Functional analysis of R651 mutations in the putative helix 6 of rat glucocorticoid receptors

https://doi.org/10.1016/S0303-7207(99)00171-9Get rights and content

Abstract

Trypsin digestion of steroid-free, but not steroid-bound, rat glucocorticoid receptor (GR) has recently been reported to occur at arginine-651 (R651). This residue is close to the affinity labeled Cys-656 and thus could be a sensitive probe of steroid binding. This hypothesis is supported by the current model of the GR ligand binding domain (LBD), which is based on the X-ray structures of several related receptor LBDs and places R651 in the middle of the putative α-helix 6 (649-EQRMS-653 of rat GR), close to the bound steroid. To test this model, R651, which could be involved in hydrophilic and/or hydrogen bonding, was mutated to alanine (A), which favors α-helices, the helix breakers proline (P) and glycine (G), or tryptophan (W). All receptors were expressed at about the same level, as determined by Western blots, but the cell-free binding activity of R651P was reduced twofold. The cell-free binding affinities were all within a factor of 10 of wild type receptors. Whole cell biological activity with transiently transfected receptors was determined with a variety of GR agonists (dexamethasone and deacylcortivazol) or antagonists (dexamethasone mesylate, RU486, and progesterone). Reporters containing both simple (GRE) and complex (MMTV) enhancers were used to test for alterations in GR interactions with enhancer/promoter complexes. Surprisingly, no correlation was observed between biological activity and ability to preserve α-helical structures for each point mutation. Finally, similar trypsin digestion patterns indicated no major differences in the tertiary structure of the mutant receptors. Collectively, these results argue that the polar/ionizable residue R651 is not required for GR activity and is not part of an α-helix in the steroid-free or bound GR. The effect of these mutations on GR structure and activity may result from a cascade of initially smaller perturbations. These LBD alterations were the most varied for interactions with deacylcortivazol and RU 486, which have recently been predicted to be sub-optimal binders due to their large size. However, further analyses of ligand size versus affinity suggest that there is no narrowly defined optimal size for ligand binding, although larger ligands may be more sensitive to modifications of LBD structure. Finally, the changes in GR activity with the various mutations seem to result from altered LBD interactions with common, as opposed to enhancer specific, transcription factors.

Introduction

The ligand binding domain (LBD) of steroid receptors effectively translates the information of the cognate ligand into the observed biological effect. Thus, steroid binding to receptors is a pivotal step in steroid receptor regulation of gene transcription. Considerable advances have recently been made in elucidating the steps following steroid binding to receptors, with a major advance coming from the discovery of the involvement of comodulators (CBP), coactivators (TIF2/GRIP-1, SRC-1, and AIB1/ACTR/RAC3/pCIP), and corepressors (SMRT and NCoR) (reviewed in Horwitz et al., 1996). These cofactors can interact with receptor LBDs in ways that are altered by agonist versus antagonist ligands. Furthermore, individual receptors can display preferential binding of specific coactivators (Eng et al., 1998, Feng et al., 1998, Kalkhoven et al., 1998, McInerney et al., 1998, Westin et al., 1998) and corepressors (Wong and Privalsky, 1998), probably with different affinities (Szapary et al., 1999). However, the receptor LBD is usually not sufficient to specify the transcriptional activity of a given steroid-receptor complex. For example, the weak activity of N-terminal truncated glucocorticoid receptors (GRs) containing amino acids 407–795 (Danielsen et al., 1987, Hollenberg et al., 1987, Szapary et al., 1996) argues that ligand-induced effects in LBD must somehow be transmitted to include interactions with the N-terminal A/B domain. Therefore, studies of ligand-induced changes in the LBD will be required for understanding this transmission phenomenon at a molecular level.

The amount of information that can be abstracted from the receptor-bound steroid would be expected to be directly proportional to the surface area of the steroid that is contacted by the receptor. In fact, early calculations on the free energy of binding suggested that the ligand was completely enveloped by the receptor protein (Wolff et al., 1978). This prediction has recently been confirmed by the X-ray crystallographic structures of the LBDs of several steroid receptors, which depict an α-helical sandwich structure with the steroid buried in the middle of the LBD (Bourguet et al., 1995, Renaud et al., 1995, Wagner et al., 1995, Wurtz et al., 1996, Brzozowski et al., 1997, Klaholz et al., 1998, Nolte et al., 1998, Shiau et al., 1998, Tanenbaum et al., 1998, Williams and Sigler, 1998). Recently, it has been proposed that ligand volume, as opposed to molecular weight, is a highly conserved property of steroid/nuclear receptors that may be useful in the design of future, high affinity analogs (Bogan et al., 1998). Nevertheless, it is still not possible to predict the final receptor protein structure that is formed after binding different ligands (Brzozowski et al., 1997, Klaholz et al., 1998, Shiau et al., 1998). Therefore, determinations of the properties of assorted receptor point mutants remain a valuable method for investigating how steroid structure influences receptor structure and biology.

