Transplant Immunology 20 (2009) 249–252

Contents lists available at ScienceDirect

Transplant Immunology

j ourna l homepage: www.e lsev ie r.com/ locate / t r im

A novel method for monitoring glucocorticoid-induced changes of the glucocorticoidreceptor in kidney transplant recipients

Yan Chen a,b,⁎, Gilbert J. Burckart a, Tariq Shah b,c, Vera Pravica a,b, Ian V. Hutchinson a,b

a University of Southern California, Los Angeles, CA, USAb National Institute of Transplantation, Los Angeles, CA, USAc St. Vincent Medical Center, Los Angeles; CA, USA

⁎ Corresponding author. 2222 Ocean View Ave., #213Tel.: +1 213 365 3703; fax: +1 213 487 4301.

E-mail address: chenyan@usc.edu (Y. Chen).

0966-3274/$ – see front matter © 2008 Elsevier B.V. Aldoi:10.1016/j.trim.2008.12.003

a b s t r a c t

a r t i c l e i n f o

Article history:

Introduction: Initial high do Received 4 October 2008Received in revised form 17 December 2008Accepted 22 December 2008

Keywords:Glucocorticoid receptor isoformsGlucocorticoid receptor haplotypeKidney transplant

se glucocorticosteroid (GC) treatment of transplant recipients induces changes inthe glucocorticoid receptor (GR) that may affect the mode of action of the drug. We developed a TaqMan one-step RT-PCR quantification method to investigate the effect of bolus GC treatment on the structure of the GRcomplex by measuring the levels of GR isoform expression and FK binding protein 5 (FKBP5) transcription.Methods: Peripheral blood mononuclear cells (PBMCs) were obtained before renal transplantation andimmediately after the initial methylprednisolone treatment. Gene expression of the GR-α, -β, -P isoforms andFKBP5 were quantified using a simplex TaqMan relative quantification assay. Genotyping was performedusing the TaqMan allele discrimination assay.Results: The gene expression assay was validated and is efficient to detect extremely low quantity of GR-βexpression. Increased expression of FKBP5 (14.01+11.52 folds), GR-α (1.47+1.00 folds), GR-β (6.21+5.71folds) and GR-P (2.72+1.90 folds) were observed after steroid bolus. The GR haplotype A-T-G is significantlyrelated to elevation of GRβ transcription (p=0.002). Patients’ hospitalization time after transplantationcorrelated with increment of GR-α (p=0.01) and FKBP5 (p=0.007) gene expression. The GR-β expressionlevel is extremely low compared with the GR-α isoform. By contrast, the GR-P isoform is relativelyabundantly expressed in PBMCs.Conclusion: The method to measure GR isoform enabled us to determine that acute high-dose GC treatmentalters the structure of the GR. The initial therapy with high dose steroid may induce a GC resistancephenotype. The modulatory role of the GR-P isoform is not fully defined but may provide another dimensionin understanding the function of GR and the role of steroid in transplant immunosuppression.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

Drug effects of glucocorticosteroids (GCs) depend on the precisefunction of the major glucocorticoid receptor (GR). For example, GR-αexists in the cytoplasm as a complex, constituted by various types ofcyclophilin, immunophilin and heat shock proteins (hsp), whichtogether are called chaperone proteins. The GR primarily resides in thecytoplasm, bound by two molecules of the hsp 90 and immunophilinslike FKBP5 [1]. Upon ligand binding, the GR complex translocatesinside the nucleus and two molecules of GR-α are dimerized. Thereceptor homodimers then recognize and interact with specific cis-acting sequences, denoted as glucocorticoid responsive elements(GREs) in the target genes. The interaction of GR-GRE is stabilized bythe recruitment of other coactivators, which in turn elicits assembly ofthe general transcription machinery consisting of RNA polymerase IIand other ancillary factors and activate gene transcription [1,2]. On the

, Los Angeles, CA, USA 90057.

l rights reserved.

other hand, GR-α monomers bound with GCs could interact withother transcriptional regulators, such as nuclear factor κB (NF-κB), tomodify gene transcription negatively (transrepression) or sometimespositively (transactivation) [3–6].

