DNA binding specificity of mutant glucocorticoid receptor DNA-binding domains

Mutation of a small number of amino acids in the DNA-binding domain of the estrogen receptor to the corresponding sequence of the glucocorticoid receptor switches the specificity of the receptor in transactivation assays (Mader, S., Kumar, V., de Verneuil, H., and Chambon, P. (1989) Nature 338, 271-274). We have made the corresponding reciprocal mutations in the context of the glucocorticoid receptor DNA-binding domain and studied the binding of wild type and mutant purified proteins to palindromic glucocorticoid and estrogen response elements as well as to elements of intermediate sequence, using gel mobility shift assays. We show here that a protein with two altered amino acids binds glucocorticoid and estrogen response elements with a low but equal affinity, whereas a protein with an additional changed residue has a high affinity for estrogen response elements but still retains a considerable affinity for glucocorticoid response elements. Using binding sites of intermediate sequence we have further characterized the interaction with DNA. The in vitro DNA binding results are confirmed by in vivo transactivation assays in yeast. Finally we suggest a testable model for amino acid/base pair interactions involved in recognition by the glucocorticoid receptor DNA-binding domain of its target sequence.     

Interaction of the glucocorticoid receptor DNA-binding domain with DNA as a dimer is mediated by a short segment of five amino acids

We have previously shown that protein-protein interactions mediate cooperative binding of the glucocorticoid receptor DNA-binding domain to a glucocorticoid response element (Dahlman-Wright, K., Siltala-Roos, H., Carlstedt-Duke, J., and Gustafsson, J.-A. (1990) J. Biol. Chem. 265, 14030-14035). The cooperativity of DNA binding is lost when the distance between the two half-sites constituting a glucocorticoid responsive element is altered or when their relative orientation is changed. We show here that mutations in the responsive element which interfere with cooperative DNA binding by the glucocorticoid receptor DNA-binding domain in vitro also abolish transactivation by the full length glucocorticoid receptor in vivo. We also identify a short segment in the proximity of one of the bound zinc ions that is required for cooperative binding of the glucocorticoid receptor DNA-binding domain to a glucocorticoid response element. We suggest that this segment is involved in dimer formation of the native glucocorticoid receptor and that it is important for correct positioning of the dimeric molecule on the double helix of DNA.     

Protein-protein interactions facilitate DNA binding by the glucocorticoid receptor DNA-binding domain

We have studied the interaction of the DNA-binding domain of the glucocorticoid receptor with a glucocorticoid response element from the tyrosine aminotransferase gene. This response element consists of two binding sites (half-sites) for the glucocorticoid receptor DNA-binding domain. The sequences of these two half-sites are not identical, and we have previously shown that binding occurs preferentially to one of the half-sites (Tsai, S.-Y., Carlstedt-Duke, J., Weigel, N. L., Dahlman, K., Gustafsson, J.-A., Tsai, M.-J., and O'Malley, B. W. (1988) Cell 55, 361-369). We show here that binding to the low affinity half-site is dependent on previous occupancy of the high affinity half-site. This facilitated binding is dependent on the distance between the two half-sites and their relative orientation but is not dependent on the integrity of the DNA backbone. This is consistent with a model where DNA binding is not only dependent on interactions between the protein and its DNA target sequence but is also influenced by interactions between the protein molecules bound.     

The mouse glucocorticoid receptor DNA-binding domain is not phosphorylated in vivo

The glucocorticoid receptor is phosphorylated, but the precise location of the phosphorylated groups is unknown. We cultured AtT-20 cells in medium containing [32P]-orthophosphate and used immunoaffinity methods to isolate the intact receptor and a tryptic fragment containing the DNA binding domain. Analysis of the intact receptor, co-labeled with the affinity ligand dexamethasone-mesylate, confirmed that the receptor was phosphorylated. Isolation of the DNA binding domain by trypsinization and immunopurification showed that it was not phosphorylated. Interestingly, a non-immunoreactive phosphorylated fragment similar in size to the DNA-binding fragment was observed. Our results suggest that phosphorylation of the DNA binding domain of the glucocorticoid receptor is not essential for hormone action.     

