INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inhaled glucocorticoids (GC) are the first-line anti-inflammatory therapy for asthma. In the last decade, our understanding of the molecular mechanisms whereby GC counteract inflammation has greatly improved. Glucocorticoids increase or inhibit gene transcription through a process known as trans-activation and trans-repression respectively.

trans-Activation is mediated by binding of the hormone- activated glucocorticoid receptor (GR) to a DNA sequence called glucocorticoid response element (GRE). Genes involved in the control of gluconeogenesis, arterial pressure, and intraocular tension contain GRE (1). Thus, trans-activation may account for some GC unwanted effects (diabetes, arterial hypertension, edema, hypokalemia, glaucoma). On the other hand, trans-activation may also result in therapeutical benefits since GC induce gene expression of the 2-adrenergic receptor (5).

The hormone-activated GR inhibits gene transcription by various mechanisms. For instance, it represses the osteocalcin gene through binding to a negative GRE (6) whereas it inhibits expression of the adrenocorticotropin gene through antagonism with the transcription factor Nur77 (7). Repression of these two genes may account for, respectively, osteoporosis and feedback adrenal suppression after GC treatment.

Although interleukin-6 expression is inhibited by GC through a negative GRE (8), the anti-inflammatory effects of GC are predominantly mediated by repression of the transcription factors activator protein-1 (AP-1) and nuclear factor kappa B (NF-B) which control many genes encoding proinflammatory mediators such as the cytokine RANTES (regulated upon activation, normal T-cell expressed and secreted). In this case, trans-repression probably results from inhibitory protein-protein interaction between the hormone-activated GR and AP-1 or NF-B (9). In addition, competition for limiting amount of the coactivator cAmp-response element binding protein-binding protein (CBP) may also account for the inhibitory action of the GR on AP-1 and NF-B activities (10). AP-1 is a dimer made of peptides belonging to the Fos and Jun families whereas NF-B is a dimer composed of proteins related to p65. Of note, asthma has been associated with elevated c-Fos or p65 immunoreactivity and increased AP-1 or NF-B DNA-binding activity (11). In addition, GC-resistant asthma is associated with increased c-fos expression in monocytes and T lymphocytes (14).

Five inhaled GCbudesonide, beclomethasone dipropionate (BDP), flunisolide, fluticasone propionate (FP), and triamcinolone acetonide (TAA)are now available for asthma treatment. These differ in their clinical efficacy and pharmacological properties (15, 16). GC are evaluated in vitro on the basis of their receptor binding characteristics and their potency to affect cell growth, cell survival, and release of inflammatory mediators. The major link between these events is the regulation of gene transcription by the ligand-activated GR. Measurement of transcriptional potencies of the various GC may help to predict their anti-inflammatory potencies as well as their capacity to produce side effects. Recently, the ability of FP and budesonide to repress the expression of an AP-1- dependent reporter gene was analyzed (17). However, the repressive potential of the three other available inhaled GC over AP-1 activity was not determined. In addition, the relative capacity of all five inhaled GC to repress NF-B activity and to trans-activate was not investigated. Even though inhaled GC are selected for a high hepatic first-pass inactivation, their trans-activation potency at concentration reached in the plasma needs to be measured to predict any possible GRE-mediated side effect.

The purpose of this study was to determine the trans-activation and trans-repression potencies of the five currently available inhaled GC. A stable transfectant of HeLa S3 cells carrying a GRE-dependent luciferase gene was used to monitor trans-activation potency, whereas A549 human lung epithelial cells transiently transfected with an AP-1- or NF-B-dependent luciferase gene were used to measure the trans-repression potencies. HeLa S3 and A549 cells are cell lines that express the GR and are well characterized for their response to GC. These are therefore appropriate to set up reproducible pharmacological tests. We demonstrate the relevance of reporter gene assays used in the present study by linking trans- repression with inhibition of RANTES production in A549 cells and trans-activation in HeLa S3 cells with induction of the gluconeogenic enzyme tyrosine aminotransferase (TAT) in hepatoma tissue culture (HTC) cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reagents

