183 Transcriptional activity of Drosophila melanogaster ecdysone receptor isoforms and ultraspiracle in Saccharomyces cerevisiae

F E Dela Cruz, D R Kirsch and J N Heinrich Molecular Genetic Screen Design, Cyanamid Agricultural Research Center, P.O. Box 400, Princeton, New Jersey 08543–0400, USA (Requests for offprints should be addressed to J N Heinrich; Email: [email protected]) (F E Dela Cruz is now at Bayer Pharmaceutical Division, Bayer Corporation, 400 Morgan Lane, West Haven, Connecticut 06516-4175, USA; D R Kirsch andJNHeinrich are now at Wyeth–Ayerst Research, CN 8000, Princeton, New Jersey 08543-8000, USA)

ABSTRACT The Drosophila melanogaster ecdysone receptor (mAR) identified one chimera, composed of the (EcR) is produced in three isoforms, which mediate mAR N-terminus and the remainder from EcR developmental processes such as metamorphosis. (mAR\EcR-CDEF) that was transcriptionally silent These isoforms were expressed in Saccharomyces and inducible by Usp. In contrast, the vertebrate cerevisiae to elucidate aspects of receptor transcrip- homologue, human retinoic acid receptor (RXR), tion activity in a highly defined genetic model showed ligand-independent transcription when system. All three EcR isoforms showed ligand- co-expressed with EcR
A/B but not mAR\ independent transcriptional activation of an EcR-CDEF. Therefore, RXR does not require its ecdysone reporter gene and the amount of activation partner to possess an N-terminal domain, yet is correlated with the size of the N-terminal A/B intolerant of a heterologous N-terminus. Similarly, (transactivation) domain present in the isoform: the human vitamin D receptor, which has a short EcR-B1>EcR-A>>EcR-B2. Upon co-expression N-terminal region, showed greater ligand- with ultraspiracle (Usp), transcriptional activation independent transcription in the presence of RXR was further increased with EcR-B1 or EcR-A, but than in the presence of Usp. These results reveal a was unchanged with EcR-B2 or a truncated EcR mechanistic basis for the differential activities lacking the A/B N-terminal domain (EcR
A/B). among the EcR isoforms, and between Usp and Thus, the enhanced activity from Usp may depend RXR. Furthermore, they provided the foundation on the presence of an N-terminal domain of EcR. for a genetic screen to identify potential insecticides Co-expression with Usp of several chimeric recep- as well as accessory proteins for Usp and EcR. tors of the EcR and the mouse androgen receptor Journal of Molecular Endocrinology (2000) 24, 183–191

INTRODUCTION (Koeller 1992, Talbot et al. 1993). The EcRs also bind DNA in the absence of ecdysone, suggesting Insect metamorphosis is in part controlled by that unliganded EcRs also regulate target genes ecdysone, a steroid hormone. Ecdysone binds to the (Buszczak & Segraves 1998, Hall & Thummel 1998 ecdysone receptor (EcR), inducing the transcription and references therein). The EcRs are differentially of a hierarchy of genes that harbor ecdysone-specific expressed, with EcR-A being present in cells that response elements (RE) upstream of their pro- differentiate to form adult tissues while EcR-B1 and moters. The Drosophila melanogaster EcR gene EcR-B2 are predominantly expressed in larval cells encodes for three isotypes (-A, -B1 or -B2) that are fated to die. Such observations suggest that cell fate typical steroid/nuclear receptors and share common is, at least in part, controlled by EcR isoform C-terminal DNA-binding domain and ligand- expression (Talbot et al. 1993). binding domain (LBD) sequences, but have The EcR has an obligatory heterodimer partner different N-terminal A/B domain sequences called ultraspiracle (Usp). Usp is a homologue of

Journal of Molecular Endocrinology (2000) 24, 183–191 Online version via http://www.endocrinology.org 0952–5041/00/024–183  2000 Society for Endocrinology Printed in Great Britain

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access 184     and others · Activity of ecdysone receptors in yeast

