Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Demonstration by Confocal Microscopy that Unliganded Overexpressed Glucocorticoid Receptors are Distributed in a Nonrandom Manner throughout All Planes of the Nucleus Vilma R. Martins*, William B. Pratt, Louis Terracio, Margaret A. Hirst, Gordon M. Ringold, and Paul R. Housley Department of Pharmacology (V.R.M, P.R.H.) Department of Anatomy, Cell Biology, and Neurosciences (L.T.) University of South Carolina School of Medicine Columbia, South Carolina 29208 Department of Pharmacology University of Michigan Medical School (W.B.P.) Ann Arbor, Michigan 48109 Institute of Cancer and Developmental Biology Syntex, Inc. (M.A.H., G.M.R.) Palo Alto, California 94304 Mouse glucocorticoid receptors (GR) that are overexpressed in Chinese hamster ovary (CHO) cells behave like progesterone receptors, in that the unliganded receptor localizes to the nucleus where it resides in a loosely bound docking complex, probably in association with the 90-kDa heat shock protein (hsp90) and hsp70. In this paper we examine the localization of the overexpressed GR within the CHO cell nucleus by confocal microscopy. In hormone-free cells the receptor distributes in a mottled pattern throughout all planes of the nucleus. The receptor is not present in nucleoli and shows no preferential localization in the periphery vs. the center of the nucleus. The mottled distribution in each plane of the nucleus demonstrates clearly that there are regions that do not contain receptor; thus, the distribution of the GR is not random. When triamcinolone acetonide is added to the CHO cells, there is no detectable change in receptor distribution. Overexpressed receptors that have either no hormonebinding activity or no DNA-binding activity because of point mutations localize in the same mottled pattern as the wild-type receptor. These observations are consistent with the proposal that the overexpressed GR can enter the nucleus in its unliganded state and proceed to loci distributed throughout the nucleus, where it is retained in an inactive docking complex until the binding of hormone triggers its 0888-8809/91 /0217-0225S03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society 217 progression to high affinity sites where the primary events in transcriptional activation occur. As there is no detectable change in localization with the addition of ligand, we suggest that the docking complex may be located very near or possibly at the site where the primary events in transcriptional activation occur. (Molecular Endocrinology 5: 217-225, 1991) INTRODUCTION Although glucocorticoid, estrogen, and progesterone receptors are all recovered from hormone-free cells in the cytosolic fraction as heteromolecular complexes in association with the 90-kDa heat shock protein (hsp90) (1), there exists a basic difference in the behavior of unliganded glucocorticoid receptors (GR) vs. estrogen and progesterone receptors in intact cells. Immunocytochemical studies have shown that unliganded estrogen and progesterone receptors are located in the nucleus (2-4). This nuclear localization is supported by the results of enucleation studies as well (5, 6). In contrast, several laboratories have used immunofluorescence and immunocytochemical localization methods to demonstrate that unliganded GR are located predominantly in the cytoplasm, and that the receptors shift to the nucleus upon exposure to hormone (7-12). It is not known how steroid receptors distribute once they have entered the nucleus. Is their distribution Vol 5 No. 2 MOL ENDO-1991 218 random, as indicated by an immunofluorescence or immunohistochemical signal that is distributed as a diffuse haze throughout the nucleus at all depths within the nucleus? Is the distribution nonrandom, with the existence of regions that do not contain receptors? Are unliganded receptors that are retained in the inactive docking state located toward the periphery of the nucleus? Are the loosely bound docking complexes located near the high affinity acceptor sites, which are presumably the sites where the primary events leading to transcriptional activation occur? Or does the receptor undergo a spatial redistribution after it binds hormone and moves from low affinity to high affinity attachments? In this manuscript, we use the technique of confocal scanning microscopy to answer some of these questions. Confocal microscopy is capable of quantitative