12-O-tetradecanoyl-phorbol-13-acetate (TPA), transferrin-polylysine, poly-L-lysine (P2636), spermine, and synthetic GC (budesonide, BDP, flunisolide, TAA) were purchased from Sigma (St. Louis, MO). FP was a gift from Glaxo Wellcome (Greenford, UK). Luciferin and dithiothreitol (DTT) were purchased from Promega (Madison, WI). Tumor necrosis factor- (TNF-) was purchased from Pharmingen (San Diego, CA). Coenzyme A was purchased from Boehringer Mannheim (Mannheim, Germany). Geneticin (G-418) was purchased from GibcoBRL (Gaithersburg, MD). GC were initially dissolved in absolute ethanol at 10³ M. Dilutions with medium were freshly made each day from original stocks to a maximal concentration of 10⁶ M. Thus, the maximal final concentration of ethanol is 0.1% and at this concentration toxic effects are not seen.

Plasmids

The reporter plasmid pMAMneo-LUC contains a GRE-dependent promoter linked to the luciferase gene and an SV40 promoter-driven aminoglycoside phosphotransferase gene conferring resistance to G-418 (Clontech, Palo Alto, CA). The luciferase reporter construct 517/+63 Coll Luc, containing part of the collagenase promoter which includes one AP-1 site, was a gift of Peter Herrlich (Institute of Genetics, Karlsruhe, Germany). The 3xIg Cona Luc plasmid, which contains three tandem repeats of the NF-B response element (NF-BRE) from the immunoglobulin  chain linked to the conalbumin minimal promoter and the luciferase gene, was obtained from Alain Israël (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 variations in transfection efficiency. The expression vector for the NF-B subunit p65 (pECEp65) and control vector (pECE) were given by Carl Scheidereit (Max Delbrück Centre for Molecular Medicine, Berlin, Germany).

Cell Culture

Two human carcinoma-derived cell lines, HeLa S3 epithelioid cervix cells and A549 lung epithelial cells, were maintained respectively in Dulbecco's modified Eagle medium (DMEM) and Ham's F12 medium containing 10% heat-inactivated fetal calf serum (FCS), 100 units/ml penicillin, 100 µg/ml streptomycin, and 2 mM glutamine. Rat HTC hepatoma cells were cultivated in DMEM completed as previously described.

Selection of a Stable Transfectant of HeLa S3 Cells Carrying a GRE-dependent Reporter Gene and Luciferase Assay

Transfection of HeLa S3 cells with the reporter plasmid pMAMneo-LUC was carried out by the calcium phosphate precipitation technique. G-418 was introduced 1 d after transfection at a concentration of 1 mg/ml in the routine medium, and cells were selected for 3 wk. G-418-resistant clones were treated with 10⁷ M dexamethasone and then with 10⁷ M of the antagonist RU486. Clones showing a strong regulation of luciferase activity were selected after addition of medium containing 3 · 10⁴ M of luciferin using a Hamamatsu ARGUS 100 photon-counting camera (Hamamatsu, Japan) coupled with an imaging analysis system. One of these clones, named HeLa-MMTV-Luc, was then expanded for further use.

For luciferase assay, HeLa-MMTV-Luc cells were seeded at 14,000 cells/well (50% of confluence) into white 96-well microtiter plates (Costar, Cambridge, MA). On the next day, cells were left untreated or incubated for 6 h at 37° C with GC at concentrations ranging from 10¹¹ M to 10⁶ M. Cells were then incubated in medium containing 3 · 10⁴ M luciferin at room temperature for 20 min and luciferase activity was measured in a Wallac 1450 MicroBeta Trilux luminescence counter (Wallac Oy, Turku, Finland).

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 by standard methods.