the vertebrate retinoid X receptor alpha (RXR) 1991). Yeast expression plasmids (YEp) and the (Oro et al. 1990), a steroid/nuclear receptor that is a derivative containing a multiple cloning site heterodimeric partner to a growing list of nuclear (YEpc) were used to insert the open reading receptors (reviewed in Mangelsdorf & Evans 1995). frames of the receptor used. Plasmids YEpcUSP Usp and RXR can substitute for one another and YEpEcR-B1 were constructed as described in forming a heterodimeric receptor: RXR can elsewhere (Dela Cruz & Mak 1997). Plasmids heterodimerize with EcR-B1 and Usp with the YEpEcR-A and YEpEcR-B2 were made from human vitamin D receptor (huVDR) (Yao et al. YEpEcR-B1, by excising an AflII-AscI fragment, 1992). However, Usp is distinctly different from and inserting polymerase chain reaction (PCR) RXR in at least two properties. First, it does not fragments containing EcR-A and EcR-B2 specific bind retinoic acid, and is probably an orphan recep- sequences. The YEpEcR
A/B plasmid was made tor, although it has been reported that Usp binds by excising from YEpEcR-B1 a BamHI-BspEI juvenile hormone III (Jones & Sharp 1997). fragment and inserting two PCR fragments. Secondly, like RXR, Usp is a global regulator of Plasmids encoding chimeras were constructed using genetic pathways, including some independent of YEpEcR-B1 and YEpmAR5 (Mak et al. 1994b). ecdysone (Segraves 1994, Sutherland et al. 1995, Plasmid YEpEcR\mAR-DE was made by excising a Hall & Thummel 1998 and references therein). SacI-BspEI fragment and replacing it with a PCR However, despite extensive searches for hetero- product derived from YEpmAR5 (mAR being dimeric partners other than EcRs, only the mouse androgen receptor). Plasmid YEpmAR\- Drosophila hormone receptor 38 (Sutherland et al. EcR-DEF was made by inserting a DraIII-BssHII 1995) has been identified. The complex roles of Usp fragment from YEpmAR5 (Mak et al. 1994b) into and EcRs in insect development may include more the SacI-BspEI site of YEpEcR-B. Plasmids diverse mechanisms than those accommodated by YEpmAR\EcR-EF and YEPmAR\EcR-E were simple models of ligand-activated gene transcription. made by cloning PCR fragments derived from Saccharomyces cerevisiae lacks steroid/nuclear EcR-B1 into YEpmAR6 (P Mak, unpublished receptors and their specific co-regulators. As a observations). Plasmid YEpmAR-D\EcR-DEF was result, yeast has been extensively employed by cloned similarly to YEpmAR\EcR-DEF except researchers in basic science and drug discovery to the DraIII-BssHII fragment was cloned into reconstitute steroid/nuclear receptor function YEpmAR6. The plasmid YEpmAR\EcR-CDEF (Garabedian 1993, Butt & Walfish 1996). Faithful was constructed by inserting a PCR product from ligand-activated target gene transcription is com- YEpmAR (Mak et al. 1994b) into YEpEcR
A/B. monly observed in yeast cells expressing steroid/ Plasmids YEpmAR5 and YEpRXR are as nuclear receptor(s) through the use of reporter described (McDonnell et al. 1989, Mak et al. 1994a), plasmids with cognate REs upstream of synthetic except the copper-responsive yeast metallothionein promoters. Co-transfecting steroid-specific co- (CUP1) promoter drives expression of Usp and activator proteins and other transcription factor RXR, and the constitutive yeast glyceraldehyde-3- accessory proteins can optimize the activity of some phosphate dehydrogenase (TDH3) promoter drives receptors. Conversely, the absence of co-regulators expression of the EcRs, the chimeras, and huVDR. in yeast provides a powerful in vivo system in which the intrinsic properties of receptors and subtle dif- ferences among receptors can be investigated. Using Construction of reporter plasmids the yeast system, we identified in this study three The YEpEcRRE2-LacZ reporter plasmid (Dela Cruz novel aspects of the EcR and Usp: differences among & Mak 1997) (RE=reporter element; subscript=in EcR isoforms; a Usp specific activity; and a distinc- duplicate) was used to make YEp-UspRE2-LacZ and tion between Usp and RXR. Additionally, we con- YEp-VDRRE2 (vitamin D reporter response ele- structed a yeast strain in which Usp activity is easily ment2). A XhoI fragment from YEpEcRRE2-LacZ monitored. This strain can be used to screen for was excised and replaced with tandem duplicate chemicals with potential insecticide activity, or for enhancer sequences separated by an arbitrary six novel Usp and EcR accessory proteins. bases. For YEp-UspRE2-LacZ, the enhancer was the 21 base pair sequence corresponding to the 64 to 44 region of the chorion s15 promoter (Shea et al. MATERIALS AND METHODS 1990). For YEp-VDRRE2-LacZ, the enhancer was a 25 base pair sequence corresponding to the 262 Construction of yeast expression plasmids to 242 region of the rat 24-hydroxylase distal gene Standard methods for manipulating DNA and yeast (r24-Hlase(d)) (Kephart et al. 1996). YRpA2 has were used (Sambrook et al. 1989, Guthrie & Fink been described previously (Mak et al. 1994b).

