| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Prince Henry’s Institute of Medical Research, Clayton, Victoria 3168 Australia
Address all correspondence and requests for reprints to: Dr. Peter J. Fuller, Prince Henry’s Institute of Medical Research, P.O. Box 5152, Clayton, Victoria 3168, Australia. E-mail: peter.fuller{at}med.monash.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although many aspects of GR function are now well characterized (4, 5, 6) the nature of the interaction between the ligand and the receptor remains to be precisely defined. Structure-function analysis of the LBD may be achieved using either artificial or natural mutations. The latter are manifest in vivo as resistance to glucocorticoids; such resistance is inevitably partial, as complete receptor inactivation results in early fetal death (1). Resistance is seen in several species, including New World primates (5, 6, 7) and the guinea pig (8), although the former reflects a generalized, rather than GR-specific, steroid resistance. We have previously reported the cloning of the guinea pig GR (9) and have demonstrated that the resistance observed in vivo (8) is conferred by the LBD in vitro (9, 10). In our original description of the guinea pig GR (9), the expression studies also revealed constitutive activity when the LBD of the human GR expression vector pRShGRNX was replaced by the guinea pig GR LBD; constitutive activity is not a feature of unliganded intact steroid receptors. The transfected receptor also responded, albeit weakly, to both dexamethasone and cortisol, and the response was blocked by the GR antagonist RU486 (10). Subsequent to publication we identified that an adenosine occurring in a run of seven adenosines at position 1458 was deleted by Taq polymerase during PCR generation of the guinea pig LBD cassette [see erratum (9)]. The new reading frame immediately encodes a stop codon; the constitutive activity reported (9) is thus the result of this truncation mutation (9, 10). Ligand-independent constitutive activity of the GR and other steroid receptors associated with truncation of the LBD has been noted previously. The apparent response to both agonist and antagonist steroids in this system remained a curious paradox (9, 10). To exclude the possibility that functional GR was being produced by "read-through," the truncated mutation was recreated with the sequence downstream of the stop codon encoding the LBD deleted from the construct. This construct also shows constitutive activity and a steroid response; when the expression vector was omitted from the transfection, a weak response to dexamethasone was still observed at high concentrations of steroid (data not shown). This response appears to be the basis of the steroid response observed with the expression vectors and is a reminder of the need for caution in the use of such cotransfection systems. Analysis with the correct guinea pig LBD revealed half-maximal induction at approximately 3 nM dexamethasone, c.f. about 0.4 nM for the human GR (9). The LBD of the guinea pig GR differs at 24 amino acids from the sequence of the human LBD. In the current studies we sought to define the residue(s) conferring the observed difference in sensitivity between these two receptors with the intention of better understanding the structural determinants of the steroid-LBD interaction.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chimeric and mutant receptors
The LBD of the guinea pig GR was substituted for the LBD in the
human GR expression vector pRShGRnx (11) as described previously (9).
As the guinea pig sequence does not include any 3'-untranslated region,
an equivalent human GR LBD insert was prepared using the primers GR-9
and GR-10 as previously described (9). Chimeric LBD were created using
a PstI restriction site at position 1768 in the guinea pig
GR sequence (9) and the human GR sequence (12). Overlap extension PCR
(13) with Pfu DNA polymerase (Stratagene, La Jolla, CA) was used to
introduce reciprocal single or double nucleotide changes into both
expression vectors. The entire LBD sequence of the chimeric and mutant
constructs was verified by DNA sequence analysis (14).
Trans-activation assays
CV-1 cells were maintained in DMEM supplemented with 1 x
nonessential amino acids (Cytosystems, Castle Hill, Australia), 1
mM L-glutamine (Cytosystems), 1%
penicillin/streptomycin/fungizone (CSL, Victoria, Melbourne,
Australia), and 5% FBS (Cytosystems) and plated at a density of 3
x 105 cells/35-mm plate the day before transfection. The
following day cells were transfected by the calcium phosphate-DNA
coprecipitation method (15) with 3 µg each of the appropriate GR
expression vector, the reporter gene pMMTV-CAT (MMTV, mouse mammary
tumor viurs; CAT, chloramphenicol acetyltransferase), and the internal
reference pSV-ßgal. All transfections were performed in duplicate
within an experiment, and all conditions were repeated in multiple
experiments. Twenty-four hours after transfection, the medium was
replaced with fresh medium containing dexamethasone or the other
steroids at the concentrations indicated. After an additional 24
h, cells were harvested, cell extracts were prepared by three cycles of
freeze/thawing, and 60 µl extract were assayed for CAT activity by
mixed phase separation (16). Expression of CAT activity was normalized
for expression of the constitutive internal control, ß-galactosidase,
which is directed by the simian virus 40 early promoter, in the same
extract. ß-Galactosidase activity was determined in 10 µl extract
by a colorimetric assay in accordance with the manufacturer’s
specifications (Promega Biotech, Madison, WI). Cotransfection assay
results are the average obtained from at least two independent
experiments performed in duplicate, with the results expressed as a
percentage of maximal normalized CAT activity.