imaging of optical sections within a particular plane of a cell and can be used to analyze the three-dimensional organization of cellular macromolecules (13). To amplify the receptor signal, we examine the localization of the mouse GR in a Chinese hamster ovary (CHO) cell line in which the receptor is expressed about 10O-fold over the level of the endogenously expressed GR in parent CHO cells (14). We have recently shown that the unliganded mouse GR in the overexpressing CHO cell lines is located in the nucleus (15) in the same manner as reported for estrogen and progesterone receptors produced from normal cellular genes (2-4). Also, as reported for progesterone receptors (16, 17), the overexpressed mouse GR is recovered in the cytosolic fraction in association with hsp70 as well as hsp90 after hypotonic cell rupture (15). The overexpressed mouse receptor in CHO cells provides an excellent system for examining the nuclear distribution of both unliganded and hormone-bound receptors by immunofluorescence, both because the receptor signal is amplified and because there is unequivocal genetic proof that the immunofluorescence signal represents the receptor. Here we show that the unliganded GR distributes in a mottled or patchy nonrandom manner throughout all planes of the nucleus. The receptor is not located in the nucleolus, and there is no preference for the periphery versus the center of the nucleus. The distribution of the immunofluorescent signal does not change with binding of hormone. RESULTS Nonrandom Distribution of the Intranuclear GR Figure 1A shows the absence of immunofluorescence in CHO cells transfected with plasmid containing the cDNA for dihydrofolate reductase but no cDNA for GR. The WCL2 cells shown in Fig. 1B are transfected with cDNA for the wild-type mouse GR. All of the panels in Fig. 1 represent transverse intranuclear section scans through the center of nuclei. The immunofluorescence pattern throughout the nucleus is mottled, with promi- nent nucleolar shadows being evident in both this and subsequent photographs. This mottled pattern is seen with either cold methanol or paraformaldehyde fixation. The WCL2 cells are a cloned cell line, and the cultures originally contained 100% GR-expressing cells. During the past 1.5 yr, the WCL2 cells used in this study have been maintained in serial culture, and approximately 30% of the cells no longer express GR, as judged by immunofluorescence. In this work we do not present fields containing the revertant cells. The cells shown in Fig. 1 were grown in hormonefree medium (without phenol red and supplemented with charcoal-stripped serum); thus, it is the unliganded wild-type GR in Fig. 1B that is localizing to the nucleus. As shown in Fig. 1C, the GR in NA cells produce the same mottled nuclear distribution as the wild-type receptor. The NA cells express mouse receptors that have a point mutation in the hormone-binding domain and do not bind steroid (14). This provides genetic proof that the mottled localization pattern cannot reflect a steroid-mediated change in receptor localization. The NB cells shown in Fig. 1D express receptors that contain a point mutation that eliminates DNA-binding activity (14). Because these cells show that the same GR distribution as the cells containing the wild-type mouse GR, the mottled distribution pattern does not require that the receptor be bound to DNA. Steroid Does not Alter GR Distribution Figure 2 shows localization of the wild-type GR in WCL2 cells determined with polyclonal anti-hGR antibody 57 as well as with the BuGR2 monoclonal antibody. The mottled nuclear distribution is seen with both antibodies, and in both cases the signal is immune specific. There is no change in the mottled distribution pattern after exposure to a glucocorticoid agonist {c.f. Fig. 2, B and E with C and F, respectively). These are single intranuclear sections, but it is important to note that we see no alteration in receptor distribution with steroid when nuclei are scanned in a series of planar views such as those shown in Fig. 3. Additionally, the mottled pattern exhibited by NA and NB cells is not altered by steroid. GR Distribution in Nuclear Sections Figure 3 shows a series of six optical sections, 2 /urn apart, through several WCL2 cell nuclei. The GR immunofluorescence is again arranged in a mottled pattern present in all sections, showing that there is no preferential localization