Transient Transfection of A549 Cells

Twenty-four hours after seeding the cells into 48-well cluster plates (50,000 cells/well), medium was replaced by 100 µl/well of serum-free medium. DNA to be transfected included 25 ng/well of the CMV-gal plasmid for normalization of transfection efficiency, 60 ng/well of the reporter plasmids 517/+63 Coll Luc or 3xIg Cona Luc, and when indicated in the figure legends, 20 ng/well of the expression vectors for the NF-B subunit p65 (pECEp65). The corresponding empty vector pECE was added so that each well contained the same total amount of DNA (600 ng).

The DNA was diluted in 20 mM Hepes pH 7.4, 150 mM NaCl, complexed with AdCMVnull adenovirus, transferrin-polylysine, and poly-L-lysine as described previously (18) and added to the cells. After transfection, the cells were treated for 20 h or as indicated in the figure legends and processed for luciferase and -galactosidase assays.

Luciferase and -galactosidase Assays in Extracts of A549 Cells

Transfected cells were lysed by one cycle of freeze-thaw in 100 µl of 15 mM Tris-HCl pH 7.8, 60 mM KCl, 15 mM NaCl, 2 mM ethylenediaminetetraacetic acid (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 µl of cell extract was measured in a luminometer after injection of 100 µl 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 µl of cell extract were mixed with 67 µl of the chemiluminescent substrate Galacton-Plus (Tropix, Bedford, MA) and processed according to the manufacturer's recommendations. -galactosidase activity was measured to correct for variations in transfection efficiency.

TAT Assay

HTC cells were seeded at 10⁶ cells/well into 6-well cluster plates. After 24 h, the medium was replaced by DMEM without serum and GC were added for 18 h. TAT activity was measured in 96-well plates as described previously (19).

RANTES Immunoassay

One day after seeding A549 cells into 48-well cluster plates (50,000 cells/well), these were left untreated or stimulated with 10 ng/ml of TNF- alone or in combination with GC at a concentration ranging from 10¹¹ M to 10⁷ M for 20 h. Stimulation of RANTES release with 10 ng/ml of TNF- was appropriate to study the inhibitory effect of GC as determined previously (20). Concentration of RANTES in supernatants was determined, using a quantitative sandwich enzyme immunoassay technique and following the manufacturer's recommendations (R&D Systems, Minneapolis, MN).

Statistical Analysis

For differences between GC at a given concentration, statistical significance was assessed using the Kruskal-Wallis nonparametric test and the Dunn's post hoc test. Statistical significance was set at p < 0.05. Correlations were analyzed using the Spearman rank test.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

trans-Activation Potency of Inhaled GC

trans-Activation potency of five inhaled GC was measured using a stable transfectant of HeLa S3 cells carrying a GRE- dependent luciferase gene. Preliminary time-course experiments indicated that optimal trans-activation occurred 24 h after stimulation with GC. A dose-response analysis was performed at this time point revealing differences between the trans-activation potencies of the various GC (Figure 1A). From 10¹¹ M to 10⁹ M, BDP, flunisolide, and TAA were inactive. Budesonide induced trans-activation 3.4 times at 10⁹ M. FP was the most potent GC at these low concentrations, increasing transcriptional activity 2.8 times at 10¹⁰ M and 11.9 times at 10⁹ M. At 10⁸ M and 10⁶ M, FP was equipotent to the other GC, whereas at 10⁷ M, this had a trans-activating effect statistically significantly lower than budesonide (p < 0.01) and TAA (p < 0.001). Maximal induction of transcription was 24-fold after GC treatment. Half-maximal effective concentration (EC₅₀) for each GC was determined from the dose-response curves (Table 1). The rank order of trans-activation potencies was as follows: FP > budesonide and TAA > BDP = flunisolide. A comparison with relative GR binding affinities, GR binding half-lives, and relative topical blanching potencies is shown in Table 2.

