discovery of glucocorticoid receptor- in mice with a … of glucocorticoid receptor- in mice with a...

Discovery of Glucocorticoid Receptor-� in Mice with aRole in Metabolism

Terry D. Hinds, Jr., Sadeesh Ramakrishnan, Harrison A. Cash,Lance A. Stechschulte, Garrett Heinrich, Sonia M. Najjar,and Edwin R. Sanchez

Center for Diabetes and Endocrine Research and the Department of Physiology and Pharmacology,University of Toledo College of Medicine, Toledo, Ohio 43614-5804

Glucocorticoid hormones control diverse physiological processes, including metabolism and im-munity, by activating the major glucocorticoid receptor (GR) isoform, GR�. However, humansexpress an alternative isoform, human (h)GR�, that acts as an inhibitor of hGR� to produce a stateof glucocorticoid resistance. Indeed, evidence exists that hGR� contributes to many diseases andresistance to glucocorticoid hormone therapy. However, rigorous testing of the GR� contributionhas not been possible, because rodents, especially mice, are not thought to express the �-isoform.Here, we report expression of GR� mRNA and protein in the mouse. The mGR� isoform arises froma distinct alternative splicing mechanism utilizing intron 8, rather than exon 9 as in humans. Thesplicing event produces a form of � that is similar in structure and functionality to hGR�. Mouse(m)GR� has a degenerate C-terminal region that is the same size as hGR�. Using a variety of newlydeveloped tools, such as a mGR�-specific antibody and constructs for overexpression and shorthairpin RNA knockdown, we demonstrate that mGR� cannot bind dexamethasone agonist, isinhibitory of mGR�, and is up-regulated by inflammatory signals. These properties are the same asreported for hGR�. Additionally, novel data is presented that mGR� is involved in metabolism.When murine tissue culture cells are treated with insulin, no effect on mGR� expression wasobserved, but GR� was elevated. In mice subjected to fasting-refeeding, a large increase of GR�

was seen in the liver, whereas mGR� was unchanged. This work uncovers the much-needed rodentmodel of GR� for investigations of physiology and disease. (Molecular Endocrinology 24:1715–1727, 2010)

Human glucocorticoid receptor (hGR) is expressed astwo major isoforms: hGR� and hGR� (1, 2). Glu-

cocorticoid hormones (GCs) control diverse physiologi-cal processes (3, 4), such as metabolism, immunity/in-flammation, development, and behavior. These responsesare a direct result of GR� activity as a hormone-activatedtranscription factor (5, 6). In contrast, the role of GR� inGC control of physiology is still poorly understood. Mostrecent studies suggest that GR� acts as an inhibitor ofGR� (7–10) to produce a state of glucocorticoid resis-

tance (1, 2). Indeed, there is indirect evidence that ele-vated expression of GR� may be responsible for a varietyof immunological diseases. Severe asthma, leukemia, ul-cerative colitis, chronic sinusitis, systemic lupus erythem-atosus, and possibly cigarette smoking all correlate withoverexpression of GR� (2, 11–13). Many patients suffer-ing from these diseases are refractory to GC treatment.Not surprisingly, increased activation of proinflamma-tory transcription factors and cytokines has also beennoted in cases of GC resistance with elevated GR� expres-

ISSN Print 0888-8809 ISSN Online 1944-9917Printed in U.S.A.Copyright © 2010 by The Endocrine Societydoi: 10.1210/me.2009-0411 Received October 5, 2009. Accepted June 17, 2010.First Published Online July 21, 2010

Abbreviations: BPS, Branch point sequences; BLAST, Basic Local Alignment Search Tool;Dex, dexamethasone; FKBP51, FK506 binding protein 51; GC, glucocorticoid hormone;GFP, green fluorescent protein; GILZ, glucocorticoid-inducible leucine zipper; G6Pase,glucose-6-phosphatase; GR, glucocorticoid receptor; hGR, human GR; MEF, mouse em-bryonic fibroblast; mGR, mouse GR; NCBI, National Center for Biotechnology Information;NF-�B, nuclear factor �B; PDK4, pyruvate dehydrogenase kinase-4; qPCR, quantitativereal-time PCR; rMGRB-Ab, rabbit polyclonal antibody to mouse GR�; shRNA, short hairpinRNA; SR, serine-arginine; TBS, Tris-buffered saline; zGR, zebrafish GR.

O R I G I N A L R E S E A R C H

Mol Endocrinol, September 2010, 24(9):1715–1727 mend.endojournals.org 1715

sion. These observations suggest an important role forGR� as a homeostatic mechanism in the normal attenu-ation of GC responses and as a possible culprit in hor-mone-resistant disease states.

The hGR gene was cloned and sequenced in 1985,revealing the expression of hGR� and hGR� (14). Addi-tional studies showed that the isoforms result from alter-native splicing to yield GRs identical through amino acid727, but which differ in their C-terminal regions. ThehGR� C terminus is composed of 50 amino acids contain-ing important sites for hormone binding, as well as helix12, which provides critical transcriptional activation ac-tivity as a site for coregulator interaction (15). In contrast,the unique and nonhomologous C terminus of hGR� is adisordered 15-amino acid region of no known function.Not surprisingly, hGR� cannot bind GC agonists (7, 16).However, binding by RU486 antagonist, although dis-puted (17), has been shown by one laboratory (18).Although hGR� contains activation function-1 andDNA-binding domains identical to those in hGR�, notranscriptional activation or repression activities in re-sponse to hormone have yet been found for this isoform.Instead, most data point to hGR� as an inhibitor of hGR�

activity, either through competition for coregulators orthrough formation of inactive �/� heterodimers. Consis-tent with this mechanism is the predominant presence ofhGR� in the nucleus of most cells, whereas hGR� residesin the cytoplasm, undergoing nuclear translocation in re-sponse to ligand (19). Thus, hGR� can be viewed as adominant-negative inhibitor of hGR�, a mechanism ofaction which may underlie the potential role of GR� inGC resistance. However, two recent studies using genearray analyses have revealed that hGR� can constitutivelyregulate genes not controlled by hGR� (17, 18). There-fore, hormone-free hGR�, in addition to its dominant-negative activity, appears to have an intrinsic gene regu-latory function important to physiological responsesdistinct from hGR�.

The only observation of GR� outside humans has been inzebrafish (20). However, when the mouse GR (mGR) wasoriginally cloned and sequenced, one active GR was discov-ered that responded to GCs (21), but two different mRNAswere found with distinct poly-A tails (22). Moreover, anintact mGR protein was identified that was unable to bindhormone (23). Curiously, the alternative isoform of mGRwas not pursued, and it is now generally accepted that ro-dents do not express GR�. This conventional wisdom owesits existence to studies designed to discover mGR� based onthe hGR� process. In humans, GR� and GR� share exons1–8 but diverge to contain exons 9� and 9�, respectively,based on alternative usage of splice acceptor sites in exon 9(24). Efforts to discover GR� based on similar splicing

events in rodents and sheep have been unsuccessful (25, 26).The recent discovery of GR� in zebrafish has shown thatsplicing can also occur, not in exon 9, but through alterna-tive donor sites in intron 8, to yield a zebrafish GR� (zGR�)with properties similar to hGR�. In this work, we demon-strate the existence of GR� mRNA and protein in themouse. Like zGR�, mGR� results from alternative us-age of donor sites, in this case, in a region previouslyassigned as intron 8 of the mGR gene. Although thesplicing mechanism differs from that which generateshGR�, mGR� contains a 15-amino acid C terminusthat is identical in length and highly similar in sequenceto hGR�. In contrast, the C terminus of zGR� is muchlarger (50 amino acids). Mouse GR� shares similarproperties with hGR�, including ubiquitous expressionin mouse organs, lack of responsiveness to the GC ag-onist dexamethasone (Dex), and an ability to inhibitmGR� activity. This study has uncovered the much-needed rodent model of GR�, providing a new tool forfuture in vivo investigations of glucocorticoid resis-tance and sensitivity in physiology and disease.

