help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2007-1683
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Becker, H.
Right arrow Articles by Bury, N. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Becker, H.
Right arrow Articles by Bury, N. R.
Endocrinology Vol. 149, No. 9 4567-4576
Copyright © 2008 by The Endocrine Society

The A/B Domain of the Teleost Glucocorticoid Receptors Influences Partial Nuclear Localization in the Absence of Hormone

Heidi Becker1, Armin Sturm1, James E. Bron, Kristin Schirmer and Nicolas R. Bury

Nutritional Sciences Division (H.B., A.S., N.R.B.), King’s College London, London SE1 9NH, United Kingdom; Department of Cell Toxicology (CellTox) (H.B., K.S.), UFZ-Centre for Environmental Research Leipzig-Halle in the Helmholtz Association, 04318 Leipzig, Germany; Institute of Aquaculture (A.S., J.E.B.), University of Stirling, Stirling FK9 4LA, United Kingdom; and Swiss Federal Institute of Aquatic Science and Technology-Eawag (K.S.), Dübendorf 8600, Switzerland

Address all correspondence and requests for reprints to: Dr. Nicolas Bury, King’s College London, Nutritional Sciences Research Division, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NH, United Kingdom. E-mail: nic.bury{at}kcl.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The glucocorticoid (GR) and mineralocorticoid receptor (MR) of extant jawed vertebrates emerged after duplication of an ancestral corticosteroid receptor. The ancestral corticosteroid receptor resembled extant MRs in hormone selectivity, and the different ligand specificity of extant GRs is a secondary derived characteristic. An additional characteristic that distinguishes the mammalian GR from the MR is the cellular distribution pattern in the absence of hormone: the naïve GR resides in the cytoplasm, whereas the naïve MR is found in both the nucleus and cytoplasm. Our results show, by the use of green fluorescent protein-tagged fusion proteins, that the GRs [rainbow trout (rt) GR1 and rtGR2] from a lower vertebrate, the teleost fish, rainbow trout (Oncorhynchus mykiss) resemble mammalian MR rather than GR in their subcellular localization pattern. The addition of cortisol caused the remaining cytoplasmic rtGR1 and rtGR2 to migrate to the nucleus. The speed of nuclear localization was cortisol concentration dependent, with rtGR2 being more sensitive than rtGR1, mimicking the transactivational properties of the receptors in which the cortisol EC50 value is an order of magnitude lower for rtGR2. By the use of chimera constructs between the trout GRs and the rat GR C656G, we show that the E domain of the trout receptors are not involved in the nucleocytoplasmic localization of naïve trout GRs, but the A/B domain, especially if linked to the corresponding trout CD region, plays a pivotal role in the cellular distribution pattern. This is unrelated to the difference in the trout GRs transactivation sensitivity, which is determined by the receptor’s E-domains.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
GLUCOCORTICOIDS ARE steroid hormones involved in the control of a vast array of cellular and physiological processes, including immune function, metabolism, development, and the stress response (1). The majority of effects of glucocorticoids are mediated through the glucocorticoid receptor (GR) which influences transcription after binding to conserved motifs in the promoter region of target genes or through protein-protein interaction with other transcription factors (2, 3). In the absence of ligand, the mammalian GR resides in the cytoplasm (4, 5), in which it is part of a large heteromeric complex that contains heat shock protein (hsp) 90, other hsps, hsp-70/hsp-90 organizing protein, and various immunophilins (6). Hormone binding facilitates GR dissociation from this complex and provokes rapid translocation of receptor into the nucleus (7, 8, 9).

There are two hormone-dependent nuclear localization signals (NLS) in the rat GR, nuclear localization (NL)-1 and NL2 (4). NL1 overlaps the C-terminal region of the DNA binding domain and the N terminus of the hinge region (9, 10), whereas NL2 maps to the ligand-binding domain of the receptor. NL1 is constitutively active in C-terminal deletion mutants of GR lacking a functional ligand-binding domain. In the context of GR, however, the ligand-binding domain represses NL1 activity in the absence of hormone (4). NL2 activity is strictly hormone dependent (9, 10). On withdrawal of the hormone, the nuclear GR quickly reassociates with the chaperone complex; however, transfer of GR back to the cytoplasm is slow (half-time of 12–24 h) (8). The mechanisms underlying nuclear export of GR are less well understood, but a nuclear export signal (NES) of 15 amino acids has been identified in the DNA-binding domain of GR (11), and a further regulatory region has recently been identified termed a nuclear retention signal (NRS) that overlaps with the NL1 site (12).

The steroid receptors, which include the progesterone, estrogen, androgen, glucocorticoid, and mineralocorticoid (MR) receptors, share a general domain architecture (13), reflecting their common evolutionary origin, and further show conserved NLS (14, 15, 16, 17) and NES sequences (11). Despite these similarities, however, steroid receptors differ markedly in their hormone binding characteristics as well as their cellular distribution in the absence of hormone: the androgen receptor is similar to the GR, being cytoplasmic, and the MR distributes evenly among the nucleus and the cytoplasm, whereas the progesterone and estrogen receptors are constitutively nuclear (14, 16, 18, 19). Recent molecular evolutionary studies suggest that the GR and its sister gene, the MR, evolved approximately 470 million years ago from an ancestral corticosteroid receptor, which possessed the broader ligand selectivity of extant MRs, i.e. was activated by both gluco- and mineralocorticoids (20). Thus, GR preference for glucocorticoids is a derived trait in the vertebrate lineage (20, 21). It is unclear when the difference in nucleocytoplasmic distribution pattern of GR and MR evolved because knowledge of the subcellular distribution is currently restricted to only a few mammalian GRs (e.g. Refs. 8, 19, 22 and 23) and has not been assessed in other vertebrate groups (e.g. the teleost fish).

