The glucocorticoid receptor gene as a candidate for gene ... - Nature

Gene Therapy (1999) 6, 245–252  1999 Stockton Press All rights reserved 0969-7128/99 $12.00 http://www.stockton-press.co.uk/gt

The glucocorticoid receptor gene as a candidate for gene therapy in asthma M Mathieu1, C Gougat1, D Jaffuel1, M Danielsen2, P Godard1, J Bousquet1 and P Demoly1 1

Institut National de la Sante´ et de la Recherche Me´dicale, U454 and the Service des Maladies Respiratoires, CHU de Montpellier, Montpellier, France; and 2Department of Biochemistry and Molecular Biology, Georgetown University Medical School, Washington, DC, USA

Glucocorticoids (GC) are commonly used as anti-inflammatory drugs in asthma, but can produce serious secondary effects and, moreover, be inefficient in corticoresistant asthmatics. After binding to the glucocorticoid receptor (GR), they repress the synthesis of proinflammatory cytokines via inhibition of the transcription factors AP-1 and NF␬B. Since qualitative and quantitative defects of the GR have been reported in corticoresistant patients, the transfer of the GR gene in the lung epithelium, the primary site of inflammation in asthma, may restore sensitivity to GC in these patients. As a prerequisite to in vivo studies, we have transfected A549 human lung epithelial cells with a GR

expression vector. Using AP-1 and NF-␬B-dependent reporter gene assays and an immunoassay for the proinflammatory cytokine RANTES, we show that the overexpressed GR significantly repressed AP-1 and NF-␬B activities in the absence of hormone and that the GC dexamethasone produced an additive inhibitory effect. The GC-independent repression of AP-1 and NF-␬B activities was further demonstrated by overexpressing a ligandbinding deficient GR mutant. Our data suggest that delivery of the GR gene in vivo may reduce inflammation without recourse to GC and may constitute an alternative therapeutic approach for corticoresistant asthma.

Keywords: glucocorticoid receptor; AP-1; NF-␬B; inflammation; asthma

Introduction Glucocorticoids (GC) are used as the major treatment of asthma and other chronic inflammatory diseases. After binding to the glucocorticoid receptor (GR), these hormones inhibit or increase gene expression through processes known as transrepression and transactivation, respectively. Transrepression results from inhibitory protein–protein interactions between the hormone-activated GR and other transcription factors such as AP-1 and NF-␬B.1 AP-1 is a dimer made of peptides belonging to the c-Fos and c-Jun families2,3 whereas NF-␬B is a dimer composed of rel-related proteins among which are p65 and p50.4,5 AP-1 response elements, also called 12-O-tetradecanoyl-phorbol-13-acetate(TPA)-response elements (TRE) because TPA increases AP-1 activity, and NF-␬B response elements (NF-␬BRE) have been found in the promoter regions of many genes encoding proinflammatory mediators and shown to be necessary for the stimulation of their gene transcription.1 Transactivation is mediated through direct binding of the hormone-activated GR to DNA sequences called glucocorticoid response elements (GRE)6 which are found in another set of genes, some of which are involved in the control of neoglucogenesis, arterial pressure and intraocular tension.7–12 Whereas transrepression of AP-1 and NF-␬B activities

Correspondence: M Mathieu, INSERM U454, Hoˆpital A de Villeneuve, 34295 Montpellier Cedex 5, France Received 30 June 1998; accepted 16 September 1998

is clearly implicated in the anti-inflammatory effect of GC, transactivation may account for some of their sideeffects (diabetes, arterial hypertension, glaucoma). These effects appear when high doses of GC are prescribed to patients with the most severe form of asthma (so-called corticodependent asthma). In addition, a subset of these asthmatic patients are resistant to GC (corticoresistant) and there is no satisfactory alternative treatment. These groups of patients, who suffer from an increase in morbidity and mortality rates, would greatly benefit from gene therapy approaches that would reduce or eliminate the recourse to GC.13 Previous studies have shown that cells of corticoresistant asthmatics contain a GR with either a decreased affinity for the hormone or a reduced number of GR,14 and high levels of AP-1,15 suggesting that the repressive capacity of the endogenous GR is overwhelmed by the excess of this proinflammatory transcription factor. Thus, it may be possible to restore a corticosensitivity in these patients by overexpressing the GR in the lung epithelium, the primary site of inflammation. Lung epithelial cells participate in and are affected by local inflammation, respond to GC treatment16 and represent a potential target for gene therapy in asthma.13,17 As a prerequisite to animal experiments and to validate the approach, we have transferred the GR gene into A549 lung epithelial cells and COS-1 kidney cells, and tested whether repression of AP-1 and NF-␬B activities by GC was then improved. This study was performed using AP1 and NF-␬B-dependent gene reporter assays and an immunoassay for the proinflammatory cytokine RANTES whose synthesis is controlled by these transcription factors.