Results

The mGR� and mGR� isoformsTo identify mGR�, the National Center for Biotechnol-

ogy Information (NCBI) website (www.ncbi.nlm.nih.gov)was used for the nuclear receptor subfamily 3, group C,member 1 (Mus musculus), and information was down-loaded for the complete GR genomic DNA sequence. In-formation regarding introns and exons in the mGR gene(ENSMUSG00000024431) was downloaded from En-sembl Mouse GeneView (www.ensembl.org). A singlemGR genomic sequence was identified. The mGR gene islocated on chromosome 18 and consists of at least nineexons (Fig. 1). The Drosophila website for splice site predic-tions (http://www.fruitfly.org/seq_tools/splice.html) wasused to characterize possible acceptor and donor alterna-tive splice sites within mGR (see Supplemental Figs. 1 and2, published on The Endocrine Society’s Journals Onlineweb site at http://mend.endojournals.org). In silico map-ping revealed that intron 8 was the most likely target ofalternative splicing, yielding several donor and acceptorsites with high splice-site prediction scores. Interestingly,no splice-site predictions were found in exon 8. Predictedsplice sites were also found in exon 9, in the distal regionof the exon where alternative splicing is known to occurin the hGR gene (7, 16). For initial screening of mGR�,primers were designed to test for utilization of the in-tron 8 and exon 9 sites, followed by RT-PCR analysisof total RNA from mouse embryonic fibroblast (MEF)cells (Fig. 1A). Consistent with prior published efforts

1716 Hinds et al. Murine GR� Mol Endocrinol, September 2010, 24(9):1715–1727

www.ncbi.nlm.nih.gov

(25), no PCR product was observed using a primerspecific to the distal region of exon 9 (Fig. 1B). How-ever, all three primers in the intron 8 region yielded

PCR products, suggesting the existence ofmGR transcripts derived from intron 8.

To determine the presence of full-lengthtranscripts, RT-PCR was performed on MEFcell total RNA using a common forwardprimer at the 5� end of exon 1 and reverseprimers to sequences within intron 8 (primerno. 1) or to the proximal region of exon 9(primer no. 2) (Fig. 2A). As expected, a full-length transcript was seen corresponding tomGR� (Fig. 2B). More importantly, a full-length transcript containing sequences derivedfrom intron 8 was also found. The intron8-containing mRNA had reduced abundancerelative to mGR�, consistent with the knownrelationship of hGR� and hGR� (16). Tosequence the novel full-length product, RT-PCR was repeated using a forward primerfrom the ATG start site in exon 2 and a re-verse primer (no. 3) specific to a more distalportion of intron 8, and the product wascloned and sequenced (GenBank accessionno. HM236293). The results (Fig. 2C) showedpresence of coding sequences that extend be-yond exon 8 into intron 8, with a stop codon atposition 46–48 base pairs within the intron(for complete mRNA sequence, please see Sup-plemental Fig. 3). Additional noncoding se-quences derived from the distal portion of in-

tron 8 were present. These results suggest that the�-specific amino acids of mGR are located in the proxi-mal portion of intron 8, in contrast to humans, where

GR� sequences are found in the distal por-tion of exon 9. It is likely, therefore, thatmGR� arises through use of alternative do-nor sites located within intron 8, as opposedto the hGR mechanism of alternative accep-tor sites in exon 9.

The translation product of mGR� spans748 amino acids (Fig. 3A). Alignment ofmGR� with the mGR� peptide sequence re-vealed 93% homology. The encoded mGR�

protein diverges from mGR� at amino acid733, beyond which is a disordered C termi-nus of 15 amino acids. In comparison,hGR� and hGR� are also 93% identical.Interestingly, both hGR� and mGR� addan additional 15 amino acids to exon 8,even though the means of alternative splic-ing are different. The two �-isoforms are anoverall 87% match and share a functionaldomain organization that appears identical(Fig. 3B). In both hGR and mGR, the point

FIG. 1. Detection of intron 8 transcripts of the mGR gene. A, Genomicorganization (not to scale) of the mGR gene, showing location of forward primer(exon 7) and reverse primers (1–5) used for RT-PCR analysis. B, Predicted PCRproduct sizes for reverse primers 1–4 are indicated. Only primers in the intron 8 (I-8)region yielded products. In the case of primer 4, two products were predicted basedon alternate utilization of a donor or acceptor site (Supplemental Fig. 1). Primers 4and 5 (data not shown) yielded no products on several attempts, consistent withpublished results (25). UTR, Untranslated region; Con, amplification withouttemplate.

FIG. 2. Isolation and sequencing of full-length mGR� mRNA. A, Genomic organization(not to scale) of the mGR gene, showing location of forward primer (exon 1) andreverse primers (1 and 2) used for RT-PCR analysis. B, Primer 3 in conjunction with theforward primer in exon 1 was used to generate a second full-length mGR� product thatwas cloned and sequenced to yield the organization of mGR� mRNA is seen in C.Primer 3 in conjunction with the forward primer in exon 2 was used to generate thecDNA cloned into pcDNA3.1� (see Fig. 5). Con, Amplification without template.


of divergence between � and � isoforms is located in thetransition region between the 10th and 11th helix of theligand-binding domain (7, 16). This results in a degener-ate helix 12 in the ligand-binding domains of hGR� andmGR�. As a consequence, mGR�, like hGR�, is unre-sponsive to the glucocorticoid agonist Dex (see below).

Expression of mGR� mRNA in vivoAs a first test of significance, expression of mGR� in

mouse organs and tissues was assayed by quantitativereal-time PCR (qPCR). Total RNA was extracted frommouse tissues and analyzed with forward primers to exon8 and reverse primers for exon 9 (mGR�) and intron 8(mGR�) indicated in Fig. 2. The qPCR results are fromtwo different C57/Bl6 mice. The data are expressed rela-tive to mGR� expression. In all tissues studied, mGR�

and mGR� mRNA appear to be coexpressed, with mGR�

mRNA levels being significantly higher than mGR� (Fig.4). The highest level of mGR� was observed in the spleen,with lesser amounts in kidney and liver. These results arein good agreement with expression profiling of GR�

mRNA in human (16) and zebrafish (20) tissues. More-over, the high levels of GR� in mouse spleen are consis-tent with the important role played by hGR� as an dom-inant-negative inhibitor of GR� in lymphoid cells (10).

Characterization of mGR� proteinA cDNA expression vector was made by cloning the

full-length mGR� PCR product (Fig. 2) into pcDNA3.1to generate the pMGR�-H57 vector (Fig. 5A). Receptor-less COS-7 cells were transfected with equal amounts ofpMGR�-H57 and pSV2Wrec encoding mGR�, followedby Western blot analysis using FiGR monoclonal anti-

FIG. 4. Tissue expression profile of mGR mRNA isoforms. Real-timePCR analysis of tissues and MEF cells was performed with primers tointron 8 (mGR�) or exon 9 (mGR�). All values were normalized to MEFcell mGR� and represent means � SEM for two independent samples/tissues. Adult, male C57/BL6 mice fed normal chow ad libitum wereused as donors.