in the nuclear periphery vs. the core. The mottled distribution of the GR is particularly clear when one views nuclear sections. Figure 4 shows a high magnification view of sections from WCL2, NA, and NB cells. These close-up views show no difference in receptor distribution between the wild-type and mutant receptors. The immunofluorescence patterns in Fig. 4 (A-C) show that there is no GR in the nucleolus and that the receptor is excluded from multiple discrete Intranuclear Distribution of GR 219 Fig. 1. Nonrandom Nuclear Distribution of Overexpressed Wild-Type and Mutant GR Cells were incubated with the BuGR2 antibody and analyzed by confocal imaging, as described in Materials and Methods. The images are transverse optical sections through the center of the cells. The cell lines shown are CHO plus dhfr (A), WCL2 (B), NA (C), and NB (D). patchy areas throughout each section. This type of pattern has been previously observed for other nuclear moieties, such as small nuclear ribonucleoprotein (snRNP) structures (18). Figure 4D shows the immunofluorescence pattern obtained when a monoclonal antibody to RNP is reacted with WCL2 cells. The pattern of U1-RNP nuclear immunofluorescence is similar to that obtained with the anti-GR probes. DISCUSSION It is clear that the nuclear immunofluorescence shown in Figs. 1-4 represents the localization of the mouse GR. The nuclear immunofluorescence is both immune specific for two antireceptor antibodies (Fig. 2) and specific for the presence of mouse GR cDNA (Fig. 1). The mottled nuclear localization is seen in the absence of steroid (Figs. 1 -4), and it occurs with the NA mutant, which is incapable of binding steroid (Figs. 1 and 4). Thus, the overexpressed mouse GR in CHO cells is delivered to the nucleus in its unliganded state. It is not clear why the overexpressed receptor is nuclear. Unfortunately, the level of endogenous hamster GR in the CHO cell is too low (1/1 OOth of the level of wild-type mouse GR in WCL2 cells) (15) to permit detection with this immunofluorescence method. It is possible that the endogenous hamster GR is distributed predominantly in the cytoplasm in its unliganded state, as has been reported for a variety of other cells (7-12). If the endogenous hamster receptor were in the cytoplasm, the nuclear localization of the mouse GR in CHO cells might be a result of receptor overexpression. Cidlowski et al. (19) have recently shown that CHO cells containing transfected human GR at a level that is increased 3-fold above that of endogenous hamster GR exhibit predominately cytoplasmic receptors, with some nuclear receptor signal in the absence of glucocorticoid agonist. As these transfected CHO cells contain approximately 5% the level of GR found in WCL2 cells, the increased receptor content of WCL2 cells may be sufficient to promote predominately nuclear localization in the absence of steroid. Using our techniques we have also examined the unliganded mouse GR in COS cells transiently transfected with the pSV2Wrec expression vector containing the mouse GR cDNA. In transfected COS cells exhibiting immunofluoresence and presum- MOL ENDO-1991 220 Vol 5 No. 2 Fig. 2. Receptors Overexpressed in CHO Cells Have the Same Nuclear Distribution in the Presence or Absence of Glucocorticoid WCL2 cells were incubated with vehicle (A, B, D, and E) or 1 HM triamcinolone acetonide (C and F) and processed for confocal imaging after incubation with nonimmune mouse IgG (A), BuGR2 antibody (B and C), normal rabbit serum (D), or antibody 57 (E and F). Intranuclear Distribution of GR 221 Fig. 3. GR Innmunofiuorescence in Successive Nuclear Sections of WCL2 Cells Optical sectioning of WCL2 cells incubated with BuGR2 antibody was carried out at 2-/xm increments, starting at a level near the coverslip (A) and progressing toward the apical surface (B-F). ably overexpressing the mouse GR, the receptor signal is localized to the nucleus (data not shown). Although several cell types containing normal amounts of endogenous GR exhibit an unliganded receptor signal that is predominately cytoplasmic (7-12, 19), it should be noted that it has been shown by both an immunocytochemical technique using the BuGR2 antibody (20) and by cell enucleation techniques (6, 20) that the majority MOL ENDO-1991 222 Vol 5 No. 2 Fig. 4. Magnified Images of GR and RNP Intranuclear Localization WCL2 (A and D), NA (B), and NB (C) cells were incubated with BuGR2 (A-C) or