior observed in nonmammalian cells is artifactual, due to factors uniquely and adventitiously present in each of those settings that subtly alter receptor function. Further experiments will be required to determine the identity of the factor or factors that modulate receptor signaling activities. Our observations in yeast suggest that it should be possible by genetic manipulation to detect those factors and to identify putative homologues from other species. Our mutants encompass both loss-of-function and gain-of-function phenotypes. The loss-of-function mutants include L581P and L584S, which fail to bind or respond to virtually all ligands tested. In principle, such lesions could affect some general aspect of protein folding. It may be significant, however, that both mutations reside within a subregion of the signaling domain that is involved in the Hsp9O interaction (Pratt et al., 1988; Howard et al., 1990); it will be interesting to assess the binding of Hsp90 to these mutant species. Other lossof-function mutants are selective in their effects on ligand utilization. Thus C661R and M664T fail to respond to many agonists that act on the wild-type receptor, but maintain their responses to a subset of ligands. Certain mutants in this class may define molecular contacts between particular receptor amino acids and specific ligands. The D659G and perhaps the E706K mutants describe another type of loss-of-function mutation. These receptors are virtually inactive in yeast, CV-1, COS 7, and HeLa cells, although they display considerable dexamethasone binding in vitro. Thus the D659G and E706K defects might compromise receptor activity at a step subsequent to hormone binding, such as the signal-meVol. 3, November 1992

diated relief of inactivation of nuclear localization, DNA binding, or transcriptional regulation. The F620S mutant represents a gain-of-function allele that produces an overall increase in ligand binding and efficacy, with particularly striking effects on dexamethasone and triamcinolone acetonide activities in yeast. Indeed, this mutation, which has no phenotype in CV1, COS 7, or HeLa cells, effectively suppresses the distinctions between receptor action in yeast and mammalian cells. We speculate that the F620S lesion may alter selectively an interaction of the receptor with a yeast nonreceptor factor, thereby altering ligand utilization in yeast without affecting the homologous interaction in mammalian cells. Aside from the species specific differences in dexamethasone and triamcinolone acetonide responsiveness, each mutant analyzed yielded qualitatively similar phenotypes in yeast and CV- 1 cells. This validates our general strategy and underscores the genetic capabilities of yeast for rapid and efficient screening of large numbers of random point mutations, which would be virtually impossible with a mammalian system. Clearly, it will be informative to characterize additional mutants with variations of the simple screens described here. Furthermore, the identification of putative nonreceptor factors that participate in glucocorticoid signaling in yeast may reveal homologous counterparts that function in mammalian cells. Thus the isolation of second-site suppressor mutations in yeast may facilitate the identification of corresponding factors in animal cells. A detailed genetic analysis of the activities residing in the signaling domain-hormone binding, nuclear localization, protein inactivation, Hsp9O binding, and putative interactions with other nonreceptor factors-should define functional subregions within this segment and illuminate further the mechanism of signal transduction by this important class of regulatory factors. ACKNOWLEDGMENTS We thank M. Schena and D. Picard for C-terminal deletion and LEX reporter clones, S. Simons and Roussel-Uclaf for DAC and RU486, J. Thomas for the yeast lysis procedure, D. Pearce, L. Hedstrom, M. Gerber, D. Morgan, and S. Logan for comments on the manuscript, and K. Mulherin and B. Maler for preparation of the text and figures, respectively. Supported by grants from the NSF and NIH; postdoctoral support (M.J.G.) was from the Damon Runyon-Walter Winchell Cancer Research Fund and the Leukemia Society of America.

REFERENCES Beato, M. (1989). Gene regulation by steroid hormones. Cell 56, 335344. Becker, D.M., Fikes, J.D., and Guarente, L. (1991). A cDNA encoding a human CCAAT-binding protein cloned by functional complementation in yeast. Proc. Natl. Acad. Sci. USA 88, 1968-1972. Benhamou, B., Garcia, T., Lerouge, T., Vergezac, A., Gofflo, D., Bigogne, C., Chambon, P., and Gronemeyer, H. (1992). A single amino