Anti-inflammation by the glucocorticoid receptor M Mathieu et al

246

Results Expression and subcellular localisation of the wild-type GR and of a hormone-binding deficient GR mutant The transfer of the GR gene into A549 human lung epithelial cells and COS-1 monkey kidney epithelial cells was analysed by Western blotting and indirect immunofluorescence microscopy. The GR was present endogenously at a higher amount in A549 cells than in COS-1 cells as shown by Western blot analysis (Figure 1, pCItransfected cells). Expression of the GR greatly increased in both cell lines after transfection with a GR expression vector (Figure 1, pCIGR-transfected cells). However, a higher level of exogenous GR was reached in A549 cells. By immunofluorescence microscopy, endogenous GR was not detectable in either cell line (Figure 2, panels A, B, G and H) at the settings used to observe exogenous GR present in cells transfected with the GR expression vector (Figure 2, panels C, D, I and J). Subcellular localisation of the overexpressed GR was found to be similar in both cell lines. In the absence of the synthetic glucocorticoid dexamethasone (DEX), the GR was detected in both the cytoplasm and nucleus of transfected cells, but expression was predominant in the latter compartment (Figure 2, panels C and I). Following DEX treatment, cytoplasmic expression of GR could no longer be detected, indicating the occurrence of a hormone-driven nuclear translocation of the GR (Figure 2, panels D and J). The GR was excluded from the nucleoli, confirming results from previous studies.18,19 Transfection of an expression vector for a ligand binding-deficient GR mutant (pCINA) into these cells further demonstrated that the majority of the GR molecules

Figure 1 Expression of the GR in A549 and COS-1 cells. The cells were transfected with 4 ␮g of either the empty vector pCI, or the expression vectors for the wild-type GR (pCIGR) or for a ligand-binding-deficient GR mutant (pCINA) as indicated. After 24 h, cell extracts were prepared, and 20 ␮g of denatured proteins were analysed by Western blotting using an anti-GR monoclonal antibody as described in Materials and methods.

accumulates in the nucleus hormone independently. Indeed, this mutant was present mainly in the cell nuclei (Figure 2, panels E and K), and its distribution was not affected by DEX treatment (Figure 2, panels F and L). The GR mutant, which harbours two amino acid exchanges in the ligand-binding domain, was transferred into the cells as efficiently as the wild-type receptor (Figure 1, pCINA-transfected cells).

Repression of AP-1 activity by overexpression of the GR To examine the effect of GR overexpression on AP-1 activity and its repression by DEX, A549 and COS-1 cells were transfected with the AP-1-dependent reporter construct 5×TRE TATA Luc in the presence and absence of DEX. AP-1 activity was induced by cotransfecting an

Figure 2 Subcellular distribution of the wild-type GR and of a ligandbinding-deficient GR mutant overexpressed in A549 and COS-1 cells. The cells were transfected with 500 ng of either the empty vector pCI, or the expression vectors for the wild-type GR (pCIGR) or for a ligand-bindingdeficient GR mutant (pCINA) as indicated, and were treated (+DEX) or not (−DEX) with 100 nm DEX. Cells were then fixed and processed for anti-GR immunofluorescence as described in Materials and methods.


247

expression vector for the AP-1 component c-Fos (pCIFos). In both cell lines, cotransfection with increasing amounts of a GR expression vector (pCIGR) improved repression of AP-1 activity by DEX. The addition of DEX inhibited AP-1 activity by about 50% through the endogenous GR and by 85% after transfection with 500 ng/well of pCIGR (Figure 3a and b). Surprisingly, in A549 cells, transcription from 5×TRE TATA Luc was also strongly repressed (by about 80%) following overexpression of the GR even in the absence of DEX (Figure 3a). In contrast, in COS-1 cells, an excess of hormone-free GR inhibited AP-1 activity by only 20% (Figure 3b). Basal transcription as well as 12-O-tetradecanoyl-phorbol-13-acetate (TPA)induced transcription from 5×TRE TATA Luc were also antagonised following overexpression of the GR in the presence and absence of DEX (data not shown). To exclude the possibility that the inhibition of AP-1 activity by the hormone-free GR was an artefact due to the presence of five tandem repeats of TREs in 5×TRE TATA Luc, we transfected the reporter plasmid −517/+63 Coll Luc which contains the natural collagenase promoter with its single TRE. After treatment with TPA, AP-1 activity increased to a maximum. Overexpression of the GR in the absence of DEX inhibited 50% of AP-1 activity in both A549 and COS-1 cells (Figure 3c and d). Inhibition of the maximal AP-1 activity by DEX through the endogenous GR was 20–25% in both cell lines, while it was above 60% following transfection with 500 ng/well of pCIGR (Figure 3c and d).