A whole genome duplication event occurred in the ray-finned fish (Actinopterygi) lineage around 355 million years ago (24). As a consequence, two genes encoding GRs exist in many teleosts (25, 26) and map to distinct chromosomes in Takifugu rubripes (pufferfish) (27). In rainbow trout (rt) the two GR isoforms, called rtGR1 and rtGR2, display a marked difference in their hormone sensitivity in transactivation and transrepression assays, with rtGR2 exerting its action at far lower concentrations of cortisol (25, 28). In line with this observation, we report here that rtGR2 nuclear transfer is more sensitive to hormone treatment than that of rtGR1. Intriguingly, hormone-free GRs from rainbow trout reside both in the cytoplasm and the nucleus, reminiscent of the behavior of tetrapod MR (18), and in stark contrast to mammalian GRs. By generating domain swaps between trout GRs and the rat GR mutant C656G (5), we were able to demonstrate that the A/B domain plays a role in the unusual nuclear distribution pattern of the trout GRs.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Plasmid
rtGRs carrying an N-terminal green fluorescent protein (GFP) tag were produced using the plasmid pEGFP-C1 (CLONTECH, Palo Alto, CA) to generate the constructs pEGFP-rtGR1 and pEGFP-rtGR2. Rat GR mutant C656G was amplified by PCR from pcI-nGFP-C656G (5) and subcloned into pEGFP-C1 to obtain pEGFP-ratGR-C656G. To generate GFP-tagged domain swap mutants between ratGR-C656G (termed rat GR in context of the mutants) and rtGR1 or rtGR2, a two-step PCR protocol based on overlap extension was used (29), followed by subcloning into pEGFP-C1 as above. Domain definitions with respect to trout and rat GR sequences (GenBank accession no. CAA90937, AAR87479, and CAA72938) were A/B (rtGR1: amino acids 2–386; rtGR2: 2–306; ratGR C656G: 2–439), combined C and D domain (rtGR1: 387–502; rtGR2: 307–419; ratGR C656G: 440–545), and E domain (rtGR1: 503–758; rtGR2: 420–669; ratGR C656G: 546–795). The constructs encoding recombinant GRs were AB(rtGR1)CDE(rat), AB(rtGR2)CDE(rat), AB(rat)CDE(rtGR1), AB(rat)CDE(rtGR2), ABCD(rtGR1)E(rat), ABCD(rtGR2)E(rat), ABCD(rat)E(rtGR1), ABCD(rat)E(rtGR2), AB(rat)CD(rtGR1)E(rat), and AB(rat)CD(rtGR2)E(rat).

Cell culture
RTG-2 cells (derived from rainbow trout gonad) and GR-deficient mammalian cell lines (COS-7, derived from African green monkey kidney, or CHO, derived from hamster ovary) were maintained in DMEM (Life Technologies, Inc., Paisley, UK) containing 4.5 g/liter glucose, 110 mg/liter pyruvate, 10% denatured fetal calf serum (FCS), 100 IU/ml penicillin, 100 µg/ml streptomycin, and 2 mM glutamine. RTG-2 cells were maintained without CO2 enrichment at 21 ± 1 C, whereas COS-7 and CHO cells were kept at 37 C and 5% CO2.

Cellular localization and nuclear transfer studies
The day before transfection, cells were seeded into Lab-Tek chamber slides (Nunc, Glostrup, Denmark) at initial densities of 1.25 x 105 / cm2 (COS-7) and 3.75 x 105/cm2 (RTG-2) in supplemented DMEM (see above) and incubated overnight. Before transfection cells were washed with PBS, and a medium devoid of serum and antibiotics was added [DMEM nutrient mix F-12 Ham (Sigma, Poole, UK) supplemented with 2 mM glutamine and 3.7 g/liter NaHCO3]. Cells were then transfected with plasmids (75 ng of GFP-tagged receptor construct and 0.94 µg of irrelevant plasmid (pBluescript SK; Stratagene, La Jolla, CA) per square centimeter) using a commercial reagent (Lipofectamine2000; Invitrogen, Paisley, UK). After 5 h, the media containing plasmids and transfection reagent were replaced with fresh serum- and antibiotic-free medium and cells subjected to a further incubation (16 h for COS-7, 40 h for RTG-2) before living cell image acquisition was performed. During image analysis, cells were placed in a temperature-controlled chamber set to the temperature appropriate for the used cell line. Nuclear identification was verified by staining with the chromatin stain Hoechst 33342. Cells were examined for 30 min using a TCS SP2 confocal microscope (Leica, Heidelberg, Germany). GFP fluorescence (500–560 nm) was recorded after argon laser excitation at 488 nm. For each picture taken, 20–30 cells were analyzed and classified into categories based on previous studies (9, 30): N = C for cells having an equal distribution of fluorescence signal between the cytoplasmic and nuclear compartment; N > C for cells showing an exclusive or predominant nuclear localization of signal; and C > N for cells with an exclusive or predominant cytoplasmic distribution of signal. An additional classification category was included in the hormone-induced nuclear transfer studies, in which cells showing an exclusively nuclear fluorescence were attributed to the category N (see Fig. 1Go for example of the cellular classification system). Experiments were repeated three times and the fidelity of the classification system confirmed by three independent investigators in double-blind encryption studies.


Figure 1
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 1. Representative confocal images of the subcellular distribution classification. All images are from transfected COS-7 cells. A, Classification C > N (example from ratGR in the absence of hormone). B, C = N [mutant AB(rat)CDE(GR2)]. C, Representatives of both C = N and N > C (cells with asterisks) [mutant ABCD(GR1)E(rat)]. D, N > C [ABCD(GR2)E(rat)]. E, N (rtGR1 in the presence of 1 µM cortisol). Scale bars, 10 µM; except 15 µM (B) and 40 µM (E).

 
Western blot analysis
Cells growing in 12-well plates were transfected as above. Forty-eight hours after transfection, cells were harvested and whole-cell extracts prepared using a lysis buffer [20 mM Tris HCl (pH 7.5), 2 mM EDTA, 150 mM NaCl2, 0.5% Triton X-100, 0.1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 1 µM leupeptin]. After centrifugation of cell extracts (5 min at 10,000 rpm and 4 C), proteins (40 µg) from the supernatant were resolved on 10% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose. Membranes were blocked in Tris-buffered saline [10 mM Tris HCl (pH 7.4), 150 mM NaCl, 0.05% Tween 20] containing 5% nonfat dry milk, washed in Tris-buffered saline containing 1% nonfat dry milk, and incubated for 1 h at room temperature with a 1:8000 dilution of a primary antibody directed against GFP (JL-8; CLONTECH). Blots were then washed in 1% Tris-buffered saline, incubated for 1 h at room temperature with a horseradish peroxidase-labeled rabbit antimouse secondary antibody, and washed in 1% Tris-buffered saline. Membranes were allowed to react with chemiluminescent reagents (Pierce, Rockford, IL) and then processed for autoradiography.