Dimerization characteristics of the DNA- and steroid-binding domains of the androgen receptor

The DNA-binding domain (DBD) of the androgen, mineralocorticoid, and glucocorticoid receptors and the steroid-binding domain (SBD) of the androgen receptor (AR) were expressed separately as fusion proteins with glutathione-S-transferase (GST) in Escherichia coli. Native polyacrylamide gel electrophoresis and gel exclusion HPLC demonstrated that the GST-ARDBD fusion protein was present as a dimer. On the other hand, the GST-ARSBD fusion protein formed a high-molecular weight oligomer, which seemed to be formed by two separate interactions, i.e. GST-GST and ARSBD-ARSBD between the fusion molecules. These findings strongly suggest that ARSBD has a potent ability to form a homodimer and that ARDBD does not. GST-ARDBD specifically interacted with the glucocorticoid response elements of the mouse mammary tumor virus long terminal repeat (GRE_MMTV). Cleavage of the fusion protein by thrombin abolished the binding, while the nonspecific DNA-cellulose binding ability was retained. Therefore, the dimeric configuration of GST-ARDBD, even if accomplished through the interaction with the GST moiety, is needed for high-affinity binding to the response element. The binding of GST-ARDBD to GRE_MMTV was strongly competed by the glucocorticoid response element of rat tyrosine aminotransferase gene, followed by the androgen response element of the rat probasin gene. A palindromic thyroid response element showed no competition. Unexpectedly, no apparent different in the binding affinity to these response elements was observed among the DBDs of androgen, mineralocorticoid and glucocorticoid receptors.     

Vitamin D receptor interaction with specific DNA. Association as a 1,25-dihydroxyvitamin D3-modulated heterodimer.

The vitamin D receptor (VDR) is a member of the steroid receptor gene family. In this report, we examine the nature of specific VDR DNA binding utilizing the vitamin D-responsive element derived from the human osteocalcin promoter. Association of the VDR with the human osteocalcin 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) responsive element (VDRE) in vitro was characterized on VDRE affinity columns by both weak and strong interactions. Weak interaction was a property of the VDR itself, monomeric in nature, and determined exclusively by the VDR's DNA-binding domain. Strong interaction, in contrast, was dependent upon an intact receptor molecule as well as a heterologous mammalian cell nuclear accessory factor (NAF). Heteromeric interaction between VDR and NAF was independent of the VDR DNA-binding domain, suggesting the presence of a functional dimerization domain separate from that for DNA binding. Direct association of NAF with immobilized VDR revealed that the interaction does not require the presence of DNA. Most importantly, while occupancy of the VDR by 1,25(OH)2D3 was not required for VDR interactions with either DNA or NAF, the presence of hormone increased the apparent relative affinity of the VDR for NAF approximately 10-fold. These studies suggest that high affinity association of the VDR with DNA requires both the DNA-binding domain as well as an additional independent structure located within the steroid-binding region. This protein subdomain interacts with NAF and is regulated by 1,25(OH)2D3.     

Phosphorylated sites within the functional domains of the approximately 100-kDa steroid-binding subunit of glucocorticoid receptors

The steroid-binding subunit of the glucocorticoid receptor is known to be a approximately 100-kDa phosphoprotein composed of an immunogenic, DNA-binding, and steroid-binding domain. When isolated from WEHI-7 cells, this protein contains between two and three phosphoryl groups per steroid-binding site (Mendel WEHI-7 cells, this protein contains between two and three phosphoryl groups per steroid-binding site (Mendel et al., 1987). To identify the domains that contain these phosphorylated sites, we have analyzed the phosphate content of selected proteolytic fragments of the approximately 100-kDa steroid-binding protein from nonactivated and activated receptors. The approximately 100-kDa steroid-binding protein from WEHI-7 cells grown in the presence of [32P]orthophosphate was covalently labeled with [3H]dexamethasone 21-mesylate, purified with the BuGR2 monoclonal antibody, digested with chymotrypsin or trypsin, and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Chymotrypsin digestion of this protein yields a approximately 45-kDa fragment containing both the steroid-binding and DNA-binding domains, which contained both 32P and 3H. Trypsin digestion of the protein yields a approximately 29-kDa fragment encompassing the steroid-binding domain but not the DNA-binding domain of the approximately 100-kDa protein, which also contained both 32P and 3H. The 32P/3H ratio of each fragment provides a measure of phosphate content per steroid-binding site and indicated that each fragment has approximately 30% of the phosphate content of the intact protein. This is sufficient to account for one of the three receptor phosphoryl groups.(ABSTRACT TRUNCATED AT 250 WORDS)     