FIG. 3. Sequence and functional domain comparisons of mGR� and hGR�. A, C-terminal regions of GR� and GR� were aligned for each species.Overall percent homologies for entire proteins are indicated. Vertical lines indicate borders between exon 8 and distal domains. Light gray boxesindicate conservative substitutions. B, The mGR� isoform exhibits a functional domain structure that is nearly identical to hGR�. Compared withmGR�, the �-isoforms of both species have reduced and distinct C-terminal regions that lack the activation function-2 (AF-2) domain (helix 12).These features account for their reduced ability to bind hormone and activate transcription. DBD, DNA-binding domain; H, hinge region; LBD,ligand-binding domain; FiGR, epitope recognized by FiGR monoclonal antibody. HSP90 binding regions of mGR� and hGR� are shown, along withputative site in mGR�.


body against mGR that recognizes a shared epitope in theN-terminal domain (see Fig. 3) (27). The results (Fig. 5A)show expression of mGR� protein with a molecular masssmaller (81.7 kDa) than mGR� (86.0 kDa). Importantly,this shows that the endogenously produced mGR�

mRNA can be translated into protein. To detect endoge-nous mGR�, a rabbit polyclonal antibody (rMGRB-Ab)was made. Specificity of rMGRB-Ab was determined bytransfecting COS-7 cells with pSV2Wrec, pMGR�-H57,or empty vector, followed by probing of Western blot-tings with rMGRB-Ab and FiGR antibodies. Results wereanalyzed with the Odyssey infrared system utilizing red-

and green-emitting counter antibodies to detect FiGR andrMGRB-Ab, respectively. As predicted, rMGRB-Ab de-tected mGR� but showed no reactivity to mGR� (Fig.5B). To demonstrate existence of endogenous mGR� pro-tein in cells, untransfected MEF cell lysates were immu-noadsorbed with rMGRB-Ab or FiGR followed by West-ern blot analysis (Fig. 5, C and D). The results showimmune-specific pull-down of mGR� by the rMGRB-Aband FiGR antibodies. As expected, the FiGR purificationresults showed expression of mGR� protein to be lowerthan mGR�. As a last test, indirect immunofluorescencewith rMGRB-Ab was performed in MEF cells (Fig. 5E),

FIG. 5. Cloning, Western blot, and indirect immunofluorescence analyses of mGR�. A, Primers spanning the ATG start site in exon 2 and thedistal region of intron 8 (see Fig. 2) were used to isolate the full-length mGR� cDNA, followed by cloning into pcDNA3.1 to yield the pMGR�-H57vector. COS-7 cells were transfected with 8 �g each of pSV2Wrec (mGR�), pMGR�-H57 (mGR�), or empty vector (mock), followed by Westernblot analysis with FiGR mouse monoclonal antibody that recognizes a common epitope on both mGR isoforms. B, A rabbit polyclonal antibodyspecific to mGR� (rMGR�-Ab) was generated using the unique 15-amino acid terminal sequence. Whole-cell extracts from COS-7 cells transfectedwith pSV2Wrec, pMGR�-H57, or empty vector (mock) were simultaneously probed with FiGR and rMGR�-Ab antibodies. The Odyssey infrareddetection system utilizing 680 nm (red) and 800 nm (green) emitting counter antibodies was used to detect FiGR (mGR�) and rMGR�-Ab,respectively. Results show that the rMGR�-Ab antibody reacts only with mGR� not with mGR�. C, MEF cell lysates were immunoadsorbed withrMGR�-Ab (I) or preimmune (PI) serum, followed by blotting with rMGR�-Ab, using enhanced chemiluminescence. Whole-cell extracts from COS-7cells transfected with pMGR�-H57 were used for comparison. D, Two separate MEF cell lysates were immunoadsorbed with FiGR (I) ornonimmune (NI) IgG, followed by sequential blotting with rMGR�-Ab and FiGR, using enhanced chemiluminescence. Relative densitometric ratiosof mGR� to mGR� are shown. E, Indirect immunofluorescence using preimmune serum, rMGR�-Ab, or rMGR�-Ab blocked with mGR� peptidewas performed on MEF cells treated or untreated with Dex (100 nM, 2 h).


showing localization of mGR� protein in the cytoplasmand nuclear foci. Interestingly, Dex treatment elevated themGR� signal in both compartments. This result is con-sistent with the ability of Dex to increase mGR� mRNAexpression (please see figure 8 below).

mGR� is a hormone-insensitive dominant-negativeinhibitor of mGR�

Because mGR� and hGR� express C-terminal do-mains that are almost identical, we reasoned that mGR�

would be unresponsive to activation by glucocorticoid

ligands. In addition, Cidlowski and co-workers (9) have shown that the ability ofhGR� to inhibit hGR� activity on geneexpression is encoded by its C-terminal15-amino acid domain. Thus, the mGR�

discovered in this work seemed likely toact as a hormone-insensitive, dominant-negative inhibitor of mGR�. To test bothof these functions, COS-7 cells were trans-fected with pGRE2E1B-Luc reporter andvarious combinations of pMGR�-H57 andpSV2Wrec. The results of Fig. 6A showedthat mGR� cannot increase pGRE2E1B-Luc activity in response to Dex or RU486.RU486 antagonist was chosen because ofone report suggesting reactivity to this li-gand by hGR� (18). Figure 6B demon-strates dose-dependent inhibition of mGR�

activity by mGR� that begins at a �:� molarratio of 0.5:1. To determine the dominant-negative activity of mGR� at endogenousgenes, real-time PCR analysis was per-formed in mGR�-expressing MEF cells af-ter transfection of pMGR�-H57 (Fig. 6C).The results showed that Dex-induced ex-pression of glucocorticoid-inducible leucinezipper (GILZ), serum- and glucocorticoid-inducible kinase, pyruvate dehydrogenase ki-nase-4 (PDK4), and glucose-6-phosphatase(G6Pase) were effectively inhibited by mGR�.In contrast, mGR� had no effect on expres-sion of FK506 binding protein 51 (FKBP51)and on the ability of Dex to down-regulatemGR� expression. The latter results suggestthat mGR� is not a global inhibitor of mGR�

actions, and that the inhibitory effect of mGR�

is not due to decreased mGR� expression.To directly test the inhibitory actions of

endogenous mGR�, short hairpin RNA(shRNA) specific to mGR� was designedbased on its unique intron 8 sequence. MEFcells were infected with lentiviral constructsexpressing mGR� shRNA or empty vector.

Figure 7A demonstrates that the mGR� shRNA effec-tively blocks expression of the mGR� full-length mRNA,whereas leaving mGR� mRNA untouched. More impor-tantly, Dex-induced expression of three endogenousgenes (PDK4, G6Pase, and GILZ) was increased in re-sponse to mGR� knockdown (Fig. 7B). These results arethe first direct evidence that endogenously expressed GR�

(mouse or human) plays a functional role in mGR� ac-tions. Consistent with the results of Fig. 6C, mGR�

knockdown had no effect on FKBP51 expression.

FIG. 6. Hormone sensitivity and dominant-negative activity of mGR�. A, COS-7 cellstransfected with pSV2Wrec (mGR�) or pMGR�-H57 (mGR�) were assayed for luciferaseactivity at the pGRE2E1B-Luc reporter after treatment with Dex (1 �M), RU486 (1 �M),or vehicle control. Values were normalized to transfection efficiency (renilla) andrepresent means � SEM for three independent treatments. B, Luciferase (pGRE2E1B-Luc)activity in COS-7 cells was measured after transfection with mGR� and increasingamounts of mGR� and treatment with 100 nM Dex (SEM, n �3; **, P � 0.01 vs. Dex,mGR� alone). C, To assess dominant-negative activity of mGR� at endogenous genes,MEF cells expressing mGR� were transfected with pMGR�-H57 or control vector andtreated with or without Dex (100 nM), followed by real-time PCR analysis. Valuesrepresent means � SEM of three independent treatments assayed in triplicate. ***, P �0.001; **, P � 0.01; *, P � 0.05 vs. Dex, mGR� alone.