Transactivation assay
Either COS-7 or CHO cells were used for the transactivation studies, previous studies have shown that the wild-type rtGR1 and rtGR2 transactivational activity in response to cortisol is the same in both cell lines (25). COS-7 or CHO cells growing in 12-well plates were transfected by the calcium precipitation method (29). Before transfection the medium was replaced with supplemented DMEM nutrient mix F-12 Ham (see above) containing 100 IU/ml penicillin, 100 µg/ml streptomycin, and 2.5% denatured, dextran/charcoal-treated FCS. Per 12-well plate, 1 µg GR receptor cDNA, 10 µg luciferase reporter pFC31Luc that contains the murine mammary tumor virus (MMTV) promoter, 2 µg control reporter pSVβ (CLONTECH), and 7 µg irrelevant DNA (pBluescript SK) were used. Twelve hours after transfection, the cells were washed with PBS, and the medium was replaced with fresh DMEM nutrient mix F-12 Ham containing antibiotics and dextran/charcoal-treated FCS (as above) and different concentrations of cortisol. The cells were incubated for further 36 h and harvested using reporter lysis buffer (Promega, Madison, WI). Luciferase activities were determined using a commercial luciferase assay (Promega). β-Galactosidase activities were measured as described before (31). All experiments were performed using triplicates and repeated three times independently. For each combination of plasmids used, a solvent control (ethanol) and a treatment with an optimal cortisol concentration (10–6 M) were included. Luciferase activity in controls lacking GR was negligible (data not shown). Luciferase activities were divided by β-galactosidase activities to correct for differences in transfection efficiency and expressed as percent of the value observed in the optimal cortisol treatment for the same receptor.

Statistical analysis
For experiments seeking to identify rtGR1 and rtGR2 regions influencing subcellular distribution, a range of statistical analyses were undertaken using Statistica (version 6.0; Statsoft Inc., Tulsa, OK). Principal components analysis (PCA) was used to reduce the dimensionality of the data and generate new variables (factors) that explained the maximal possible variability within the data. Before PCA analysis, data expressed as proportions (i.e. proportion of cells showing N > C, N = C, and C > N localization) were subjected to an arcsine transformation. The two best new explanatory factors generated by the PCA (i.e. those explaining the majority of the data set’s variability) were then used as the independent variables for general linear models (GLMs), which assessed whether domain had a significant effect (P < 0.05), in terms of a given factor for each combination of GR and cell type. After the GLM, differences between receptors were assessed in more detail using a Bonferroni post hoc test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cellular localization of GFP-tagged rtGR1 and rtGR2
The subcellular distribution of rainbow trout rtGR1 and rtGR2 was assessed using recombinant proteins in which GFP was added to the N terminus of the receptor. A construct encoding GPF-tagged rat GR mutant C656G (termed rGRC656G) (5) served as a mammalian GR control. Immunoblots of COS-7 cells transfected with the GFP-tagged receptors showed single bands of GFP-immunoreactive proteins of the expected molecular masses and similar expression levels among receptors (Fig 2AGo). The GFP-tagged trout glucocorticoid receptors induced murine mammary tumor virus (MMTV)-promoter activity in a cortisol-dependent fashion, albeit somewhat less efficiently than the wild-type receptors (Fig. 2BGo). The GFP-tagged receptors retained the same cortisol sensitivity as the wild-type receptors (25) (Fig. 2CGo). Upon expression of rGRC656G in COS-7 and in the absence of hormone treatment, the majority of cells showed a mainly cytoplasmic distribution of fluorescence (C > N ~65%) with the remaining cells having similar intensities of signal in the cytoplasm and nucleus (N = C ~35%) (Fig. 2DGo). The results are similar to those observed in previous studies using transient expression of wild-type rat GR in COS-7 cells (9). A markedly different pattern was observed for the trout receptors (Fig. 2DGo), in which hormone-free rtGR1 and rtGR2 was seldom found in the cytoplasm, with GFP-rtGR1 distribution being N > C 48 ± 1%, C = N 51 ± 2%, C > N 1 ± 1%, and GFP-rtGR2 being N > C 80 ± 4%, C = N 20 ± 3%, C > N 0 ± 0%).


Figure 2
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 2. Characterization of localization, function, and expression levels of GFP-tagged rtGR1 and rtGR2. A, Western Blot analysis and size marker of protein (kilodaltons) of the levels of GFP-tagged rat and trout glucocorticoid receptors transiently transfected into COS-7 cells. B, Transcriptional activation of MMTV-luciferase reporter gene in COS-7 cells cotransfected with rtGR1, GFP-rtGR1, rtGR2, or GFP-rtGR2 expression vectors as well as pSVβ plasmid and treated with either ethanol or 1 µM cortisol. Transactivation activity was normalized to the internal β-galactosidase control. C, Cortisol dose-response curve for rtGR1 ({blacksquare}), GFP-rtGR1 ({square}), rtGR2 ({blacktriangledown}) or GFP-rtGR2 ({triangledown}) transactivation. Values expressed as a percent of the respective construct activity measured in the presence of 1 µM cortisol. Values represent means ± SEM of three independent experiments, with each experiment performed in duplicate. Subcellular localization of GFP-rtGR1, GFP-rtGR2, and GFP-rGRC6565G fusion proteins transiently expressed in COS-7 cells (D) or the rainbow trout gonad cell lines RTG-2 (E) using Lipofectamin2000, in the absence of steroids, before and after a 30-min treatment with 1 µM cortisol determined by fluorescence microscopy. Quantifications are compilations of at least three independent experiments. Error bars, means ± SE.