1 255


acid that determines the sensitivity of progesterone receptors to RU 486. Science 252, 206-209. Bresnick, E.H., Dalman, F.C., Sanchez, E.R., and Pratt, W.B. (1989). Evidence that the 90-kDa heat shock protein is necessary for the steroid binding conformation of the L cell glucocorticoid receptor. J. Biol. Chem. 264, 4992-4997. Buratowski, S., Hahn, S., and Sharp, P.A. (1988). Function of a yeast TATA element binding protein in a mammalian transcription system. Nature 334, 37-42. Byravan, S., Milhon, J., Rabindran, S.K., Olinger, B., Garabedian, M.J., Danielsen, M., and Stallcup, M.R. (1991). Two point mutations in the hormone-binding domain of the mouse glucocorticoid receptor that dramatically reduce its function. Mol. Endocrinol. 5, 752-758. Carlstedt-Duke, J., Stromstedt, P.E., Persson, B., Cederlund, E., and Gustafsson, J.-A. (1988). Identification of hormone-interacting amino acid residues within the steroid-binding domain of the glucocorticoid receptor in relation to other steroid hormone receptors. J. Biol. Chem. 263, 6842-6846. Chakraborti, P.K., Garabedian, M.J., Yamamoto, K.R., and Simons, Jr., S.S. (1991). Creation of "super" glucocorticoid receptors by point mutations in the steroid binding domain. J. Biol. Chem. 233, 2207522078. Cormack, B.P., Strubin, M., Ponticelli, A.S., and Struhl, K. (1991). Functional differences between yeast and human TFIID are localized to the highly conserved region. Cell 65, 341-348. Dalman, F.C., Bresnick, E.H., Paresh, P.D., Perdew, G.H., Watson, Jr., S.J., and Pratt, W.B. (1989). Direct evidence that the glucocorticoid receptor binds to hsp90 at or near the termination of receptor translation in vitro. J. Biol. Chem. 264, 19815-19821. Dalman, F.C., Scherrer, L.C., Taylor, L.P., Akil, H., and Pratt, W.B. (1991). Localization of the 90-kDa heat shock protein-binding site within the hormone-binding domain of the glucocorticoid receptor by peptide competition. J. Biol. Chem. 266, 3482-3490. Danielian, P.S., White, R., Lees, J.A., and Parker, M.G. (1992). Identification of a conserved region required for hormone dependcent transcriptional activation by steroid hormone receptors. EMBO J 11, 10251033. Eilers, M., Picard, D., Yamamoto, K.R., and Bishop, J.M. (1989). Chimeras of MYC oncoprotein and steroid receptors cause hormone-dependent transformation of cells. Nature 340, 66-68. Evans, R. (1988). The steroid and thyroid hormone receptor superfamily. Science 240, 889-895. Gametchu, B., and Harrison, R.W. (1984). Characterization of a monoclonal antibody to the rat liver glucocorticoid receptor. Endocrinology 114, 274-279. Giguere, V., Hollenberg, S.M., Rosenfeld, M.G., and Evans, R.M. (1986). Functional domains of the human glucocorticoid receptor. Cell 46, 645-652. Gill, G., and Tjian, R. (1991). A highly conserved domain of TFIID displays species specificity in vivo. Cell 65, 333-340. Godowski, P.J., and Picard, D. (1989). Steroid receptors-how to be both a receptor and a transcription factor. Biochem. Pharmacol. 38, 3135-3143. Godowski, P.J., Picard, D., and Yamamoto, K.R. (1988). Signal transduction and transcriptional regulation by glucocorticoid receptor-LexA fusion proteins. Science 241, 812-816. Godowski, P.J., Rusconi, S., Miesfeld, R., and Yamamoto, K.R. (1987). Glucocorticoid receptor mutants that are constitutive activators of transcriptional enhancement. Nature 325, 365-368. Hope, T.J., Huang, X., McDonald, D., and Parslow, T.G. (1991). Steroid-receptor fusion of the human immunodeficiency virus type 1