Repression of NF-␬B activity by overexpression of the GR Regulation of NF-␬B activity was investigated by transfecting the reporter construct 3×Ig␬ Cona Luc which contains three tandem repeats of a NF-␬BRE in the presence or absence of DEX. NF-␬B activity was increased by cotransfecting the cells with a NF-␬B subunit p65 expression vector (pECEp65). In the absence of hormone, overexpression of the GR inhibited NF-␬B activity by 50% in A549 cells and by 30% in COS-1 cells. Following DEX treatment, an additional repression of NF-␬B activity occurred in both cell lines (Figure 4a and b). Repression of AP-1 and NF-␬B activities by overexpression of a ligand-binding and transactivationdeficient GR mutant Since the repression of AP-1 and NF-␬B activities observed in A549 cells in the absence of DEX could be due to activation of the GR by traces of GC present in the growth medium, we tested if this inhibition still occurred when a hormone-binding deficient GR mutant (pCINA) was overexpressed instead of the wild-type receptor. This point mutant, in which a glutamic acid residue has been replaced by a glycine residue at position 546 in the ligand-binding domain, is transactivation deficient (see below). As shown in Figure 5, this mutant antagonised AP-1 and NF-␬B activities, confirming that hormone binding is not absolutely required for transrepression to occur. However, due to the presence of endogenous GR in A549 cells, DEX treatment further inhibited these activities when low doses of pCINA were transfected (Figure 5a).

Figure 3 Repression of AP-1 activity by overexpression of the GR. A549 and COS-1 cells were transfected with either 5×TRE TATA Luc (a and b), or −517/+63 Coll Luc (c and d), and the expression vectors for c-Fos (pCIFos), and the GR (pCIGR) at the indicated amounts. Cells were left untreated (striped bars) or were stimulated either by 100 nm DEX (black bars), or 150 nm TPA (shaded bars), or a combination of both (open bars) for 20 h, and transcriptional activity was measured by luciferase and ␤galactosidase assays as described in Materials and methods. Data are shown as the percentage of the maximum activity. Values represent the means ± s.e.m. of four independent experiments conducted in triplicates. * P ⬍ 0.05 versus maximum activity.


248

Figure 4 Repression of NF-␬B activity by overexpression of the GR. A549 cells (a) and COS-1 cells (b) were transfected with 3×Ig␬ Cona Luc and the expression vectors for the NF-␬B subunit p65 (pECEp65), and the GR (pCIGR) at the indicated amounts. Cells were left untreated (striped bars) or were stimulated by 100 nm DEX (black bars) for 20 h, and transcriptional activity was measured by luciferase and ␤-galactosidase assays as described in Materials and methods. Data are shown as the percentage of the maximum activity. Values represent the means ± s.e.m. of four independent experiments conducted in triplicates. * P ⬍ 0.05 versus maximum activity.

Strict hormone dependence for transactivation by the GR In contrast to GR-mediated repression of AP-1 and NF␬B activities, transactivation from the GRE-containing reporter plasmid pHHLuc was strictly hormone-dependent in both cell lines (Figure 6a and b). Transactivation did not increase after overexpression of the GR unless DEX was present. Overexpression of the ligand-bindingdeficient GR mutant (pCINA) did not alter basal and DEX-induced transcription from this promoter, confirming the absolute hormone requirement for transactivation by the GR. The absence of transcriptional induction by DEX in COS-1 cells transfected with the control vector pCI (Figure 6b) was probably due to their lower endogenous GR content as compared to A549 cells (see Figure 1). It appears though that this low amount of endogenous receptor was sufficient to trigger a significant repression of AP-1 and NF-␬B activities after hormone addition (Figure 3b, d, and Figure 4b). Inhibition of RANTES production by overexpression of the GR and of a hormone-binding-deficient GR mutant Since our previous observations were made using transfected promoter–gene constructs, we investigated whether the expression of an endogenous gene containing AP-1 and NF-␬B sites in its promoter and implicated in the pathophysiology of asthma would also be affected by overexpression of the GR. One such gene codes for the

Figure 5 Repression of AP-1 and NF-␬B activities by overexpression of a ligand-binding-deficient GR mutant. A549 cells were transfected with either 5×TRE TATA Luc (a), −517/+63 Coll Luc (b), or 3×Ig␬ Cona Luc (c), and the expression vectors for c-Fos (pCIFos), the NF-␬B subunit p65 (pECEp65), and for a ligand-binding-deficient GR mutant (pCINA) at the indicated amounts. Cells were left untreated (striped bars) or were stimulated either by 100 nm DEX (black bars), or 150 nm TPA (shaded bars), or a combination of both (open bars) for 20 h, and transcriptional activity was measured by luciferase and ␤-galactosidase assays as described in Materials and methods. Data are shown as the percentage of the maximum activity. Values represent the means ± s.e.m. of four independent experiments conducted in triplicates. * P ⬍ 0.05 versus maximum activity.