anti-RNP (D) antibodies and scanned at high magnification. of the unliganded GR is located in nuclei of GH3 pituitary tumor cells. Thus, it is possible that in some cell types the endogenous (nontransfected) GR is localized predominantly to the nucleus. Picard and Yamamoto (21) have identified two nuclear localization signals in the GR. One signal (NL1) is located just to the COOH-terminal side of the DNAbinding domain, and the other (NL2) is located within the COOH-terminal one third of the receptor that contains the hormone-binding domain. Nuclear localization of fusion proteins containing NL2 is hormone dependent, whereas nuclear localization of fusion proteins directed by NL1 is not (21). In that the mottled distribution of the overexpressed mouse GR shown in this paper is independent of hormone, it is possible that the passage of GR into the nucleus and to the sites that make up this pattern is directed by NL1. We have tested the simple model in which exposure of the NL1 sequence is correlated with nuclear localization of the unliganded GR. The overexpressed mouse GR in WCL2 cytosol and the normally expressed mouse GR in L-cell cytosol were reacted with the AP64 antibody directed against the NL1 sequence. In both cases, about 20% of the untransformed non-DNA-binding receptors reacted with the AP64 antibody, and this value increased to 85-90% after receptor transformation (15). Thus, we found no differential accessibility of the NL1 sequence on receptors in cytosol prepared from cells in which the unliganded GR is localized entirely to the nucleus (WCL2 cells) and in cytosol prepared from cells where a substantial portion of the receptor is located in the cytoplasm (L-cells) (20). This result could represent a difference between the in vivo and in vitro accessibility of NL1, or the ability of the NL1 sequence to interact with the nuclear localization apparatus may not be reflected by immunoadsorption of the GR with AP64. Alternatively, it is possible that the NL1 sequence is not required for the nuclear localization process we are observing in WCL2 cells, and exposure of NL1 might be required only for the tight nuclear association seen after ligand binding. Given our current primitive understanding of the systems that determine transport to and within the nucleus, it is reasonable to propose that the attachment of the overexpressed mouse GR via a nuclear localization signal to some cytoskeletal-based transport system permits its movement from its cytoplasmic site of synthesis to multiple foci located within all planes of the nucleus. The mottled pattern of GR distribution in the nucleus indicates that there are regions (in addition to Intranuclear Distribution of GR nucleoli) to which the receptor is not transported. The wild-type mouse receptors in WCL2 cells are transcriptionally active in a hormone-dependent manner (14), and after hypotonic cell rupture, the unliganded receptors are recovered in cytosol in a complex that contains both hsp90 and hsp70 (15). It seems likely that unliganded GR in WCL2 cells are transported to the loci represented by the mottled distribution within the nucleus, and once they are there, the receptors remain in a docking complex until the binding of hormone triggers their conversion to high affinity association with nuclear acceptor sites. Although the elements that make up the proposed docking complex have not been characterized, hsp90 and hsp70 are potential components. Given a generalized mottled distribution pattern, one cannot conclude from our failure to see any change in GR distribution on exposure to hormone (Fig. 2) that there is no intranuclear movement of the GR as it proceeds from low affinity (docking) to high affinity association with the nucleus. It is clear, however, that no gross rearrangement (e.g. from the periphery to more central regions) occurs. Perrot-Applanat et al. (22) examined the effect of hormone on localization of the rabbit progesterone receptor at the electron microscopic level using monoclonal antibody and the proteinA-gold technique. They found that the unliganded receptor was randomly scattered over clumps of condensed chromatin, but after hormone administration, it was mainly detected in the border regions between condensed chromatin and nucleoplasm. A very local redistribution of this nature could probably not be detected with the confocal techniques we have employed in this work. The thyroid