Previous studies have demonstrated that FKBP5 maintains GRconformation and regulates GR stability. GC sensitivitymay be affectedby the function of FKBP5, since the in vitro overexpression of humanFKBP5 attenuates hormone binding affinity and nuclear translocationactivity of GR [7]. NewWorldMonkeys represent a natural model withoverexpressed FKBP5 protein and manifest GC insensitivity, withraised plasma cortisol levels [7]. Although it is unlikely that FKBP5plays a significant role in FK506-mediated immunosuppression inhumans, gene expression for FKBP5 is unambiguously induced by GCsin a dose-dependentmatter [8]. Thereby, FKBP5 andGRmay constitutea short negative feedback loop, by which the FKBP5 gene expression isdirectly regulated by GC and in turn affects hormone sensitivitythrough the GR [7].

Studies have also indicated that an individual response to GCtreatment is associated with number and affinity of GR-α [9].However to complicate current understanding, various GR isoforms

Table 1

A) The validation experiment between 18 S and GR-α

Input mRNA (ng) Log of input (X) Ave Ct (18 s) Ave Ct (GRα) ΔCt (GRα-18 s) (Y)

60.00 1.78 12.75 24.15 11.4012.00 1.08 14.80 26.67 11.872.40 0.38 16.95 28.72 11.770.48 −0.32 19.50 31.14 11.64Y=0.09X+11.7

B) The validation test between GR-α and -P isoform detection

Input mRNA (ng) Log of input (X) Ave Ct (GRα) Ave Ct (GRp) ΔCt GRp-GRα (Y)

480.00 2.68 20.62 23.58 2.97240.00 2.38 21.11 24.54 3.44120.00 2.08 22.66 25.68 3.0260.00 1.78 23.82 26.86 3.0430.00 1.48 24.66 27.90 3.24Y=0.01X+3.1

C) The validation test between GR-P and –β isoform detection

Input mRNA (ng) Log of input (X) Ave Ct (GRP) Ave Ct (GRβ) ΔCt (GRβ-GRP) (Y)

480.00 2.68 23.58 27.48 3.90240.00 2.38 24.54 28.19 3.65120.00 2.08 25.68 29.64 3.9660.00 1.78 26.86 30.86 4.0030.00 1.48 27.9 31.76 3.86Y=0.03X+3.79

250 Y. Chen et al. / Transplant Immunology 20 (2009) 249–252

have been identified so far, possibly defining GC sensitivity coopera-tively through their ratios with GR-α. Due to the mechanism ofalternative splicing, GR-α (777 AA) and the GR-β isoform (742 AA) aretranscribed using a different segment of exon 9 [10]. GR-β does notbind hormone but rather GR-α itself and its coactivators, and thusmay hinder the activity of GR-α through forming a heterodimer withGR-α and competing with GR-α for coactivator binding [5,11,12]. Thelack of exon 8 in GR-P (or GR-δ), replaced by intron G gives rise to atruncated protein (676 AA) without ligand binding function [1]. Byitself, GR-P has been exhibited with a lower transactivation activitythan GR-α, however with synergistic activities when co-transfectingGR-P with the -α isoform in certain cell types [1]. Expression of GRisoforms has been implicated subject to changes following GCsexposure, namely upregulation of GR-β and downregulation of GR-α[7,8], and hence may potentially alter the individual response to GCs.

2. Objective or hypothesis

In organ transplantation, patients are ubiquitously exposed to thehigh doses of bolus GC treatment. Here we investigated the effects ofsuperphysiological GC dosages on the structure of the GR complex,immediately post transplantation, by measuring the levels of GRisoform and FKBP5 transcription, using the innovative one-stepTaqMan gene expression quantification method. The objective of thestudy was to assess the association between single nucleotidepolymorphisms (SNPs), and GR and FKBP5 gene transcription. More-over, clinical data among the patients were collected in order toinitiate a pilot study on the potential clinical consequences followingconstitutional changes of the GR complex.

3. Methods and materials

3.1. Patients and sample collection

There were 26 kidney transplant recipients enrolled and followedfor up to one year after transplantation in this study. All participantshad signed written consent approved by the Institutional ReviewBoard of St. Vincent Medical Center (IRB No. 04-033).