proinflammatory cytokine RANTES, which is positively regulated by TNF-␣ in A549 cells.20,21 Thus, A549 cells were transfected with the expression vector for the wildtype GR (pCIGR) or for the hormone-binding and transactivation-deficient mutant (pCINA). Cells transfected with the empty vector and treated with TNF-␣ produced


249

Figure 7 Inhibition of RANTES production by overexpression of the wildtype GR and a ligand-binding-deficient GR mutant. A549 cells were either transfected with the empty vector pCI, the expression vector for the wildtype GR (pCIGR), or the expression vector for a ligand-binding deficient GR mutant (pCINA) at 500 ng per well. The cells were then treated or not treated with 1 ng/ml TNF-␣ or 100 nm DEX or a combination of both as indicated. Forty-eight hours after transfection, concentration of RANTES in supernatant was determined using a sandwich enzyme immunoassay. Data are shown as the percentage of the highest concentration of RANTES (58 ± 7 pg/57 ␮g total protein per well) which was found in pCI-transfected cells treated with TNF-␣. Relative values represent the means ± s.e.m. of four independent experiments conducted in duplicates. * P ⬍ 0.05 versus maximum activity. Figure 6 Strict hormone dependence for transactivation by the GR. A549 (a) and COS-1 cells (b) were transfected with 600 ng of pHHLuc, and with increasing amounts of the expression vector for wild-type GR (pC1GR) or for the hormone-binding-deficient GR mutant (pCINA) as indicated. Transcriptional activity in the absence (striped bars) or presence (black bars) of 100 nm DEX was measured by luciferase and ␤-galactosidase assays as described in Materials and methods. Data are shown as the ratio of luciferase activity to ␤-galactosidase activity in arbitrary units. Values represent the means ± s.e.m. of four independent experiments conducted in triplicates. * P ⬍ 0.05 versus activity in DEX-treated cells transfected with the empty vector pCI.

the highest amount of RANTES. Results show that overexpression of the wild-type and mutant GR strongly repressed (by about 80%) both basal and TNF-␣-induced RANTES production in the absence of DEX (Figure 7). Hormonal treatment produced an additive inhibitory effect which was most likely mediated through the endogenous or transfected wild-type GR.

Discussion In order to evaluate the possible benefits of a GR-based gene therapy in corticoresistant asthma, we have measured in vitro, at the cellular level, the effects of transferring the gene encoding the GR on AP-1 and NF-␬B proinflammatory activities. We show that overexpression of the GR in A549 human lung epithelial cells repressed AP1 and NF-␬B activities even in the absence of added GC. As opposed to our findings, previous studies have reported that inactivation of AP-1 and NF-␬B by the GR was hormone dependent.22–27 Of note, in one of these studies, a repression of AP-1 activity by the GR occurred in HeLa cells in the absence of added hormone.24 However, this effect was attributed to the putative presence

of GC in the growth medium that would have bound and activated the GR. This possibility was ruled out in our study, since repression of AP-1 and NF-␬B activities was also triggered by a hormone-binding and transactivationdeficient GR mutant. The divergent data may also result from the use of different cell lines. Indeed, we show that the efficiency with which the ligand-free GR antagonises AP-1 and NF-␬B activities is cell type dependent, being higher in A549 cells than in COS-1 cells. In addition, we have observed that AP-1 activity was also inhibited by the hormone-free GR in HeLa cells, but with an efficiency in between that found in A549 and COS-1 cells (data not shown). Immunofluorescence and Western blot analyses indicated that this cell type-dependent efficiency of repression was not due to a difference in subcellular distribution of the GR but rather to a difference in the level of GR overexpression. Indeed, the amount of exogenous GR was higher in A549 cells than in COS-1 cells. Alternatively, various endogenous factors interacting with the GR molecule28 may influence its repressive capacity. Quantitative and qualitative studies will be required to determine if these factors play a role in the cell type-dependent transcriptional repression by the hormone-free GR. Nevertheless, the mechanism of this repression is likely to involve physical association between the steroid-free GR and AP1 or NF-␬B since this interaction has been reported by a number of investigators.24–27 We observed that glucocorticoid treatment triggered an additive repression of AP-1 and NF-␬B activities in all cells that we studied and even after overexpression of the ligand-binding-deficient GR mutant. Since the GR is an


