HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rogerson, F. M. Articles by Fuller, P. J. Search for Related Content PubMed PubMed Citation Articles by Rogerson, F. M. Articles by Fuller, P. J. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 51, 36305-36311, December 17, 1999

Structural Determinants of Aldosterone Binding Selectivity in the Mineralocorticoid Receptor^*

Fraser M. Rogerson, Nektaria Dimopoulos, Pavel Sluka, Simon Chu, Andrea J. Curtis, and Peter J. Fuller

From Prince Henry's Institute of Medical Research, Clayton, Victoria, 3168, Australia

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The structural determinants of aldosterone binding specificity in the mineralocorticoid receptor (MR) have not been determined. The MR has greatest sequence identity with the better characterized glucocorticoid receptor (GR), which is reflected in their overlapping ligand binding specificities. There must be subtle sequence differences that can account for the MR-specific binding of aldosterone and the shared binding of cortisol. To characterize ligand binding specificity, chimeras were made between the human MR and GR ligand-binding domains (LBDs). Three points were chosen as break points to generate a total of 16 different constructs. These chimeric LBDs were placed in a human GR expression vector containing the GR DNA-binding and N-terminal domains and assayed by co-transfection into CV-1 cells with the mouse mammary tumor virus-luciferase reporter plasmid. Binding of [³H]aldosterone and [³H]dexamethasone was also measured. All of the constructs that are potently activated by aldosterone contain amino acids 804-874 of the MR. The results of the ligand binding experiments using [³H]aldosterone were consistent with the transactivation assay. Cortisol activation of the chimeras was surprisingly complex. Constructs that are activated by cortisol contain either amino acids 804-874 and 932-984 of the MR or amino acids 598-668 and 726-777 of the GR. However, all of the chimeras retained the ability to bind the synthetic glucocorticoid [³H]dexamethasone, and cortisol was able to displace [³H]dexamethasone binding, suggesting that the differential effects of cortisol on transcriptional activation are caused by an effect that occurs downstream of ligand binding. These results identify a subregion of the MR LBD that confers specificity of aldosterone binding, which contrasts with cortisol binding where differential effects between chimeras appear to be mediated by interactions distal to ligand binding.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The steroid hormone aldosterone is important for the control of blood pressure through the promotion of sodium reabsorption in the kidney and colon (1). Aldosterone has other roles besides sodium balance; it has direct central effects on blood pressure (2) and has been implicated in the pathogenesis of cardiac fibrosis (3). The importance of aldosterone in blood pressure control and hypertension is underlined by the fact that the three known monogenetic causes of hypertension, glucocorticoid remediable hyperaldosteronism (4), apparent mineralocorticoid excess (5), and Liddle's syndrome (pseudoaldosteronism) (6), all involve this steroid, its receptor, or sodium reabsorption. Despite the importance of aldosterone there is only limited knowledge of the molecular mechanisms that underlie its action.

Aldosterone binds to, and acts through, the mineralocorticoid receptor (MR).¹ This protein belongs to the steroid hormone receptor (SHR) family of ligand-dependent transcription factors. As with other members of the family the receptor structure can be divided into three major domains (7). The first of these is the N-terminal domain, which contains an "activation function" involved in transcriptional activation (8, 9). In the middle of the protein is the DNA-binding domain, which binds to specific DNA sequences on target genes. C-terminal to the DNA-binding domain is the ligand-binding domain (LBD), which has sequences involved in ligand binding (10), transcriptional activation (8), and heat-shock protein (HSP) binding (11, 12). The mineralocorticoid receptor is the least characterized of the SHRs.

The critical function of the LBD is the binding of ligand. In the absence of ligand the MR exists predominantly in the cytoplasm in a complex with a number of HSPs (12). The role of HSP binding is two-fold: to keep the receptor in an inactive state and to maintain the LBD in a structural conformation that promotes high affinity ligand binding (12). The binding of ligand alters the conformation of the receptor and displaces the HSPs, which exposes sequences involved in receptor dimerization, nuclear localization, and DNA binding. Thus ligand binding activates the receptor.