 
To exclude the possibility that the unusual distribution of the trout receptors was related to the heterologous expression in COS-7, we repeated the experiments in RTG-2 cells, derived from rainbow trout gonads. GFP-rGRC656G expressed in RTG-2 cells exhibited a similar pattern of distribution as that observed in COS-7 cells (Fig. 2EGo). GFP-rtGR1 transiently expressed in RTG-2 cells showed very little cytoplasmic distribution (N > C 12 ± 8%, C = N 88 ± 8%, C > N 0 ± 0%) (Fig. 2EGo). The subcellular localization of the GFP-rtGR2 fusion protein differed from that of GFP-rtGR1 with a greater proportion of RTG-2 cells having a nuclear distribution in the absence of hormone (N > C 91 ± 9%, C = N 8 ± 7%, C > N 1 ± 1%) (Fig. 2DGo).

Upon steroid treatment (1 µM cortisol for 30 min), all GRs investigated transferred completely to the nucleus in both cellular systems (Fig. 2Go, D and E).

Temporal and dose-dependent pattern of GFP-rtGR1 and GFP-rtGR2 nuclear transfer
A key difference between the two trout GR isoforms is their distinct sensitivity to cortisol in transactivation (Fig. 2CGo) and transrepression assays (28). Thus, we evaluated whether a difference in dose-dependent and temporal kinetics of the nuclear translocation exists between GFP-rtGR1 and GFP-rtGR2 after treatment with the natural ligand cortisol. The results are expressed as a percentage of cells showing a mostly nuclear (N > C) or entirely nuclear (N) localization of GFP-GR. Nuclear translocation of the cytoplasmatic GFP-rtGR1, transiently expressed in COS-7 cells, occurred over a concentration range of cortisol (1 nM to 1 µM). After 5 min of treatment with 1 µM cortisol, all cells transfected with GFP-rtGR1 showed a mostly nuclear localization of signal, whereas with 1 nM cortisol, only 60% of cells displayed a mostly nuclear localization of GFP-rtGR1 after 30 min (Fig. 3AGo). By comparison, cortisol induced the nuclear translocation of GFP-rtGR2 more efficiently than of GFP-rtGR1, with a majority of cells (80%) showing a mainly nuclear localization of fluorescence after 5 min of treatment with 1 nM cortisol (Fig. 3BGo). A similar picture arises when only cells with an entirely nuclear localization of signal are taken into account. Thirty minutes after addition of 1 nM cortisol, the signal was entirely nuclear in about 70% of cells expressing GFP-rtGR1 but in only 25% of cells expressing GFP-rtGR2 (Fig. 3BGo). Similar differences in sensitivity to cortisol between the two receptors was seen in RTG-2 cell lines (data not shown). However, there appeared to be a difference between the two cell lines, with the nuclear localization response to cortisol being slower and slightly less sensitive in the RTG-2 cells. An explanation for this difference between the two cell lines may be because they are cultured at different temperatures (COS-7 cells at 37 C; RTG-2 cells at 21 C). In addition, RTG-2 cells also possess cortisol-binding sites (32) and an endogenous GR population, as determined by quantitative PCR, which may compete with GFP-tagged receptors for the hormone (Becker, H., and K. Schirmer, unpublished data).


Figure 3
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 3. Kinetics of GFP-rtGR1 and GFP-rtGR2 nuclear transfer in transiently transfected COS-7 cells after treatment with different concentrations of cortisol. Cells were transiently transfected with GFP-rtGR1 (A) and GFP-rtGR2 (B). Nuclear transfer of the receptors was induced on addition of 1 µM ({blacksquare}), 100 nM ({blacktriangleup}), 10 nM ({blacktriangledown}), 1 nM ({diamondsuit}), 100 pM (bullet) cortisol, and the solvent ethanol ({square}). For 30-min changes in subcellular distribution was evaluated by confocal laser-scanning microscopy. Of each time point analyzed, an average of 20–80 cells were quantified for the cellular localization of the receptor. Values represent mean ± SEM from three experiments.

 
Identification of rtGR1 and rtGR2 regions influencing nuclear distribution and transactivation hormone sensitivity
To determine the receptor regions causing the unexpected nucleocytoplasmic distribution of the trout GRs in the absence of hormone, we generated a number of domain swaps between GFP-rGRC656G (termed ratGR when discussed in context of the mutants) and GFP-rtGR1 or GFP-rtGR2 and investigated their subcellular distribution in COS-7 and RTG-2 cells. All GFP-tagged mutant receptors were fully nuclear after 5 min of treatment with 1 µM cortisol (data not shown). To ascertain that the recombinant receptors were functional, we investigated their potential to activate the MMTV promoter in a cortisol-dependent fashion, using a transient cotransfection transactivation assay in CHO-cells, a hamster cell line that lacks functional endogenous corticosteroid receptors. The transactivation activity of wild-type and recombinant GRs was expressed in terms of percent of the response of the same receptor at a cortisol concentration of 1 µM (Fig. 4Go). For a number of receptors, transactivation activities with intermediary cortisol concentrations were higher than those observed at 1 µM cortisol, resulting in higher than 100% values after normalization. Similar curves have been observed with human GR and MR in different cell types (33, 34), and the data possibly reflect effects of the higher cortisol concentrations on receptor expression. However, the ascending part of the dose-response curves is indicative of the hormone sensitivity of a given receptor, and we confirmed that rtGR2 is transcriptionally activated at lower cortisol concentrations than rtGR1, irrespective of the presence of a GFP-tag or cell type [Figs. 2Go and 4AGo (25)]. The C656G mutation of the rat GR increases its sensitivity to glucocorticoids about 500-fold (35). Consequently, the transactivation activity of GFP-ratGRC656G was increased above background at cortisol levels of 0.1 nM (10–10 M) and above (Fig. 4AGo). All domain-swap mutant GRs were capable of activating the MMTV promoter in a hormone-dependent fashion, confirming that the mutagenesis yielded functional GR receptors (Fig. 4Go, B and C). A comparison of curves for the cortisol-dependent receptor transactivation between original (Fig. 4AGo) and chimeric receptors derived from rGRC656G and rtGR1 (Fig. 4BGo) or rGRC656G and rtGR2 (Fig. 4CGo) revealed that the cortisol sensitivity of transactivation potential was mainly determined by the origin of the E domain of the chimera.