Journal of Molecular Endocrinology (2000) 24, 183–191 www.endocrinology.org

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access Activity of ecdysone receptors in yeast ·     and others 185

S. cerevisiae host strains 195·2 Ci/mmol (New England Nuclear, Boston, MA, USA), in the absence or presence of 1000-fold Plasmid YEpUspRE -LacZ was used in the protease 2 excess of muristerone A (Mur A). After incubation deficient strain BJ2168 which has the following for 1 h at 25 C with constant agitation, the reaction genotype: MATa leu2 trp1 ura3–52 prb1–1122 was passed through an Inotech A21 glass fiber filter pep4–3 prc1–407 gal2. Plasmid YEpEcRRE -LacZ 2 attached to the Inotech Cell Harvester (Inotech and YRpA were used in a canavanine permease 2 Biosystems Int., Rockville, MD, USA) maintained (can1) mutant yeast strain, CGY44(DC45): MAT under vacuum, washed with three volumes of ste11-
1 his4–519 leu2 trp1 ura3 can1–101. Trans- TEDG, and subjected to the vacuum for 1 formations of yeast cells were performed according additional min. The filters were placed in 3 ml to standard procedures and transformants were scintillation cocktail (Ready Safe; Beckman Coulter grown in the appropriate selective media for Inc., Fullerton, CA, USA) and radioactivity was plasmid maintenance (Guthrie & Fink 1991). measured as c.p.m. in a Taurus Automated Liquid Scintillation Counter (ICN Biochemicals Inc., Preparation of cytosolic extract Irvine, CA, USA). The readings from assays Yeast cells were grown at 30 C overnight with performed in the absence and presence of Mur A constant agitation in minimal media supplemented represent total and non-specific c.p.m., and the difference is specific counts or specific binding of with 2% glucose and 100 µM CuSO4 as needed, 3 and subcultured into fresh media in the absence H-Pon A. Binding was done in triplicate and or presence of steroid agonists, until the cell the data were analyzed using GraphPad Prism density reached mid-log phase (optical density (GraphPad Software Inc., San Diego, CA, USA). (OD) =1·0). To prepare cytosolic extracts, cells The standard transcription assay consisted of 600 ff maintained at 4 C were harvested by centrifuga- preincubating 800 µl Z-bu er containing 5–20 µg  tion, washed twice, and lysed in 1/100 volume by cytosolic extract for 10 min at 28 C and then adding  five cycles of 1-min vortexing with glass beads 200 µl O-nitrophenyl - -galactopyranoside (4 mg/ (0·5 mm; B. Braun Instruments, Kronberg, ml dissolved in water) for an additional 10 min. The Germany) followed by 1 min on ice and high speed reaction was stopped by adding 500 µl 1 M sodium centrifugation. The protein concentration of the carbonate and the OD420 was measured. Each supernatant or cytosolic extract, determined by the treatment was done in triplicate and the average Bradford protein assay (Bio-Rad, Richmond, CA, values were converted to Miller units (Mu) mg USA), was typically between 5 and 10 mg/ml. For protein (Miller 1972). hormone-binding assays, the cells were resuspended in ice-cold TEDG (10 mM Tris, pH 8·0, 0·1mM EDTA, 2 mM dithiothreitol, 10% glycerol) contain- RESULTS ing 0·4 M NaCl and the protease inhibitors (150 µg/ml phenylmethylsulfonyl fluoride, 4 µg/ml Behavior of EcR isoforms in yeast cells leupeptin, 3 µg/ml chymostatin and 1 µg/ml pep- statin). An equal volume of glass beads was added, Dela Cruz & Mak (1997) expressed the EcR-B1 the cells were lysed by vortexing and the suspension isoform in yeast cells. Extracts from cells expressing ffi was centrifuged at 1000 g for 10 min to remove the EcR-B1 and Usp showed high-a nity ligand unbroken cells and then at 100 000 g for 30 min to binding in these studies. However, in vitro -Gal pellet the particulate fraction and recover the assays showed that EcR-B1 activated lacZ,a cytosolic extract. For the -galactosidase (-Gal) reporter plasmid that contained an ecdysone RE. assays, the cells were resuspended in 500 µl Z-buffer Neither co-expression of EcR-B1 with Usp, nor (60 mM Na HPO 7H O, 40 mM NaH PO H O, growth of these cells in the presence of ecdysone 2 4 2 2 4 2 further increased expression of the reporter gene. 10 mM KCl, 1 mM MgSO47H2O, 50 mM -mercaptoethanol, pH 7·0), lysed in a microfuge This -Gal assay was used here to compare tube containing a one-third volume of glass beads, EcR-B1, EcR-A and EcR-B2. In the absence and microfuged for 10 min at 12 000 g. of Usp, EcR-B1 showed significant ligand- independent transactivation (Table 1), while EcR-A displayed approximately 60% the activity of Bioassays EcR-B1, and EcR-B2 showed about one-tenth the The standard binding assay was performed in a activity of EcR-B1. Co-expression with Usp total reaction volume of 200 µl TEDG, includ- increased by over fourfold the activation levels with ing 200–400 µg cytosolic extract and 2·5nM EcR-B1, about threefold with EcR-A, and there was [24,25,26,27-3H(N)]Ponasterone A (3H-Pon A), no detected increase with EcR-B2. www.endocrinology.org Journal of Molecular Endocrinology (2000) 24, 183–191