Of the steroid hormone receptors, the MR has the most sequence identity with the glucocorticoid receptor (GR) (7). This is reflected in the overlapping binding specificities of the two receptors. Both receptors bind cortisol and corticosterone with high affinity (13), whereas only the MR binds aldosterone with high affinity. In vivo binding specificity is maintained by a number of mechanisms, the most of important of which involves the type 2 isoform of the enzyme 11-hydroxysteroid dehydrogenase that converts cortisol/corticosterone to inactivate 11-keto metabolites in MR-containing cells (5). The synthetic GR agonist dexamethasone also binds to the MR (7), albeit with a lower affinity. This is reflected in functional data showing dexamethasone to be a less potent activator of the MR than of the GR (14). It is likely that there are sequences shared between the two receptors that allow both to bind cortisol with high affinity and sequences that differ between the two receptors that permit specific binding of aldosterone to the MR. However, the amino acids involved in aldosterone binding specificity remain to be determined. In this report we have used chimeras between the LBDs of the MR and GR as a way of determining sequences important in the specific binding of aldosterone to the MR. Our laboratory has previously used this strategy to investigate cortisol resistance in the guinea pig GR (15, 16). A similar strategy has also been used previously to determine sequences important to ligand specificity to the androgen receptor (17) and to identify determinants of RU486 binding in the progesterone receptor (18). By examining these chimeras using a transactivation assay and by direct ligand binding we have identified a region in the LBD that is responsible for the specificity of aldosterone binding to the MR.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Creation and Construction of Chimeras-- To create the chimeras between the MR and GR, the LBD's three break points were selected in regions with the highest degree of amino acid sequence identity between the two receptors, following the rationale that this would most likely maintain the basic structural integrity of the proteins. These break points are shown in Fig. 1. The restriction enzyme sites ScaI and NsiI were chosen as two of the break points. The GR gene contains both of these sites at positions corresponding to amino acids 598 (ScaI) and 726 (NsiI). In contrast the MR gene only contains the corresponding NsiI site at amino acid 932. The ScaI site, at the position corresponding to amino acid 804, had to be generated in the MR cDNA by PCR mutagenesis. A natural NsiI site at amino acid 841 in the MR had to be removed by PCR mutagenesis. In neither case was the amino acid sequence changed. There were no convenient restriction enzyme sites for the middle break point, corresponding to amino acid 678 of the GR and 874 of the MR, so these chimeras were generated using overlap extension PCR. In all 14 different MR/GR chimeras were created. The chimeras are identified by a four-letter code in which the identity of each particular section of the construct is indicated in order from N terminus to C terminus; "G" indicates the GR sequence and "M" indicates the MR sequence. The full complement of 16 constructs includes the intact MR and GR LBDs.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Schematic representation of the break points used to create the MR/GR LBD chimeras. The N-terminal break point corresponds to a ScaI site, and the C-terminal break point corresponds to an NsiI site. The middle break point was created by overlap extension PCR. All of the chimeras were placed into an expression vector (pRShGR) containing the DNA-binding and N-terminal domains of the GR. DBD, DNA-binding domain.

The GR LBD was derived from the plasmid pRShGR_BX (19). This construct was derived from pRShGR_NX (8) by truncating the 3' untranslated region. For generating the chimeras a XhoI/BamHI fragment, encompassing the entire LBD, was ligated into the pSP72 vector (Promega Corporation, Madison, WI). In all cases PCRs were performed using Pfu polymerase (Stratagene, La Jolla, CA). The primers used are as follows (restriction enzyme sites are underlined): MR1, 5' GTG GAG TCA TGG AAA TCA CAC GGC 3' (MR +1669-1692 forward); MR2, 5' GGC AAA GAT CTT CTG GGC AGC GGG CAG TCA 3' (MR +2952-2982 reverse, incorporating the BglII site); GR1, 5' GGA ATG AAC CTC GAG GCT CGA A 3' (GR +1453-1474 forward); GR2, 5' AGG GAT CCA TTC TTA TTA AGG CAG TCA 3' (GR +2332-2358 reverse); MR3, 5' CCC TAA TCC AGT ACT CTT GGA TGT GT 3' (MR +2399-2424 forward, incorporating the ScaI site); MR4, 5' ACA CAT CCA AGA GTA CTG GAT TAG 3' (MR +2401-2424 reverse, incorporating the ScaI site); MR5, 5' GAG AAG ATG CAC CAG TCT GCC ATG 3' (MR +2512-2535 forward, removing the NsiI site); MR6; 5' CAT GGC AGA CTG GTG CAT CTT CTC 3' (MR +2512-2535 reverse, removing the NsiI site); GGMM1, 5' TAT GAA AAC CTT GCT GCT ACT AAG CAC AAT TCC A 3' (GR +1995-2004/MR +2623-2646 forward); GGMM2, 5' GTA GCA GCA AGG TTT TCA TAC AGA GAT ACT CTT C 3' (GR +1981-2004/MR +2623-2632 reverse); MMGG1, 5' CAT GAA AGT TTT ACT GCT TCT CTC TTC AGT TCC T 3' (MR +2613-2622/GR +2005-2028 forward); MMGG2, 5' GAA GCA GTA AAA CTT TCA TGA TGG TGT ATT CTT C 3' (MR +2599-2622/GR +2005-2014 reverse).

MR LBD-- The MR LBD was amplified by PCR from the plasmid pRShMR_NX (7), using the primers MR1 and MR2, which incorporates a BglII site downstream of the stop codon. The PCR product was digested with XhoI/BglII and ligated into the pSP72 vector and sequenced. The MR LBD sequence was then modified to create a ScaI site at position +2410 and to remove an NsiI site at position +2522.

MR LBD(+ScaI)-- To create the ScaI site two PCRs were performed using the primers MR1 and MR4 and MR3 and MR2. The template was MR LBD-pSP72. The two PCR products were joined together by overlap extension PCR using the primers MR1 and MR2.