Hormonal and dietary control of mGR� expressionAs an agonist to mGR�, Dex not only causes activation

or repression of GR-regulated genes, it also causes auto-logous down-regulation of GR� expression in many spe-cies, as a means by which to attenuate overstimulation byGC ligands (28, 29). Because GR� can be viewed as analternative mechanism of GR� attenuation, and becauseGR� cannot bind Dex, we determined what effect Dexhad on mGR� expression. MEF cells were treated with100 nM Dex for 2 h, followed by measurement mGR� andmGR� mRNA expression via qPCR (Fig. 8A). As ex-pected, a significant decrease in mGR� mRNA was seenin response to hormone treatment. Interestingly, mGR�

mRNA was increased. This suggests the existence of anegative feedback loop, most likely mediated by GR�,that up-regulates GR� expression to control sensitivity ofcells to glucocorticoids.

GCs acting through hGR� are potent antiinflamma-tory drugs that inhibit the ability of nuclear factor �B

(NF-�B) to activate expression of proinflammatory cyto-kine genes (30). In turn, activation of NF-�B by TNF-�leads to reciprocal attenuation of hGR� transcriptionalactivity (31). Although NF-�B can inhibit hGR� throughdirect protein-protein binding (30), an additional mech-anism was recently discovered, in which NF-�B causesselective up-regulation of hGR� expression (10). To testwhether this mechanism applies to mGR�, we measuredmGR� and mGR� expression in RAW 264.7 monocyticmacrophage cells subjected to TNF-� treatment (Fig. 8B).The results show a modest approximately 1.5-fold induc-tion of mGR� at the 8-h time point but a greater approx-imately 2.5-fold induction of mGR�. These data are ingood agreement with those of Webster et al. (10) andsuggest that mGR�, like its human cognate, serves toinduce a state of GC resistance to maximize inflammatoryresponses.

As a last test of relevance, we asked whether mGR�

might play a role in GC control of metabolism. Glucocor-ticoids are well-known antagonists to insulin that pro-mote gluconeogenesis over glycogenesis, especially in theliver (4, 32). Conversely, insulin acts to inhibit gluconeo-genesis, in part, by blocking GR activity at the hepaticphosphoenolpyruvate carboxykinase promoter (33, 34).Because mGR� can similarly block mGR� activity at thePDK4 and G6Pase genes (Fig. 6C), we tested whetherinsulin might achieve this effect on gluconeogenic genesby up-regulating expression of mGR� (Fig. 8C). The re-sults show a short-lived increase of mGR� immediatelyafter insulin treatment of MEF cells, but a more dramaticincrease of mGR� with longer treatment. As a furthertest, we assayed for mGR isoform expression in the liversof mice subjected to a fasting-refeeding regimen (Fig. 8D).During refeeding, there is a well-documented and robuststimulation of hepatic metabolism by insulin (35, 36).The results show a small increase in mGR� during theearly stages of refeeding but a much larger increase inmGR� at the later stages. Taken as a whole, these resultsrepresent a potential new role for GR� (mouse or human)in the GC-insulin axis, in which up-regulation of GR�

may serve to maintain insulin sensitivity.

Discussion

Due to underlying pathology or drug treatment, glucocor-ticoid resistance can develop and is now a major concernin many disease states. Thus, the need for a mammalianGR� model is imperative. In humans, GC resistance canoccur by two major mechanisms: loss-of-function muta-tions in GR� (37) or by increased expression of GR�,which acts as a dominant-negative inhibitor of GR�. Al-though GR� mutations can result in a type of GC resis-

FIG. 7. Gene silencing of mGR�. Lentiviral delivery of mGR� shRNAwas use to make a MEF cell line with stable down-regulation of mGR�.A control cell line was infected with lentivirus expressing empty vector.A, Complete mRNA constructs for mGR� or mGR� were amplified viaPCR to show mGR� knockdown. B, mGR� shRNA and vector MEF cellswere treated with 100 nM Dex for 2 h or vehicle, followed by real-timePCR analysis. Values represent means � SEM of three independenttreatments assayed in triplicate. **, P � 0.01; *, P � 0.05 vs. Dex,vector control.


tance that is both systemic and severe, these mutations arerare. In contrast, the evolving evidence suggests that GCresistance based on GR� is much more common andlikely to be tissue specific in nature. To date, GC resis-tance based on GR� has been principally characterized inimmunological diseases and drug-resistant states. Im-mune system homeostasis is balanced by glucocorticoids,which regulate immune cell turnover by suppressing cy-tokine production and promoting apoptosis. GC insensi-tivity due to elevated hGR� expression increases proin-flammatory cytokines, leading to escalated cell growthand reduced cell death (38). Superantigens, such as staph-ylococcal enterotoxin B and toxic shock syndrome toxin,have been demonstrated to cause increased GR� expres-sion and GC resistance (39). Also, proinflammatory cy-tokines, such as TNF� and IL-1, increase expression ofGR� via the NF-�B pathway (10). Although cytokineproduction is increased in all asthma patients, somesubjects do not benefit from GC therapy because ofelevated GR� (40, 41). Indeed, fatal asthma has beenlinked to extremely high levels of GR� in the airways(42) and a complete loss of GC drug response (43).Other inflammatory diseases linked to high levels of GR�

include: ulcerative colitis, ankylosing spondylitis (44),cigarette smoking (45), leukemia (12), and systemic lupuserythematosus (46).

GC resistance via GR� is almost unknown in diseasesof metabolism. Only one study on this topic has beenpublished. It showed that exposure of skeletal muscle toGCs leads to a decline in GR� expression and a concom-itant increase of GR� (47). Because GC hormones arepotent regulators of glucose and lipid metabolism, andbecause GCs are broad and chronic antagonists to theactions of insulin (48), we reasoned that GR� may play arole as a modulator of GR� actions, similar to insulinantagonism. In this work, we provide evidence thatmGR� is indeed involved in metabolic processes. GC in-duction of the gluconeogenic enzymes PDK4 and G6Pasewas inhibited by mGR�, suggesting that one function ofthe GR� isoform, like insulin, is to block glucose produc-tion. Not surprisingly, in cells treated with insulin, mGR�

expression was unchanged, but mGR� went up. In cellstreated with Dex, mGR� was also increased, suggestingthat both GCs and insulin share up-regulation of mGR�

as a common mechanism for antagonism of mGR�. Sim-ilarly, in mice subjected to fasting refeeding, a large in-crease of GR� was seen in the liver, whereas mGR� wasonce again unchanged. The fasting-refeeding regimen em-ployed is known to produce a robust stimulation of he-patic metabolism by insulin (35, 36). Taken as a whole,these data are the first evidence that GR� up-regulation

FIG. 8. Hormonal and dietary control of mGR� expression. Real-time PCR analysis of mGR isoforms in (A) MEF cells treated with 100 nM Dex(***, P � 0.001), (B) RAW 264.7 monocytic macrophage cells treated with 10 nM TNF� (**, P � 0.01), (C) MEF cells treated with 100 nM insulin(***, P � 0.001), and (D) livers of adult male C57/BL6 mice subjected to fasting refeeding (**, P � 0.01). All values were normalized to 18S RNAand represent means � SEM of three independent treatments assayed in triplicate.


may be an important mechanism for maintaining organsensitivity to insulin.

In this work, we present data that mice express a GR�

isoform that derives from alternative splicing of intron 8,similar to the mechanism in zebrafish (20). The sequenceencoding the GR�-specific amino acids is located in themiddle portion of exon 9 in the human gene but is foundin intron 8 in the zebrafish gene. In zGR, exon 9 is skippedor silenced as a result of alternative splicing, and intron 8is retained. Therefore, hGR� and hGR� mRNA are pro-duced through alternative usage of a splice acceptor site inexon 9, whereas alternative use of a splice donor site inintron 8 appears to be the underlying mechanism in bothmice and zebrafish GR. This mechanism is often referredto as intron retention and is not unique to mGR�. Indeed,C-terminal isoform variants of vitamin D, peroxisomeproliferator-activated-� and peroxisome proliferator-ac-tivated-� receptors can be generated by intron retention(24). It has long been thought that GR� does not exist inrodents, in large part because one high-profile study con-cluded that the alternative splicing event does not occur inmice (25). The study assumed that the splicing event inmice must be similar to humans and used primers thatfocused on the distal portion of exon 9. It is now clear thatmice express a GR� isoform derived from intron 8. Incontrast to zGR�, which has a 50-amino acid C-terminalregion, mGR� has a protein structure in which the C-terminal region is the same size (15 amino acids) as thehuman �-isoform. Moreover, for the properties so fartested, mGR� is highly similar to hGR�.