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access 186     and others · Activity of ecdysone receptors in yeast

 1. The EcR isotypes and Usp show ligand-independent transcriptional activity. -Gal activity in yeast strains containing the reporter plasmid YEpEcRRE2-LacZ, and expression plasmids for YEpEcR-B1,YEpEcR-A or YEpEcR-B2, and YEpc ()orYEpUSP (+). The activity of untreated () and 10 µM ligand (Mur A)-treated cells are compared. Induction over the absence of nuclear receptor is calculated from a basal mean of 58. The results show the mean.. from four independent clones, for each condition, and are presented in standard Miller units (Mu/mg)

Nuclear receptors
-Gal activity (Mu/mg) Induction EcR isoform Usp () Mur A Activity/basal activity* Fold activation by Usp

—  2352 5769 1 — +8864 6372 1 1 EcR-B1  4103709 4135716 719 EcR-B1 + 18 0071967 15 691548 31034 4·40·3 EcR-A  2523710 2890729 4313 EcR-A + 66943187 61442198 11637 3·01·5 EcR-B2  480150 431107 82 EcR-B2 + 459120 5173620·91·5

*Basal mean for all experiments is 58.

These experiments were repeated using cells that (data not shown). Therefore, an approach employ- had been treated with Mur A, a high-affinity ing receptor chimeras was used to assess for ecdysone receptor agonist. Mur A produced no transactivation by Usp. significant increase in reporter activity in any of the ecdysone receptor expressing cells (Table 1). To eliminate the possibility that Mur A does not Behavior of EcR-mAR chimeras in yeast cells diffuse into the yeast cell, a number of structurally distinct ecdysone agonists, including RH5849 (a Steroid/nuclear receptors are composed of non-steroidal, diacylhydrazine ecdysone receptor structurally and functionally conserved domains. ligand), were also tested and failed to induce Transactivation is produced by the A/B domain, transcription (data not shown). Radioligand binding DNA binding is mediated by the C domain, the D studies with 3H-Pon A were performed with cells domain is required for nuclear localization, the E expressing Usp and each receptor isoform to domain determines ligand binding as well as determine whether the lack of responsiveness to heterodimerization, and the F domain (present in Mur A was due to the inability of the ligand to bind some and absent from other receptors) has a less the receptors. In vitro ligand competition curves defined function (reviewed in Mangelsdorf & Evans (Fig. 1A–D) showed apparent high-affinity binding 1995). In yeast, the mAR displayed ligand- to 3H-Pon A, and an order of potency for ecdysone dependent transcription activation (Mak et al. agonist competitors, Pon A>Mur A>ecdysone> 1994b). We therefore constructed receptor chimeras RH5849, that was similar for the three isoforms. containing complementary portions of the EcR-B1 Since the three receptor isoforms showed similar and the mAR (Fig. 2) to identify regions responsible binding kinetics, the differences in activity among for the observed ligand-independent transactivation them was not the result of differential ligand of the EcR-B1. binding. In addition, the absence of transactivation As expected (Table 2), an EcR with a truncated in yeast cells expressing EcR-B2 cannot be A/B domain (EcR
A/B) produced significantly less explained by poor receptor expression since extracts (44%) transcriptional activation compared with from all three cells showed similar ligand binding EcR-B1, and the activation of EcR
A/B was not (approximately 15 000 c.p.m./200 µg protein). further increased with Usp. Since ligand binding is The enhanced reporter expression observed in dependent on heterodimerization between Usp and cells co-expressing EcR-B1 or EcR-A suggested EcR, radioligand-binding experiments were per- that Usp might have intrinsic ligand-independent formed with extracts from cells expressing both Usp transcription activation activity. To test this, a and EcR
A/B. Saturation binding curves with these reporter plasmid containing a Usp RE was used to extracts containing both EcR
A/B and Usp were measure Usp-specific transactivation. However, this similar to those with EcR-B1 and Usp (Fig. 1E and reporter was transcriptionally active in the absence F). In addition, Scatchard analysis showed nearly ffi of Usp, and was not further transactivated by Usp identical high-a nity dissociation constants (Kd)