MR LBD(+ScaI/-NsiI)-- To remove the NsiI site two PCRs were performed using the primers MR1 and MR6 and MR5 and MR2. The template was MR LBD-pSP72(+ScaI). The two PCR products were joined together by overlap extension PCR using the primers MR1 and MR2. The final PCR product, MR LBD(+ScaI/-NsiI) was then fully sequenced to confirm its identity. The chimeras MMGG and GGMM were generated by overlap extension PCR.

GGMM-- The GR LBD-pSP72 was amplified using the primers GR1 and GGMM2. The MR LBD(+ScaI/-NsiI)-pSP72 was amplified using the primers GGMM1 and MR2. The two PCR products were joined together by overlap extension PCR using the primers GR1 and MR2. The final PCR product was ligated into the pSP72 vector and sequenced.

MMGG-- The MR LBD(+ScaI/-NsiI) PCR product was amplified using the primers MR1 and MMGG2. GR LBD-pSP72 was amplified using the primers MMGG1 and GR2. The two PCR products were joined together by overlap extension PCR using the primers MR1 and GR2. The final PCR product was ligated into the pSP72 vector and sequenced.

The other chimeras were created in the pSP72 vector using restriction enzyme digests of GR LBD, MR LBD, MMGG, and GGMM, using whichever restriction sites were appropriate for the construct. The chimeras were sequenced to confirm their identities.

The chimeric LBD was then swapped in place of the native LBD in the pRShGR_BX vector digested with XhoI/BamHI. When the 3' end of the chimera was comprised of GR sequence the construct was removed from pSP72 using XhoI/BamHI, and when the 3' end of the chimera was comprised of MR sequence the construct was removed from pSP72 using XhoI/BglII. In each case the chimera was sequenced to confirm its identity.

Tissue Culture and Transactivation Assay-- 1 µg of the chimera expression plasmid, 1 µg of MMTV-LUC reporter plasmid, and 0.25 µg of pRSV--galactosidase were co-transfected into cultured CV-1 African green monkey kidney fibroblast cells using the calcium phosphate precipitation method. Cells were grown at 37 °C in DMEM (supplemented with 1 mM glutamine, non-essential amino acids, and 1% penicillin-streptomycin) and either 5 or 10% fetal bovine serum. The cells were trypsinized and replated in 48-well plates at a density of 2 × 10⁴ cells/well. After 20-24 h the cells were transfected. Steroids were added 20-24 h after transfection and then harvested 22-24 h later. Measurement of transactivation was performed using the Dual-Light kit (Tropix, Bedford, MA) as per the manufacturer's instructions. The cells were incubated with 100 µl of lysis buffer for 30 min at room temperature. A 10-µl aliquot was removed for assay. Steroid-dependent luciferase activity was measured, the tubes were then incubated for 1 h at room temperature to allow the luciferase protein to degrade, and -galactosidase activity, which was used to determine transfection efficiency, was finally measured. Light measurements were made in a Berthold luminometer (Berthold, Wildbad, Germany). All measurements were performed in duplicate for two separate experiments. Results are expressed as luciferase activity/-galactosidase activity.

Ligand Binding Assay-- Ligand binding assays were performed in COS-1 cells using either [³H]aldosterone or [³H]dexamethasone. Cells were grown at 37 °C in DMEM supplemented with 1 mM glutamine, non-essential amino acids, 1% penicillin-streptomycin, and 10% fetal bovine serum. The cells were trypsinized and replated in 6-well plates at a density of 2 × 10⁵ cells/well. After 20-24 h the cells were transfected with the expression plasmid using the DEAE-dextran method and then incubated overnight in DMEM + 10% fetal bovine serum. The medium was replaced with DMEM 1 h before the ligand binding assay. The cells were washed 3 times with ice-cold phosphate-buffered saline and then [³H] steroid was added in DMEM. To measure MR-type binding 2 nM [³H]aldosterone was used, and nonspecific binding was assessed by adding a 500-fold excess of non-radioactive aldosterone. Scatchard analysis was performed on the MR LBD, GMGG, and GMGM chimeras using [³H]aldosterone at concentrations of 0.26, 0.64, 1.6, 4, 10, and 20 nM. Each point was assayed in triplicate, and data was analyzed using the EBDA program (20). To measure GR-type binding 10 nM [³H]dexamethasone was used, and nonspecific binding was assessed by adding a 500-fold excess of non-radioactive dexamethasone. Cortisol binding to the chimeras was assessed by incubating the [³H]dexamethasone with 10-, 50- and 500-fold excesses of cold cortisol. The cells were incubated at 37 °C for 2 h. Cells were then washed 3 times with ice-cold phosphate-buffered saline and then scraped from the wells in phosphate-buffered saline. The suspension was added to scintillant, and radioactivity was measured in a Packard 2500 TR liquid scintillation counter.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Transactivation Assays-- The chimeras were initially examined using a functional assay as this method is more sensitive than measuring ligand binding. The chimeric LBDs were placed into the pRShGR_BX expression vector in place of the native GR LBD. The transcriptional activity of the chimeras was determined by measuring luciferase expression driven by the MMTV promoter in transiently transfected CV-1 cells (21). It has been shown previously that the ligand-independent AF-1 of the GR N-terminal domain is primarily responsible for transactivation in this system (8). Thus, assuming the constructs are expressed and are transcriptionally active, transactivation should correspond to ligand binding. The GR, rather than the MR, expression vector was chosen because the transcriptional effect is larger (8) and therefore provides a more sensitive read-out.