Each of 8 ml peripheral blood samples was collected byvenipuncture from patients at 1–3 days before transplantation and1–2 days after 4-day methylprednisolone pulse (i.e. 500 mg duringtransplantation, day 1; 200 mg, day 2; 160 mg and day 3; 120 mg aftertransplantation), respectively. 7 ml whole blood sample was appliedfor PBMC separation, using HISTOPAQUE®-1077 (Sigma-Aldrich, Inc.,USA) by density gradient centrifugation. PBMC was processed formRNA extraction using the QIAamp RNA blood Mini Kit (Qiagen,GmbH, Germany), and followed by clean-upwith the RNeasyMinEluteCleanup kit (Qiagen, GmbH, Germany).

For DNA extraction, 1 ml peripheral whole blood was processed onthe KINGFISHER ml equipment (Thermo Fisher Scientific, Inc, USA),using BioSprint 15 DNA kits (QIAGEN GmbH, Germany).

3.2. TaqMan gene expression assays

GR-α, GR-β and GR-P isoforms and FKBP5 were quantified by thesimplex TaqMan relative quantification assay, using the one-stepAB7900HT RT-PCR system [Applied Biosystems (ABI)]. 18 S ribosomalRNA (18 S rRNA) (ABI) served as an endogenous control and pre-transplant patients’ samples as calibrator. FKBP5 mRNA expressionwas quantified by a validated assay (Hs01561009_m1, ABI). Wedeveloped our own TaqMan quantification assay for the GR isoforms,partially referring to thework by Hagendorf et al [13]. The primers andprobe (5'-N3') are shown as follows: the generic probe: 6FAM-CCTTACTGCTTCTCTCT-MGBNFQ; the generic forward primer: 5'-GTTTCCTCTGAGTTACACAGGCTTC-3'; the GR-α reverse primer: 5'-GGGAATTCAATACTCATGGTCTTATCCAA-3'; the GR-β reverse primer:

5'-TTAATGTGTGAGATGTGCTTTCTGGTT-3'; and the GR-P reverseprime: 5'-ACCTCTCTGTTTCTGCCATACCTATT-3'. The assays havepassed the validation test in order to render equivalent RT-PCR effi-ciency among assays.

For FKBP5 and 18 S rRNA, RT-PCR was performed in 25 µl reactionmixture containing 1.25 µl 20x primer and probe mix (ABI) and 0.5 µlROX Passive Reference Dye (Bio-Rad) using iScript™ One-Step RT-PCRkit (Bio-Rad). For the GR isoforms, the 25 µl reaction mixture contains10 µM reverse primer, 10 µM forward primer and 6 µM probe, instead.Apart from applying 1 ng of total mRNA in the 18 S rRNA assay, 500 ngof total mRNAwere used in the others. RT-PCR reaction steps include:1) 50 °C for 10 min, 2) 95 °C for 5 min, 3)40 cycles consisting 95 °C for15 s and 58 °C for 1 min.

3.3. The polymorphism study of the GR and FKBP5 genes

Validated TaqMan allele discrimination assays (ABI) were used todetect the SNPs, namely FKBP5 intron B⁎C/T (rs1360780), GR33388⁎A/T(rs33388) andGR766⁎T/C (rs6196). However, GR3669⁎A/G (rs6198)wasdetected by ABI assay-on-demand (VIC probe; 5'-CTTTTATTTTTTCATT-TAAATTT-3', FAM probe; 5'-TTTATTTTTTCGTTTAAATTT-3', forward pri-mer; 5'-TCAGACTGTAAAACCTTGTGTGGAA-3' and reverse primer 5'-CCAATTCGGTACAAATGTGTGGTT-3'). DNA samples were processedusing the AB7900HT RT-PCR instrument. GR haplotypes were deter-mined by alleles of the GR33388⁎A/T, 766⁎T/C and 3669⁎A/G poly-morphisms according to Stevens et al [14] and using the Arlequin 3.01software.

4. Results

In the validation experiment, we tested how the ΔCT (CT target−CT reference) of RT-PCRvaries with mRNA template dilution to determine if the efficiency of the targetamplification (i.e. the assays we designed) and the efficiency of the referenceamplification (i.e. the validated assay) are approximately equal. The ΔCT valuesgenerated were plotted against log input amount of mRNA to create a semi-logregression line (Table 1A–B). The absolute value of the slope resulted from theregression line is below 0.1, as suggested by ABI.