View larger version (17K): [in this window] [in a new window]   Figure 1.   (A) trans-Activation elicited by inhaled GC. HeLa-MMTV-Luc cells were stimulated for 24 h with the various GC at a concentration ranging from 1011 M to 106 M. trans-Activation was then measured by a luciferase assay as described in METHODS. Data are shown as fold inductions over basal activity and represent the mean ± SEM of three independent experiments performed in triplicates. (B) Dose-response curves of TAT induction by GC. HTC cells were treated by increasing concentrations of the various GC for 18 h. TAT activity was measured as described in METHODS and is expressed in fold inductions over basal activity. Data represent the mean ± SEM of three independent experiments. (C ) Correlation between the trans-activating effect of inhaled GC in HeLa cells and their capacity to induce TAT activity in HTC cells. Data obtained with all GC except BDP were analyzed using the Spearman rank test.

View larger version (29K): [in this window] [in a new window]   Figure 2.   Repression of AP-1 activity by GC. A549 cells were transiently transfected with the reporter plasmid 517/+63 Coll Luc and stimulated for 20 h by 150 nM TPA in the presence or absence of GC as indicated. Transcriptional activity was measured by luciferase and -galactosidase assays as described in METHODS. Data are shown as a percentage of the TPA-induced activity and represent the mean ± SEM of three independent experiments performed in triplicates.

View larger version (21K): [in this window] [in a new window]   Figure 3.   (A) NF-B activity is induced either by TNF- treatment or overexpression of the NF-B subunit p65. A549 cells were transfected with the reporter plasmid 3xIg Cona Luc. These were then left untreated for 20 h (UT) or treated during the last 4 h with 10 ng/ml of TNF- as indicated. Alternatively, cells were cotransfected with an expression vector of the NF-B p65 subunit (pECEp65) or the empty vector pECE as indicated and cultivated for 20 h. Transcriptional activity was measured by luciferase and -galactosidase assays as described in METHODS. Data are shown as fold inductions over basal activity and represent the mean ± SEM of three independent experiments performed in triplicate. (B) Repression of NF-B activity by GC. A549 cells were transfected with the reporter plasmid 3xIg Cona Luc and an expression vector of the NF-B p65 subunit (pECEp65). These were then stimulated for 20 h in the presence of increasing concentrations of GC. Transcriptional activity was measured by luciferase and -galactosidase assays as described in METHODS. Data are shown as a percentage of the p65-induced activity and represent the mean ± SEM of three independent experiments performed in triplicates.

View larger version (16K): [in this window] [in a new window]   Figure 4.   (A) RANTES release was induced either by TNF- treatment or overexpression of the NF-B subunit p65. A549 cells were left untreated (UT) or stimulated with 10 ng/ml of TNF- as indicated. Alternatively, cells were transfected with 300 ng/well of an expression vector of the NF-B p65 subunit (pECEp65) or the empty vector pECE as indicated. Twenty hours later, concentration of RANTES in the supernatant was determined using a sandwich enzyme immunoassay as described in METHODS. Data represent the mean ± SEM of three independent experiments performed in triplicates. (B) Inhibition of RANTES release by GC. A549 cells were stimulated for 20 h with 10 ng/ml TNF- in the absence or presence of increasing concentrations of GC as indicated. Concentration of RANTES in the supernatant was determined using a sandwich enzyme immunoassay as described in METHODS. Data are shown as a percentage of the TNF--induced RANTES production and represent the mean ± SEM of two independent experiments performed in triplicates. *p < 0.05 for FP versus BDP, flunisolide, and TAA; **p < 0.05 for FP versus BDP. (C ) Correlation between inhibition of RANTES release and trans-repression of NF-B activity by GC. Data from the previous experiments were analyzed using the Spearman rank test.

Induction of TAT by Inhaled GC

The previous experiments were made with a transfected GRE-dependent reporter gene. We next investigated whether the various inhaled GC would show the same relative potencies to trans-activate the endogenous TAT gene, which encodes an enzyme involved in the GC-dependent stimulation of gluconeogenesis (1). TAT activity was measured in rat hepatic HTC cells treated with increasing concentrations of GC (Figure 1B). A maximal 10.5-fold induction was observed at GC concentration of 10⁶ M. EC₅₀ for each GC was determined from the dose-response curves (Table 1). The rank order of TAT induction potencies was FP > budesonide = TAA > flunisolide >> BDP. Between 10¹⁰ M and 10⁹ M, only FP and budesonide induced TAT activity. trans-Activation potency in HeLa cells and capacity to induce TAT in HTC cells were correlated (r = 0.84, p < 0.0001) for all GC except BDP (Figure 1C). A comparison with relative GR binding affinities, GR binding half-lives, and relative topical blanching potencies is shown in Table 2.