Journal of Molecular Endocrinology (2000) 24, 183–191 www.endocrinology.org

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access Activity of ecdysone receptors in yeast ·     and others 187

 1. Competition with agonists of 3H-Pon A binding to yeast cytosolic extracts containing Usp with EcR-B1, EcR-A and EcR-B2 or EcR
A/B (A–D) and saturation binding and Scatchard analysis with Usp and either EcR-B1 or EcR
A/B (E–F). The standard competition binding assay containing 2·5nM3H-Pon A was done with increasing tenfold concentrations from 10 pM to 100 M of the competitors, Pon A ( ), Mur A ( ), 20-hydroxyecdysone (Ec; ) and RH5849 (Hydra; M). The results are graphs showing the percentage bound compared with the amount bound in the absence of a competitor. The standard saturation binding assay was done with concentrations of 3H-Pon A of 0·1, 0·5, 1·0, 2·5, 5, 10, 20 and 40 nM and in the absence and presence of unlabeled Pon A. The results are shown in graph form for total binding ( ), non-specific binding (), and their difference, the

specific binding ( ). Scatchard plots were used to obtain apparent Kd values of 3 nM for the EcR and 2 nM for the EcRA
B respectively. for both. Therefore Usp and EcR
A/B associate with the EcR mAR chimeras. These chimeras were together, and the inability of Usp to enhance designed to substitute mAR domains throughout EcR
A/B transactivation is due to the absence of the EcR and included virtually all contiguous the EcR-B1 N-terminal domain. combinations of functional domains from the EcR To further characterize the contribution of the and mAR. Usp increased reporter activity in all N-terminal domain of the EcR to the ligand- chimeras (Table 3) except one, mAR\EcR-DEF. independent activity of Usp, Usp was co-expressed Since no single EcR domain was required for this www.endocrinology.org Journal of Molecular Endocrinology (2000) 24, 183–191

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access 188     and others · Activity of ecdysone receptors in yeast

None of the chimeras containing the ecdysone LBD responded to Mur A treatment or bound 3H-Pon A. In contrast, a receptor chimera containing the androgen LBD (EcR\mAR-DE) responded to testosterone, indicating that domain swapping was permissible in this system. Further- more, reporter induction by EcR\mAR-DE with Usp (2400 Mu/mg) and by EcR\mAR-DE with testosterone (3500 Mu/mg), when combined, approximated the induction by Usp and testoster- one together (6489+144 Mu/mg). This apparent additivity suggested that testosterone and Usp induction of this receptor might occur by independent mechanisms.

 2. Schematic comparison of the EcR (open RXR and Usp are selectively boxes), mAR (hatched boxes), the N-terminal truncated interchangeable in yeast EcR
A/B, the chimeras between the EcR and mAR and Usp (checkered boxes). The recombinant proteins are Since RXR can substitute for Usp in binding and depicted by the standard nomenclature designating the transactivation in yeast (Dela Cruz & Mak 1997), we functional domains, A through F, of nuclear receptors assayed RXR for ligand-independent transcrip- and are aligned by their C domains. Protein N-termini tional activity. We compared RXR and Usp are presented to the left and the C-termini to the right. activity with (1) mAR\EcR-CDEF, (2) the N- The first amino acid and the number of the amino acid terminally deleted EcR
A/B, and (3) huVDR, at the C-terminal of each domain are indicated. Also which has one of the shortest N-terminal domains indicated are the addition of three amino acids (solid of the known steroid receptors. RXR, interest- boxes) or two amino acids (horizontally lined box; in Usp, at the beginning on the left) to a number of the ingly, produced responses opposite to those EcR forms and to Usp respectively. The names of the observed with Usp: RXR did not activate mAR\ chimeras begin with the nuclear receptor which EcR-CDEF but did activate EcR
A/B (Table 3A). contributes the N-terminal domain(s), followed by a The transactivation of huVDR by RXR was slash ‘\’ indicating the junction between two nuclear approximately tenfold higher than transactivation receptors, and ends with the second nuclear receptor by Usp (Table 3B). Treatment with vitamin D ff plus the domains which it contributes. The last chimera (1,25-dihydroxyvitamin D3) had no e ect, in accord contains D domains from two nuclear receptors and this with previous observations (Kephart et al. 1996). situation is indicated by a -D after the name of the first These results demonstrated that Usp and RXR nuclear receptor (mAR-D). shared the ability to induce ligand-independent transcription, but that Usp was dependent upon the activity, there is a large degree of flexibility in the presence of an A/B domain in its heterodimeric ability of Usp to increase the activity of a receptor partner while RXR was not. partner. This result suggested the possibility that the transactivation was due to the inherent activity of Usp. One of the receptor chimeras that DISCUSSION responded to Usp, mAR\EcR-CDEF, produced no detectable ligand-independent reporter activation, In S. cerevisiae, expression of each of the three EcR indicating that Usp transcription is not dependent isoforms resulted in a different level of transcription on a high basal level of reporter expression. The of a reporter gene containing an ecdysone RE. The single non-responsive chimera, mAR\EcR-DEF, level was proportional to the length of the consisted of the A/B and C domains from mAR and N-terminal domain of the EcR isoform: EcR-B1 the D, E and F domains from EcR. This result (226 amino acids)>EcR-A (197 amino acids)>> suggested incompatibility between the C domain of EcR-B2 (17 amino acids). A second correlation with the mAR and the D domain of the EcR, because the the N-terminal domain was enhancement of extremely similar chimera mAR-D\EcR-DEF, transactivation by Usp, EcR-B1>EcR-A and none which has an additional D domain from the with EcR-B2. Consistent with these properties, an mAR, and the ‘reverse chimera’ EcR\mAR-DE N-terminal truncated EcR (EcR
A/B) behaved both showed increased reporter expression when similarly to EcR-B2. Since the receptors were co-expressed with Usp. expressed at similar levels (similar amount of ligand