Constitutional FKBP5 and GR gene expression was first studied among pre-transplant samples. GR-β expression is extremely low compared with the GR-α isoform[GR-β/GR-α=(1.96±1.89)×10−3:1, Mean±SD], whereas the GR-P expression is rela-tively abundant (GR-P/GR-α=0.24±0. 09:1). PBMC samples after GC exposurewere thencomparedwith samples before exposure to generate fold increase of gene expression. Ingeneral, FKBP5 and GR-β gene expression were significantly increased. On the otherhand, GR-α and GR-P were less affected following the GC treatment (Fig. 1).

Table 2Hospitalization time is linked to fold increase in GR and FKBP5 gene expression

Hospital stay (d) N aGR-α mRNA GR-β mRNA GR-p mRNA bFKBP5 mRNA

≤6 22 1.21±0.66 5.80±1.18 2.21±0.96 10.65±7.98N6 4 2.88±1.48 8.49±3.26 5.54±1.57 32.51±11.04

ap=0.01 bp=0.007 by Kruskal–Wallis test.

Fig. 1. Increment of gene expression of the GR complex after the GC treatment.

251Y. Chen et al. / Transplant Immunology 20 (2009) 249–252

Five GR haplotypes were identified among samples, i.e. T-T-A (n=29), A-T-A (n=10),A-C-A (n=7), A-T-G (n=5), and T-C-A (n=1) based on the tagging SNPs, namelyGR33388⁎A/T, GR766⁎T/C and GR3669⁎A/G (Ref [14]). Increment of GR-β expressionpost transplant is tightly related to the GR haplotype A-T-G (Fig. 2). However, FKBP5genotype was not associated with the magnitude of expression increment induced byGC.

Fromall the confounding factors, patients' hospitalization time after transplantationis significantly associated with increase of GR-α and FKBP5 gene expression (Table 2).This association was not found with other post transplant observations [acute/chronicrejection, serum creatinine (Cr) levels at discharge, surgical complications, time to thenormal Cr level]. Also, demographic differences (age, race and gender), concommitantmedications (Simulect, mycophenolate, cyclosporine, tacrolimus), and pre-transplantconditions (HLAmismatching, dialysis, original disease and previous steroid treatment),did not significantly influence gene expression regulation by GCs.

5. Discussion

In this study, a common forward primer and probe, paired with anisoform-specific reverse primer, were designed and applied to measuregene expression of GR-α, -β, -P isoforms. Due to extremely lowexpressed GR-β, a series of stringent verification tests was performedto confirmaunified experimental efficiencyamong the customized (GR-α, β and P) and the endogenous control (18 s) assays. The ΔCT valuesgenerated from TaqMan RT-PCR amplification were plotted against loginput amount ofmRNA to create a semi-log regression line. The absolutevalue of the slope resulted from the regression line is below 0.1 todetermine that the efficiency of assay that we developed is approxi-mately equal to the validated assay by ABI.

A number of studies have shown that decrease in GR-α numberwithin circulating lymphocytes, induced by the long-term low dose ofGC treatment and short-termbolus, may be associatedwith lymphocyteresistance to GCs, thereby a risk of acute allograft rejection and graft lossunderGC therapy [8,9,15–17]. For example, Berki et al has shown that 1 g

Fig. 2. The effect of GCs on GR-β expression is related to GR haplotype, i.e. foldincrement of GR-β gene expression is 2.98±1.67 (mean±SD) in A-T-G negativeindividuals (n=21), whereas 10.66±7.34 in positive individuals (n=5), p=0.002 usingMann–Whitney Rank Sum analysis.

methylprednisolone booster during the operation resulted in atemporary decrease of GR-α level in most renal transplant patients.The receptor level normalized after 1 week of surgery in 50% of thepatients, detected by using FITC-labeled anti-GCR mAb within theperipheral lymphocyte population [17]. Here we did not find significantalteration inGR-αnumber, nevertheless theGR-β and FKBP5 expressionlevel was unambiguously up-regulated following the initial GC treat-ment within PBMCs. As for GR-α, discrepancy between our study andothers could bedue to difference of cell population studied, i.e. PBMCs inthis study and peripheral lymphocytes in some other studies. Inaddition, we are also aware that transplant patients in our center wereadministrated with a lower initial methylprednisolone dosage, i.e.500 mg, compared with other patient groups in the related studies.