The mechanism controlling alternative splicing of theGR gene is poorly understood but is generally thought toinvolve the generation of a spliceosome composed of ri-bonucleoproteins and serine-arginine (SR)-rich proteins,among others, that bind structures found in both intronsand exons, such as branch point sequences (BPS) andpolypyrimidine tracts (ror review, see Ref. 49). Intron 8 inmice is nearly twice as large as intron 8 in humans (1061vs. 526 bp, respectively). Because intron size correlates tothe likelihood of alternative splicing, this reason alonemay account for why murine species utilize intron 8 forisoform control. Interestingly, we have identified two BPSsites within murine intron 8, as well as a single polypyri-midine tract (Supplemental Fig. 1). We speculate thatmGR� and mGR� are generated through a mechanismthat uses the separate, but distinct, BPS sites to initiatesplicing. Targeting of these sites or the spliceosome pro-teins that bind them may eventually form the basis bywhich to inhibit mGR� expression, resulting in cells andtissues that are more sensitive to glucocoritocoids. Recentadvances lend credence to this hypothesis. Bombesin, aligand for G protein-coupled receptors, is known to up-

regulate expression of SR protein p30c, causing elevatedexpression of GR� in prostate cancer cells (50). A relatedprotein, SRp40, regulates splicing of GR� in HeLa and293T cells (51). As can be seen by these examples, majorfactors that regulate spliceosome action on the GR geneare only now being discovered. Our newly discoveredmouse model of GR� can now be used to establish feasi-bility of these targets for eventual alteration of hGR�

expression.Our data show that mGR� expressed in receptor-less

COS cells cannot respond to Dex or RU486 by activatingexpression of a pGRE-Luc reporter. This result is in goodagreement with most studies of hGR� showing that itcannot bind GC agonists nor activate reporter or endog-enous gene expression in response to hormone (7–9, 16).However, there is one report demonstrating hGR� bind-ing by RU486 with consequent induction of nucleartranslocation (18). In contrast to this unresolved issue, itis now clear that hGR� can exert constitutive control ongene expression (17, 18). These studies used gene arrayapproaches after overexpression of hGR� and showedthat, in addition to its ability to suppress hGR�-regulatedgenes, hGR� exerted both positive and negative controlover a unique set of genes not regulated by hGR�. Inaddition, a constitutive ability of hGR� to induce histonedeacetylation has been found (52, 53), providing a possi-ble mechanism for hGR�-mediated repression of geneexpression. Taken as a whole, these results show thathormone-free GR� has an unexpected constitutive andintrinsic gene expression function that may regulate cel-lular and physiological responses distinct from GR�. Al-though it remains to be seen whether mGR� can replicatethe latter property, the mGR� mouse model describedhere is likely to foster study of glucocorticoid resistanceand sensitivity in diverse disease states, such as inflamma-tion, hematological cancers, diabetes, and obesity.

Materials and Methods

In silico prediction of mGR�To identify mGR�, the NCBI website (www.ncbi.nlm.nih.gov)

was used for the nuclear receptor subfamily 3, group C, member1 (M. musculus), and gene information was downloaded for thecomplete GR genomic DNA sequence. Information regardingintrons and exons (ENSMUSG00000024431) was downloadedfrom Ensembl Mouse GeneView (www.ensembl.org). A singlemGR genomic sequence was identified. The NCBI website wasused to search for expressed sequence tag and cDNA sequencesderived from mGR transcripts. Text searches and Basic LocalAlignment Search Tool (BLAST) searches were performed.BLAST searches of the mouse genomic sequence were carriedout using the BLAST nucleotide tool at the Sanger InstituteEnsembl server.


www.ncbi.nlm.nih.gov

Characterization of alternative splice sites withinmGR gene

Recognition of alternative splice sites in exons 7–9 and in-tron 8 of the mGR gene were investigated using the Drosophilawebsite for splice site predictions (http://www.fruitfly.org/seq_tools/splice.html). Sequences with the highest scores in exon 8,exon 9, and intron 8 were identified as potential targets foralternative splice sites that could lead to production of mGR�(Supplemental Fig. 1). Primers were developed for these pro-spective splice sites (see below), and an initial screen for mGR�was performed via RT-PCR.

Initial screening of mGR�Total RNA was isolated from MEF cells using 5-Prime Per-

fectPure RNA Cell kit (Fisher Scientific Co., LLC, Auburn, AL)according to the manufacturer’s instructions. Total RNA con-centration and purity was determined by measuring absorbanceat 260/280 nm and confirmed on an RNA denaturing formal-dehyde gel. Purified RNA (1 �g) was used to produce comple-mentary strands of DNA (cDNA) using a 1st strand synthesis kit(Roche Applied Science, Indianapolis, IN). Newly synthesizedDNA (3 �l) was amplified by RT-PCR using forward primerscontaining sequences from exon 7 (GCAGAGAATGACTCT-ACCCTGCA) and reverse primers based on prospective splicesites ratings from the Drosophila website. Three different re-verse primers for intron 8 (TAAAGGCATCTGCCACCACC,CTGTCTTTGGGCTTTTGAGATAGG, and CTTTGGGCT-TTTGAGATAGGATC) and two different reverse primers forthe latter part of exon 9 (TCCCAGCTCCCTCTCCCTAG andTCCCTCTCCCTAGCTTAGAG) were used to identify the lo-cation of mGR� (see Fig. 1). 18S RNA was amplified as aninternal control. PCR conditions used were: 95 C for 5 min, 95C for 1 min, 60 C for 1 min, 72 C for 40 sec, and 72 C for 10 min.PCR products were electrophoresed in a 1% agarose gel andvisualized with ethidium bromide. The 1 Kb Plus DNA Ladder(Invitrogen, Carlsbad, CA) was used as a size standard.

Generation of complete mGR� transcriptPrimers for exon 1 (GTAGAGACGAAGTCCCCAGCA)

and reverse primers were based on sequences from intron 8(GR�) (TAAAGGCATCTGCCACCACC) and exon 9 (GR�)(AGCTAAGGAGATTTTCAACCACA) of mGR and used todemonstrate the complete mGR� and mGR� mRNA con-structs. The expected mGR� and mGR� products were 2251and 2361 bp, respectively. PCR conditions used were: 95 C for5 min, 95 C for 1 min, 60 C for 1 min, 72 C for 3.5 min, and 72C for 10 min.

Cloning and sequencing of mGR�Cloning and sequencing of mGR� from MEF cells was per-

formed as follows. After total RNA isolation, cDNA synthesiswas achieved using KOD Xtreme Hot Start DNA Polymerase(Novagen, Madison, WI) and a forward primer for the ATGstart site of the open reading frame in exon 2 (CGGGATCCAT-GGGACTGTATATGGGAGAG) and a reverse primer to thedistal portion of intron 8 (GCTCTAGAGTAATGTATCTT-GATTGTGGC). The expected product was 2852 bp. GR� PCRproducts were ligated to pcDNA 3.1� vector using BamHI andXbaI and transformed into One Shot INV F cells (Invitrogen).Plasmid DNA from positive clones was determined by RT-PCRand further extracted using the QIAGEN Spin Miniprep Kit

(QIAGEN, Crawley, UK). BamHI and XbaI sites flank the po-sition on the vector at which the mGR� gene is inserted. Thepresence of GR� in mice was confirmed by restriction digestionwith BamHI and XbaI to determine the size of the insert (Boehr-inger Mannheim, Mannheim, Germany) and further digestedwith HindIII to determine sequence specificity. Sequencing wasperformed by the University of Iowa DNA Facility (Iowa City,IA) using T7 forward and BGH reverse primers that flank thegene insertion site of the plasmid. The sequence confirmed plas-mid was named pMGR�-H57. The mGR� sequence has beendeposited to GenBank under accession no. HM236293.