trans-Repression of AP-1 Activity by Inhaled GC

To measure AP-1 activity, A549 cells were transiently transfected with the reporter plasmid 517/+63 Coll Luc which contains the natural collagenase promoter with its single AP-1 binding site. The AP-1 binding site is called TPA-response element (TRE) because TPA is a potent inducer of AP-1 activity. AP-1 activity was induced 4.5-fold by TPA treatment (95% confidence limits 3.5 to 5.6) and was given the nominal value of 100%. The various GC repressed this activity in a dose-dependent manner but with different potencies (Figure 2). AP-1 activity was maximally inhibited by 44% after GC treatment. Half-maximal inhibitory concentration (IC₅₀) for each GC was determined from the dose-response curves (Table 1). The rank order of trans-repression potencies on AP-1 activity was as follows: FP > budesonide > BDP and TAA > flunisolide. A comparison with relative GR binding affinities, GR binding half-lives, and relative topical blanching potencies is shown in Table 2.

trans-Repression of NF-B Activity by Inhaled GC

To measure NF-B activity, A549 lung epithelial cells were transiently transfected with the NF-BRE-dependent reporter plasmid 3xIg Cona Luc. NF-B activity can be induced either by TNF- treatment or overexpression of the NF-B subunit p65 (Figure 3A). Overexpression of the NF-B p65 subunit resulted in an 8.9-fold induction of transcriptional activity (95% confidence limits 1.9 to 17.2) which was given the nominal value of 100%. The various GC repressed this activity in a dose-dependent manner but with different potencies (Figure 3B). NF-B activity was maximally inhibited by 42.5% after GC treatment. IC₅₀ for each GC was determined from the dose- response curves (Table 1). The rank order of trans-repression potencies on NF-B activity was as follows: FP > budesonide > TAA > BDP = flunisolide. A comparison with relative GR binding affinities, GR binding half-lives, and relative topical blanching potencies is shown in Table 2.

Inhibition of RANTES Production by Inhaled GC

Because the previous observations were made with transfected reporter genes, we investigated whether the various inhaled GC would show the same relative potencies to repress the release of RANTES, a proinflammatory cytokine whose gene is controlled by AP-1 and NF-B (21). RANTES release can be induced either by TNF- treatment or overexpression of the NF-B subunit p65. TNF- had a greater effect than overexpression of p65 on RANTES release probably because TNF- acted on all cells whereas only a fraction of these cells were transfected by p65 (Figure 4A). Upon TNF- stimulation, concentration of RANTES in cell supernatants increased from 0.90 ± 0.15 pg/ml to 535 ± 149 pg/ml. Cotreatment with GC decreased the production of RANTES in a dose-dependent manner (Figure 4B). Release of RANTES was maximally inhibited by 85% after GC treatment. IC₅₀ for each GC was calculated from the dose-response curves (Table 1). The rank order of inhibitory potencies on RANTES production was as follows: FP > budesonide > BDP, flunisolide, and TAA. A comparison with relative GR binding affinities, GR binding half-lives, and relative topical blanching potencies is shown in Table 2. Statistically significant differences in the inhibitory effect were found at 10¹⁰ M and 10⁹ M for FP versus BDP, flunisolide and TAA (p < 0.05), and at 10⁸ M for FP versus BDP (p < 0.05).