1256

Rev transactivator: mapping cryptic functions of the arginine-rich motif. Proc. Natl. Acad. Sci. USA 87, 7787-7791. Howard, K.J., and Distelhorst, C.W. (1988). Evidence for intracellular association of the glucocorticoid receptor with the 90-kDa heat shock protein. J. Biol. Chem. 263, 3474-3481. Howard, K.J., Holley, S.J., Yamamoto, K.R., and Distelhorst, C.W. (1990). Mapping the Hsp90 binding region of the glucocorticoid receptor. J. Biol. Chem. 265, 11928-11935. Jones, E.W. (1977). Proteinase mutants of Saccharomyces cerevisiae. Genetics 85, 23-33. LaBaer, J. (1989). A detailed analysis of the DNA-binding affinity and sequence specificity of the glucocorticoid receptor DNA-binding domain. Ph.D. Dissertation, University of California, San Francisco. Lee, M.G., and Nurse, P. (1987). Complementation used to clone a human homologue of the fission yeast cell cycle control gene cdc2. Nature 327, 31-35. Leopold, P., and O'Farrell, P.H. (1991). An evolutionary conserved cyclin homolog from Drosophila rescues yeast deficient in Gl cyclins. Cell 66, 1207-2119. Lew, D.J., Dulic V., and Reed, S.I. (1991). Isolation of three novel human cyclins by rescue of G-1 (cln) function in yeast. Cell 66, 11971206. Mak, P., McDonnell, D.P., Weigel, N.L., Schrader, W.T., and O'Malley, B.W. (1989). Expression of functional chicken oviduct progesterone receptors in yeast (Saccharomyces cerevisiae). J. Biol. Chem. 264, 2161321618. Metzger, D., White, J.H., and Chambon, P. (1988). The human oestrogen receptor functions in yeast. Nature 334, 31-36. Miesfeld, R., Rusconi, S., Godowski, P.J., Maler, B.A., Okret, S., Wikstrom, A.-C., Gustafsson, J.-A., and Yamamoto, K.R. (1986). Genetic complementation of a glucocorticoid receptor deficiency by expression of cloned receptor cDNA. Cell 46, 389-399. Miller, N.R., and Simons, Jr., S.S. (1988). Steroid binding to hepatoma tissue culture cell glucocorticoid receptors involves at least two sulfhydryl groups. J. Biol. Chem. 263, 15217-15225. Moguilewski, M., and Raynaud, J.P.J. (1980). Evidence for a specific mineralocorticoid receptor in rat pituitary and brain. J. Steroid Biochem. 12, 309-314. Myers, R.M., Lerman, L.S., and Maniatis, T. (1985). A general method for saturation mutagenesis of cloned DNA fragments. Science 229, 242-247. Nemoto, T., Oharo-Nemoto, Y., Dennis, M., and Gustafsson, j.-A. (1990). The transformed glucocorticoid receptor has a lower steroidbinding affinity than the nontransformed receptor. Biochemistry 29, 1880-1886. Nitsis, J., and Wang, J.C. (1989). DNA topoisomerase-targeting antitumor drugs can be studied in yeast. Proc. Natl. Acad. Sci. USA 85, 7501-7505. Picard, D., Khursheed, B., Garabedian, M.J., Fortin, M.G., Lindquist, S., and Yamamoto, K.R. (1990a). Reduced levels of hsp9o compromise steroid receptor action in vivo. Nature 348, 166-168. Picard, D., Salser, S.J., and Yamamoto, K.R. (1988). A movable and regulable inactivation function within the steroid binding domain of the glucocorticoid receptor. Cell 54, 1073-1080. Picard, D., Schena, M., and Yamamoto, K.R. (1990b). An inducible expression vector for both fission and budding yeasts. Gene 86, 257261. Picard, D., and Yamamoto, K.R. (1987). Two signals mediate hormonedependent nuclear localization of the glucocorticoid receptor. EMBO. J. 6, 3333-3340.