250

ubiquitous protein, the hormone-driven nuclear translocation of the endogenous wild-type GR may partly account for this effect, but other ligand-facilitated processes may be involved, such as the sequestration of the CREB-binding protein, a common coactivator of transcription.29 Alternatively, hormone-activated GR may have induced c-Jun22 and I␬B␣ expression,30,31 two factors known to reduce AP-1 and NF-␬B activities, respectively.32–34 Importantly, the negative regulation of AP-1 and NF␬B activities by the hormone-free GR affected the expression of transfected reporter genes but also that of an endogenous gene which is involved in the pathophysiology of asthma,35 contains AP-1 and NF-␬B sites,20 and is induced by the physiological ligand TNF-␣.21,36 This gene encodes the proinflammatory cytokine RANTES which is chemotactic for monocytes, T cells and eosinophils.37–39 Although the maximal fraction of A549 cells that could be transfected was estimated to be 50%, an 80% drop in RANTES production was observed. This amplified effect may result either from an underestimation of the transfection efficiency or from the inhibition of other proinflammatory cytokines induced by TNF-␣ such as IL-640 which would act in a autocrine/paracrine fashion to stimulate AP-1 and NF-␬B. In conclusion, we report that overexpression of the GR in the absence of hormone inhibited AP-1 and NF-␬B proinflammatory activities in a lung epithelial cell line whereas it did not induce transcription through a GRE. Following hormone treatment, a further repression of AP-1 and NF-␬B activities occurred while transactivation increased. We extrapolate that overexpression of the GR in vivo may reduce inflammation with no or little recourse to GC, and may therefore constitute a complementary or an alternative therapy for corticoresistant asthma. In addition, the possibility that the natural GC triggers GRE-mediated side-effects in vivo could be avoided by transferring the gene encoding the hormonebinding-deficient GR mutant instead of the wild-type GR. This potential gene therapy approach now needs to be tested on an animal model of asthma.

Materials and methods Materials 12-O-tetradecanoyl-phorbol-13-acetate (TPA), transferrinpolylysine, polylysine (P2636), spermine and the synthetic glucocorticoid dexamethasone (DEX) were purchased from Sigma, St Louis, MO, USA. Tumour necrosis factor-␣ (TNF-␣) was purchased from PharMingen (San Diego, CA, USA). Luciferin and DTT were purchased from Promega (Madison, WI, USA). Coenzyme A was purchased from Boehringer Mannheim (Mannheim, Germany). Analytical grade chemicals were purchased from Merk (Darmstadt, Germany) or Carlo Erba (Milan, Italy). Plasmids The plasmid pHHLuc contains sequences from −223 to +105 of the mouse mammary tumour virus promoter (including one functional GRE) fused to the luciferase gene and was purchased from the American Type Culture Collection (Rockville, MD, USA). The luciferase reporter constructs 5xTRE TATA Luc, containing five

copies of the AP-1 binding site (TRE) from the collagenase gene upstream of a TATA element, and −517/+63 Coll Luc, containing part of the collagenase promoter which includes the AP-1 site, were a gift from Peter Herrlich (Institute of Genetics, Karlsruhe, Germany).22 The 3×Ig␬ Cona Luc plasmid which contains three tandem repeats of the NF-␬BRE from the immunoglobulin ␬ chain linked to the conalbumin minimal promoter and the luciferase gene was obtained from Alain Israe¨l (Institut Pasteur, Paris, France). The plasmid CMV␤-gal consisting of the cytomegalovirus early promoter linked to the ␤-galactosidase gene was used as a control vector to correct for variations in transfection efficiency. The mouse GR cDNA was isolated as a BglII– XbaI fragment from pSV2Wrec,41 made blunt-end by Klenow treatment and inserted into the SmaI site of the pCI expression vector (Promega) to create pCIGR. pCINA contains the cDNA of a ligand-binding-deficient GR mutant (point mutations E546G and V437G, the latter being a silent polymorphism). To construct pCINA, pSV2NArec41 was digested by BglII, made blunt-end by Klenow treatment and further cut by XbaI. The cDNA fragment was then ligated into pCI that had been cut by EcoRI, made blunt-end and further digested by XbaI. pCIbased expression vector for human c-Fos (pCIFos)42 was provided by Dany Chalbos (INSERM U148, Montpellier, France). The expression vector for the NF-␬B subunit p65 (pECEp65) and the control vector (pECE) were given by Carl Scheidereit (Max Delbrueck Centre for Molecular Medicine, Berlin, Germany).43

