The c-fos serum response element (SRE) confers negative response to glucocorticoids.

Ligand activated Glucocorticoid Receptor (GR), specifically inhibited the serum induced c-fos promoter activation in NIH3T3 fibroblasts. The negative control was mediated by the c-fos SRE and correlated with the relative abundance of active GR. Serum activated SRE was repressed 3-4-fold by glucocorticoids irrespective of the promoter context (heterologous or authentic). The suppressing ability of GR was absolutely dependent on its DNA binding domain (DBD), since deletion of this region left the serum induction unimpaired. The methylation interference pattern of GR revealed two distinct binding sites within the SRE and identified the GR contact bases, important also for binding and function of SRE targeted transcription factors, such as the Serum Response Factor (SRF) and the p62 Ternary Complex Factor (TCF). We conclude that GR binds to c-fos SRE and inhibits c-fos promoter activation by antagonizing the function of positive transcription factors targeting to overlapping or identical sites. Since the c-fos SRE is activated by multiple mitogenic signalling pathways, inactivation by GR could explain, at least in part, the growth inhibitory response of fibroblasts to glucocorticoid hormones.

Insulin potentiates the transactivation potency of the glucocorticoid receptor.

A single copy of a glucocorticoid-responsive element (GRE) is sufficient in mediating the combinatorial response of a promoter to both glucocorticoids and insulin in HepG2 cells. This requires the presence of active glucocorticoid receptor (GR) since the response is significantly inhibited by the anti-glucocorticoid RU30406. The N'- and C'-terminal parts of the GR protein are not involved in mediating the response. Insulin had no effect on GR binding to GRE but it affected both the level and the phosphorylation state of nuclear-bound GR. Thus, insulin alters the GR transactivation potency while, concomitantly, modifies the molecule at the posttranslational level.

Interaction of cytosol fractions containing activated glucocorticoid-receptor complexes from rat liver and thymus with heterologous nuclei: effects on transcription.

Two rat liver cytosol fractions containing activated glucocorticoid-receptor complexes are able to stimulate the transcriptional activity of rat liver nuclei; the respective fractions from the cytosol of thymocytes inhibit the capacity of thymus nuclei for RNA synthesis. A similar inhibitory effect on thymus nuclei is exerted by the presence of rat liver cytosol fractions. Spot hybridization using a tyrosine aminotransferase (TAT) probe demonstrates that TAT gene expression is stimulated by the liver cytosol fractions acting on homologous nuclei whereas it is inhibited, in thymus nuclei, by the addition of thymus cytosol fractions. No effect on transcription is observed if the liver or thymus cytosol is heat activated in the presence of the glucocorticoid antagonist, cortexolone. Treatment of liver nuclei, previously subjected to the action of thymus cytosol fractions with the respective liver ones, restores transcriptional activity to control or higher levels. We conclude that rat thymocyte nuclei and cytosol contain transcriptional factors, which in the presence of the glucocorticoid-receptor complex, irrespective of its source, inhibit gene expression, whereas in the absence of such factors, the glucocorticoid-receptor complex positively regulates the respective genes.

The mitochondrion as a primary site of action of glucocorticoids: the interaction of the glucocorticoid receptor with mitochondrial DNA sequences showing partial similarity to the nuclear glucocorticoid responsive elements.

Six mitochondrial genome sequences, showing strong similarity to the glucocorticoid responsive element consensus sequence (GRE), four localized within the cytochrome c oxidase (COX) subunit I and II genes (GREs I-IV) and two within the D-loop region (GREs a and b) have been examined as binding sites of glucocorticoid receptor (GR) from rat liver cytosol. Purified GR from rat liver cytosol binds with high specificity to all potential mitochondrial GREs, as shown by filter retention and gel shift assays. Specific binding of protein(s), present in a mitochondrial extract from dexamethasone-induced mice, to all six putative mitochondrial GREs was also documented by the same methodology. Both purified GR and protein(s) from mitochondrial extract give the same band in the gel retardation assay. Using monospecific anti-glucocorticoid receptor polyclonal antibody (EP), a supershift of the gel retarded protein-DNA band was obtained. These results demonstrate that the mitochondrial genome sequences examined have characteristics of GREs, since they show the capacity to specifically bind the respective receptor protein. These findings support the hypothesis that the mitochondrial genome is a primary site of action of steroid and thyroid hormones (Sekeris C.E.: The mitochondrial genome: a possible primary site of action of steroid hormones, In vivo 4 (1990) 317-320).