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 4. Transactivation activities of mutant constructs. Cortisol concentration transactivation activity curves for wild-type and GFP-tagged rtGR1, rtGR2, and rat GR-C6565G receptors (A); rtGR1 mutants (B); and rtGR2 mutants (C). Transfection normalization procedure follows those described in the legend of Fig. 2Go, except CHO cells were used, and error bars have been omitted for clarity. See figure for description of the symbols that represents a specific mutant. Values represent mean ± SEM from three experiments.

 
The different GFP-tagged recombinant and wild-type receptors were transiently expressed in COS-7 and RTG-2 cells and their nucleocytoplasmic localization determined in the absence of hormone (Fig. 5Go, A–D). PCA was performed to facilitate the comparison between the receptors and revealed three factors, of which the first factor accounted for 59% and the second factor for 36% of the variation (Fig. S1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org). The variables N > C and C > N were best discriminated by the first factor, whereas C = N was best discriminated by the second factor (supplemental data Fig. S2). To compare receptors, GLMs were constructed using the first and second PCA factors as independent variables. For each of the four experiments shown in Fig. 5Go, A–D, the differences among receptors were highly significant for both factor one (P < 0.001) and two (P < 0.01). Individual receptors within each panel were further subjected to post hoc comparisons. As expected from the scree plot (supplemental data Fig. S1), comparisons using factor one were more informative than those using factor two. Where significant differences between receptors were indicated for factor two, these were in most cases also detected for factor one. Consequently, we report the post hoc comparisons using factor one in Fig. 5Go, and, for completeness, report the post hoc tests for factor two in supplemental data Tables S1 and S2. The results from the factor one analysis will be considered from here on, except in one case in which a significant difference was detected by factor 2 (supplemental data Table S1) and not detected by factor one (Fig. 5Go).


Figure 5
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 5. Subcellular localization of rtGR1, rtGR2, and rat GR-C6565G domain swap mutants. Domain-swap mutants were created by introducing A/B, CD, or E domain of rtGR1 or rtGR2 into the context of the rat GR-C656G (termed rat in the context of the mutants). The left panel provides a visual representation of the mutants generated, and the rat and trout GR domains are represented by the white and hatched rectangles, respectively. Mutants derived from rtGR1 were transiently transfected into COS-7 cells (A) and RTG-2 cells (C) and mutants derived from rtGR2 in COS-7 cells (B) and RTG-2 cells (D). Subcellular localization of receptor mutants was determined using fluorescence microscopy. Of each field of view, 40–60 cells were evaluated for subcellular distribution. All quantifications are compilations of at least three independent experiments, and data represent the means ± SE. PCA was used to determine the factors that explained the majority of the data set’s variability, and these factors were used as the independent variables in GLMs, followed by Bonferroni post hoc test, to establish interreceptor/interchimera differences (P < 0.05); see text for more details.{dagger}, Significant difference from the wild-type trout GR1 nuclear distribution pattern; {dagger}, Significant differences from the wild-type trout GR1 nuclear distribution patern; §, significant differences from the rGR C656G nuclear distribution pattern; {ddagger}, significant differences from the wild-type trout GR2 nuclear distribution pattern; three symbols, P < 0.001; two symbols, P < 0.01; and one symbol, P < 0.05.

 
When the A/B domain in rGRC656G was replaced with that of rtGR1, the resulting chimera AB(GR1)CDE(rat) was significantly shifted toward the nucleus when compared with rGRC656G in both cell types (Fig. 5Go, A and C), suggesting a role of the rtGR1 A/B domain in determining the unusual localization of naïve rtGR1. In line with this, the complementary mutant AB(rat)CDE(GR1) did not differ significantly from rGRC656G in distribution (Fig. 5Go, A and C). Receptor AB(rat)CD(GR1)E(rat), in which the C and D domains of rGRC656G were substituted for that of rtGR1, resembled rGRC656G in distribution in both cell types (Fig. 5Go, A and C). However, the slight shift of AB(rat)CD(GR1)E(rat) to the nuclear compartment when compared with rGRC656G was deemed significant in RTG-2 cells by post hoc tests on factor two (supplemental data Table S1), suggesting that the combined C and D regions might also contribute to the partially nuclear localization of naive rtGR1. A role of the central region of rtGR1 was also suggested by the distribution of the chimera ABCD(GR1)E(rat), which was more nuclear than AB(GR1)CDE(rat) in both cell types (Fig. 5Go, A and C). However, the effect of the combined rtGR1 C and D regions was context dependent because AB(rat)CDE(GR1) equaled rGRC656G in distribution (Fig 5Go, A and C). The localization of ABCD(rat)E(GR1) was indistinguishable from that of rGRC656G (Fig. 5Go, A and C), arguing against an effect of the E domain of rtGR1 in the partially nuclear localization of the naïve rtGR1.

Results obtained for chimeras of rtGR2 and rGRC656G showed similar trends to those observed in rtGR1 derived mutants. As observed with rtGR1, the introduction of the A/B domain of rtGR2 into the context of rGRC656G induced a more nuclear distribution than observed for rGRC656G [Fig. 5Go, B and D; AB(GR2)CDE(rat)], whereas the replacement of the E domain of rGRC656G with that of rtGR2 had no effect [Fig. 5Go, B and D; ABCD(rat)E(GR2)]. However, in rtGR2-derived chimera with rGRC656G, the effects of the combined C and D regions appeared to be clearer than in corresponding mutants derived from rtGR1. For instance, the subcellular localization of AB(rat)CD(GR2)E(rat) was significantly different from that of rGRC656G in both cell types (Fig. 5Go, B and D). Similarly, combining the A/B, C, and D domains of rtGR2 with the E domain of rGRC656G resulted in a receptor indistinguishable from rtGR2 in distribution (Fig. 5Go, B and D), whereas the corresponding rtGR1 derived mutant, (ABCD(GR1)E(rat), still differed from rtGR1 in phenotype (Fig. 5Go, A and C). Finally, the mutant AB(rat)CDE(GR2) was more nuclear than rGRC656G in RTG-2 (Fig. 5DGo), whereas AB(rat)CDE(GR1) showed a rGRC656G-like distribution in both cell types.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present study is the first to investigate the subcellular localization of GRs in a lower vertebrate, the teleost rainbow trout. To our surprise, a significant fraction of both trout GR isoforms (rtGR1 and rtGR2) was located in the nucleus in the absence of hormone. This nucleocytoplasmic distribution pattern differs markedly from other vertebrate GRs studied so far, which reside in the cytoplasm in the hormone-free inactive state [e.g. rat (5), mouse (22), human (19), and squirrel monkey (23)], and the distribution pattern of the trout GRs resembles that of mammalian MRs (18, 30).