hormone receptor is a member of the steroid receptor family (23). In contrast to glucocorticoid, estrogen, and progesterone receptors, the thyroid hormone receptor does not bind hsp90 (24) and does not enter a loosely bound docking association with the nucleus (25, 26). In its unliganded state, the thyroid hormone receptor rapidly becomes tightly associated with the nucleus (25, 26) and can in some instances act as a transcriptional inhibitor (27). Thus, it would seem that the unliganded thyroid hormone receptor makes its way directly to appropriate regions of the genome containing thyroid hormone response elements with which the receptor directly interacts. The absence of any gross movement of the overexpressed GR when hormone is added to WCL2 cells suggests that the docking complexes containing unliganded receptor may be located at multiple termini of the nuclear transport pathway for the receptor. Although the unliganded GR is presumably not bound to a glucocorticoid response element (28) in the same manner as the unliganded thyroid hormone receptor is probably bound to a thyroid hormone response element, the docking complexes may, nevertheless, be located very close to, or even at, the loci where the primary events in transcriptional activation occur. It is interesting to note that snRNP particles, which exhibit an immunofluorescence pattern similar to that of the overexpressed mouse GR (Fig. 4), 223 appear to be localized within the nucleus to a reticular network that extends between the nucleolar surface and the nuclear envelope (29). Often, the nuclear localization of a steroid hormone receptor has been thought to be determined by its DNA-binding activity (see Ref. 30 for a review of nuclear binding models). Some models have been based on the assumption that hormone-mediated transformation of a steroid receptor to the DNA-binding state results in its passive trapping in the nucleus by binding to DNA. As we show that the NB mutant GR, which has no DNA-binding activity, is localized to the nucleus and has the same mottled distribution pattern within the nucleus as the wild-type receptor (Figs. 1 and 4), it is clear that DNA-binding activity is irrelevant to the nuclear localization process we are observing. MATERIALS AND METHODS Materials Triamcinolone acetonide, polylysine, methotrexate, nonimmune mouse immunoglobulin G (IgG), 1,4-diazabicyclo[2.2.2.] octane, and fluorescein isothiocyanate-conjugated antirabbit IgG were obtained from Sigma Chemical Co. (St. Louis, MO). Dulbecco's Modified Eagle's Medium without phenol red was purchased from Hazleton (Lenexa, KS). Iron-supplemented bovine calf serum was obtained from Hyclone Laboratories (Logan, UT). The BuGR2 monoclonal antibody (31) was kindly provided by Drs. William J. Hendry and Robert W. Harrison (University of Arkansas for Medical Sciences, Little Rock, AK). Polyclonal antibody 57 was elicited in a rabbit against a peptide corresponding to amino acids 346-367 of the human GR (19) and was kindly provided by Dr. John A. Cidlowski (University of North Carolina, Chapel Hill, NC). The corresponding amino acid sequence of the mouse GR differs by only one amino acid, a conservative substitution of Val for lie. We have previously determined that polyclonal antibody 57 reacts with the mouse GR on blots and in vitro. Bodipy-conjugated antimouse IgG was from Molecular Probes (Eugene, OR). The monoclonal antibody to U1-RNP (32) was kindly provided by Dr. Jeffrey R. Patton (University of South Carolina, Columbia, SC). Cell Lines Construction of the four CHO cell lines used in this work has been previously described (14). Essentially, CHO cells were transfected with a plasmid containing the cDNA for dihydrofolate reductase (dhfr) and a plasmid containing wild-type or mutant mouse GR cDNA, and then selected for varying degrees of amplification by growth in methotrexate (MTX). The cell lines are: 1) CHO plus dhfr, CHO cells transfected with only the cDNA for dhfr and maintained in 10 MM MTX; 2) WCL2 cells, a CHO subline transfected with wild-type mouse GR and maintained in 3 ^M MTX; 3) NA cells, CHO cells that were transfected with a mutant mouse GR that does not bind steroid because of a point mutation in the hormone-binding domain where Glu546 has been changed to Gly (14); NA cells were maintained in 100 ^ M MTX; and 4) NB cells, CHO cells expressing a mutant mouse GR that binds hormone but is unable to bind to DNA because