Import of the glucocorticoid receptor into rat liver mitochondria in vivo and in vitro.

Administration of inducing doses of dexamethasone (10 microg/100 g) to adrenalectomized rats results, within 2-5 min, in import of the glucocorticoid receptor from liver cytoplasm into mitochondria, as demonstrated by Western blotting and by electron microscopy. Furthermore, glucocorticoid receptor (GR) synthesized in an in vitro reticulocyte system programmed with the respective mRNA, enters within minutes to added rat liver mitochondria in the form of intact GR, as demonstrated by Western blotting using either monoclonal or polyclonal antibodies against different domains of GR. In vitro studies show that the import is dependent on temperature and/or activation of the hormone-GR complex. These results, in connection with the presence in the human and rodent mitochondrial genome of sequences showing partial homology to the nuclear glucocorticoid response elements, support the hypothesis that the well documented effects of glucocorticoids on mitochondrial functions result from a direct interaction of the GR complex with the mitochondrial genome.

Presence of glucocorticoid responsive elements in the mitochondrial genome.

Glucocorticoids rapidly affect nuclear RNA synthesis by binding, as hormone-receptor complexes, to Glucocorticoid Responsive Elements (GRE), differently positioned in glucocorticoid inducible genes. Glucocorticoids also affect, within minutes, mitochondrial RNA synthesis. We therefore searched for GREs in the mitochondrial genome of human (H), rat (R) and mouse (M) and found a number of such potential elements as follows: one within the 12s - rRNA gene (H1, R1 and M1) one within (H2), or at the start (R2, M2), of the presumptive protein 1 gene and three within the mouse cytochrome oxidase subunit 1 gene (COX1, COX2 and COX3). The nucleotide sequence of H1, R1 and M1 reveals the possibility of the formation of hairpin structures, stabilized by hydrogen bond formation, between three or four consecutive bases. The presence of potential GREs in the mitochondrial genome suggests an adjustment of mitochondrial metabolic control to the general control mechanisms of the cell.

Insulin enhances glucocorticoid induction of gene expression in a sequence specific manner.

Glucocorticoid receptor (GR) induces transcriptional activation of specific genes by binding to cis-acting enhancer sequences termed glucocorticoid responsive elements (GREs). Insulin regulates gene transcription part of which is mediated sequences dominated by GREs. By transient expression in HTC cells of fusion gene consist MMTV 5' flanking sequences containing the GREs on the thymidine kinase promoter (TK) and the coding region of the chloramphenicol acetyl transferase gene (CAT), we demonstrate that insulin increases 2-fold the produced induction from dexamethasone, while insulin alone does not exceed any significant effect. This synergistic effect of insulin is sequence specific since, expression of the glucocorticoid unresponsive gene RSV-CAT at the same cell line is affected neither by insulin nor dexamethasone or both. Transient expression of CAT constructs with progressive 5' deletion mutants of mouse mammary tumor virus (MMTV) promoter, fail to detect a cis-acting insulin responsive DNA element, that participates in increased dexamethasone mediated induction. We believe that insulin activates an intracellular mediator-transcription factor, that stimulates gene expression indirectly, by interaction with the glucocorticoid receptor.

The effect of partially purified dexamethasone receptor on RNA synthesis in rat thymus and liver nuclei.