To identify the regions of the receptor responsible for the unusual subcellular distribution of the naive trout GRs, we investigated the behavior of chimeras of rtGR1 and rtGR2 with rGRC656G. Replacing the A/B domain of rGRC656G with that of either rtGR1 or rtGR2 provoked a shift of the resulting mutant receptor to the nucleus, albeit not to the extent observed in unadulterated rtGR1 and rtGR2 (Fig. 5Go). The converse mutants, substituting the A/B domain of either rtGR1 or rtGR2 for that of rGRC656G yielded a subcellular distribution indistinguishable from rGRC656G in the absence of hormone. This result demonstrates a role of the A/B domain in the partially nuclear localization of ligand-free rtGR1 and rtGR2. But the situation is more complex because mutants possessing the A/B, C, and D domains from the trout GRs show a nuclear distribution phenotype pattern similar to that of GFP tagged wild-type trout receptors, which suggests that in certain contexts the central CD region also influence nucleocytoplasmic localization.

A role for the A/B domain in the subcellular localization has only been reported for the rat MR (30). The rat MR A/B domain does not contain classical basic NL signals but possesses a motif (562KPHGDLLSSRR571) reminiscent of a nonconventional NL signals characterized in Borna virus protein p10 (30, 36); however, this region is not responsible for the nucleocytoplasmic distribution pattern of naïve MR (30). Instead, the MR A/B domain harbors a serine/threonine rich NL signal called NL0 that is believed to be regulated by phosphorylation and influence the proteins cellular distribution (30). The nature of the putative NL region(s) in the A/B domain of the trout GRs is (are) uncertain. The rtGR1 A/B domains possess a basic motif (274KQENDRR280) that resembles the NLS of Borna disease virus protein p10 (6RLTLLELVRR15) (35), but these motifs are lacking in the A/B domain of rtGR2. The A/B domains of both rtGR1 and rtGR2 contain several S/T rich regions, and based on NetPhos 2.0 server (http://www.cbs.dtu.dk/services/NetPhos/), these are potential phosphorylation sites, but the role of phosphorylation in naïve trout receptor cellular localization is still to be determined.

The A/B domain, however, is not solely responsible for the nuclear distribution pattern in the absence of hormone, and in certain contexts the central region of the receptors, defined by the combined C and D domains, was found to influence the localization of chimeric receptors with rGRC656G (see Fig. 5Go, A–D). For example, the introduction of the combined C and D domains of rtGR2 into the context of rGRC656G provoked a significant shift of subcellular distribution to the nucleus in both COS-7 cells and RTG-2 cells. Moreover, the rtGR2 A/B, C, and D domains were together capable of conferring a phenotype indistinguishable from that of rtGR2 when combined with the rGRC656G E domain. In contrast, when the central regions of rtGR1 were combined with the remaining domains of rGRC656G, the resulting chimera behaved like rGRC656G, but, if introduced in combination with the rtGR1 A/B domain, the resulting chimera showed a greater nuclear distribution than those that possessed only the rtGR1 A/B domain, although none of the chimeras showed the full extent of nuclear localization displayed by wild-type rtGR1.

The central C and D domains of GR contain a number of motifs that regulate nucleocytoplasmic shuttling in steroid receptors: the NL1 site (37), the NES signal between the linker region of the two zinc fingers of the DNA binding domain (11), and an NRS that has been mapped to amino acids 500–525 of the D region of rGR and overlaps with the C-terminal end of NL1 (12). The trout GRs show anomalies in the amino acid sequences of each of these regions: the NL1 site is comprised of a tripartite motif that is conserved in both trout GRs and the rat MR, except for a two-amino acid insertion at the same position in the core region of NL1 (rtGR1: 471LI472; rtGR2: 387LN388; rMR: 679LL680); the zinc-fingers linker region in trout is characterized by amino acid insertions (25, 38). The NRS motif, like the NL1, is imperfectly conserved in rtGR1 and rtGR2. Thus, future studies will be needed to unravel the extent to which these motifs participate in regulating nuclear localization of the trout GRs in the absence of hormone.

The cytoplasmic fractions of naïve rtGR1 and rtGR2 transfer rapidly into the nucleus after cortisol treatment in both mammalian and rainbow trout cell lines (Fig. 3Go). rtGR1 and rtGR2 showed distinct differences in their sensitivity in terms of nuclear transfer on exposure to cortisol, a characteristic that mimics exactly those observed in transactivational studies [Figs. 2Go (COS-7 cells) and 4 (CHO cells)]. The molecular basis for this difference in transactivational sensitivity is unrelated to the A/B domain or CD region but is clearly controlled by the E domain. This is because all mutant constructs possessing a rtGR2 E domain showed a similar transactivational sensitivity profile to the wild-type rtGR2 or GFP-tagged rtGR2 (Fig. 4CGo); similarly, mutants containing the E domain of rtGR1 or rGRC656G show the same transactivation profile of the respective wild-type receptors (Fig. 4Go). This observation may have been expected because the E domain is known to mediate ligand-induced GR transcription (2) and possesses the NL2 signal that is pivotal in hormone-induced translocation of the mammalian GR to the nucleus (4).