The results of the transactivation assays for all of the chimeras are summarized in Table I. Both aldosterone and cortisol potently activate the MR LBD construct (Fig. 2), whereas the GR is activated potently by cortisol but only very weakly by aldosterone (Fig. 3). Lower concentrations of cortisol are required to activate the MR LBD construct than are required for the GR. This is consistent with previous studies using the same system (22-25).

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Dose-response curves of the wild-type MR LBD and chimeras in which one section of the MR has been replaced with the corresponding sequence of the GR. Constructs were transfected, together with the MMTV-LUC reporter plasmid, into CV-1 monkey kidney fibroblast cells and treated with both aldosterone () and cortisol (). Results of luciferase expression were calculated relative to a constitutively expressed -galactosidase reporter and graphed as a percentage of the maximal response to steroid for each chimera. Each point represents the mean ± S.E. of duplicates of two different experiments. M refers to the MR sequence, and G refers to the GR sequence.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Comparison of wild-type GR with GMGG to demonstrate the importance of the second region of the MR in the response to aldosterone. Constructs were transfected, together with the MMTV-LUC reporter plasmid, into CV-1 monkey kidney fibroblast cells and treated with both aldosterone () and cortisol (). Results of luciferase expression were calculated relative to a constitutively expressed -galactosidase reporter and graphed as a percentage of the maximal response to steroid. Each point represents the mean ± S.E. of duplicates of two different experiments. M refers to the MR sequence, and G refers to the GR sequence.

Aldosterone Activation of Chimeras-- Fig. 2 demonstrates the effect of replacing each section of the MR LBD with the corresponding region of the GR LBD. Three of the four chimeras, GMMM, MMGM, and MMMG are potently activated by aldosterone. In contrast MGMM is not activated by the mineralocorticoid (Table I). This suggests that the MR sequence in the second region of the chimeras, corresponding to amino acids 804-874 of the MR LBD, is critical for aldosterone binding. Fig. 3 compares the wild-type GR with GMGG, the reciprocal of the inactive MGMM chimera, and again demonstrates the importance of this region. The GMGG chimera is activated by nanomolar concentrations of aldosterone, representing a 2-order of magnitude increase in sensitivity over the wild-type GR. The importance of this region is further emphasized by comparing the results using GGMM (Table I) and GMMM (Fig. 2 and Table I). The GGMM chimera is not activated by aldosterone at concentrations up to 300 nM, whereas the EC₅₀ of aldosterone for the GMMM chimera is approximately 1 nM. This represents at least a 3-order of magnitude increase in sensitivity. Examination of the results as a whole (Table I) reveals that all of the chimeras that are activated by aldosterone contain amino acids 804-874 of the MR. The only exception is the GMMG construct. In contrast any chimera that contains the GR sequence corresponding to this region of the MR is not activated by aldosterone.

Analysis of the chimeras (Table I) reveals another characteristic of aldosterone activation. If the chimera contains the MR sequence in the fourth region as well as in the second (GMMM, MMGM) then the EC₅₀ value for aldosterone is approximately 1 nM, similar to wild-type MR LBD. If only the second region is present (GMGG, MMGG, MMMG) then the EC₅₀ value is higher. This suggests that although they are not critical, amino acids 932-984 of the MR LBD, corresponding to this fourth region of the chimeras, are also involved in ligand binding. The exception is the GMGM chimera for which the EC₅₀ value is 50 nM.

Cortisol Activation of Chimeras-- Because both the MR and GR bind cortisol with high affinity (13) it was expected that all of the chimeras would be activated by cortisol. However, this steroid activates only GR, MR, GMMM, MMGM, MGMG, GGMG, and MGGG (Table I). All of the chimeras activated by cortisol contain sequences in the second and fourth sections derived from the same receptor; that is, they contain both amino acids 804-874 and 932-984 of the MR or amino acids 598-668 and 726-777 of the GR. The exception is GMGM, which is only very weakly activated by cortisol.