Journal of Molecular Endocrinology (2000) 24, 183–191 www.endocrinology.org

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access www.endocrinology.org

 2. Different forms of the EcR show Usp-specific transcriptional activity. -Gal activity in yeast strains containing a LacZ reporter gene, a form of  the EcR, and lacking ( ) or containing Usp (+). The RE-LacZ reporter plasmid contained either EcRRE2 -LacZ or A2 -LacZ and was selected based on domain C of the EcR form listed in the first column. The yeast was grown in the absence (), or the presence of either 10 µM Mur A or 1 µM testosterone. The results are presented in Miller units (Mu/mg) as a fold difference between the EcR in the absence as compared with the presence of Usp (Usp (fold induction)). The results are from two independent experiments

-Gal activity (Mu/mg) Nuclear receptors No treatment Mur A Testosterone EcR chimera Usp Mu/mg Usp (fold induction) Mu/mg Usp (fold induction) Mu/mg Usp (fold induction)

Tested with EcR response element —  4914 4311 5213 — +7017 1·40·46651·50·57312 1·40·3 EcR-B1  3199889 3172700 32341063 yeast in receptors ecdysone of Activity EcR-B1 + 10 5581246 3·30·710212913 3·20·7 9846724 3·01·0 EcR
A/B  1411395 1502389 1305250 EcR
A/B + 1435190 1·001705167 1·10·2 1521102 1·10·2 EcR\mAR-DE  31977 32096 3548289 EcR\mAR-DE + 2349149 7·51·52489289 7·71·9 6489144 1·80·3 mAR\EcR-CDEF  4212 5112 3311 mAR\EcR-CDEF + 1988378 4712 2536106 5012 3030173 9233 ora fMlclrEndocrinology Molecular of Journal Tested with mAR response element —  10125 14424 8970 — +9691 0·90·7 19123 1·30·13720 0·70·8 mAR\EcR-DEF  31551 25838 ND Downloaded fromBioscientifica.com at10/01/202102:02:13AM mAR\EcR-DEF + 18320 0·60 22912 0·90·1ND mAR\EcR-EF  30752 30452 ND mAR\EcR-EF + 2029122 6·71·5205414 6·91·1ND

mAR\EcR-E  3044 28171 ND ·     mAR\EcR-E + 2781 214 9·2 0·63751286 8·8 3·2ND     mAR-D\EcR-DEF  32224 24662 ND mAR-D\EcR-DEF + 2091247 6·50·32022267 8·52·0ND (2000) ND, not determined. 24, n others and 183 – via freeaccess 191 189 190     and others · Activity of ecdysone receptors in yeast

 3. Usp and RXR show differences in their ligand-independent transcriptional activity. -Gal activity (Miller units; Mu) in yeast strains containing YEpRXR or YEpcUsp and (A) YEpEcR
A/B or YEpEcR-CDEF with YEpEcRRE -LacZ, or (B) YEpV5 with 2  YEpVDRRE2-LacZ. Strains were grown in the absence ( ) or presence (+) of vitamin D (Vit D)