Molecular Biology of the Cell

Genetics of Steroid Receptor Signaling Pratt, W.B., Jolly, D.J., Pratt, D.V., Hollenberg, S.M., Giguere, V., Cadepond, F.M., Schweizer-Groyer, G., Catelli, M-G., Evans, R.M., and Baulieu, E.-E. (1988). A region in the steroid binding domain determines formation of the non-DNA-binding, 9S glucocorticoid receptor complex. J. Biol. Chem. 263, 267-273. Privalsky, M.L., Sharif, M., and Yamamoto, K.R. (1990). The viral erbA oncogene protein, a constitutive repressor in animal cells, is a hormone-regulated activator in yeast. Cell 63, 1277-1286. Rusconi, S., and Yamamoto, K.R. (1987). Functional dissection of the hormone and DNA binding activities of the glucocorticoid receptor. EMBO. J. 6, 1309-1315. Rousseau, G.G., Baxter, J.D., and Tomkins, G.M. (1972). Glucocorticoid receptors: relations between steroid binding and biological effects. J. Mol. Biol. 67, 99-115. Rousseau, G.G., and Schmit, J.-P. (1977). Structure-activity relationships for glucocorticoids-I: determination of receptor binding and biological activity. J. Steroid Biochem. 8, 911-919. Sakai, D.D., Helms, S., Carlstedt-Duke, J., Gustafsson, J.-A., Rottman, F.M., and Yamamoto, K.R. (1988). Hormone-mediated repression: a negative glucocorticoid response element from the bovine prolactin gene. Genes Dev. 2, 1144-1154. Samuels, H.H., and Tomkins, G.M. (1970). Relation of steroid structure to enzyme induction in hepatoma tissue culture cells. J. Mol. Biol. 52, 57-74. Schena, M., Freedman, L.P., and Yamamoto, K.R. (1989). Mutations in the glucocorticoid receptor zinc finger region that distinguish interdigitated DNA binding and transcriptional enhancement activities. Genes Dev. 3, 1590-1601. Schena, M., Picard, D., and Yamamoto, K.R. (1991). New vectors for constitutive and inducible gene expression in yeast. Methods Enzymol. 194, 389-398. Schena, M., and Yamamoto, K.R. (1988). Mammalian glucocorticoid receptor derivatives enhance transcription in yeast. Science 241, 965967. Scherrer, L.C., Dalman, F.C., Massa, E., Meshinchi, S., and Pratt, W.B. (1990). Structural and functional reconstitution of the glucocorticoid receptor-hsp90 complex. J. Biol. Chem. 265, 21397-21400.

Vol. 3, November 1992

Simons, Jr., S.S., Pumphrey, J.G., Rudikoff, S., and Eisen, H.J. (1987). Identification of cysteine 656 as the amino acid of hepatoma tissue culture cell glucocorticoid receptors that is covalently labeled by dexamethasone 21-mesylate. J. Biol. Chem. 262, 9676-9680. Simons, Jr., S.S., Thompson, E.B., and Johnson, D.F. (1979). Antiinflammatory pyrazolo-steroids: potent glucocorticoids containing bulky A-ring substituents and no C3-carbonyl. Biochem. Biophys. Res. Commun. 86, 793-800. Superti-Furga, G., Bergers, G., Picard, D., and Busslinger, M. (1991). Hormone-dependent transcriptional regulation and cellular transformation by Fos-steroid receptor fusion proteins. Proc. Natl. Acad. Sci. USA 88, 5114-5118. Umek, R.M., Friedman, A.D., and McKnight, S.L. (1991). CCAATenhancer binding protein: a component of a differentiation switch. Science 251, 288-292. Vanderbilt, J.N., Miesfeld, R., Maler, B.A., and Yamamoto, K.R. (1987). Intracellular receptor concentration limits glucocorticoid-dependent enhancer activity. Mol. Endocrinol. 1, 68-74. Vegeto, E., Allan, G.F., Schrader, W.T., Tsai, M.-J., McDonnell, D.P., and O'Malley, B. (1992). The mechanism of RU 486 antagonism is dependent on the conformation of the carboxy-terminal tail of the human progesterone receptor. Cell 69, 703-713. Wright, A.P.H., Carlstedt-Duke, J., and Gustafsson, J.-A. (1990). Ligand-specific transactivation of gene expression by a derivative of the human glucocorticoid receptor expressed in yeast. J. Biol. Chem. 265, 14763-14769. Yamamoto, K.R. (1985). Steroid receptor regulated transcription of specific genes and gene networks. Annu. Rev. Genet. 19, 209-252. Yamamoto, K.R., Godowski, P.J., and Picard, D. (1988). Ligand-regulated nonspecific inactivation of receptor function: a versatile mechanism for signal transduction. Cold Spring Harbor Symp. Quant. Biol. 53, 803-811. Yoshinaga, S.K., and Yamamoto, K.R. (1991). Signaling and regulation by a mammalian glucocorticoid receptor in Drosophila cells. Mol. Endocrinol. 5, 844-853.

1257