Using the reporter gene assays described previously, we found that TNF- stimulated the activity of NF-B (Figure 3A) but not that of AP-1 (data not shown) in A549 cells. Accordingly, inhibition of RANTES production was strongly correlated with trans-repression of NF-B activity by GC (r = 0.93, p < 0.0001) (Figure 4C).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inhaled GCs are the first-line anti-inflammatory therapy for asthma. We have compared the potencies of the five available inhaled GC to regulate transcription, which is the key molecular event that mediates their physiological and pharmacological effects (22). We report that the low capacity of inhaled GC to activate transcription at concentrations found in the plasma is in agreement with their restricted systemic side effects in vivo, whereas their relative abilities to trans-repress AP-1 or NF-B activity are in agreement with their relative clinical efficacy.

trans-Activation Potencies of Inhaled GC

trans-Activation may account for the following GC side effects: diabetes, arterial hypertension, edema, hypokalemia, and glaucoma. Repressive mechanisms rather than trans-activation may underlie GC-induced osteoporosis and inhibition of the hypothalamic-pituitary-adrenal axis (6, 7).

Because of local retention in the airways and a high hepatic first-pass inactivation, inhaled GC concentrations in plasma are less important than in lungs. Systemic availability of FP results only from absorption via the lungs, whereas for the other inhaled GC, oral bioavailability has to be taken into account (23). After an inhalation of 1 mg flunisolide, 2 mg TAA, 1 mg FP, or 1.6 mg budesonide, the peak concentration reached in plasma is, respectively, 9.2 × 10¹⁰ M, 4.1 × 10¹⁰ M, 2.2 × 10¹⁰ M, and 6.3 × 10¹⁰ M (24). At these respective concentrations, potencies of inhaled GC to induce the GRE- dependent reporter gene and the gluconeogenic enzyme TAT were null except for FP and budesonide. These data are in agreement with the absence of GRE-mediated side effects during inhaled GC therapy. Indeed, no cases of diabetes, arterial hypertension, edema, or hypokalemia associated with the prescription of inhaled GC have yet been reported. A risk of glaucoma has been observed, but the implication of inhaled GC was questioned (27, 28). Nevertheless, our observation that FP is the most potent inducer of a gluconeogenic enzyme supports a recent case report implicating high doses (1,000 and 2,000 µg/d) but not a medium dose (500 µg/d) of this GC in the loss of diabetic control (29).

Concentrations of inhaled GC at pulmonary level have only been reported for FP and budesonide. After inhalation of 1.6 mg budesonide or 1 mg FP, concentrations of these GC in lung tissues are subject to strong interindividual variations and are estimated to be at least in the nanomolar range (25, 26). Our data indicate that at 10⁹ M, FP triggered a higher trans-activating effect than budesonide whereas at 10⁸ M or 10⁶ M, these GC were equipotent. Moreover, at 10⁷ M, FP was statistically significantly less potent than budesonide as measured by the GRE-dependent reporter gene assay, but this observation was not confirmed on TAT induction. Therefore, because of the variability in their pulmonary concentrations and in their trans-activating effects, it is not possible to predict to which relative extent these GC might upregulate the 2-adrenergic receptor in the lung in vivo. Direct analyses of smooth muscle cells of the lung after GC inhalation would be the sole approach to answer this question.

trans-Repressive Potencies of FP and Budesonide

In a recent study, the ability of budesonide and FP to repress AP-1 activity was measured using a rat fibroblast cell line stably transfected with an AP-1-dependent -galactosidase gene (17). In this system, FP was 6 times more potent than budesonide in repressing AP-1 activity after a continuous exposure of 24 h. Using A549 pulmonary epithelial cells, transiently transfected with an AP-1-dependent luciferase gene, we found exactly the same relative potency in favor of FP after 20 h of treatment (Table 2), suggesting that measurement of relative transcriptional potencies of GC is not influenced by the type of cell line, transfection, and reporter gene. Others have reported that the inhibitory potential of FP was 10 times higher than that of budesonide on the E-selectin promoter (30). Because this promoter contains binding sites for a number of transcription factors including AP-1, activating transcription factor (ATF), NF-B, and high mobility group protein HMG-I(Y), the relative capacity of FP and budesonide to inhibit the activity of these transcription factors individually remained to be determined.