DEAE Sephadex-A-50 chromatography of rat liver and thymus cytosol charged with dexamethasone (0.1 microM) and heated at 25 degrees C for 30 min, separates two main hormone--receptor complexes, one in the flow through (DE-1) and one eluting with 150 mM [Cl-] (DE-2). Incubation of these two fractions with homologous nuclei results in a 20% stimulation and a 20-25% inhibition of transcription of liver and thymus nuclei, respectively. No difference was observed between the action of DE-1 and DE-2. If the DE-1 and DE-2 fractions of cytosol charged with the glucocorticoid antagonist, cortexolone, are used under the same experimental conditions, no effect on the transcriptional activity of liver or thymus nuclei can be observed. The fact that from nuclei only DE-1 can be isolated, not DE-2, suggests that DE-1 is the intranuclearly active form and that DE-2, with a still unknown mechanism, is transformed to DE-1. A linear relation exists between the number of GR-complexes (DE-1 or DE-2) translocated and the change in RNA synthesis. Transcriptional effects start when a critical number of acceptor sites are occupied in the nucleus (approximately 200-300 sites/nucleus) for both receptor forms.

The detection of glucocorticoid receptors in breast cancer by immunocytochemical and biochemical methods.

Glucocorticoid receptors have been detected in 90 human breast tumors and tumor-like conditions by the immunoperoxidase method using a specific antibody against the glucocorticoid receptor isolated from rat thymocytes. In some of the specimens the [3H]-dexamethasone binding assay was also performed and the results obtained were compared with those of the immunoperoxidase method. When the biochemical method was used, no strict correlation between the degree of binding of [3H]-dexamethasone and malignancy on the basis of histological findings could be demonstrated. In contrast, the immunoperoxidase method was in full agreement with the histological type of the tumor. Thus, nearly all malignant breast tumors (carcinomas) were positive by the immunoperoxidase method to a vary degree. Most of the examined benign tumor-like conditions (fibrocystic disease) were found to be negative. Intermediate situations, such as atypical duct of lobular hyperplasia, papillomatosis etc, were mostly positive. These findings suggest that ther immunoperoxidase method, part from its usefulness for the detection of glucocorticoid receptors in breast tissue, may be used as an early biological marker to detect early conversion of normal to hyperplastic tissue and/or malignancy of the mammary gland.

Dexamethasone-binding proteins in cytosol and nucleus of rat thymocytes. Purification of three receptor proteins.

Dexamethasone-binding proteins from the cytosol and the nucleus of rat thymocytes were analyzed by ion-exchange chromatography on DEAE-cellulose. Three dexamethasone-binding proteins were revealed in cytosol, one in the flow-through (DE-1) and two (DE-2 and DE-3) eluting from the column with 0.13 M and 0.23 M NH4Cl, respectively. In nuclear extracts only one receptor fraction, present in the flow-through, could be detected. By a combination of affinity chromatography on Cl-Sepharose to which dexamethasone 21-methanesulfonate was linked through a disulufide bond and DEAE-cellulose chromatography, three receptor proteins were highly purified from cytosol, with molecular weights of 45 000, 72 000 and 90 000 and one from nuclear extracts with molecular weight of 72 000. Antibodies to the 45 000-Mr and 90 000-Mr proteins were elicited in rabbits. The antibodies to the 45 000-Mr protein cross-react with the 90 000-Mr. Similarly, the antibodies to the 90 000-Mr protein cross-react with the 45 000-Mr protein. Antibodies to either of the two proteins immunoprecipitate 60--70% of the dexamethasone-binding activity of rat thymus cytosol. Immunoaffinity chromatography of cytosol and nucleosol on columns of Sepharose linked to the IgG against either the 45 000-Mr or the 90 000-Mr protein leads to binding of these proteins on the columns but not of the 72 000-Mr species. Two nuclear polypeptides with molecular weights of 36 000 and 38 000 remain attached to the immunoaffinity column; these polypeptides may represent degradation products of the cytoplasmic receptor upon entrance into the nucleus. Antibodies against two dexamethasone-binding proteins from rat liver cytosol immunoprecipitate the 45 000-Mr cytosol receptors from rat thymus.