The duplication of the GR in the teleost lineage occurred around 355 million years ago (24), and retention of the two GR isoforms suggests they are not redundant and have undergone sub- or neofunctionalization. But presently there is no clear evidence that the two teleost GRs control different regulatory pathways. The proposed molecular mechanisms [e.g. ligand exploitation (17) or molecular exploitation (20)] that enabled the evolution of the steroid receptor family are dependent on changes in receptor hormone binding preference and are thus unlikely to explain the retention of the two teleost GRs because both use a single hormone cortisol (25, 26). The observation that the nuclear localization of rtGR2 was more pronounced than that of rtGR1, that migrations of the remaining cytoplasmic rtGR2 proteins occurred at lower hormone concentrations, compared with rtGR1, and that components within the E domain determine the differences in the cortisol-induced transactivational sensitivity profiles suggests alternative divergent characteristics that may have been beneficial and allowed the two receptors to flourish after duplication.

In summary, components within the A/B domain of the receptor, especially if linked to the corresponding trout CD region, are responsible for the unexpected partial nuclear localization of hormone-free trout GRs, a nucleocytoplasmic distribution pattern that resembles that of mammalian MRs. Further analysis of other teleost GR cellular distribution patterns will ascertain how widespread a GR nuclear distribution pattern in the absence of hormone is in this vertebrate lineage and whether cytoplasmic localization of GR in the absence of hormone is a derived trait in mammals.


   Acknowledgments
 
We thank different colleagues for kindly placing plasmids at our disposal. Professor G. Hager provided pcI-nGFP-C656G, Dr. B. Ducouret supplied pCMrtGR, Dr. P. Prunet gave pCMrtGR2, and pFC31Luc was a gift from Dr. F. Gouilleux.


   Footnotes
 
This work was supported by Grant S18960 from the Biotechnology and Biological Sciences Research Council (to N.R.B.).

Disclosure Summary: H.B., A.S., J.E.B., K.S., and N.R.B. have nothing to declare.

First Published Online May 15, 2008

1 H.B. and A.S. contributed equally to this work. Back

Abbreviations: FCS, Fetal calf serum; GFP, green fluorescent protein; GLM, general linear model; GR, glucocorticoid receptor; hsp, heat shock protein; MMTV, murine mammary tumor virus; MR, mineralocorticoid receptor; NES, nuclear export signal; NL, nuclear localization; NLS, nuclear localization signals; NRS, nuclear retention signal; PCA, principal components analysis; rt, rainbow trout.

Received December 5, 2007.

