| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Biomedicum Helsinki (T.R., J.J.P., O.A.J.), Institute of Biomedicine, University of Helsinki, FIN-00014 Helsinki; Institute of Biotechnology (J.J.P.), University of Helsinki, FIN-00014 Helsinki; Department of Pediatrics (S.K., R.V.), Kuopio University Hospital, FIN-70211 Kuopio; and Department of Clinical Chemistry (O.A.J.), University of Helsinki, and Helsinki University Central Hospital, FIN-00014 Helsinki, Finland
Address all correspondence and requests for reprints to: Taneli Raivio, M.D., Ph.D., Biomedicum Helsinki, Institute of Biomedicine/Physiology, P.O. Box 63 (Haartmaninkatu 8), FIN-00014 University of Helsinki, Finland. E-mail: . taneli.raivio{at}helsinki.fi
Abstract
We have developed a mammalian cell (COS-1) bioassay, which measures glucocorticoid bioactivity (GBA) directly from a small amount of human serum. The assay is based on the expression of human glucocorticoid receptor (GR) together with a coactivator protein and reporter plasmid containing GR response elements upstream of the luciferase gene. Ten microliters of human serum, in duplicate, are added directly to the cell culture medium, and GBA is derived from reporter gene activity. The assay differentiates between biopotencies of synthetic steroids, and importantly, mifepristone (RU486) is able to block glucocorticoid-induced response. The assay is sensitive (<15.6 nM cortisol in fetal calf serum) and precise, with the within- and between-assay coefficients of variation less than 8% and 10%, respectively. We measured serum GBA (bioassay) and cortisol (RIA) levels in 34 asthmatic children (age range, 5.7–14.2 yr) at baseline and after treatment with either inhaled budesonide (800 µg/d, n = 14), fluticasone propionate (500 µg/d, n = 14), or cromones (control group, n = 6). Pretreatment serum GBA and cortisol levels correlated strongly (r = 0.90, P < 0.0001, n = 34). Two months of treatment with inhaled budesonide resulted in excess GBA in circulation, which was not attributable to endogenous cortisol (P < 0.001). In the fluticasone propionate group, the presence of serum excess GBA was at the borderline of statistical significance (P < 0.08) after 2 months of inhalation therapy, and no excess GBA was detected in the cromone group. In conclusion, our bioassay enables measurement of mammalian cell response to bioactive glucocorticoids in circulation and provides a novel means to investigate patients receiving drugs acting through the GR.
GLUCOCORTICOIDS ARE STEROID hormones, which are essential for survival, necessary in mediating stress responses and have profound effects on immunological and inflammatory responses (1, 2). Synthetic derivatives of these hormones are widely used in the treatment of inflammatory diseases. For example, inhaled glucocorticoids have been shown to be beneficial in terms of lung function, quality of life indices, and reducing the severity of asthma attacks (3, 4). On the other hand, children may respond to this treatment by displaying suppression of adrenal cortisol production, gaining extra weight, and/or decreasing growth velocity (5, 6, 7, 8, 9, 10, 11). The bioactivities of synthetic glucocorticoids, however, are difficult to evaluate biochemically, because the drugs show different systemic bioavailabilities (12). Moreover, their affinities to glucocorticoid receptor (GR) differ (12, 13), which is not revealed by employing conventional methods for measurement of serum steroids.
The biological effects of glucocorticoids are mediated through GR, a ligand-activated transcription factor that belongs to the nuclear receptor superfamily (14). According to the classical activation model, GR dimerizes after binding of the ligand, enters the nucleus, binds to the regulatory region of the target gene as a homodimer, and interacts with coregulator proteins of transcription, which ultimately leads to activated or repressed target gene expression. In this work, we developed a mammalian cell bioassay, which is based on the measurement of GR-dependent reporter gene activity elicited by a small amount of human serum. This assay was subsequently employed to investigate glucocorticoid bioactivity (GBA) in pediatric patients before and during their treatment with inhaled steroids.
Materials and Methods
Glucocorticoid bioassay
Plasmids, cell culture, and transfection. pSG5-human (h)GR, containing full-length hGR cDNA (15), pFLAG-ARIP3, a steroid receptor coactivator (16), and the luciferase reporter pARE4tk-LUC (17), have been described previously. pCMVß was obtained from CLONTECH Laboratories, Inc. (Palo Alto, CA). COS-1 cells (American Type Culture Collection, Manassas, VA) were cultured in phenol red-free DMEM (Life Technologies, Inc., Santa Clara, CA), which contained penicillin (25,000 U/liter), streptomycin (25,000 U/liter), and 10% (vol/vol) fetal calf serum (FCS; Life Technologies, Inc., Paisley, UK). Twenty-four hours before transfection, the cells were divided onto a 96-well plate (NUNC, Roskilde, Denmark) at a density of 5,000 cells/well. The plates were then incubated overnight at 37 C in a humidified atmosphere of 5% CO2/air. The cell culture medium was replaced by DMEM containing 8% of charcoal-stripped FCS without detectable cortisol, as measured by RIA, 3 h before transfection. The cells were transfected using FuGene reagent (Roche Molecular Biochemicals, Mannheim, Germany) according to the instructions provided by the manufacturer. Each well received a total of 68 ng DNA (pARE4tk-LUC, 45.2 ng; pSG5-hGR, 4.5 ng; pFLAG-ARIP3, 13.6 ng; and pCMVß, 4.5 ng), which was delivered as a single mastermix of plasmids.