-Gal activity (Mu/mg) Nuclear receptor Vector Usp RXR

(A) EcR form (with YEpEcRRE2) — 9353 10753 550 EcR
A/B 2085509 1350291 43201230 mAR\EcR-CDEF 8325 726246 12894

(B) VDR Vit D (with VDRRE2)  4822 5921 4423  +6334 3414 4730 +  6234 279138 2752700 ++ 5521 600148 2891700

binding), the differences in transactivation potential by testosterone, then Usp would have contributed was due to the intrinsic properties of the EcR the additional transcriptional activity. Secondly, isoforms and Usp. The ligand-independent activity Usp’s vertebrate homologue, RXR, produced of these receptors may be explained by the lack of transactivation when co-expressed with EcR
A/B. Drosophila accessory proteins and steroid/nuclear Since the A/B domain is the major ligand- receptor coregulators, i.e. co-repressors (reviewed independent transcriptional activation domain, the in Mangelsdorf & Evans 1995). Evidence for observed activation was likely dependent upon co-repressor association with EcR, but not Usp or RXR. However, RXR’s ability to transactivate the Usp and EcR complex, was recently reported EcR
A/B, as well as its greater ability than Usp with the vertebrate co-repressors nuclear receptor to activate huVDR and inability to activate co-repressor (N-CoR), silencing of mediator of EcR\mAR-CDE, also emphasized a difference from retinoid and thyroid receptors (SMRT), and Alien, Usp: that RXR does not require the N-terminal and the Drosophila homologue of Alien (Dressel domain of its partner and is less tolerant of a et al. 1999, Thormeyer et al. 1999). Interestingly, in heterologous N-terminal on its partner. Interest- mammalian cells, Usp, but not RXR, showed ingly, in mammalian cultured cells, ligand- ligand-independent activity when co-expressed with independent activation by RXR was reported to EcR–B1 (Thomas et al. 1993). These results suggest occur through the E domain of the orphan nuclear that Usp ligand-independent activity might be receptor 1 (Weibel & Gustafsson 1997). physiologically relevant. A property of the yeast strain containing the Two approaches were used to discriminate mAR\EcR-CDEF and Usp is its potential ability to between the contribution of Usp and the EcR to the detect compounds that regulate Usp activity ligand-independent transcriptional activity. First, a (Heinrich et al. 1997). Such compounds could be reporter plasmid containing a Usp RE was not used as probes of Usp activity or could provide transactivated by Usp. In the second approach a clues to the chemical structure of the Usp ligand. receptor chimera that lacked constitutive activity, Furthermore, since Usp is essential for insect mAR\EcR-CDEF, yet responded to the presence of development and distinctive from RXR receptors, Usp was identified. Since mAR\EcR-CDEF failed such compounds would serve as leads to develop to bind 3H-Pon A, it was possible that it was also novel selective insecticides. Another potential use of transcriptionally uninducible by Usp, and merely this yeast strain, as well as others described in this provided the anchor for Usp, which was the source report, is to identify nuclear receptor co-regulators of the transcriptional activity (N-terminal A/B and accessory proteins from insect cDNA libraries domain of Usp). This hypothesis was consistent by expression cloning, since expression of such with other observations. First, the EcR\mAR-CDE proteins should also alter Usp and EcR activity. chimera was additively induced by Usp and The information obtained by using yeast to testosterone; if the chimera was maximally induced delineate the mechanism of function for EcR