Accepted for publication May 6, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Charmandari E, Tsigos C, Chrousos G 2005 Endocrinology of the stress response. Annu Rev Physiol 67:259–284[CrossRef][Medline]
  2. Bamberger CM, Schulte HM, Chrousos GP 1996 Molecular determinants of glucocorticoid receptor function and tissue sensitivity to glucocorticoids. Endocr Rev 17:245–261[Abstract/Free Full Text]
  3. Glass CK, Rosenfeld MG 2000 The coregulator exchange in transcription of nuclear receptors. Genes Dev 14:121–141[Free Full Text]
  4. Picard D, Yamamoto K 1987 Two signals mediate hormone-dependent nuclear localization of the glucocorticoid receptor. EMBO J 6:3333–3340[Medline]
  5. Htun H, Barsony J, Renyi I, Gould DL, Hager GL 1996 Visualization of glucocorticoid receptor translocation and intranuclear organization in living cells with a green fluorescent protein chimera. Proc Natl Acad Sci USA 93:4845–4850[Abstract/Free Full Text]
  6. Tissing, WJE, Meijerink JPP, den Boer ML, Brinkhof B, Pieters R 2005 mRNA expression levels of (co)chaperone molecules of the glucocorticoid receptor are not involved in glucocorticoid resistance in pediatric ALL. Leukemia 19:727–733[CrossRef][Medline]
  7. Pratt WB, Toft D 2003 Regulation of signaling protein function and trafficking by the hsp90/hsp70-based chaperone machinery. Exp Biol Med 228:111–133[Abstract/Free Full Text]
  8. Sackey FN, Haché RJ, Reich T, Kwast-Welfeld J, Lefebvre YA 1996 Determinants of subcellular distribution of the glucocorticoid receptor. Mol Endocrinol 10:1191–1205[Abstract/Free Full Text]
  9. Savory JG, Hsu B, Laquian IR, Giffin W, Reich T, Hache RJ, Lefebvre YA 1999 Discrimination between NL1- and NL2-mediated nuclear localization of the glucocorticoid receptor. Mol Cell Biol 19:1025–1037[Abstract/Free Full Text]
  10. Freedman ND, Yamamoto KR 2004 Importin 7 and importin {alpha}/importin β are nuclear import receptors for the glucocorticoid receptor. Mol Biol Cell 15:2276–2286[Abstract/Free Full Text]
  11. Black BE, Holaska JM, Rastinejad F, Paschal BM 2001 DNA binding domains in diverse nuclear receptors function as nuclear export signals. Curr Biol 11:1749–1758[CrossRef][Medline]
  12. Carrigan A, Walther RF, Salem HA, Wu D, Atlas E, Lefebvre YA, Hache RJ 2007 An active nuclear retention signal in the glucocorticoid receptor functions as a strong inducer of transcriptional activation. J Biol Chem 282:10963–10971[Abstract/Free Full Text]
  13. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895[Abstract/Free Full Text]
  14. Picard D, Kumar V, Chambon P, Yamamoto KR 1990 Signal transduction by steroid hormones: nuclear localization is differentially regulated in estrogen and glucocorticoid receptors. Cell Regul 1:291–299[Medline]
  15. Simental JA, Sar M, Lane MV, French FS, Wilson EM 1991 Transcriptional activation and nuclear targeting signals of the human androgen receptor. J Biol Chem 266:510–518[Abstract/Free Full Text]
  16. Ylikomi T, Bocquel MY, Berry M, Gronemeyer H, Chambon P 1992 Cooperation of proto-signals for nuclear accumulation of estrogen and progesterone receptors. EMBO J 11:3681–3694[Medline]
  17. Thornton JW 2001 Evolution of vertebrate steroid receptors from an ancestral estrogen receptor by ligand exploitation and serial genome expansions. Proc Natl Acad Sci USA 98:5671–5676[Abstract/Free Full Text]
  18. Fejes-Toth G, Pearce D, Naray-Fejes-Toth A 1998 Subcellular localization of mineralocorticoid receptors in living cells: effects of receptor agonists and antagonists. Proc Natl Acad Sci USA 95:2973–2978[Abstract/Free Full Text]
  19. Wan Y, Coxe KK, Thackray VG, Housley PR, Nordeen SK 2001 Separable features of the ligand-binding domain determine the differential subcellular localization and ligand-binding specificity of glucocorticoid receptor and progesterone receptor. Mol Endocrinol 15:17–31[Abstract/Free Full Text]
  20. Bridgham JT, Carroll SE, Thornton JW 2006 Evolution of hormone-receptor complexity by molecular exploitation. Science 312:97–101[Abstract/Free Full Text]
  21. Ortlund EA, Bridgham JT, Redinbo MR, Thornton JW 2007 Crystal structure of an ancient protein: evolution by conformational epistasis. Science 317:1544–1548[Abstract/Free Full Text]
  22. Galigniana MD, Radanyi C, Renoir JM, Housley PR, Pratt WB 2001 Evidence that the peptidylprolyl isomerase domain of the hsp90-binding immunophilin FKBP52 is involved in both dynein interaction and glucocorticoid receptor movement to the nucleus. J Biol Chem 276:14884–14889[Abstract/Free Full Text]
  23. Her S, Patel PD, Schatzberg AF, Lyons DM 2005 Mutations in squirrel monkey glucocorticoid receptor impair nuclear translocation. J Steroid Biochem Mol Biol 94:319–326[CrossRef][Medline]
  24. Jaillon O, Aury JM, Brunet F, Petit JL, Stange-Thomann N, Mauceli E, Bouneau L, Fischer C, Ozouf-Costaz C, Bernot A, Nicaud S, Jaffe D, Fisher S, Lutfalla G, Dossat C, Segurens B, Dasilva C, Salanoubat M, Levy M, Boudet N, Castellano S, Anthouard V, Jubin C, Castelli V, Katinka M, et al. 2004 Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype. Nature 431:946–957[CrossRef][Medline]
  25. Bury NR, Sturm A, Le Rouzic P, Lethimonier C, Ducouret B, Guiguen Y, Robinson-Rechavi M, Laudet V, Rafestin-Oblin ME, Prunet P 2003 Evidence for two distinct functional glucocorticoid receptors in teleost fish. J Mol Endocrinol 31:141–156[Abstract]
  26. Greenwood AK, Butler PC, White RB, DeMarco U, Pearce D, Fernald RD 2003 Multiple corticosteroid receptors in a teleost fish: distinct sequences, expression patterns, and transcriptional activities. Endocrinology 144:4226–4236[CrossRef][Medline]
  27. Stolte EH, van Kemenade BM, Savelkoul HF, Flik G 2006 Evolution of glucocorticoid receptors with different glucocorticoid sensitivity. J Endocrinol 190:17–28[Abstract/Free Full Text]
  28. Bury NR, Sturm A 2007 Evolution of the corticosteroid receptor signalling pathway in fish. Gen Comp Endocrionol 153:47–56[CrossRef]
  29. Sambrook J, Russell DW 2001 Molecular cloning. 3rd ed. Cold Spring Harbor, NY: CSHL
  30. Walther RF, Atlas E, Carrigan A, Rouleau Y, Edgecombe A, Visentin L, Lamprecht C, Addicks GC, Haché RJ, Lefebvre YA 2005 A serine/threonine-rich motif is one of three nuclear localization signals that determine unidirectional transport of the mineralocorticoid receptor to the nucleus. J Biol Chem 280:17549–17561[Abstract/Free Full Text]
  31. Sturm A, Bury N, Dengreville L, Fagart J, Flouriot G, Rafestin-Oblin ME, Prunet P 2005 11-Deoxycorticosterone is a potent agonist of the rainbow trout (Oncorhynchus mykiss) mineralocorticoid receptor. Endocrinology 146:47–55[CrossRef][Medline]
  32. Lee LE, Bols NC 1989 The corticosteroid receptor and the action of various steroids in rainbow trout fibroblasts. Gen Comp Endocrinol 74:85–95[CrossRef][Medline]
  33. Lim-Toi SS, Keightly M-C, Fuller PJ 1997 Determinants of specificity of transactivation by the mineralocorticoid or glucocorticoid receptor. Endocrinology 138:2537–2543[Abstract/Free Full Text]
  34. Fuller PJ, Lim-Toi SS, Brennan FE 2000 Specificity in mineralocorticoid verus glucocorticoid action. Kidney Int 57:1256–1264[CrossRef][Medline]
  35. Kucera T, Waltner-Law M, Scott DK, Prasad R, Granner DK 2002 A point mutation of the AF2 transactivation domain of the glucocorticoid receptor disrupts its interaction with steroid receptor coactivator 1. J Biol Chem 277:26098–26102[Abstract/Free Full Text]
  36. Wolff T, Unterstab G, Heins G, Richt JA, Kann M 2002 Characterization of an unusual importin {alpha} binding motif in the Borna disease virus p10 protein that directs nuclear import. J Biol Chem 277:12151–12157[Abstract/Free Full Text]
  37. Tang Y, Ramakrishnan C, Thomas J, DeFranco DB 1997 A role for HDJ-2/HSDJ in correcting subnuclear trafficking, transactivation, and transrepression defects of a glucocorticoid receptor zinc finger mutant. Mol Biol Cell 8:795–809[Abstract]
  38. Ducouret B, Tujague M, Ashraf J, Mouchel N, Servel N, Valotaire Y, Thompson EB 1995 Cloning of a teleost fish glucocorticoid receptor shows that it contains a deoxyribonucleic acid-binding domain different from that of mammals. Endocrinology 136:3774–3783[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Becker, H.
Right arrow Articles by Bury, N. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Becker, H.
Right arrow Articles by Bury, N. R.

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