Preparation of AdCMVnull adenovirus AdCMVnull adenovirus, a replication-defective strain of human adenovirus type 5 in which E1A and E1B sequences were replaced by the CMV early promoter, was propagated on the complementing 293 cell line as described.44 Cell culture and transfection A549 human lung carcinoma cells and COS-1 green monkey kidney cells were maintained in Ham’s F12 medium containing 10% FCS, 100 U/ml penicillin, 100 ␮g/ml streptomycin and 2 mm glutamine. Six-well cluster plates (500 000 cells per well) were used for Western blot analysis, 24-well cluster plates (100 000 cells per well) with laminin-coated coverslips for immunofluorescence microscopy, and 48-well cluster plates (50 000 cells per well) for reporter and RANTES assays. On the next day, just before transfection, medium was replaced by serumfree medium. The DNA to be transfected was diluted in 20 mm Hepes pH 7.4, 150 mm NaCl. For reporter assays, it included 60 ng of either 5×TRE TATA Luc, −517/+63 Coll Luc, and 3xIg␬ Cona Luc, or 600 ng of pHHLuc, 25 ng of an internal control plasmid (CMV␤-gal) and various amounts of the expression vectors for c-Fos (pCIFos), the NF-␬B subunit p65 (pECEp65), the wild-type GR (pCIGR) and the ligand-binding-deficient GR mutant (pCINA) per well. The corresponding empty vectors (pCI and pECE) were added so that each well contained the same amount of DNA. For reporter assays and Western blotting, A549 and COS-1 cells were transfected with a system composed of the AdCMVnull adenovirus, transferrin-polylysine and polylysine as described previously.45–47 For immunofluorescence microscopy, A549 cells were transfected by the same method whereas COS-


1 cells were transfected with the polycation polyethylenimine (600 000–1 000 000 g/mol) at 2 × 10−8 mol/␮g DNA.48 After transfection, fresh medium containing 10% FCS and the various treatments was added.

Western blot Cells were washed with cold PBS and lysed in 10 mm Tris-HCl, pH 7.4, 50 mm NaCl, 5 mm EDTA, 1% Nonidet P-40, 10 g/ml aprotinin, 10 g/ml leupeptin and 10 g/ml PMSF. Cell extracts were transferred into microcentrifuge tubes, mixed and left on ice for 10 min. After one cycle of freeze–thaw, they were centrifuged at 12 000 g for 5 min at 4°C. A sample of the supernatant was taken for protein estimation and the remainder adjusted to 1 × Laemmli dissociation buffer. Twenty microgrammes of total protein were subjected to SDS-PAGE on 4–12% gradient gels (Novex, San Diego, CA, USA) and blotted on to nitrocellulose. The blot was then probed with the anti-GR monoclonal antibody NCL-GCR (Novocastra, Newcastle-upon-Tyne, UK) at a 1:200 dilution. Revelation was performed by using a peroxidase- conjugated antimouse IgG (Sigma) and an enhanced chemiluminescence system (NEN, Boston, MA, USA). Subcellular distribution of the GR Transfected cells were fixed by 4% paraformaldehyde and permeabilized by 0.5% (v/v) Triton X-100 in PBS. After blocking in PBS containing 3% BSA and 0.1% (v/v) Tween 20, the cells were incubated with anti-GR rabbit polyclonal antibodies P-20 (Santa Cruz, CA, USA) at 1 ␮g/ml in the same buffer. P-20 antibodies are directed against an epitope at the carboxy terminus corresponding to amino acids 729–748 of mouse GR and to amino acids 750–769 of human or monkey GR. A 1:50 dilution of goat serum was then added for 15 min at 22°C, removed and replaced by tetramethylrhodamine isothiocyanate (TRITC)-coupled goat anti-rabbit immunoglobulin G diluted 1:80 for 30 min at 22°C. Extensive washes were made after fixation, permeabilization and incubation with antibodies. Cells were mounted and observed with a Nikon fluorescence microscope (Nikon, Tokyo, Japan) at a magnification of ×500. Luciferase and ␤-galactosidase assays Transfected cells were lysed by one cycle of freeze–thaw in 100 ␮l of 15 mm Tris-HCl pH 7.8, 60 mm KCl, 15 mm NaCl, 2 mm EDTA, 0.15 mm spermine and 1 mm DTT. After a brief centrifugation, the supernatant was recovered for the reporter assays. Luciferase activity in 40 ␮l of cell extract was measured in a luminometer after injection of 100 ␮l of luciferin mix (25 mm Tris-acetate pH 7.8, 41 mm DTT, 0.125 mm EDTA, 4.67 mm MgSO4, 0.34 mm coenzyme A, 0.66 mm ATP, 0.59 mm luciferin). For ␤galactosidase assay, 10 ␮l of cell extract were mixed with 67 ␮l of the chemiluminescent substrate Galacton-Plus and processed according to the manufacturer’s recommendations (Tropix, Bedford, MA, USA). ␤-Galactosidase activity was measured to correct for differences in transfection efficiency. The coefficient of variation for ␤galactosidase activity was 27% in A549 cells and 37% in COS-1 cells, indicating the absolute necessity to correct luciferase data for differences in transfection efficiency. RANTES immunoassay Transfected A549 cells were treated or not treated with 1 ng/ml of TNF-␣ or 100 nm DEX or a combination of

both. Concentration of RANTES in supernatants collected at 48 h was determined using a quantitative sandwich enzyme immunoassay technique and following the manufacturer’s recommendations (R&D Systems, Oxon, UK).