Binding studies
COS-7 cells were maintained as described for CV-1 cells. They
were transiently transfected with the expression vectors at
concentrations of 1 µg/35-mm plate. Transfections were performed
using the diethylaminoethyl-dextran method (15). The cells were
harvested 24 h later, incubated with
[3H]dexamethasone (Amersham, Aylesbury, UK; SA, 83
Ci/mmol) in DMEM for 1 h at 37 C with the following concentrations
of dexamethasone in triplicate: 2.5 x 10-8
M, 1.25 x 10-8 M, 6.25
x 10-9 M, 3.125 x 10-9
M, and 1.5625 x 10-9 M. The
cells were then washed three times in ice-cold PBS, and activity was
measured using a liquid scintillation assay. Nonspecific binding was
measured in parallel by the concurrent addition of cold dexamethasone
in a 500-fold excess, and specific binding was calculated by
subtracting nonspecific counts from total counts. The binding affinity
was calculated by Scatchard analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) to examine trans-activation
of both the human and guinea pig receptors. In no case was there any
evidence of an altered selectivity with the steroids examined. Both
aldosterone and corticosterone produced much lower responses in the
guinea pig GR at a steroid concentration of 100 nM. The
guinea pig GR responds to higher concentrations of steroids, as shown
for corticosterone in Fig. 1, yielding a pattern of response similar to
that for dexamethasone (9, 10) in that the responses of guinea pig GR
appear to parallel those of human GR, albeit with a decrease in
sensitivity (right shift), although the maximum responses are
equivalent in the trans-activation assay (data not
shown).
|
). The human GR has a Kd for
dexamethasone of about 1.6 nM, whereas for the guinea pig
GR, the Kd is approximately 6 nM. This finding
is consistent with previous ex vivo binding studies using
whole tissues (8). The result with the human GR is in good agreement
with those derived previously under similar conditions for the human GR
(17).
|
) that the upstream region of the LBD
confers the resistance phenotype seen in the intact guinea pig LBD.
Ligand binding studies using these chimeras further confirm that
differences in affinity are also conferred by the upstream region of
the LBD (Fig. 4).
|
|
). All five occur in a region of high
sequence conservation. This conservation is less marked when the
Xenopus and trout sequences are included, although certain
sequences are still relatively invariant. When the sequence of the
equivalent region of the other steroid receptors is compared with the
GR (18), the conservation is less marked; however, the alanine at
human GR position 546 is conserved across the closely related
androgen, progesterone, glucocorticoid, and mineralocorticoid
receptors.
|
).
The alanine for serine substitution had a small effect on the
sensitivity of both receptors (Fig. 6A). When the adjacent tyrosine for
histidine substitutions were examined with the alanine-serine change as
double mutants, the guinea pig LBD was little altered, whereas the
sensitivity of the human LBD decreased significantly when both a
histidine and a serine were introduced (Fig. 6B). On the basis that the
critical change may thus be that of the tyrosine-histidine alone, this
single substitution was introduced, yielding a marked decrease in the
sensitivities of both LBD (Fig. 6C). This suggested that the altered
sensitivity was likely to be the result of several interrelated
substitutions. The roles of the other pairs of substitutions were
similarly examined, and again, although the sensitivity of the guinea
pig LBD to dexamethasone was little altered, introduction of the
tyrosine-serine pair into the human LBD again caused a marked decrease
in sensitivity (Fig. 6D). Finally, both paired mutations were combined,
leaving only the leucine-isoleucine substitution unaltered (Fig. 6E); again, the sensitivities of both LBD were decreased. The details
of the mutations and the resulting sensitivity (DE50) are
tabulated in Fig. 6.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To define which of the 24 amino acid differences between the guinea pig and human GR LBD sequences might confer these differences, a chimeric receptor approach was adopted to isolate the critical region/residues in the LBD. The initial chimera identified the first third of the LBD as being critical. This region contains 6 changes, of which 1, a lysine for arginine substitution, is clearly conservative. The other 5 differences cluster in a region that is highly conserved across all mammalian GR species sequenced (18). Of the 5 residues, the isoleucine appears to be a relatively conservative substitution, the histidine and serine are unique to the guinea pig (at least when compared with other mammalian receptors), the threonine substitution is also seen in New World monkeys, and the last serine is unique, although a leucine occurs at this position in the tupaia sequence (25). If a teleost fish GR (26) and the Xenopus GR (27) sequence are included in the comparison, then several differences emerge at these positions, although overall, there is still significant sequence conservation. When the comparison is made between receptor types, then the conservation of sequence is less marked. The alanine that is substituted for a serine in the guinea pig was the most strictly conserved of the residues, so our mutagenesis started with this residue alone or in combination with the tyrosine/histidine