[³H]aldosterone Binding-- The results of [³H]aldosterone binding are shown in Fig. 4A. The concentration of tracer used, 2 nM, should exhibit maximal binding to the MR and very little binding to the GR. In general there is a good correlation between the ability to bind [³H]aldosterone and the ability to activate the receptor. The wild-type MR, GMMM, MMGM, MMMG, and MMGG all show high levels of specific [³H]aldosterone binding. The wild-type GR, MGGG, GGMG, GGGM, GGMM, MGMM, MGMG, and MGGM bind very little [³H]aldosterone, consistent with their poor ability to be activated by aldosterone. Three chimeras demonstrate anomalous behavior. At 2 nM [³H]aldosterone the GMGG and GMGM chimeras bind very little tracer, despite both being activated by aldosterone. In the functional assay the GMGG and GMGM chimeras demonstrate higher EC₅₀ values than the other aldosterone-activated chimeras (Table I). Subsequent Scatchard analysis of binding confirmed that aldosterone has a lower affinity for these two chimeras than for the wild-type MR LBD. The dissociation constant (k_d) values for MR LBD, GMGG, and GMGM were 0.7, 6.6, and 80 nM, respectively. These values are consistent with the EC₅₀ values for aldosterone in the functional assay. The other chimera to behave anomalously is GMMG, which binds 2 nM [³H]aldosterone but is not activated by aldosterone at concentrations of steroid up to 300 nM.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Binding of [3H] steroids to the constructs. Constructs were transfected into COS-1 transformed monkey kidney fibroblast cells, and ligand binding studies were performed in whole cells. A shows the results of the binding of 2 nM [3H]aldosterone to the constructs. Nonspecific binding was assessed using a 500-fold excess of aldosterone. B shows the binding of 10 nM [3H]dexamethasone to the constructs. Nonspecific binding was assessed using a 500-fold excess of dexamethasone. C shows the displacement of [3H]dexamethasone binding by cortisol from selected chimeras. The results are expressed as mean ± S.E. (n = 5 for A and B, and n = 3 for C). M refers to the MR sequence and G refers to the GR sequence.

[³H]dexamethasone Binding-- To test glucocorticoid-type binding to the chimeras [³H]dexamethasone was used at a concentration (10 nM) that should bind to both the GR and MR. As shown in Fig. 4B there is binding of [³H]dexamethasone to both wild-type LBDs. All of the chimeras also bind [³H]dexamethasone (Fig. 4B). This result is particularly significant for the chimeras that were not activated by either aldosterone or cortisol (GGGM, GGMM, MGMM, GMMG, and MGGM). It shows that the negative results in the functional assay were not caused by a lack of expression of the chimeric protein. All of the chimeras are expressed and can bind [³H]dexamethasone.

To assess whether the chimeras that are not activated by cortisol (GGMM, GGGM, MMGG, MMMG, GMGG, MGMM, GMMG, and MGGM) are still able to bind this steroid, experiments were performed in which [³H]dexamethasone binding was displaced by 10-, 50-, and 500-fold excesses of cortisol. As shown in Fig. 4C cortisol was able to displace [³H]dexamethasone from the wild-type GR and the transactivation-deficient chimeras with a similar efficacy, demonstrating that cortisol does indeed bind to these receptors. The GGGM chimera appears to bind cortisol with even greater affinity than the wild-type GR. Overall the results of this experiment suggest that, in contrast to the results using aldosterone, the differential effects of cortisol on the chimeras are not due to their differing abilities to bind the ligand but to their differing abilities to induce transactivation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The nature and determinants of ligand binding to the MR are only partially understood. The characterization of ligand binding in other nuclear hormone receptors has been assisted by the study of steroid insensitivity syndromes where natural amino acid mutations/substitutions alter binding. To date, however, only nonsense and splice-site mutations have been found in the MR in the autosomal dominant form of aldosterone resistance (pseudohypoaldosteronism) (26). Conclusions about the nature of ligand binding to the MR have therefore been drawn largely from analogies with other receptors in the SHR family. In this study we have used a different approach to determining sequences important for the specificity of aldosterone binding to the MR, by taking advantage of the high degree of amino acid sequence identity between the MR and GR (7) to create chimeras between the LBDs of the two receptors. The binding of both aldosterone and cortisol, the two natural ligands for the MR in humans, was then examined. The chimeras were analyzed by a sensitive functional assay using luciferase expression driven by the MMTV promoter, and the results of this assay were then compared with the results of direct ligand binding experiments. Importantly for the interpretation of the results, all of the constructs bound [³H]dexamethasone, confirming that all were expressed and able to bind ligand.

Aldosterone Binding-- There was a good correlation between the results of the functional assay and the direct ligand binding experiments, which attests to the validity of the chimeric approach to the investigation of the determinants of aldosterone binding specificity. Aldosterone only activates chimeras in which the second section contains MR sequence, corresponding to amino acids 804-874 of the MR LBD. Inserting this region into the GR (the GMGG chimera) leads to an approximate 2-order of magnitude increase in affinity for aldosterone compared with wild-type GR. This region of the MR LBD must therefore be critical for aldosterone binding specificity. It was also observed that aldosterone activates the MR LBD, GMMM, and MMGM more potently than MMGG, GMGG, and MMMG, suggesting that the second region alone is not sufficient for high affinity binding of aldosterone and that although they are not critical, sequences in the fourth region of the MR LBD (between amino acids 932-984) also contribute. This is consistent with analysis of the crystal structures of the estrogen receptor- (27, 28) and progesterone receptor (PR) (29) LBDs. In these structures the regions corresponding to the second and fourth regions of the chimeras form the ligand-binding pocket. The same approach of using LBD chimeras has been used to examine the binding specificities of the androgen receptor and PR (17). Inserting the region between amino acids 766-799 of the androgen receptor into the PR leads to a dramatic increase in the affinity of the chimera for testosterone. Interestingly this region overlaps the MR (804-874) region that we find determines aldosterone binding specificity.