Monoclonal antibody to the rat glucocorticoid receptor. Relationship between the immunoreactive and DNA-binding domain

The region of the glucocorticoid receptor that reacted with a monoclonal antibody (BUGR-1) was identified. In order to identify the immunoreactive region, the rat liver glucocorticoid receptor was subjected to limited proteolysis; immunoreactive fragments were identified by Western blotting. The monoclonal antibody reacted with both the undigested Mr approximately 97,000 receptor subunit and a Mr approximately 45,000 fragment containing the steroid-binding and DNA-binding domains. Digestion by trypsin also produced two steroid-binding fragments of Mr approximately 27,000 and 31,000 which did not react with the antibody and an immunoreactive Mr approximately 16,000 fragment. This Mr approximately 16,000 fragment was shown to bind to DNA-cellulose, indicating that it contained a DNA-binding domain of the receptor. The undigested receptor must have steroid associated with it to undergo activation to a DNA-binding form. However, the Mr approximately 16,000 immunoreactive fragment binds to DNA-cellulose even if it is obtained by digestion of the steroid-free holoreceptor which does not itself bind to DNA.     

The steroid-binding properties of recombinant glucocorticoid receptor: a putative role for heat shock protein hsp90

The steroid-binding domain of the human glucocorticoid receptor was expressed in Escherichia coli either as a fusion protein with protein A or under control of the T7 RNA polymerase promoter. The recombinant proteins were found to bind steroids with the normal specificity for a glucocorticoid receptor but with reduced affinity (Kd for triamcinolone acetonide approximately 70 nM). Glycerol gradient analysis of the E. coli lystate containing the recombinant protein indicated no interaction between the glucocorticoid receptor fragment and heat shock proteins. However, synthesis of the corresponding fragments of glucocorticoid receptor in vitro using rabbit reticulocyte lystate resulted in the formation of proteins that bound triamcinolone acetonide with high affinity (Kd 2nM). Glycerol gradient analysis of these proteins, with and without molybdate, indicated that the in vitro synthesised receptor fragments formed complexes with hsp90 as previously shown for the full-length rat glucocorticoid receptor. Radiosequence analysis of the recombinant steroid-binding domain expressed in E. coli and affinity labelled with dexamethasone mesylate identified binding of the steroid to Cys-638 predominantly. However, all cysteine residues within the steroid-binding domain were affinity labelled to a certain degree indicating that the recombinant protein has a structure similar to the native receptor but more open and accessible.     

Determinants for DNA-binding site recognition by the glucocorticoid receptor.

The glucocorticoid receptor binds with high specificity to glucocorticoid response elements, discriminating them from other closely related binding sites. Three amino acids in the recognition alpha-helix of the DNA-binding domain of the receptor are primarily responsible for this specific DNA binding activity. In this study we analyze in detail how these residues determine the specific DNA binding by studying a series of mutant glucocorticoid receptor DNA-binding domains containing all combinations of glucocorticoid and estrogen receptor-specific residues at these positions. Statistical analysis of the results enables us to create models describing the association between amino acids and base pairs. Several strategies appear to be used in accomplishing discrimination between the glucocorticoid and estrogen response elements. Single residues (i.e., Val-443 in the glucocorticoid receptor and Glu-439 in the estrogen receptor) appear to form both positive contacts with specific base pairs in the cognate binding site and negative contacts in the non-cognate site. In the glucocorticoid receptor Ser-440 is pleiotropically negative for all sites tested but the negative effect is stronger for the estrogen response element thus contributing to binding site discrimination. Furthermore, combinations of amino acids appear to act synergistically, most often causing a reduction in binding to non-cognate sites.     