of a point mutation in the DNAbinding domain in which Arg484 has been changed to His (14); NB cells were maintained in 10 ^M MTX. WCL2, NA, and NB are all cloned cell lines that contain approximately equal amounts of immunoreactive GR protein (15). All CHO cell lines were grown in monolayer culture in Dulbecco's modified Eagle Vol 5 No. 2 MOL ENDO-1991 224 medium supplemented with 40 supplemented calf serum. proline and 10% iron- Confocal Microscopy In preparation for confocal studies, all cells were grown on polylysine-coated glass coverslips in medium without phenol red containing charcoal-stripped serum for 2 days, with daily medium changes. For the experiment shown in Fig. 2, cells were further incubated in medium containing vehicle or 1 MM triamcinolone acetonide for 30 min at 37 C. Coverslips were washed with PBS, immersed in cold (-20 C) methanol for 10 min, and washed with PBS. Cells were permeabilized with 0.2% Triton X-100 in PBS at room temperature for 10 min and washed again with PBS. Cells were then incubated for 1 h at 37 C with one of the following primary antibodies diluted in PBS containing 0.5% BSA: BuGR2 or normal mouse IgG (60 ^g/ml), antibody 57 or normal rabbit serum (1:100), or antiRNP (1:100). After washing with 0.5% BSA in PBS, cells were subsequently incubated for 1 h at 37 C with one of the following diluted in 0.5% BSA in PBS: Bodipy-conjugated antimouse IgG (1:40) or fluorescein isothiocyanate-conjugated antirabbit IgG (1:40). Coverslips were inverted onto depression slides containing 10 mg/ml 1,4-diazabicyclo[2.2.2.]octane in PBS-glycerol (1:3) to retard laser bleaching and submitted to fluorescence microscopy using a Bio-Rad MRC-600 laser scanning confocal microscope (Richmond, CA). Serial optical sections were obtained at 0.5- to 2-^m increments. Acknowledgments We thank Drs. William J. Hendry III and Robert W. Harrison III for providing the BuGR2 monoclonal antibody, Dr. John A. Cidlowski for providing polyclonal antibody 57 and for communicating results before publication, and Dr. Jeffrey R. Patton for providing the monoclonal antibody to U1-RNP. We are grateful to Nancy Vinson in the Integrated Microscopic Analysis Facility at the University of South Carolina School of Medicine for technical assistance with the confocal instrument. Received October 16, 1990. Revision received November 26,1990. Accepted November 29,1990. Address requests for reprints to: Paul R. Housley, Department of Pharmacology, University of South Carolina School of Medicine, Room D-6, Building 1, Columbia, South Carolina 29208. This work was supported by a fellowship grant from the Interamerican Development Bank and Universidade de Sao Paulo (to V.R.M.), American Cancer Society Grant IN-107 (to P.R.H.), and NIH Grants DK-36905 (to P.R.H.), CA-28010 (to W.B.P.), HL-40424 (to L.T.), and GM-25821 (to G.M.R.). * Present address: Faculdade de Medicina, Disciplina de Oncologia, Universidade de Sao Paulo, Sao Paulo, Brazil. REFERENCES 1. Pratt WB 1987 Transformation of glucocorticoid and progesterone receptors to the DNA-binding state. J Cell Biochem 35:51-68 2. King WJ, Greene GL 1984 Monoclonal antibodies localize oestrogen receptor in the nuclei of target cells. Nature 307:745-747 3. Perrot-Applanat M, Logeat F, Groyer-Picard MT, Milgrom E 1985 Immunocytochemical study of mammalian progesterone receptor using monoclonal antibodies. Endocrinology 116:1473-1484 4. Gasc JM, Delahaye F, Baulieu EE 1989 Compared intracellular localization of the glucocorticosteroid and progesterone receptors: an immunocytochemical study. Exp Cell Res 181:492-504 5. Welshons WV, Lieberman ME, Gorski J 1984 Nuclear localization of unoccupied oestrogen receptors. Nature 307:747-749 6. Welshons WV, Krummel BM, Gorski J 1985 Nuclear localization of unoccupied receptors for glucocorticoids, estrogens, and progesterone in GH3 cells. Endocrinology 117:2140-2147 7. Papamichail M, Tsokos G, Tsawdaroglou N, Sekeris CE 1980 Immunocytochemical demonstration of glucocorticoid receptors in different cell types and their translocation from the cytoplasm to the cell nucleus in the presence of dexamethasone. Exp Cell Res 125:490-493 8. Govindan MV 1980 Immunofluorescence microscopy of the intracellular translocation of glucocorticoid-receptor complexes in rat hepatoma (HTC) cells. Exp Cell Res 127:293-297 9. Antakly T, Eisen HJ 1984 Immunocytochemical localization of glucocorticoid receptor in target