Quantitative real-time PCR analysisTotal RNA was extracted from mouse tissues using 5-Prime

PerfectPure RNA Tissue kit (Fisher Scientific Co., LLC). TotalRNA from MEF cells was extracted as described above. cDNAwas synthesized using iScript cDNA Synthesis kit (Bio-Rad,Hercules, CA). PCR amplification of the cDNA was performedby qPCR using qPCR Core kit for SYBR Green I (Applied Bio-systems, Foster City, CA). The thermocycling protocol consistedof 10 min at 95 C, 40 cycles of 15 sec at 95 C, 30 sec at 61 C, and20 sec at 72 C and finished with a melting curve ranging from 60to 95 C to allow distinction of specific products. Primers weredesigned using Primer Express 3.0 software (Applied Biosys-tems) to amplify a region in intron 8 that was revealed in theinitial screening for mGR�. A common forward primer in exon8 (AAAGAGCTAGGAAAAGCCATTGTC) was used in con-junction with a reverse primer in intron 8 for mGR� (CT-GTCTTTGGGCTTTTGAGATAGG) or a reverse primer inexon 9 for mGR� (TCAGCTAACATCTCTGGGAATTCA).Normalization was performed in separate reactions with prim-ers to 18S mRNA (TTCGAACGTCTGCCCTATCAA and AT-GGTAGGCACGGCGACTA). To study whether genomic se-quences were amplified, a control sample was used, in which noreverse transcriptase was added (non-RT control). The follow-ing primers were used for endogenously expressed genes: serum-and glucocorticoid-inducible kinase, forward GAGAA-GGATGGGCCTGAACGAT and reverse CGGACCCAGGTT-GATTTGTTGA; GILZ, forward AATGCGGCCACGGATGand reverse GGACTTCACGTTTCAGTGGACA; PDK4, for-ward TTTCTCGTCTCTACGCCAAG and reverse GATA-CACCAGTCATCAGCTTCG; G6Pase, forward TGCAAGG-GAGAACTCAGCAA and reverse GGACCAAGGAAGCCA-CAATG; and FKBP51, forward GCTGGCAAACAACACG-AGAG and reverse GAGGAGGGCCGAGTTCATT. Allprimer sequences were uploaded to the PrimerFinder data-base at http://www.primerfinder.com/.

Generation of mGR� antibodyA rabbit polyclonal antibody to mGR� via the method used

to produce antibody to hGR� (33). A peptide corresponding toamino acids 733–748 (VSTKHKSKTTAKKKK) at the C termi-nus of the mGR� protein was synthesized and purified by PacificImmunology (Ramona, CA). An N-terminal cysteine was addedto the peptide as a linker, followed by conjugation to a peptidecarrier protein keyhole limpet hemocyanin and adjuvant-basedimmunization in a female New Zealand White rabbit. Preim-mune serum was collected before injecting the rabbits withmGR� conjugate peptide. The rabbits were boosted 2 wk afterinjection with the mGR� peptide with Complete Freund’s Ad-juvant and subsequently boosted two more times with Incom-plete Freund’s Adjuvant every 2 wk. Serum was collected at 2


months and analyzed via ELISA for mGR� specific antibodies.Serum of high titer was obtained and subjected to one round ofaffinity purification using the mGR� peptide.

Generation of mGR� shRNA lentiviral constructTo identify an small interfering RNA to knockdown mGR�,

a free Web-based tool (http://www.genelink.com/sirna/shRNAi.asp) was used to design a putative small interfering RNAagainst the mGR� and to design oligonucleotides that encode acorresponding shRNA. The resulting shRNA recognized a se-quence beginning at exon 8 within the mGR mRNA and ex-tended into intron 8. XbaI and XhoI restriction sites were addedto flanking regions of the sequence. Oligonucleotides were:GGACTCCATGCATGATGTAAGTACCAAACATCAAGAG-TGTTTGGTACTTACATCATGCATGGAGTCTTTTTT andthe homologous sequence. Synthetic oligonucleotides were an-nealed, digested with restriction enzymes, and then ligated intothe XbaI/XhoI sites of the FG12 vector that has an independentgreen fluorescent protein (GFP) marker and transformed inDH5� cells (Invitrogen). Clones were selected and tested bytransient transfection to determine knockdown of mGR�. Afterconfirmation of knockdown, the construct was cotransfectedtogether with vectors expressing gag-pol, REV, and VSV-G into293FT cells (Invitrogen) to generate a third generation lentiviralconstruct. Transfection was achieved using Lipofectamine 2000(Invitrogen) using 100 ng total DNA per cm2 of the growth plateor well. The supernatants were harvested, and the cell debriswas removed by centrifugation at 2000 � g. The supernatantwas used to infect MEF cells after addition of polybrene (5ng/ml; Sigma Chemical Co., St. Louis, MO) to establish cell lineswith stable down-regulation of mGR� mRNA or expressingempty vector. After 72 h the cells were sorted by flow cytometryfor GFP by the Flow Cytometry Core Facility at the University ofToledo Health Science Campus. GFP positive cells were used forall experiments.

Transfection and reporter assaysExpression vector for mGR� (pMGR�-H57) was con-

structed as described above. The Ringold laboratory had al-ready developed a plasmid for mGR�, pSV2Wrec (21). Bothplasmids were transiently transfected into COS-7 cells (Africangreen monkey kidney cells lacking an endogenous GR), andprotein expression was measured via Western blot analysis.Dominant-negative activity was measured by luciferase assayusing the GR-responsive minimal reporter pGRE2EIB-Luc (54)and pRL-CMV Renilla reporter for normalization to transfec-tion efficiency. Transient transfection was achieved using Lipo-fectamine 2000. Twenty-four-hour posttransfected cells weretreated with vehicle or 1 �M Dex or 1 �M RU486 for an addi-tional 24 h until harvest. Cell lysates and assay were performedusing the Promega luciferase assay system (Promega, Madison,WI). Statistical analyses employed the Student’s t test orANOVA using GraphPad Prism version 5.0a for Mac (Graph-Pad Software, San Diego, CA).

Gel electrophoresis and Western blot analysisWhole-cell extracts were prepared from COS-7 cells that

were transiently transfected for 48 h with either pMGR�-H57or pSV2Wrec using Lipofectamine 2000. Control cells were un-transfected COS-7 cells that do not express GR. Protein contentwas determined by BCA method of Pierce. Protein samples wereresolved by SDS-PAGE and electrophoretically transferred to

Immobilon-FL membranes. Membranes were blocked at roomtemperature for 1 h in Tris-buffered saline (TBS) [10 mM Tris-HCl (pH 7.4) and 150 mM NaCl] containing 3% BSA. Subse-quently, the membrane was incubated overnight at 4 C withFiGR antibody for total GR or rMGR� antibody for mGR� ata dilution of 1:1000 in TBS. After three washes in TBS with0.1% Tween 20, the membrane was incubated with an infraredantirabbit (IRDye 800, green) or antimouse (IRDye 680, red)secondary antibody labeled with IRDye infrared dye (LI-CORBiosciences, Lincoln, NE) (1:15,000 dilution in TBS) for 2 h at 4C. Immunoreactivity was visualized and quantified by infraredscanning in the Odyssey system (LI-COR Biosciences).