The function GR-β has been proposed as a biologically relevantinhibitor of GR-α, participating in defining the sensitivity of the targettissue to GCs [11]. GR-β is transcriptionally inactive, and does not haveGCs binding capability. By itself, GR-β does not possess any transcrip-tional activity [18]. The mechanism of GR-β in GC resistance is possiblydue to specific repression of transactivation on endogenous GR-α bybinding to GR-α itself and its transcription coactivators [5,11,12].Upregulated GR-β expression has been documented in PBMC of severesteroid resistant rheumatoid arthritis (RA), and inflammatory boweldisease (IBD) patients [11,19,20]. However, no conclusion has beendrawnas for association ofGR-βwithGC resistance. In the current study,we have observed average increment of GR-β expression followingmethylprednisolone pulse. To significantly inhibit glucocorticoid-mediated gene expression, a number of transfection experiments haveshown that the GR-β isoform has to be expressed in at least a 5- to 10-fold excess relative to GR-α [4,21]. Thus, it is unlikely that the very lowamounts of GR-β can exert a trans-dominant negative effect [4], sincethe expression of GR-β in all tissues is far less than the amount of GR-αtranscripts as examined in this study and others.

Interestingly,we found a strong correlationbetween theGR33388⁎A-GR766⁎T-GR3669⁎G haplotype and GR-β gene expression withinsubjects. Such an association may be due to the presence of theGR3669⁎A-NG SNP, which has previously been documented linked toelevated GR-β mRNA stability and receptor protein expression in vitro.The SNP is located at the first A in an AUUUA motif of 3' untranslatedregion of GR-βmRNA AUUUAmotifs in mRNA, which are often found inAU-rich elements [22] in the 3'UTR of short-lived messengers ofcytokines, transcription factors and proto-oncogenes, having an essentialfunction in destabilization of mRNA [23]. Therefore, we were able tosubstantiate phenotypic association of the GR3669⁎A-NG SNP in an invivo study.

In addition to genetic predisposition, proinflammatory cytokines,e.g. IL-8, TNF-α and IL-1, have been demonstrated to increase GR-βexpression [21], so do IL-2 and IL-4 through distorting the GR-β-bearing population [20]. From our data, increment of GR-β and theGR33388⁎A-GR766⁎T-GR3669⁎G haplotype were not in an apparentassociation with the steroid resistant status, demonstrated byepisodes of acute/chronic rejection. Therefore, it is hard to determineif increased GR-β expression is merely a secondary result of thesystemic inflammatory reaction, in many GC-treated diseases whereinflammation frequently takes place [4].

To explore the potential application ofmonitoring theGR complex inrenal transplant patients, we followed up our patients one year aftersurgery. Although acute rejection episodes were not associated witheither FKBP5 or GR isoform expression levels, recipients' hospitalization

252 Y. Chen et al. / Transplant Immunology 20 (2009) 249–252

time was related to the gene up-regulation of FKBP5 and GR-α.Preliminary causes vary among four patients with more than 6 days ofhospitalization. Among them, two patients had a positive cross-matchbefore transplantation and experienced 11 and 12 days of hospital stay,respectively. One transplant patient, staying in hospital for 8 days,received an extended criteria kidney. The fourth patient with 12 days ofstay was noted with ischemic areas in the transplanted kidney duringthe surgery. All these patients experienced delayed urine productionafter transplantation, and also large increment of FKBP5 (average 32fold) and a statistically significant increase of GR-α (around 3 fold) geneexpression were found. Although the identified factors could directlycontribute to patients' delayed graft function, evident increase of FKBP5expression may affect GC treatment sensitivity by attenuating hormonebinding affinity and nuclear translocation of GR [7], thus furtherimpeding organ accommodation. It is alsoworthy to note that alterationof GR-α and FKBP5 might be a consequence due to the transplanttreatment in addition to the immunosuppressive regimen. Previousstudies have shown that mitogen stimulation, inflammatory and anti-inflammatory cytokines could increase binding affinity and density ofGR [24]. Thereby, wound, hematoma and specific responses to donorantigens may affect regulation of GR-α gene expression as well as thepivotal components of the GR complex. More thorough studies arenecessary in order to draw further conclusions. However, prolongedhospitalization related to delayed graft functionmay be associatedwithattenuated organ survival rates, and hence post-transplant monitoringof the GR complex might provide valuable assessment for long-termtransplant outcomes.