It is well established that ligand binding effects a conformational change in the receptor LBD. A particularly easy method of documenting these changes is to follow the trypsin digestion patterns of receptors±ligand (Simons et al., 1989, Allan et al., 1992, Modarress et al., 1997). Most ligand-free receptors are completely degraded by trypsin but yield several resistant,≈30 kDa fragments if first bound by ligand (Simons et al., 1989, Allan et al., 1992, Beekman et al., 1993, Leng et al., 1993, Modarress et al., 1997, Simons, 1998). The differences in these digestion products, at least for GR, result from cleavage at either ends of the LBD and do not correlate with the agonist or antagonist activity of the steroid (Modarress et al., 1997). Consequently, the cause and nature of these conformational changes is not yet clear. The steroid-free GR appears to be unique in affording a 16 kDa fragment after trypsin digestion. This 16 kDa fragment, which can be covalently labeled by Dex-Mes and retains significant binding affinity and specificity (Simons et al., 1989, Chakraborti et al., 1992), has recently been identified as amino acids 652–795 of the rat GR. In solution, however, this 16 kDa species exists as a non-covalent complex with the other half of the LBD (amino acids 518–651) (Xu et al., 1999).

The fact that steroid-bound GR no longer permits trypsin cutting at R651 of rat GR, as seen by the absence of the 16 kDa fragment (Simons et al., 1989, Modarress et al., 1997), argues that steroid binding has a major effect on the accessibility of this amino acid. One possibility, given the general lack of cleavage of amino acids present in an α-helix (Fontana et al., 1986), is that the short sequence containing R651, and predicted to be present as an α-helix in steroid-bound GR (Wurtz et al., 1996), does not attain the α-helical structure until after steroid binds to the receptor. In this case, prevention of the formation of helix 6 might have significant repercussions for the biological activity of GR. Other reasons for suspecting that R651 might be important in GR action are that R651 of rGR is close to C656, which is affinity labeled by Dex-Mes (Simons et al., 1987), and that R651 is relatively near the steroid in models of steroid-bound GR (Wurtz et al., 1996, Williams and Sigler, 1998).

The purpose of this study was to investigate the hypothesis that R651 plays a significant role in, and is part of a sequence involved in a steroid-induced conformational change that influences, steroid-regulated gene expression by GR. Further interest derived from the fact that little information exists on structure versus function for this region of GR ( Simons, 1994, Martinez et al., 1998). The fact that R651 is in the middle of a predicted short α-helix (Wurtz et al., 1996, Williams and Sigler, 1998) makes it ideal for mutational analyses. We therefore examined the consequences of several mutations on ligand binding and the biological activity of agonist and antagonist steroids with two different enhancer/promoter constructs. Trypsin digestion was performed to probe the structural consequences of the mutations. The results argue that mutations of R651 influence GR activity in ways that are inconsistent with the presence of helix 6 in either steroid-free or steroid-bound GR. The significance of these results with two different enhancers is discussed. The results with steroids of very different sizes were consistent with a computer analysis suggesting that bulky ligands can be useful, active hormones.

Section snippets

Materials and methods

Unless otherwise indicated, all operations were performed at 0°C.

Properties of mutant receptors

Four different mutations were selected for the basic amino acid, arginine, at position 651 of the rat GR: alanine, glycine, proline, and tryptophan. All mutations were to neutral amino acids in order to assess the importance of the lone basic amino acid in the putative helix 6. Alanine is considered to constitute a neutral replacement (Cunningham, 1989) while tryptophan is the most hydrophobic. Proline, and to a lesser extent glycine, are disrupters of α-helices (Chakrabartty et al., 1991,

Discussion

This study examined the existence and importance of helix 6 in GR functions. The extensive homology between the LBDs of GR and progesterone receptors (PRs) led to the suggestion that both proteins may also have highly similar tertiary structures (Wurtz et al., 1996, Williams and Sigler, 1998). In this case, it would be predicted that R651 would lie in the middle of a small α-helix, helix 6, and near both C656, which is affinity labeled by Dex-Mes (Simons et al., 1987), and the bound ligand (

Acknowledgements

We thank Bernd Groner (Frankfurt am Main, Germany), Gordon Hager (NCI/NIH), Keith Yamamoto (University of California, San Francisco, CA) for their generous gifts of research materials, Keith Yamamoto for sharing unpublished data, Paul Sigler (Yale University) for discussions about the PR X-ray structure, David Wheeler (NLM/NIH) for assistance with RasMac, Susan Chacko (NIH) for invaluable help with retrieving structures from the Cambridge Structural Database and with volume calculations by

References (77)

  • H.-M Jantzen et al.

    Cooperativity of glucocorticoid response elements located far upstream of the tyrosine aminotransferase gene

    Cell

    (1987)
  • E Langley et al.