Two of the chimeras, GMGM and GMMG, behaved anomalously. Given the results of the other chimeras it would be predicted that the GMGM chimera should bind aldosterone with high affinity, yet in both the functional and ligand binding assays this construct has a decreased affinity for the ligand as compared with the MR LBD. The GMMG chimera is not activated by aldosterone in the functional assay by concentrations up to 300 nM despite containing the MR sequence in the second region. It does however bind [³H]aldosterone at a concentration of 2 nM. The GMMG chimera therefore appears to be able to bind aldosterone but is unable to then activate transcription. It is possible that in these two chimeras the secondary and tertiary structures of the proteins have been perturbed enough to affect ligand binding and receptor function. The design of the chimeras was begun before publication of a consensus structure of the nuclear hormone receptor LBD (30) based on the crystal structures of the retinoid acid receptor  and retinoid X receptor  LBDs. As a result the design of the chimeras does not take into account the putative -helical nature of the LBD structure. The dissociation between ligand binding and transcriptional activity observed with the GMMG chimera may, for example, be caused by structural perturbations leading to a failure of nuclear localization of the ligand-bound receptor. The importance of nuclear localization is demonstrated in a study showing that certain PR antagonists are poor GR antagonists, despite having a high affinity for the GR, because of an inability of these compounds to promote nuclear localization of the receptor (31).

The region between amino acids 804-874 in the MR (second region in the chimeras) corresponds to -helices 5-8 of the PR LBD crystal structure (29) (labeled helices 6-9 in the estrogen receptor- LBD structure (27)). There is 48% amino acid identity between this region of the MR and the corresponding region of the GR, which is lower than the overall 57% sequence identity between the two proteins in their LBDs. Fig. 5 shows the position of the region of the MR between amino acids 804-874 based on the crystal structure of the PR LBD. Given the position of the ligand in the PR LBD crystal structure (29) it is likely that the amino acid or amino acids that govern aldosterone binding specificity lie in helix 5, the -turn, and/or helix 7. Previous studies on the structures of the LBDs of the MR and other SHRs provide very few clues as to the identity of these critical residues. In the crystal structure of the PR holo-LBD (29) interactions were observed between two residues (Gln⁷²⁵ and Arg⁷⁶⁶) and the 3-keto group of progesterone. However, this region of the steroid molecule is identical in aldosterone, cortisol, corticosterone, and progesterone, and the two residues identified are conserved in the MR, GR, and PR, so it is very unlikely that they are involved in conferring the specificity of aldosterone binding. In the estrogen receptor- holo-LBD crystal structure (27) a residue (His⁵²⁴) was identified that interacts with the 17-hydroxyl moiety of estradiol. The structures of aldosterone and cortisol do differ at C₁₇, but the analogous residue in the MR lies outside the 804-874 region. The secondary and tertiary structures of the MR LBD, bound to aldosterone, have been modeled based on the retinoid acid receptor  crystal structure (10). The amino acids Gln⁷⁷⁶ and Arg⁸¹⁷ were found to interact with the 3-keto group of aldosterone, and Asn⁷⁷⁰ and Thr⁹⁴⁵ were found to interact with the 20-keto/21-hydroxyl groups of the ligand. Site-directed mutagenesis of these residues confirmed their importance in ligand binding and validated the model. All four residues are conserved between the MR and GR, consistent with the structures of aldosterone, cortisol, and corticosterone being identical at the C₃, C₂₀, and C₂₁ positions, so again it is highly unlikely that these residues are involved in MR ligand binding specificity. In another study it was observed that mutation of either Cys⁸⁴⁹ or Cys⁹⁴² of the MR LBD dramatically decreases or abolishes ligand binding (32). Cys⁸⁴⁹ lies within the region of 804-874, but both of these residues are conserved in the GR and are therefore again unlikely to be involved in MR ligand specificity. Consistent with this, mutation of these two residues produced identical effects on cortisol and aldosterone binding (32).

View larger version (67K): [in this window] [in a new window]   Fig. 5.   The position of amino acids 804-874 of the MR LBD modeled onto the PR LBD crystal structure. The region corresponding to amino acids 804-874 of the MR LBD is shown in dark gray. The helices are numbered as they are in the PR LBD crystal structure (29).