Binding linkage in a telomere DNA-protein complex at the ends of Oxytricha nova chromosomes

Alpha and beta protein subunits of the telomere end binding protein from Oxytricha nova (OnTEBP) combine with telomere single strand DNA to form a protective cap at the ends of chromosomes. We tested how protein-protein interactions seen in the co-crystal structure relate to DNA binding through use of fusion proteins engineered as different combinations of domains and subunits derived from OnTEBP. Joining alpha and beta resulted in a protein that bound single strand telomere DNA with high affinity (K(D-DNA)=1.4 nM). Another fusion protein, constructed without the C-terminal protein-protein interaction domain of alpha, bound DNA with 200-fold diminished affinity (K(D-DNA)=290 nM) even though the DNA-binding domains of alpha and beta were joined through a peptide linker. Adding back the alpha C-terminal domain as a separate protein restored high-affinity DNA binding. The binding behaviors of these fusion proteins and the native protein subunits are consistent with cooperative linkage between protein-association and DNA-binding equilibria. Linking DNA-protein stability to protein-protein contacts at a remote site may provide a trigger point for DNA-protein disassembly during telomere replication when the single strand telomere DNA must exchange between a very stable OnTEBP complex and telomerase.     

Urea-induced transformation of native estrogen receptor and evidence for separate DNA-and ATP-binding sites

We have investigated the involvement of hydrophobic receptor domains during transformation of the native estrogen receptor to a form(s) with high affinity for immobilized DNA and ATP. In the presence of 6 M urea the intact estrogen-receptor complex was completely (greater than 90%, n = 12) transformed into a DNA-binding configuration but only partially (35-45%, n = 8) transformed into an ATP-binding state. Similar experiments performed with unliganded receptor preparations further distinguished the receptor's DNA and ATP binding properties. While the urea-induced increase in receptor affinity for DNA-agarose was estrogen-dependent, the urea-induced increase in affinity for ATP-agarose was steroid-independent. This is the first direct evidence that hydrophobic receptor domains may be involved in the steroid-dependent exposure of the DNA binding site. This event is partially reversible and suggests that electrostatic interactions alone may not be sufficient to accurately describe receptor recognition of specific DNA acceptor sites.     

Expression of active hormone and DNA-binding domains of the chicken progesterone receptor in E. coli.

Bacterially-expressed fusion proteins containing the DNA-(region C) or hormone-binding (region E) domains of the chicken progesterone receptor (cPR) fused to the C terminus of Escherichia coli beta-galactosidase were analysed for the specificity of interaction with natural and synthetic hormone-responsive elements (HREs) and progestins, respectively. The purified fusion protein containing the progestin-binding domain bound progesterone with an apparent Kd of 1.0-1.5 nM and was specifically photocross-linked with the synthetic progestin R5020 in crude bacterial lysates. Labelling of intact bacterial cells with [3H]R5020 revealed that the majority, if not all, of the bacterially produced hormone-binding domain was active. No differences in the binding to a synthetic palindromic glucocorticoid/progestin-responsive element (GRE/PRE) were found when the bacterially produced cPR DNA-binding domain was compared in methylation interference assays with the full-length chicken progesterone receptor form A expressed in eukaryotic cells. The study of dissociation kinetics, however, revealed differences in the half-life of the complexes formed between the palindromic GRE/PRE and either the receptor form A or the fusion protein containing the cPR DNA-binding domain. DNase I protection experiments demonstrated that the bacterially produced region C of the cPR generated specific 'footprints' on the mouse mammary tumour virus long terminal repeat (MMTV-LTR) which were nearly identical to those previously reported for the rat glucocorticoid receptor.