Statistical analysis All data represent the mean ± s.e.m. Data were analysed using the Mann–Whitney U test. Statistical significance was set at P ⬍ 0.05.

Acknowledgements We are grateful to P Herrlich, A Israe¨l, C Scheidereit and D Chalbos for providing plasmids. We thank DT Curiel, P Boulanger and G Salazar for advice. MD is supported by a grant from the American Cancer Society (DB157). This work was supported by grants from Zeneca France, the Institut Europe´en des Allergies-Fondation de France, and the Conseil Re´gional du Languedoc-Roussillon.

References 1 Cato A, Wade E. Molecular mechanisms of anti-inflammatory action of glucocorticoids. Bioessays 1996; 18: 371–378. 2 Angel P, Karin M. The role of Jun, Fos and the AP-1 complex in cell-proliferation and transformation. Biochim Biophys Acta 1991; 1072: 129–157. 3 Abate C, Curran T. Encounters with Fos and Jun on the road to AP-1. Semin Cancer Biol 1990; 1: 19–26. 4 Kawakami K, Scheidereit C, Roeder R. Identification and purification of a human immunoglobulin-enhancer-binding protein (NF-kappa B) that activates transcription from a human immunodeficiency virus type 1 promoter in vitro. Proc Natl Acad Sci USA 1988; 85: 4700–4704. 5 Baeuerle P, Baltimore D. A 65-kD subunit of active NF-kB is required for inhibition of NF-kB by I kB. Genes Dev 1989; 3: 1689–1698. 6 Beato M. Gene regulation by steroid hormones. Cell 1989; 56: 335–344. 7 Ming M et al. Beta-adrenoceptors and dexamethasone synergistically stimulate the expression of the angiotensinogen gene in opossum kidney cells. Kidney Int 1996; 50: 94–101. 8 Guo D et al. Identification of a cis-acting glucocorticoid responsive element in the rat angiotensin II type 1A promoter. Circ Res 1995; 77: 249–257. 9 Stone E et al. Identification of a gene that causes primary open angle glaucoma. Science 1997; 275: 668–670. 10 Jantzen H et al. Cooperativity of glucocorticoid response elements located far upstream of the tyrosine aminotransferase gene. Cell 1987; 49: 29–38. 11 Ogawa H et al. Isolation and nucleotide sequence of the cDNA for rat liver serine dehydratase mRNA and structures of the 5⬘ and 3⬘ flanking regions of the serine dehydratase gene. Proc Natl Acad Sci USA 1988; 85: 5809–5813. 12 Imai E et al. Glucocorticoid receptor-cAMP response elementbinding protein interaction and the response of the phosphoenolpyruvate carboxykinase gene to glucocorticoids. J Biol Chem 1993; 268: 5353–5356. 13 Demoly P et al. Gene therapy strategies for asthma. Gene Therapy 1997; 4: 507–516. 14 Sher ER et al. Steroid-resistant asthma: cellular mechanisms contributing to inadequate response to glucocorticoid therapy. J Clin Invest 1994; 93: 33–39. 15 Adcock IM et al. Abnormal glucocorticoid receptor-activator protein 1 interaction in steroid-resistant asthma. J Exp Med 1995; 182: 1951–1958. 16 Wang J et al. Expression of RANTES by human bronchial epithelial cells in vitro and in vivo and the effects of corticosteroids. Am J Respir Cell Mol Biol 1996; 14: 27–35.