The inability of studies described above to identify or provide clues to amino acids involved in aldosterone binding specificity may reflect a limitation of the techniques of crystallography and molecular modeling, both of which determine the final binding conformation of the ligand-receptor complex. The differences between the aldosterone and cortisol/corticosterone molecules lie at the C₁₁ and C₁₈ positions. Therefore, it is highly likely that interactions between these regions of the steroid and the receptor determine binding specificity. However, in the MR LBD model no amino acids were identified as interacting with the C₁₁ and C₁₈ positions of aldosterone (10). Ligand binding to SHRs is a dynamic process, as evidenced by the dramatic changes in conformation involving the helix 12 region of the LBD (33, 34), and binding specificity may be governed by the initial interactions between ligand and receptor that occur before the resulting conformational changes (29). Based on those results, it may therefore not be possible to determine which amino acids are critical for aldosterone binding specificity from crystal structures or molecular models. The use of receptor chimeras has proven to be very useful in gaining insights into ligand binding to other SHRs (15-18) and complements these other techniques.

Cortisol Binding and Transactivation-- We had assumed that, as cortisol binds with high affinity to both the MR and GR, this steroid would activate all of the chimeras. Surprisingly, however, only the GR, MR, GMMM, MMGM, MGGG, GGMG, and MGMG constructs were activated by cortisol. One common feature of these constructs is that the second and fourth sections are derived from the same receptor; they contain either amino acids 804-874 and 932-984 of the MR or amino acids 598-668 and 726-777 of the GR. The only exception is the GMGM chimera, which also shows anomalous aldosterone binding behavior. It was subsequently found that cortisol can displace [³H]dexamethasone binding from the other chimeras, thereby demonstrating that all of the chimeras retain the ability to bind this ligand. This suggests that the differential effects of cortisol in the functional assay are because of an effect downstream of ligand binding. There is a complex cascade of events involved in MR and GR function after ligand binding: heat-shock protein dissociation, nuclear localization, dimerization, DNA binding, and transcriptional activation. Presently we do not know which of these aspects of receptor function is being affected. Both the MR and GR LBDs harbor a ligand-dependent AF-2 region involved in transcriptional activation (8), and one possibility is that this region is responsible for the dissociation between ligand binding and transactivation observed with cortisol. However, in the system used in these experiments the AF-1 region in the N-terminal domain of the GR, which is common to all of the constructs, is predominantly responsible for the transcriptional effect (8), so it is highly unlikely that dysfunction of the AF-2 region can account for the results. It is also unlikely that ligand binding affects the AF-1 region itself, which functions independent of ligand binding (8).

The absolute requirement of the second and fourth regions of the MR for activation by cortisol is not a characteristic shared by aldosterone. The study of other nuclear hormone receptors has shown a relationship between the nature of ligand binding, the resulting conformational changes in the receptor structure, and the downstream effect this has on receptor function. It is therefore possible that the two ligands bind to the MR differently and may therefore have different effects on transcription. There is some evidence for such differences. One consistent finding in many cells lines is that, although both aldosterone and cortisol bind with the same affinity to the receptor, aldosterone activates the receptor at lower concentrations than are required by cortisol (22-24). This implies a functional difference in the binding of the two steroids. It has also been shown that intracerebroventricularly administered aldosterone increases blood pressure in the unilaterally nephrectomized rat, and this effect is not mimicked by corticosterone. In fact, corticosterone antagonizes the action of aldosterone (35).

Summary-- We have identified a region of the MR LBD, between amino acids 804-874, that is crucial for aldosterone binding specificity. Our approach of using receptor chimeras complements both crystallography and molecular modeling to assist in our understanding of ligand binding to the MR. We believe that the initial interactions between ligand and receptor, before the resulting conformation changes in the protein structure, determine binding specificity. The techniques of crystallography and molecular modeling can only identify amino acid residues interacting with the ligand in its final "resting position" after the conformational changes have occurred. The results described in this study are the first step in a process that should allow us to identify individual amino acids that are crucial for aldosterone binding specificity. Results from this analysis of aldosterone binding to the MR may prove very useful in the design of new classes of MR antagonists for use in treating conditions in which high levels of aldosterone contribute to the pathogenesis.

	ACKNOWLEDGEMENTS

We thank Professor R. M. Evans, Dr. S. Nordeen, and Dr. W. Tilley for generous gifts of the plasmids pRShGR_NX, pRShMR_NX, MMTV-LUC, and pRSV--GAL.

	FOOTNOTES

^* This work was supported by a postdoctoral fellowship from the National Heart Foundation of Australia, by the National Health and Medical Research Council, and by a generous donation from Eva and Les Erdi.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Prince Henry's Inst. of Medical Research, P. O. Box 5152, Clayton, Victoria 3168, Australia. Tel.: 61-3-9594-4380; Fax: 61-3-9594-6125; E-mail: fraser.rogerson@med.monash.edu.au.