Finally, the abundance of GR-P gene expression in transplantpatients' PBMCs was examined from our data. According to literature,GR-P/δwas first isolated from a GC-refractory myeloma patient and ismost often found in hematological malignancies at high levels [25].The expression of GR-P is diverse among different types of diseases,for example, ranging from 23 to 54% of total GR mRNA in multiplemyeloma (MM) patients, and 10–20% in normal peripheral lympho-cytes [25]. Nevertheless, quantification of GR-P mRNA has not yetshown a clear relevance with steroid sensitivity from various studiesin different diseases. Intriguingly, several transient transfection assayshave shown synergy when GR-P is co-transfected with the GR-αisoform [1]. In our study, we found patients who stayed in hospitallonger than 6 days tend to have a larger increment of GR-P geneexpression than the others, although statistical significance was notreached (p=0.201) and the possible clinical association is unclear.However, abundance of this GR isoform and its up-regulationfollowing the GC treatment in transplantation raise questions aboutthe physiological importance of GR-P that has a ubiquitous appear-ance in human tissues [25].

In conclusion, we have developed a reliable and time efficient (one-step RT-PCR) method to monitor several markers in the GR complex,including GR alternative splicing variants and the chaperon proteinFKBP5. This technique identified GC treatment-induced alteration inGR gene expression. Increment of FKBP5 and GR-β expression may berelated to a GR resistance phenotype. However, clinical consequences,such as acute rejection, may not be obvious in association with thepivotal molecules of drug metabolism pathways within the transplantsetting, simply because of concomitant administration of other potentimmunosuppressants such as the cacineurin inhibitors. Associationbetween GR genotypes and GR-β phenotypes was also demonstratedhere, and a larger number of patients are being enrolled to confirm theresult. In this study, we found the GR-P isoform is relatively abundant(GR-P/GR-α ratio=0.24±0.09:1), compared with the GR-β expressionlevel [GR-β/GR-α=(1.96±1.89)x10-3:1]. So the three-fold post-treat-ment increasemay be functionally relevant, and study of GR-P functionmay provide another dimension in understanding the mechanism ofGC resistance in transplantation.

Acknowledgement

The authors would like to acknowledge the assistance of Dr. YongCho in statistical analysis.

References

[1] Zhou J, Cidlowski JA. The human glucocorticoid receptor: one gene, multipleproteins and diverse responses. Steroids 2005;70(5–7):407–17.

[2] Chrousos GP, Kino T. Intracellular glucocorticoid signaling: a formerly simplesystem turns stochastic. Sci STKE 2005;2005(304):e48.

[3] Bray PJ, Cotton RG. Variations of the human glucocorticoid receptor gene (NR3C1):pathological and in vitro mutations and polymorphisms. Hum Mutat 2003;21(6):557–68.

[4] Farrell RJ, Kelleher D. Glucocorticoid resistance in inflammatory bowel disease;2003. p. 339–46.

[5] DeRijk RH, Schaaf M, de Kloet ER. Glucocorticoid receptor variants: clinicalimplications. J Steroid Biochem Mol Biol 2002;81(2):103–22.

[6] Webster JC, Cidlowski JA. Mechanisms of Glucocorticoid-receptor-mediatedrepression of gene expression. Trends Endocrinol Metab 1999;10(10):396–402.

[7] Binder EB, Salyakina D, Lichtner P, Wochnik GM, Ising M, Putz B, et al.Polymorphisms in FKBP5 are associated with increased recurrence of depressiveepisodes and rapid response to antidepressant treatment. Nat Genet 2004;36(12):1319–25.

[8] Vermeer H, Hendriks-Stegeman BI, van der Burg B, van Buul-Offers SC, Jansen M.Glucocorticoid-induced increase in lymphocytic FKBP51 messenger ribonucleic acidexpression: a potential marker for glucocorticoid sensitivity, potency, and bioavail-ability. J Clin Endocrinol Metab 2003;88(1):277–84.