substitution. Neither substitution completely reciprocally reversed the phenotype, although they did alter the sensitivity. The serine/alanine substitutions increased the sensitivity of the guinea pig LBD, as did the combined substitution, although only to an intermediate phenotype. The same is true of the combined substitution in the human GR, which does not achieve the full resistance seen in the wild-type guinea pig GR. To fully recapitulate the phenotypes, a series of other changes were made. No single substitution or simple combination of substitutions was able to switch the ligand sensitivity as seen with the original PstI chimera. Although it became unrealistic to test every combination and permutation of the five residues, it was clear that at least four and probably all five residues contribute to defining the sensitivity. It is also of interest that for several chimeras, the sensitivity became much less than that of the guinea pig wild type, arguing for disruption of a critical interaction(s) within the cluster of amino acids. In several cases there was also a change in the shape of the curve, rather than a simple leftward or rightward shift, which also argues for some disruption of intramolecular interactions.
These findings identify this region of the GR as having a role in defining sensitivity to glucocorticoids. Mutations of this region have not been reported in glucocorticoid resistance syndromes, although in 1 kindred a missense mutation, leading to a substitution at position 559 in the human GR, lies close to the region of interest (19). Milhon et al. (28) characterized a series of alanine or glycine substitutions in the 9 residues up to and including the valine at human GR position 543. They concluded that the side-chains of several amino acids in this region are specifically involved in the function of normal binding (28). The interpretation of these and other such studies depends on being able to define the tertiary structure of the ligand-binding pocket. Xu et al. (29) have argued that 3 independent regions are required for generating the ligand-binding form of the receptor: 1) sequences upstream of the binding domain, which convey stability; 2) sequences downstream; which are required for correct folding of the steroid-binding domains; and 3) a core of 124 amino acids starting at position 550 in the rat GR (position 532 in the human GR), which after correct folding of the LBD may be sufficient for steroid binding. The region of interest in the guinea pig GR thus lies upstream of the core region and in this model is unlikely to be in direct contact with the ligand.
A range of ligands was examined to establish whether there was any change in the specificity of binding. Altered specificity has been observed due to point mutations of various receptors, including the estrogen receptor in MCF-7 breast cancer cells (30), the androgen receptor in LNCaP cells (31), and the progesterone receptor for RU486 binding (32). Of particular interest was the relative binding of cortisol and corticosterone, given that the physiological steroid for rats and mice is corticosterone, whereas in the guinea pig, which is arguably a rodent (24), and humans, cortisol is the physiological circulating glucocorticoid. Residues reported to be involved in conferring specificity of binding generally lie in the C-terminal half of the LBD, which is consistent with our observations. These findings further support the concept (29) that this region lies outside of the ligand-binding pocket.
Unlike the DBD of the GR, the LBD has eluded crystallographic or NMR
definition of its tertiary structure. The crystal structures of the
unliganded retinoid X, the liganded retinoic acid-
, and liganded
thyroid 
-receptors have recently been reported (33). Wurtz et
al. (33) extrapolated from these structures to produce a model of
the GR structure. The basic structure of the LBD predicted from the
crystal structures is of a single structural domain packed in 3 layers
composed of 12 -helixes (H1–12) and a mixed ß-sheet (33).
Comparison of the apo-retinoid X receptor and holo-retinoic acid
receptor structures suggests that upon ligand-binding helix 12 swings
up, "trapping" the ligand in the LBD. The model of the GR
constructed by Wurtz et al. (33) predicts that the residues
of interest in the guinea pig GR lie close to H1 in the region H2,
which varies between receptors. Adjacent residues in H1 make key
contacts with the LBD core (33). Thus, these residues lie outside the
core, but conformational changes or altered interactions may affect
ligand binding through a distant packing effect (Moras, D., personal
communication).
The LBD has other functions, including dimerization, hsp90 binding, and trans-activation. The difference between the two species of GR fundamentally involves ligand binding, and therefore, despite the residues lying in a putative activating region of GR designated tau2 (2), this seems unlikely to be relevant. Similarly, dimerization will affect trans-activation, but not binding. Although the regions of the GR responsible for hsp90 binding have not been defined, most studies suggest a more central location for this function in the LBD (34). It does, however, remain possible, given the pivotal role of hsp90 binding in maintaining the GR in a transcriptionally inactive, but high affinity, state (34) that a perturbation of this interaction could alter binding affinity.