251


252

17 Curiel D, Pilewski J, Albelda S. Gene therapy approaches for inherited and acquired lung diseases. Am J Respir Cell Mol Biol 1996; 14: 1–18. 18 Martins V et al. Demonstration by confocal microscopy that unliganded overexpressed glucocorticoid receptors are distributed in a nonrandom manner throughout all planes of the nucleus. Mol Endocrinol 1991; 5: 217–225. 19 Htun H et al. Visualization of glucocorticoid receptor translocation and intranuclear organization in living cells with a green fluorescent protein chimera. Proc Natl Acad Sci USA 1996; 93: 4845–4850. 20 Nelson P et al. Genomic organization and transcriptional regulation of the RANTES chemokine gene. J Immunol 1993; 151: 2601–2612. 21 Kwon OJ et al. Glucocorticoid inhibition of RANTES expression in human lung epithelial cells. Am J Respir Cell Mol Biol 1995; 12: 488–496. 22 Jonat C et al. Antitumor promotion and antiinflammation: down-modulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone. Cell 1990; 62: 1189–1204. 23 Schu¨le R et al. Functional antagonism between oncoprotein cJun and the glucocorticoid receptor. Cell 1990; 62: 1217–1226. 24 Yang-Yen H-F et al. Transcriptional interference between c-jun and the glucocorticoid receptor: mutual inhibition of DNA binding due to direct protein–protein interaction. Cell 1990; 62: 1205–1215. 25 Ray A, Prefontaine K. Physical association and functional antagonism between the p65 subunit of transcription factor NF-kappa B and the glucocorticoid receptor. Proc Natl Acad Sci USA 1994; 91: 752–756. 26 Scheinman RI et al. Characterization of mechanisms involved in transrepression of NF-␬B by activated glucocorticoid receptors. Mol Cell Biol 1995; 15: 943–953. 27 Caldenhoven E et al. Negative cross-talk between RelA and the glucocorticoid receptor: a possible mechanism for the antiinflammatory action of glucocorticoids. Mol Endocrinol 1995; 9: 401–412. 28 Bodine P, Litwack G. The glucocorticoid receptor and its endogenous regulators. Receptor 1990; 1: 83–119. 29 Kamei Y et al. A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 1996; 85: 403–414. 30 Auphan N et al. Immunosuppression by glucocorticoids: inhibition of NF-kappa B activity through induction of I kappa B synthesis. Science 1995; 270: 286–290. 31 Scheinman R, Cogswell P, Lofquist A, Baldwin AJ. Role of transcriptional activation of I kappa B alpha in mediation of immunosuppression by glucocorticoids. Science 1995; 270: 283– 286. 32 Halazonetis TD, Georgopoulos K, Greenberg ME, Leder P. cJun dimerizes with itself and with c-Fos, forming complexes of different DNA binding affinities. Cell 1988; 55: 917–924.

33 Nakabeppu Y, Ryder K, Nathans D. DNA binding activities of three murine Jun proteins: stimulation by Fos. Cell 1988; 55: 907–915. 34 Baeuerle P, Baltimore D. I kappa B: a specific inhibitor of the NF-kappa B transcription factor. Science 1988; 242: 540–546. 35 Teran L et al. Eosinophil recruitment following allergen challenge is associated with the release of the chemokine RANTES into asthmatic airways. J Immunol 1996; 157: 1806–1812. 36 Stellato C et al. Expression of the chemokine RANTES by a human bronchial epithelial cell line. Modulation by cytokines and glucocorticoids. J Immunol 1995; 155: 410–418. 37 Schall T, Bacon K, Toy K, Goeddel D. Selective attraction of monocytes and T lymphocytes of the memory phenotype by cytokine RANTES. Nature 1990; 347: 669–671. 38 Rot A et al. RANTES and macrophage inflammatory protein 1 alpha induce the migration and activation of normal human eosinophil granulocytes. J Exp Med 1992; 176: 1489–1495. 39 Kameyoshi Y et al. Cytokine RANTES released by thrombinstimulated platelets is a potent attractant for human eosinophils. J Exp Med 1992; 176: 587–592. 40 Zhang Y, Lin J, Vilcek J. Interleukin-6 induction by tumor necrosis factor and interleukin-1 in human fibroblasts involves activation of a nuclear factor binding to a kappa B-like sequence. Mol Cell Biol 1990; 10: 3818–3823. 41 Danielsen M, Northrop JP, Ringold GM. The mouse glucocorticoid receptor: mapping of functional domains by cloning, sequencing and expression of wild-type and mutant receptor proteins. EMBO J 1986; 5: 2513–2522. 42 Roux P et al. Retrovirus-mediated gene transfer of a human cfos cDNA into mouse bone marrow stromal cells. Oncogene 1991; 6: 2155–2160. 43 Naumann M, Wulczyn F, Scheidereit C. The NF-kappa B precursor p105 and the proto-oncogene product Bcl-3 are I kappa B molecules and control nuclear translocation of NF-kappa B. EMBO J 1993; 12: 213–222. 44 Curiel D, Agarwal S, Wagner E, Cotten M. Adenovirus enhancement of transferrin-polylysine-mediated gene delivery. Proc Natl Acad Sci USA 1991; 88: 8850–8854. 45 Wagner E, Cotten M, Foisner R, Birnstiel M. Transferrin-polycation-DNA complexes: the effect of polycations on the structure of the complex and DNA delivery to cells. Proc Natl Acad Sci USA 1991; 88: 4255–4259. 46 Cotten M et al. High-efficiency receptor-mediated delivery of small and large (48 kilobase gene constructs) using the endosome-disruption activity of defective or chemically inactivated adenovirus particles. Proc Natl Acad Sci USA 1992; 89: 6094–6098. 47 Kohout T et al. Novel adenovirus component system that transfects cultured cardiac cells with high efficiency. Circ Res 1996; 78: 971–977. 48 Boussif O et al. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc Natl Acad Sci USA 1995; 92: 7297–7301.