Expression level-dependent contribution of glucocorticoid receptor domains for functional interaction with STAT5.

The action of the glucocorticoid receptor (GR) on beta-casein gene transcription serves as a well-studied example of a case where the action of the GR is dependent on the activity of another transcription factor, STAT5. We have investigated the domain-requirement of the GR for this synergistic response in transfection experiments employing GR mutants and CV-1 or COS-7 cells. The results were influenced by the expression levels of the GR constructs. At low expression, STAT5-dependent transactivation by mutants of the GR DNA binding domain or N-terminal transactivation domain was impaired and the antiglucocorticoid RU486 exhibited a weak agonistic activity. When the N-terminal region of the GR was exchanged with the respective domain of the progesterone receptor, STAT5-dependent transactivation was reduced at low and high expression levels. Only at high expression levels did the GR exhibit the properties of a coactivator and enhanced STAT5 activity in the absence of a functional DNA binding domain and of GR binding sites in the proximal region of the beta-casein gene promoter. Furthermore, at high GR expression levels RU486 was nearly as efficient as dexamethasone in activating transcription via the STAT5 dependent beta-casein gene promoter. The results reconcile the controversial issue regarding the DNA binding-independent action of the GR together with STAT5 and provide evidence that the mode of action of the GR depends not only on the type of the particular promoter at which it acts but also on the concentration of the GR. GR DNA binding function appears to be mandatory for beta-casein gene expression in mammary epithelial cells, since the promoter function is completely dependent on the integrity of GR binding sites in the promoter.

Synergistic and antagonistic interactions of transcription factors in the regulation of milk protein gene expression. Mechanisms of cross-talk between signalling pathways.

The stage and tissue specific expression of milk protein genes in the mammary gland is controlled by modular response regions with multiple binding sites for distinct classes of transcription factors, which either co-operate or are antagonistic. In addition, the activity of some of these factors is individually control-led by diverse extracellular signals. A well studied paradigm for a synergistic co-operation is the activation of beta-casein gene transcription by prolactin and glucocorticoids mediated by the signal transducer and activator of transcription STAT5 and the glucocorticoid receptor (GR). As an example for an antagonistic interaction we can demonstrate inhibition of prolactin signalling by TNF-alpha, which is mediated by NF-kappa B. In both cases, the interactions occur at several levels: For GR and STAT5, the synergy is discussed to be promoted by protein-protein interactions. Furthermore, we can demonstrate a co-operation between GR and STAT5 in DNA binding by a mechanism, which is dependent on the integrity of the DNA binding domain of the GR and on the existence of half-palindromic GR binding sites in the hormone response region. Indirect effects of glucocorticoids by modulation of the expression of secondary genes are also important. They might account for the observed enhancement of prolactin induced tyrosine phosphorylation of STAT5 by glucocorticoids. For NF-kappa B and STAT5, one component of the antagonism is the inhibition of STAT5 tyrosine phosphorylation by activation of NF-kappa B. Another potential mechanism is the inhibition of DNA binding of STAT5 due to overlapping binding sites for STAT5 and NF-kappa B in the beta-casein gene promoter. Thus, synergistic and antagonistic interactions between GR, NF-kappa B, and STAT5 involve (a) cross-talk mechanisms influencing the activation of STAT5 and (b) promoter-dependent interactions modulating the DNA binding activity of the transcription factors.

Activation of NF-kappaB p50/p65 is regulated in the developing mammary gland and inhibits STAT5-mediated beta-casein gene expression.

The NF-kappaB family of transcription factors regulates diverse cellular functions such as immune response and cell growth and development, and has been reported to be constitutively active in a variety of mammary carcinoma cell lines. However, its role in normal mammary gland development has not been addressed. In our study, we detected developmentally regulated NF-kappaB activity in the mammary gland of mice. During pregnancy, DNA binding activity of NF-kappaB p50/p65 increased until day 16 postcoitum and decreased with the onset of lactation, most likely due to reduced p50 and p65 protein levels in the nucleus. Cotransfection experiments performed with 293 cells revealed an inhibition of the prolactin receptor/JAK2/STAT5 pathway by NF-kappaB. In HC11 cells, NF-kappaB p50/p65 activity was inversely correlated with prolactin-induced STAT5 tyrosine phosphorylation, expression of endogenous beta-casein gene, and of a transfected beta-casein gene promoter reporter construct. This indicates a negative cross talk between NF-kappaB and the prolactin receptor/JAK2/STAT5 activation pathway, which occurs at the level of STAT5 tyrosine phosphorylation. Our results provide evidence for a role of NF-kappaB in normal mammary gland development, and indicate its function as a negative regulator of beta-casein gene expression during pregnancy by interfering with STAT5 tyrosine phosphorylation.

Transcription factors of the Sp1 family: interaction with E2F and regulation of the murine thymidine kinase promoter.

Promoters of growth and cell cycle regulated genes frequently carry binding sites for transcription factors of the E2F and Sp1 families. We have demonstrated recently that direct interaction between Sp1 and a subgroup of the E2F factors is essential for the regulation of certain promoters. We show here that the amino acids necessary for this interaction in both cases are located within the DNA binding domain. This is in line with the assumption, that the interaction between E2F and Sp-factors contributes to promoter-specificity. Cyclin A, which binds to E2F-1 in close vicinity to Sp1 does not interfere with this interaction. Moreover we have investigated the ability of other members of the Sp1 family to interact with E2F-1 and to regulate the activity of the E2F and Sp1 dependent murine thymidine kinase promoter. All four factors of the Sp1 family are able to bind E2F-1 in co-immunoprecipitation and GST-pull down experiments. Mobility shift assays with oligonucleotides comprising the Sp1, or both the Sp1 and the E2F binding site suggest that Sp1 and Sp3 supply most if not all activity binding to the GC-box of the thymidine kinase promoter in murine fibroblasts. Reporter gene assays in Drosophila melanogaster SL2 cells and murine fibroblast 3T6 cells demonstrate that the thymidine kinase promoter is activated strongly by Sp1 and Sp3, weakly by Sp4, and not at all by Sp2. Co-expression of E2F-1 results in synergistic activation in 3T6 but not in SL2 cells.

A distal enhancer region in the human beta-casein gene mediates the response to prolactin and glucocorticoid hormones.

The 5' flanking region of the human beta-casein gene was investigated for the presence of regulatory sequences mediating the action of the lactogenic hormones prolactin and dexamethasone. DNA encompassing 9389 base pairs of the flanking region was isolated and a sequence comparison performed with regulatory regions previously identified in the beta-casein gene of rodents and ruminants. The analysis revealed the presence of a distal region between -4700 and -4550 with a high percentage of identity to the bovine beta-casein enhancer region, and a proximal region between -1 and -200 similar to the proximal promoter regions found in rodents and ruminants. Reporter gene constructs under the control of the distal or the proximal region of the human beta-casein gene were tested for their responsiveness to prolactin and dexamethasone. In transfection experiments, the distal region functioned as a lactogenic hormone inducible enhancer, whereas the proximal region exhibited low activity. In electromobility shift assays, multiple binding sites for Stat5, CCAAT/enhancer-binding proteins, and Ets domain proteins were identified in the distal human enhancer. These transcription factors have already been demonstrated as important regulators of the transcription of milk protein genes in rodents. Thus, a common set of transcription factors appears to be required for the expression of the human beta-casein gene and of milk protein genes in other species.