In conclusion, analysis of the basis of the decreased sensitivity of the guinea pig GR to glucocorticoids reveals a cluster of amino acids within the N-terminal portion of the LBD, which, despite not being directly involved in ligand binding, contributes to determining ligand affinity. In addition, these studies suggest that critical intramolecular interactions occur within this region as well as with other regions in the LBD. It should not come as a surprise that the changes in the guinea pig GR are complex and involve a series of interrelated substitutions, as these changes have presumably evolved over time in the relative isolation of the New World in response to some yet to be defined evolutionary pressure (24). Formal definition of the interactions within this region must await the description of a crystal structure for the GR LBD.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Present address: Department of Cell Biology, Baylor College of
Medicine, Houston, Texas 77030.
Received August 27, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. H Stolte, B M L. V. van Kemenade, H. F J Savelkoul, and G. Flik Evolution of glucocorticoid receptors with different glucocorticoid sensitivity. J. Endocrinol., July 1, 2006; 190(1): 17 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Arampatzis, B. Kadereit, D. Schuster, Z. Balazs, R. A S Schweizer, F. J Frey, T. Langer, and A. Odermatt Comparative enzymology of 11{beta}-hydroxysteroid dehydrogenase type 1 from six species J. Mol. Endocrinol., August 1, 2005; 35(1): 89 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. B. Denny, V. Prapapanich, D. F. Smith, and J. G. Scammell Structure-Function Analysis of Squirrel Monkey FK506-Binding Protein 51, a Potent Inhibitor of Glucocorticoid Receptor Activity Endocrinology, July 1, 2005; 146(7): 3194 - 3201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Necela and J. A. Cidlowski A Single Amino Acid Change in the First Zinc Finger of the DNA Binding Domain of the Glucocorticoid Receptor Regulates Differential Promoter Selectivity J. Biol. Chem., September 17, 2004; 279(38): 39279 - 39288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chu, Y. Nishi, T. Yanase, H. Nawata, and P. J. Fuller Transrepression of Estrogen Receptor {beta} Signaling by Nuclear Factor-{kappa}B in Ovarian Granulosa Cells Mol. Endocrinol., August 1, 2004; 18(8): 1919 - 1928. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Owen, E. Setiawan, A. Li, L. McCabe, and S. G. Matthews Regulation of N-Methyl-D-Aspartate Receptor Subunit Expression in the Fetal Guinea Pig Brain Biol Reprod, August 1, 2004; 71(2): 676 - 683. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Banjanin, A. Kapoor, and S. G. Matthews Prenatal glucocorticoid exposure alters hypothalamic-pituitary-adrenal function and blood pressure in mature male guinea pigs J. Physiol., July 1, 2004; 558(1): 305 - 318. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Andrews, A. Kostaki, E. Setiawan, L. McCabe, D. Owen, S. Banjanin, and S. G. Matthews Developmental regulation of the 5-HT7 serotonin receptor and transcription factor NGFI-A in the fetal guinea-pig limbic system: influence of GCs J. Physiol., March 15, 2004; 555(3): 659 - 670. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Shafqat, B. Elleby, S. Svensson, J. Shafqat, H. Jornvall, L. Abrahmsen, and U. Oppermann Comparative Enzymology of 11beta -Hydroxysteroid Dehydrogenase Type 1 from Glucocorticoid Resistant (Guinea Pig) Versus Sensitive (Human) Species J. Biol. Chem., January 10, 2003; 278(3): 2030 - 2035. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Yudt and J. A. Cidlowski The Glucocorticoid Receptor: Coding a Diversity of Proteins and Responses through a Single Gene Mol. Endocrinol., August 1, 2002; 16(8): 1719 - 1726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Liu, A. Li, and S. G. Matthews Maternal glucocorticoid treatment programs HPA regulation in adult offspring: sex-specific effects Am J Physiol Endocrinol Metab, May 1, 2001; 280(5): E729 - E739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Rogerson, N. Dimopoulos, P. Sluka, S. Chu, A. J. Curtis, and P. J. Fuller Structural Determinants of Aldosterone Binding Selectivity in the Mineralocorticoid Receptor J. Biol. Chem., December 17, 1999; 274(51): 36305 - 36311. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) is about 1.6 nM, and that for the guinea pig
GR (
) it is approximately 6 nM.
)
and guinea pig-human (•) chimeric receptors and Scatchard analysis
yielded Kd values of about 2 and about 8.1 nM,
respectively.