	ABBREVIATIONS

The abbreviations used are: MR, mineralocorticoid receptor; SHR, steroid hormone receptor(s); LBD, ligand-binding domain(s); HSP, heat-shock protein(s); GR, glucocorticoid receptor; PCR, polymerase chain reaction(s); MMTV, mouse mammary tumor virus; LUC, luciferase; DMEM, Dulbecco's modified Eagle's medium; PR, progesterone receptor.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. B Pippal and P. J Fuller Structure-function relationships in the mineralocorticoid receptor J. Mol. Endocrinol., December 1, 2008; 41(6): 405 - 413. [Abstract] [Full Text] [PDF]

				F. M. Rogerson, Y.-Z. Yao, R. E. Elsass, N. Dimopoulos, B. J. Smith, and P. J. Fuller A Critical Region in the Mineralocorticoid Receptor for Aldosterone Binding and Activation by Cortisol: Evidence for a Common Mechanism Governing Ligand Binding Specificity in Steroid Hormone Receptors Mol. Endocrinol., April 1, 2007; 21(4): 817 - 828. [Abstract] [Full Text] [PDF]

				N. Z. Lu, S. E. Wardell, K. L. Burnstein, D. Defranco, P. J. Fuller, V. Giguere, R. B. Hochberg, L. McKay, J.-M. Renoir, N. L. Weigel, et al. International Union of Pharmacology. LXV. The Pharmacology and Classification of the Nuclear Receptor Superfamily: Glucocorticoid, Mineralocorticoid, Progesterone, and Androgen Receptors Pharmacol. Rev., December 1, 2006; 58(4): 782 - 797. [Full Text] [PDF]

				S. Sola, J. D. Amaral, P. M. Borralho, R. M. Ramalho, R. E. Castro, M. M. Aranha, C. J. Steer, and C. M. P. Rodrigues Functional Modulation of Nuclear Steroid Receptors by Tauroursodeoxycholic Acid Reduces Amyloid {beta}-Peptide-Induced Apoptosis Mol. Endocrinol., October 1, 2006; 20(10): 2292 - 2303. [Abstract] [Full Text] [PDF]

				R. E. Schweppe, T. H. Cheung, and N. G. Ahn Global Gene Expression Analysis of ERK5 and ERK1/2 Signaling Reveals a Role for HIF-1 in ERK5-mediated Responses J. Biol. Chem., July 28, 2006; 281(30): 20993 - 21003. [Abstract] [Full Text] [PDF]

				L. Pascual-Le Tallec and M. Lombes The Mineralocorticoid Receptor: A Journey Exploring Its Diversity and Specificity of Action Mol. Endocrinol., September 1, 2005; 19(9): 2211 - 2221. [Abstract] [Full Text] [PDF]

				T. Kodama, N. Shimizu, N. Yoshikawa, Y. Makino, R. Ouchida, K. Okamoto, T. Hisada, H. Nakamura, C. Morimoto, and H. Tanaka Role of the Glucocorticoid Receptor for Regulation of Hypoxia-dependent Gene Expression J. Biol. Chem., August 29, 2003; 278(35): 33384 - 33391. [Abstract] [Full Text] [PDF]

				W. Sun, X. Wei, K. Kesavan, T. P. Garrington, R. Fan, J. Mei, S. M. Anderson, E. W. Gelfand, and G. L. Johnson MEK Kinase 2 and the Adaptor Protein Lad Regulate Extracellular Signal-Regulated Kinase 5 Activation by Epidermal Growth Factor via Src Mol. Cell. Biol., April 1, 2003; 23(7): 2298 - 2308. [Abstract] [Full Text] [PDF]

				N. Yoshikawa, Y. Makino, K. Okamoto, C. Morimoto, I. Makino, and H. Tanaka Distinct Interaction of Cortivazol with the Ligand Binding Domain Confers Glucocorticoid Receptor Specificity. CORTIVAZOL IS A SPECIFIC LIGAND FOR THE GLUCOCORTICOID RECEPTOR J. Biol. Chem., February 8, 2002; 277(7): 5529 - 5540. [Abstract] [Full Text] [PDF]

				N. Farman and M.-E. Rafestin-Oblin Multiple aspects of mineralocorticoid selectivity Am J Physiol Renal Physiol, February 1, 2001; 280(2): F181 - F192. [Abstract] [Full Text] [PDF]

				G. Auzou, J. Fagart, A. Souque, C. Hellal-Lévy, J.-M. Wurtz, D. Moras, and M.-E. Rafestin-Oblin A Single Amino Acid Mutation of Ala-773 in the Mineralocorticoid Receptor Confers Agonist Properties to 11beta -Substituted Spirolactones Mol. Pharmacol., October 1, 2000; 58(4): 684 - 691. [Abstract] [Full Text]

				W. Sun, K. Kesavan, B. C. Schaefer, T. P. Garrington, M. Ware, N. L. Johnson, E. W. Gelfand, and G. L. Johnson MEKK2 Associates with the Adapter Protein Lad/RIBP and Regulates the MEK5-BMK1/ERK5 Pathway J. Biol. Chem., February 9, 2001; 276(7): 5093 - 5100. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rogerson, F. M. Articles by Fuller, P. J. Search for Related Content PubMed PubMed Citation Articles by Rogerson, F. M. Articles by Fuller, P. J. Social Bookmarking                      What's this?