    Intermolecular NH2-/carboxyl-terminal interactions in androgen receptor dimerization revealed by mutations that cause androgen insensitivity

    J. Biol. Chem.

    (1998)
  • X Leng et al.

    Ligand-dependent conformational changes in thyroid hormone and retinoic acid receptors are potentially enhanced by heterodimerization with retinoic X receptor

    J. Steroid Biochem. Molec. Biol.

    (1993)
  • K.J Modarress et al.

    Steroid-induced conformational changes at ends of the hormone binding domain in the rat glucocorticoid receptor are independent of agonist versus antagonist activity

    J. Biol. Chem.

    (1997)
  • A Munck et al.

    Glucocorticoid-receptor complexes and the earliest steps in the action of glucocorticoids on thymus cells

    J. Steroid Biochem.

    (1972)
  • I Nilsson et al.

    Breaking the camel’s back: proline-induced turns in a model transmembrane helix

    J. Mol. Biol.

    (1998)
  • H Oshima et al.

    Sequence-selective interactions of transcription factor elements with tandem glucocorticoid-responsive elements at physiological steroid concentrations

    J. Biol. Chem.

    (1993)
  • N.J Sarlis et al.

    Quantity of partial agonist activity for antiglucocorticoids complexed with mutant glucocorticoid receptors is constant in two different transactivation assays but not predictable from steroid structure

    J. Steroid Biochem. Mol. Biol.

    (1999)
  • A.K Shiau et al.

    The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen

    Cell

    (1998)
  • D Szapary et al.

    Induction properties of a transiently transfected glucocorticoid-responsive gene vary with glucocorticoid receptor concentration

    J. Biol. Chem.

    (1996)
  • H.H Thole et al.

    The ligand-binding site of the estradiol receptor resides in a non-covalent complex of two consecutive peptides of 17 and 7 kDa

    Biochem. Biophys. Res. Commun.

    (1994)
  • M Truss et al.

    Interplay of steroid hormone receptors and transcription factors on the mouse mammary tumor virus promoter

    J. Steroid Biochem. Molec. Biol.

    (1992)
  • M Xu et al.

    Modular structure of glucocorticoid receptor domains is not equivalent to functional independence. Stability and activity of the steroid binding domain are controlled by sequences in separate domains

    J. Biol. Chem.

    (1996)
  • T.K Archer et al.

    Transcription factor loading on the MMTV promoter: a bimodal mechanism for promoter activation

    Science

    (1992)
  • J.M Beekman et al.

    Transcriptional activation by the estrogen receptor requires a conformational change in the ligand binding domain

    Mol. Endo.

    (1993)
  • P.A Bell et al.

    Steroid-binding properties and stabilization of cytoplasmic glucocorticoid receptors from rat thymus cells

    Biochem. J.

    (1973)
  • H Berglund et al.

    Structure and dynamics of the glucocorticoid receptor DNA-binding domain: comparison of wild type and a mutant with altered specificity

    Biochemistry

    (1997)
  • Berrevoets, C.A., Doesburg, P., Steketee, K., Trapman, J., Brinkmann, A.O., 1998. Functional interactions of the AF-2...
  • E Bignon et al.

    Effect of triphenylacrylonitrile derivatives on estradiol-receptor binding and on human breast cancer cell growth

    J. Med. Chem.

    (1989)
  • A.A Bogan et al.

    Natural ligands of nuclear receptors have conserved volumes

    Nat. Struct. Biol.

    (1998)
  • W Bourguet et al.

    Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha

    Nature

    (1995)
  • E.H Bresnick et al.

    Glucocorticoid receptor-dependent disruption of a specific nucleosome on the mouse mammary tumor virus promoter is prevented by sodium butyrate

    Proc. Natl. Acad. Sci. USA

    (1990)
  • A.M Brzozowski et al.

    Molecular basis of agonism and antagonism in the oestrogen receptor

    Nature

    (1997)
  • A Chakrabartty et al.

    Large differences in the helix propensities of alanine and glycine

    Nature

    (1991)
  • B.C.a.W.J.A Cunningham

    High-resolution epitope mapping of hGH-receptor interactions by alanine-scanning mutagenesis

    Science

    (1989)
  • M Danielsen et al.

    Domains of the glucocorticoid receptor involved in specific and nonspecific deoxyribonucleic acid binding, hormone activation, and transcriptional enhancement

    Mol. Endo.

    (1987)
  • W Feng et al.

    Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors

    Science

    (1998)
  • A Fontana et al.

    Correlation between sites of limited proteolysis and segmental mobility in thermolysin

    Biochemistry

    (1986)
  • Cited by (5)

    The X-ray coordinates, and the thermal parameters, for the human PR LBD were obtained from the Protein Data Bank (ID No. 1A52).

    1

    Present address: Genetic therapy, Novartis, 938 Clopper Road, Gaithersburg, MD, USA

    View full text