Immunoadsorption of GR complexesCells were harvested in HEMG [10 mM HEPES, 3 mM EDTA,

20 mM sodium molybdate, and 10% glycerol (pH 7.4)] plusprotease inhibitor cocktail and set on ice for 10 min followed byDounce homogenization. Supernatants (cytosol) were collectedproceeding a 10 min 4 C centrifugation at 20,800 � g, thenprecleared with protein A or G-Sepharose nutating for 1 h at 4C. Samples were spun down, split into equal aliquots of cytosol,and immunoadsorbed overnight with FiGR antibodies againsttotal GR, rMGR� antibody for mGR�, and appropriate con-trols (nonimmune mouse IgG or preimmune rabbit serum) at 4C under constant rotation. Pellets were washed five to seventimes with TEG [10 mM Tris, 3 mM EDTA, 10% glycerol, 50 mM

NaCl, and 20 mM sodium molybdate (pH 7.4)], and complexeswere eluted with 6� sodium dodecyl sulfate sample buffer.

AnimalsAdult, male C57/BL6 mice maintained on a normal diet ad

libitum or subjected to a fasting-refeeding regimen were used astissue donors. Fasting encompassed 16 h (including the over-night 12-h dark cycle), followed by 8 h of refeeding ad libitumwith normal chow at the start of the light cycle. All procedureswere approved by the Institutional Animal Care and UtilizationCommittee of The University of Toledo.

Acknowledgments

We thank Rudel Saunders and Qiong Wu for advice on plasmidconstruction.

Address all correspondence and requests for reprints to:Edwin R. Sanchez, Department of Physiology and Pharmacol-ogy, University of Toledo College of Medicine, 3035 Arling-ton Avenue, Toledo, Ohio 43614-5804. E-mail: [email protected].

This work was supported by National Institutes of HealthGrants DK70127 (to E.R.S.), DK54254, and DK83850 (toS.M.N.) and by the United States Department of AgricultureGrant 38903-02315 (to S.M.N.). T.D.H. was supported by thePredoctoral National Institutes of Health National ResearchService Award F31DK84958.

Disclosure Summary: The authors have nothing to disclose.

References

1. Chrousos GP, Kino T 2005 Intracellular glucocorticoid signaling: aformerly simple system turns stochastic. Sci STKE 2005:pe48


2. Lewis-Tuffin LJ, Cidlowski JA 2006 The physiology of human glu-cocorticoid receptor � (hGR�) and glucocorticoid resistance. AnnNY Acad Sci 1069:1–9

3. Sapolsky RM, Romero LM, Munck AU 2000 How do glucocorti-coids influence stress responses? Integrating permissive, suppres-sive, stimulatory, and preparative actions. Endocr Rev 21:55–89

4. Vegiopoulos A, Herzig S 2007 Glucocorticoids, metabolism andmetabolic diseases. Mol Cell Endocrinol 275:43–61

5. Tata JR 2002 Signalling through nuclear receptors. Nat Rev MolCell Biol 3:702–710

6. Kumar R, Johnson BH, Thompson EB 2004 Overview of the struc-tural basis for transcription regulation by nuclear hormone recep-tors. Essays Biochem 40:27–39

7. Bamberger CM, Bamberger AM, de Castro M, Chrousos GP 1995Glucocorticoid receptor �, a potential endogenous inhibitor of glu-cocorticoid action in humans. J Clin Invest 95:2435–2441

8. de Castro M, Elliot S, Kino T, Bamberger C, Karl M, Webster E,Chrousos GP 1996 The non-ligand binding �-isoform of the humanglucocorticoid receptor (hGR�): tissue levels, mechanism of action,and potential physiologic role. Mol Med 2:597–607

9. Oakley RH, Jewell CM, Yudt MR, Bofetiado DM, Cidlowski JA1999 The dominant negative activity of the human glucocorticoidreceptor � isoform. Specificity and mechanisms of action. J BiolChem 274:27857–27866

10. Webster JC, Oakley RH, Jewell CM, Cidlowski JA 2001 Proinflam-matory cytokines regulate human glucocorticoid receptor gene ex-pression and lead to the accumulation of the dominant negative �

isoform: a mechanism for the generation of glucocorticoid resis-tance. Proc Natl Acad Sci USA 98:6865–6870

11. Leung DY, de Castro M, Szefler SJ, Chrousos GP 1998 Mechanismsof glucocorticoid-resistant asthma. Ann NY Acad Sci 840:735–746

12. Longui CA, Vottero A, Adamson PC, Cole DE, Kino T, Monte O,Chrousos GP 2000 Low glucocorticoid receptor �/� ratio in T-celllymphoblastic leukemia. Horm Metab Res 32:401–406

13. Zhang X, Ognibene CM, Clark AF, Yorio T 2007 Dexamethasoneinhibition of trabecular meshwork cell phagocytosis and its modu-lation by glucocorticoid receptor �. Exp Eye Res 84:275–284

14. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Oro A, Lebo R,Thompson EB, Rosenfeld MG, Evans RM 1985 Primary structureand expression of a functional human glucocorticoid receptorcDNA. Nature 318:635–641

15. Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signaturemotif in transcriptional co-activators mediates binding to nuclearreceptors. Nature 387:733–776

16. Oakley RH, Sar M, Cidlowski JA 1996 The human glucocorticoidreceptor � isoform. Expression, biochemical properties, and puta-tive function. J Biol Chem 271:9550–9559

17. Kino T, Manoli I, Kelkar S, Wang Y, Su YA, Chrousos GP 2009Glucocorticoid receptor (GR)� has intrinsic, GR�-independent tran-scriptional activity. Biochem Biophys Res Commun 381:671–675

18. Lewis-Tuffin LJ, Jewell CM, Bienstock RJ, Collins JB, CidlowskiJA 2007 Human glucocorticoid receptor � binds RU-486 and istranscriptionally active. Mol Cell Biol 27:2266–2282

19. Oakley RH, Webster JC, Sar M, Parker Jr CR, Cidlowski JA 1997Expression and subcellular distribution of the �-isoform of the hu-man glucocorticoid receptor. Endocrinology 138:5028–5038

20. Schaaf MJ, Champagne D, van Laanen IH, van Wijk DC, MeijerAH, Meijer OC, Spaink HP, Richardson MK 2008 Discovery of afunctional glucocorticoid receptor �-isoform in zebrafish. Endocri-nology 149:1591–1599

21. Danielsen M, Northrop JP, Ringold GM 1986 The mouse glucocor-ticoid receptor: mapping of functional domains by cloning, se-quencing and expression of wild-type and mutant receptor proteins.EMBO J 5:2513–2522

22. Northrop JP, Danielsen M, Ringold GM 1986 Analysis of glucocor-ticoid unresponsive cell variants using a mouse glucocorticoid re-ceptor complementary DNA clone. J Biol Chem 261:11064–11070

23. Northrop JP, Gametchu B, Harrison RW, Ringold GM 1985 Char-

acterization of wild type and mutant glucocorticoid receptors fromrat hepatoma and mouse lymphoma cells. J Biol Chem 260:6398–6403

24. van der Vaart M, Schaaf MJ 2009 Naturally occurring C-terminalsplice variants of nuclear receptors. Nucl Recept Signal 7:e007

25. Otto C, Reichardt HM, Schutz G 1997 Absence of glucocorticoidreceptor-� in mice. J Biol Chem 272:26665–26668

26. Root B, Abrassart J, Myers DA, Monau T, Ducsay CA 2008 Ex-pression and distribution of glucocorticoid receptors in the ovinefetal adrenal cortex: effect of long-term hypoxia. Reprod Sci 15:517–528