Journal of Molecular Endocrinology (2000) 24, 183–191 www.endocrinology.org

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access Activity of ecdysone receptors in yeast ·     and others 191 isoforms and Usp should provide insights into their transcription unit in Saccharomyces cerevisiae. Molecular and complex roles in insect biology and advance the Cellular Biology 9 3517–3523. MakP,FuernkranzHA,GeR&Karathanasis SK 1994a understanding of the function of steroid/nuclear Retinoid X receptor homodimers function as transcriptional receptors in general. activators in yeast. Gene 145 129–133. Mak P, Young CY-F & Tindall DJ 1994b A novel yeast expression system to study androgen action. Recent Progress ACKNOWLEDGEMENTS in Hormone Research 49 347–352. Mangelsdorf DJ & Evans RM 1995 The RXR heterodimers and orphan receptors. Cell 83 841–850. We are grateful to William R Baumbach for helpful Miller JH 1972 Assays for -galactosidase. In Experiments in discussions and suggestions in this study. We thank Molecular Genetics,pp352–355. Ed. JH Miller. New York: Deborah Chaleff for the can1 strain. Cold Spring Harbor Laboratory. Oro AE, McKeown M & Evans RM 1990 Relationship between the product of the Drosophila ultraspiracle locus and the vertebrate retinoid X receptor. Nature 347 298–301. REFERENCES Sambrook J, Fritsch EF & Maniatis T 1989 Molecular Cloning: A Laboratory Manual. New York: Cold Spring Harbor Buszczak M & Segraves WA 1998 Drosophila metamorphosis: Laboratory. the only way is USP? Current Biology 8 R879–R882. Segraves WA 1994 Steroid receptors and other transcription ButtTR&Walfish PG 1996 Human nuclear receptor factors in ecdysone response. Recent Progress in Hormone heterodimers: opportunities for detecting targets of Research 49 167–195. transcriptional regulation using yeast. Gene Expression 5 Shea MJ, King DL, Conboy MJ, Mariani BD & Kafatos FC 255–268. 1990 Proteins that bind to Drosophila chorion cis-regulatory DelaCruzF&MakP1997 Drosophila ecdysone receptor elements: a new C2H2 zinc finger protein and a C2H2 steroid functions as a constitutive activator in yeast. Journal of receptor-like component. Genes and Development 4 Steroid Biochemistry and Molecular Biology 62 353–359. 1128–1140. Dressel U, Thormeyer D, Altincicek B, Paululat A, Eggert M, Sutherland JD, Kozlova T, TzertzinisG&Kafatos FC 1995 SchneiderS,TenbaumSP,RenkawitzR&BaniahmadA Drosophila hormone receptor 38: a second partner for 1999 Alien, a highly conserved protein with characteristics of Drosophila USP suggests an unexpected role for nuclear a corepressor for members of the nuclear hormone receptor receptors of the nerve growth factor-inducible protein B superfamily. Molecular and Cellular Biology 19 3383–3394. type. Proceedings of the National Academy of Sciences of the Garabedian MJ 1993 Genetic approaches to mammalian USA 92 7966–7970. nuclear receptor function in yeast. A Companion to Methods Talbot WS, Swyryd EA & Hogness DS 1993 Drosophila ff in Enzymology 138–146. tissues with di erent metamorphic responses to ecdysone 5 ff GuthrieC&FinkGR1991 Guide to yeast genetics and express di erent ecdysone receptor isoforms. Cell 73 molecular biology. In Methods in Enzymology, vol. 194. Eds 1323–1337. C Guthrie & GR Fink. New York: Academic Press. Thomas HE, Stunnenberg HG & Stewart AF 1993 Hall BL & Thummel CS 1998 The RXR homolog ultraspiracle Heterodimerization of the Drosophila ecdysone receptor with is an essential component of the Drosophila ecdysone retinoid X receptor and ultraspiracle. Nature 362 471–475. receptor. Development 125 4709–4717. Thormeyer D, Tenbaum SP, Renkawitz R & Baniahmad A Heinrich JN, Dela CruzF&Kirsch DR 1997 Screen for 1999 EcR interacts with corepressors and harbours an ultraspiracle inhibitors. Patent Application WO 97/45737. autonomous silencing domain functional in both Drosophila Jones G & Sharp PA 1997 Ultraspiracle: an invertebrate and vertebrate cells. Journal of Steroid Biochemistry and receptor for juvenile hormones. Proceedings of the National Molecular Biology 68 163–169. Academy of Sciences of the USA 94 13499–13503. Weibel FF & Gustafsson J-Å 1997 Heterodimeric interaction Kephart DD, Walfish PG, DeLuca H & Butt TR 1996 between retinoid X receptor  and orphan nuclear receptor Retinoid X receptor isotype identity directs human vitamin OR1 reveals dimerization-induced activation as a novel D receptor heterodimer transactivation from the mechanism of nuclear receptor activation. Molecular and 24-hydroxlyase vitamin D response elements in yeast. Cellular Biology 17 3977–3986. Molecular Endocrinology 10 408–419. Yao T-P, Segraves WA, Oro AE, McKeown M & Evans RM Koeller MR 1992 Molecular analysis of the Drosophila 1992 Drosophila ultraspiracle modulates ecdysone receptor ecdysone receptor complex. PhD Thesis. Stanford University, function via heterodimer formation. Cell 71 63–72. Stanford, California, USA. McDonnell DP, Pike JW, Drutz DJ, Butt TR & O’Malley BW 1989 Reconstitution of the vitamin D-responsive osteocalcin    19 August 1999

www.endocrinology.org Journal of Molecular Endocrinology (2000) 24, 183–191

Downloaded from Bioscientifica.com at 10/01/2021 02:02:13AM via free access