Wieslander and coworkers (17) considered that after inhalation therapy, cells in vivo are exposed to a pulse of GC. Accordingly, they exposed the cells to GC for 6 h, transferred them to new culture plates after extensive washes, and found that inhibitory potency of FP on AP-1 activity was then 2 times lower than that of budesonide. However, the situation in vivo is more likely to be mimicked by exposing the cells continuously to FP because it was reported that after a single inhalation of 1 mg of FP, this GC remained in central lung tissue at high concentration for more than 17 h (26). This finding is compatible with the long receptor binding half-life (over 10 h) of FP (Table 2) (31). In addition, at continuous exposure to GC for 20 h, we found that FP was the most potent inhibitor of AP-1 activity but also of NF-B activity and RANTES release (Table 2). Other investigators have recently confirmed that NF-B activity is repressed more efficiently by FP than by budesonide (32). Although, these data were obtained on cell lines, they are in good accordance with the medical practice, FP being the most potent available inhaled glucocorticoid in 1999 (16).

Comparison of the Various Tests Designed to Evaluate the Potency of GC

A similar ranking of inhaled GC was found by the different assays performed in this study and when analyzing reported values for GR binding affinity, GR binding half-life, and topical skin blanching potency. FP was more potent than budesonide which had a greater or equivalent potency than BDP, TAA, and flunisolide (Table 2). However, among the latter compounds, there was no strict relation between the various potencies. For example, despite having the same topical skin blanching potency, flunisolide was less potent than TAA in all other assays. One explanation is that the former test may not be as discriminating as the others. Clinical relevance of these small differences between BDP, flunisolide, and TAA remains to be established.

Of note, BDP was active in A549 lung cells but not in HTC hepatic rat cells. This may be due to a rapid metabolism of BDP into an inactive product, such as beclomethasone 21-monopropionate, in the latter cell type. In A549 cells, BDP and flunisolide were equipotent even though BDP has a 4-fold lower relative GR binding affinity. In this regard, it was shown that BDP is hydrolyzed in human lung cells to the more active metabolite beclomethasone 17-monopropionate (33). Thus, the potency of BDP in A549 cells probably results from the combinatorial effect of BDP and beclomethasone 17-monopropionate.

Reporter assays have potential advantages over current binding tests: (1) they are functional assays, which allow discrimination between high-affinity agonists' and antagonists' molecules, (2) they do not require use of a radiolabeled ligand, (3) using stably transfected cells, they are amenable to automation for high-throughput screening of pharmocological compounds. An advantage over immunoassays is that specific antibodies are not necessary. In this regard, inhibition of RANTES production and trans-repression of NF-B activity by GC were strongly correlated (see Figure 4 and Table 2). Moreover, IC₅₀ values were very similar between both assays (see Table 1), suggesting that GC decrease the expression of RANTES through inhibition of NF-B activity. Because RANTES is a proinflammatory cytokine involved in the pathophysiology of asthma, measure of trans-repression potency may be used to predict anti-inflammatory effects of glucocorticoids.

Maximal repression of RANTES production by the various GC was greater than that of NF-B activity. Several hypotheses can be made regarding this increased inhibitory effect. First, because the endogenous RANTES gene is integrated in the chromosome, whereas the reporter gene is transiently transfected and episomal, it is likely that their chromatin structures and their localization inside the nucleus are different (34). Therefore, the RANTES gene is possibly more accessible to regulatory factors. Second, organization of the RANTES promoter differs from that of the reporter promoter by several aspects: four NF-B binding sites are present instead of three, surrounding sequences are different, and recognition sequences for other transcription factors which may be sensitive to GC action are present (21). Third, although RANTES messenger RNA (mRNA) stability was unaffected by GC (20), further downregulation may occur through other post-transcriptional mechanisms as recently described for inducible nitric oxide synthase (35).

In conclusion, our data suggest that trans-repression and trans-activation potencies of inhaled GC may help to predict their anti-inflammatory effects as well as some of their adverse effects. These tests could serve as an in vitro screening system to select and evaluate new GC as well as compounds with suspected GC-like activities.