cells. Endocrinology 115:1984-1989 10. Fuxe K, Wikstrom AC, Okret S, Agnati LF, Harfstrand A, Yu ZY, Granholm L, Zoli M, Vale W, Gustafsson JA 1985 Mapping of glucocorticoid receptor immunoreactive neurons in the rat tel- and diencephalon using a monoclonal antibody against rat liver glucocorticoid receptor. Endocrinology 117:1803-1812 11. Wikstrom AC, Bakke O, Okret S, Bronnegard M, Gustafsson JA 1987 Intracellular localization of the glucocorticoid receptor: evidence for cytoplasmic and nuclear localization. Endocrinology 120:1232-1242 12. Qi M, Hamilton BJ, DeFranco D 1989 V-mos oncoproteins affect the nuclear retention and reutilization of glucocorticoid receptors. Mol Endocrinol 3:1279-1288 13. Shotton D, White N 1989 Confocal scanning microscopy: three-dimensional biological imaging. Trends Biol Sci 14:435-440 14. Hirst MA, Northrop JP, Danielsen M, Ringold GM 1990 High level expression of wild type and variant mouse glucocorticoid receptors in Chinese hamster ovary cells. Mol Endocrinol 4:162-170 15. Sanchez ER, Hirst M, Scherrer LC, Tang HY, Welsh MJ, Harmon JM, Simons SS, Ringold GM, Pratt WB 1990 Hormone-free mouse glucocorticoid receptors overexpressed in Chinese hamster ovary cells are localized to the nucleus and are associated with both hsp70 and hsp90. J Biol Chem 265:20123-20130 16. Kost SL, Smith DF, Sullivan W, Welch WJ, Toft DO 1989 Binding of heat shock proteins to the avian progesterone receptor. Mol Cell Biol 9:3829-3838 17. Smith DF, Faber LE, Toft DO 1990 Purification of unactivated progesterone receptor and identification of novel receptor-associated proteins. J Biol Chem 265:39964003 18. Verheijen R, Kuijpers H, Vooijs P, Van Venrooij W, Ramaekers F 1986 Distribution of the 70K U1 RNA-associated protein during interphase and mitosis: correlation with other U RNP particles and proteins of the nuclear matrix. J Cell Sci 86:173-190 19. Cidlowski JA, Bellingham DL, Powell-Oliver FE, Lubahn DB, Sar M 1990 Novel antipeptide antibodies to the human glucocorticoid receptor: recognition of multiple receptor forms in vitro and distinct localization of cytoplasmic and nuclear receptors. Mol Endocrinol 4:14271437 20. LaFond RE, Kennedy SW, Harrison RW, Villee CA 1988 Immunocytochemical localization of glucocorticoid receptors in cells, cytoplasts, and nucleoplasts. Exp Cell Res 175:52-62 21. Picard D, Yamamoto KR 1987 Two signals mediate hormone-dependent nuclear localization of the glucocorticoid receptor. EMBO J 6:3333-3340 225 Intranuclear Distribution of GR 22. Perrot-Applanat M, Groyer-Picard MT, Logeat F, Milgrom E 1986 Ultrastructural localization of the progesterone receptor by an immunogold method: effect of hormone administration. J Cell Biol 102:1191-1199 23. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889-895 24. Dalman FC, Koenig RJ, Perdew GH, Massa E, Pratt WB 1990 In contrast to the glucocorticoid receptor, the thyroid hormone receptor is translated in the DNA binding state and is not associated with hsp90. J Biol Chem 265:36153618 25. Samuels HH, Tsai JS, Casanova J, Stanley F1974 Thyroid hormone action: in vitro characterization of solubilized nuclear receptors from rat liver and cultured GH, cells. J Clin Invest 54:853-865 26. Casanova J, Horowitz ZD, Copp RP, Mclntyre WR, Pascual A, Samuels HH 1984 Photoaffinity labeling of thyroid hormone nuclear receptors: influence of n-butyrate and analysis of the half-lives of the 57,000 and 47,000 molec- 27. 28. 29. 30. 31. 32. ular weight receptor forms. J Biol Chem 259:1208412901 Koenig RJ, Lazar MA, Hodin RA, Brent GA, Larson PR, Chin WW, Moore DD 1989 Inhibition of thyroid hormone action by a non-hormone binding c-erbA protein generated by alternative mRNA splicing. Nature 337:659-661 Becker PB, Gloss B, Schmid W, Strahle U, Schutz G 1986 In vivo protein-DNA interactions in a glucocorticoid response element require the presence of hormone. Nature 324:686-688 Spector DL1990 Higher order nuclear organization: threedimensional distribution of small nuclear ribonucleoprotein particles Proc Natl Acad Sci USA 87:147-151 Walters MR 1985 Steroid hormone receptors and the nucleus. Endocr Rev 6:512-543 Gametchu B, Harrison RW 1984 Characterization of a monoclonal antibody to the rat liver glucocorticoid receptor. Endocrinology 114:274-279 Billings PB, Allen RW, Jensen FC, Hoch SO 1982 AntiRNP monoclonal antibodies derived from a mouse strain with lupus-like autoimmunity. J Immunol 128:1176-1180