[9] Ribarac-Stepic N, Isenovic E, Naumovic R, Koricanac G, Vulovic M, Zakula Z, et al.Glucocorticoid receptors in lymphocytes and stability of kidney graft function. ClinExp Med 2001;1(4):179–86.

[10] Flood L, Lofberg R, Stierna P, Wikstrom AC. Glucocorticoid receptor mRNA inpatients with ulcerative colitis: a study of responders and nonresponders toglucocorticosteroid therapy. Inflamm Bowel Dis 2001;7(3):202–9.

[11] Lu NZ, Cidlowski JA. Glucocorticoid receptor isoforms generate transcriptionspecificity. Trends Cell Biol 2006;16(6):301–7.

[12] Oakley RH, Jewell CM, Yudt MR, Bofetiado DM, Cidlowski JA. The dominantnegative activity of the human glucocorticoid receptor beta isoform. Specificityand mechanisms of action. J Biol Chem 1999;274(39):27857–66.

[13] Hagendorf A, Koper JW, de Jong FH, Brinkmann AO, Lamberts SWJ, Feelders RA.Expression of theHumanGlucocorticoid Receptor Splice Variants {alpha}, {beta}, andP in Peripheral Blood Mononuclear Leukocytes in Healthy Controls and in Patientswith Hyper- and Hypocortisolism; 2005. p. 6237–43.

[14] Stevens A, Ray DW, Zeggini E, John S, Richards HL, Griffiths CEM, Donn R.Glucocorticoid Sensitivity Is Determined by a Specific Glucocorticoid ReceptorHaplotype; 2004. p. 892–7.

[15] Baughman G, Wiederrecht GJ, Chang F, Martin MM, Bourgeois S. Tissuedistribution and abundance of human FKBP51, and FK506-binding protein thatcan mediate calcineurin inhibition. Biochem Biophys Res Commun 1997;232(2):437–43.

[16] Shahidi H, Vottero A, Stratakis CA, Taymans SE, Karl M, Longui CA, et al. Imbalancedexpression of the glucocorticoid receptor isoforms in cultured lymphocytes from apatient with systemic glucocorticoid resistance and chronic lymphocytic leukemia.Biochem Biophys Res Commun 1999;254(3):559–65.

[17] Berki T, Tavakoli A, Nagy KK, Nagy G, Nemeth P. Alterations of glucocorticoid receptorexpression during glucocorticoid hormone therapy in renal transplant patients.Transpl Int 2002;15(2–3):132–8.

[18] Li LB, Leung DY, Hall CF, Goleva E. Divergent expression and function ofglucocorticoid receptor beta in human monocytes and T cells. J Leukoc Biol 2006;79(4):818–27.

[19] Wilkstrom AC. Nuclear receptors and their role in regulation of inflammation. Aspecial section consisting of proceedings from the EUROSTERONE meeting,Huddinge, Sweden, 27 September 2000. J Endocrinol 2001;169(3):425–8.

[20] Honda M, Orii F, Ayabe T, Imai S, Ashida T, Obara T, et al. Expression ofglucocorticoid receptor [beta] in lymphocytes of patients with glucocorticoid-resistant ulcerative colitis. Gastroenterology 2000;118(5):859–66.

[21] Yudt MR, Cidlowski JA. The Glucocorticoid Receptor: Coding a Diversity of Proteinsand Responses through a Single Gene; 2002. p. 1719–26.

[22] Mungall AJ, Palmer SA, Sims SK, Edwards CA, Ashurst JL, Wilming L, et al. The DNAsequence and analysis of human chromosome 6. Nature 2003;425(6960):805–11.

[23] Schaaf MJ, Cidlowski JA. AUUUA motifs in the 3'UTR of human glucocorticoidreceptor alpha and beta mRNA destabilize mRNA and decrease receptor proteinexpression. Steroids 2002;67(7):627–36.

[24] Yeager MP, Guyre PM, Munck AU. Glucocorticoid regulation of the inflammatoryresponse to injury; 2004. p. 799–813.

[25] de Lange P, Segeren CM, Koper JW, Wiemer E, Sonneveld P, Brinkmann AO, et al.Expression in hematological malignancies of a glucocorticoid receptor splicevariant that augments glucocorticoid receptor-mediated effects in transfectedcells. Cancer Res 2001;61:3937–41.