27. Bodwell JE, Ortí E, Coull JM, Pappin DJ, Smith LI, Swift F 1991Identification of phosphorylated sites in the mouse glucocorticoidreceptor. J Biol Chem 266:7549–7555

28. Burnstein KL, Jewell CM, Sar M, Cidlowski JA 1994 Intragenicsequences of the human glucocorticoid receptor complementaryDNA mediate hormone-inducible receptor messenger RNA down-regulation through multiple mechanisms. Mol Endocrinol 8:1764–1773

29. Schaaf MJ, Cidlowski JA 2002 Molecular mechanisms of glucocor-ticoid action and resistance. J Steroid Biochem Mol Biol 83:37–48

30. Scheinman RI, Gualberto A, Jewell CM, Cidlowski JA, Baldwin JrAS 1995 Characterization of mechanisms involved in transrepres-sion of NF-�B by activated glucocorticoid receptors. Mol Cell Biol15:943–953

31. Caldenhoven E, Liden J, Wissink S, Van de Stolpe A, RaaijmakersJ, Koenderman L, Okret S, Gustafsson JA, Van der Saag PT 1995Negative cross-talk between Rel A and the glucocorticoid receptor:a possible mechanism for the anti-inflammatory action of glucocor-ticoids. Mol Endocrinol 9:401–412

32. Saltiel AR, Kahn CR 2001 Insulin signalling and the regulation ofglucose and lipid metabolism. Nature 414:799–806

33. Hall RK, Wang XL, George L, Koch SR, Granner DK 2007 Insulinrepresses phosphoenolpyruvate carboxykinase gene transcriptionby causing the rapid disruption of an active transcription complex:a potential epigenetic effect. Mol Endocrinol 21:550–563

34. Pierreux CE, Ursø B, De Meyts P, Rousseau GG, Lemaigre FP 1998Inhibition by insulin of glucocorticoid-induced gene transcription:involvement of the ligand-binding domain of the glucocorticoidreceptor and independence from the phosphatidylinositol 3-kinaseand mitogen-activated protein kinase pathways. Mol Endocrinol12:1343–1354

35. Kim JB, Sarraf P, Wright M, Yao KM, Mueller E, Solanes G, LowellBB, Spiegelman BM 1998 Nutritional and insulin regulation offatty acid synthetase and leptin gene expression through ADD1/SREBP1. J Clin Invest 101:1–9

36. Najjar SM, Yang Y, Fernstrom MA, Lee SJ, Deangelis AM, RjailyGA, Al-Share QY, Dai T, Miller TA, Ratnam S, Ruch RJ, Smith S,Lin SH, Beauchemin N, Oyarce AM 2005 Insulin acutely decreaseshepatic fatty acid synthase activity. Cell Metab 2:43–53

37. Kino T, Chrousos GP 2001 Glucocorticoid and mineralocorticoidresistance/hypersensitivity syndromes. J Endocrinol 169:437–445

38. Strickland I, Kisich K, Hauk PJ, Vottero A, Chrousos GP, KlemmDJ, Leung DY 2001 High constitutive glucocorticoid receptor � inhuman neutrophils enables them to reduce their spontaneous rate ofcell death in response to corticosteroids. J Exp Med 193:585–593

39. Hauk PJ, Goleva E, Strickland I, Vottero A, Chrousos GP, KisichKO, Leung DY 2002 Increased glucocorticoid receptor � expres-sion converts mouse hybridoma cells to a corticosteroid-insensitivephenotype. Am J Respir Cell Mol Biol 27:361–367

40. Sousa AR, Lane SJ, Cidlowski JA, Staynov DZ, Lee TH 2000 Glu-cocorticoid resistance in asthma is associated with elevated in vivoexpression of the glucocorticoid receptor �-isoform. J Allergy ClinImmunol 105:943–950

41. Goleva E, Li LB, Eves PT, Strand MJ, Martin RJ, Leung DY 2006Increased glucocorticoid receptor � alters steroid response in glu-cocorticoid-insensitive asthma. Am J Respir Crit Care Med 173:607–616


42. Christodoulopoulos P, Leung DY, Elliott MW, Hogg JC, Muro S,Toda M, Laberge S, Hamid QA 2000 Increased number of glu-cocorticoid receptor-�-expressing cells in the airways in fatalasthma. J Allergy Clin Immunol 106:479–484

43. Leung DY, Hamid Q, Vottero A, Szefler SJ, Surs W, Minshall E,Chrousos GP, Klemm DJ 1997 Association of glucocorticoid insen-sitivity with increased expression of glucocorticoid receptor �. JExp Med 186:1567–1574

44. Lee CK, Lee EY, Cho YS, Moon KA, Yoo B, Moon HB 2005Increased expression of glucocorticoid receptor � messenger RNAin patients with ankylosing spondylitis. Korean J Intern Med 20:146–151

45. Livingston E, Darroch CE, Chaudhuri R, McPhee I, McMahon AD,Mackenzie SJ, Thomson NC 2004 Glucocorticoid receptor �:�ratio in blood mononuclear cells is reduced in cigarette smokers. JAllergy Clin Immunol 114:1475–1478

46. Piotrowski P, Burzynski M, Lianeri M, Mostowska M, WudarskiM, Chwalinska-Sadowska H, Jagodzinski PP 2007 Glucocorticoidreceptor � splice variant expression in patients with high and lowactivity of systemic lupus erythematosus. Folia Histochem Cytobiol45:339–342

47. Whorwood CB, Donovan SJ, Wood PJ, Phillips DI 2001 Regulationof glucocorticoid receptor � and � isoforms and type I 11�-hydroxy-steroid dehydrogenase expression in human skeletal muscle cells: akey role in the pathogenesis of insulin resistance? J Clin EndocrinolMetab 86:2296–2308

48. Weinstein SP, Wilson CM, Pritsker A, Cushman SW 1998 Dexa-methasone inhibits insulin-stimulated recruitment of GLUT4 to thecell surface in rat skeletal muscle. Metabolism 47:3–6

49. Black DL 2003 Mechanisms of alternative pre-messenger RNAsplicing. Annu Rev Biochem 72:291–336

50. Zhu J, Gong JY, Goodman Jr OB, Cartegni L, Nanus DM, Shen R2007 Bombesin attenuates pre-mRNA splicing of glucocorticoidreceptor by regulating the expression of serine-arginine proteinp30c (SRp30c) in prostate cancer cells. Biochim Biophys Acta 1773:1087–1094

51. Yan XB, Tang CH, Huang Y, Fang H, Yu ZQ, Wu LM, Liu RY2010 Alternative splicing in exon 9 of glucocorticoid receptor pre-mRNA is regulated by SRp40. Mol Biol Rep 37:1427–1433

52. Kelly A, Bowen H, Jee YK, Mahfiche N, Soh C, Lee T,Hawrylowicz C, Lavender P 2008 The glucocorticoid receptor �

isoform can mediate transcriptional repression by recruiting his-tone deacetylases. J Allergy Clin Immunol 121:203–208.e1

53. Kim SH, Kim DH, Lavender P, Seo JH, Kim YS, Park JS, Kwak SJ,Jee YK 2009 Repression of TNF-�-induced IL-8 expression by theglucocorticoid receptor-� involves inhibition of histone H4 acety-lation. Exp Mol Med 41:297–306

54. Allgood VE, Oakley RH, Cidlowski JA 1993 Modulation by vita-min B6 of glucocorticoid receptor-mediated gene expression re-quires transcription factors in addition to the glucocorticoid recep-tor. J Biol Chem 268:20870–20876

Renew Your Subscription Now!Don’t miss a single issue of our highly-cited,

high-impact factor journals.www.endo-society.org


discovery of glucocorticoid receptor- in mice with a … of glucocorticoid receptor- in mice with a...

Documents