Twenty-four hours after transfection, medium in each well was replaced by 90 µl phenol red-free DMEM without FCS, and 10 µl cortisol-containing FCS in triplicate (standard) or 10 µl human serum (unknown GBA) in duplicate was added. After an overnight incubation at 37 C, in humidified atmosphere of 5% CO2/air, the wells were aspirated empty, the cells were lysed in 30 µl diluted reporter lysis buffer (Promega Corp., Madison, WI), and 10 µl cell lysates was transferred to 96-well plates for measurements of ß-galactosidase (18) and luciferase (19) activities. The sensitivity of the bioassay was defined as mean + 2 SD of multiple relative luciferase activities brought about by charcoal-stripped FCS without added glucocorticoid.
Steroids.
Cortisone, prednisolone, 6
-methylprednisolone, 11-deoxycortisol, 17-hydroxyprogesterone (Sigma-Aldrich Corp., Steinheim, Germany), cortisol (Sigma-Aldrich Corp.), progesterone (Sigma), 5-dihydrotestosterone (DHT) (Steraloids Inc., Wilton, NH), and dexamethasone (RdH Laborchemikalien GmbH & Co, Seelze, Germany) were dissolved in ethanol or dimethyl sulfoxide (Sigma-Aldrich Corp.), added to charcoal-stripped FCS, and tested in the bioassay at multiple concentrations. The antiglucocorticoid effect of mifepristone (RU486; Sigma-Aldrich Corp.) was tested in the bioassay by increasing mifepristone concentrations in cell culture medium containing 10% of human serum; the cortisol concentration in this serum pool measured with RIA was 260 nM.
Preparation of standards and patient sera for the bioassay. Cortisol was dissolved, serially diluted in ethanol, added to charcoal-stripped FCS (Hyclone Laboratories, Inc., Logan, UT), and divided into aliquots that were stored at -70 C for future use as standards in the bioassay. Sixty microliters of serum from each subject (see below) was centrifuged briefly, filtered through a 0.22-µm Spin-X centrifuge filter unit (Corning, Inc., Corning, NY), and added to the cells of the bioassay. Samples were analyzed blinded to serum cortisol levels. The within-assay coefficient of variation (CV) was determined by repeated measurements of the same sample (pooled sera of study participants) in the bioassay.
Subjects and study protocol
The study group consisted of 34 asthmatic children (17 boys and 17 girls) with a mean age of 9.6 yr (range, 5.7–14.2 yr) selected among the children of a previous study (11). In short, 29 children had newly diagnosed asthma and started their first period of maintenance medication. One child had used cromone (CROM) medication continuously during the preceding 3 months, and 4 subjects had used on-demand ß2 agonists. None of the subjects had used inhaled or oral glucocorticoid therapy during the preceding 12 months. Fourteen subjects (7 boys and 7 girls) received inhaled budesonide (BUD) (800 µg/d) administered with a dry powder inhaler (Turbuhaler; AstraZeneca, Södertälje, Sweden), and another 14 subjects (7 boys and 7 girls) received fluticasone propionate as a dry powder (Diskus; Glaxo Wellcome Inc., Hertfordshire, UK) with a dose of 500 µg/d divided into 2 daily doses (11). Six patients (3 boys and 3 girls) received either cromolyn or nedocromil (the CROM group; 11) and acted as a nonsteroidal control group; the clinical efficacy of these drugs is, at least in part, the result of their ability to inhibit histamine release from mast cells (20). The children were clinically investigated by one of the authors (S.K.) at the beginning of the study and at 2 months (fluticasone propionate and BUD) or at 4 months (CROM) of treatment. A venous blood sample was drawn between 0800–1100 h. For individual patients, the blood sampling was performed at the same time on both occasions. Serum cortisol level was measured using a commercially available RIA kit (Cortisol 125I RIA; Orion Diagnostica, Espoo, Finland), with within- and between-assay CVs being less than 5% and 9%, respectively. An aliquot of serum was frozen for subsequent measurement of transcortin (corticosteroid-binding globulin) and GBA levels (see above). Serum transcortin levels were measured by using a commercially available immunoradiometric assay kit (Immuno-Biological Laboratories, Hamburg, Germany) according to the instructions provided by the manufacturer. The study was approved by the research ethics committee of Kuopio University Hospital, and the parents gave informed written consent for the study.
Data analyses
Relative luciferase activities were calculated by dividing the luciferase activity by the respective ß-galactosidase activity to correct for differences in transfection efficiency. Standard curves for the bioassay were fitted using the AssayZap program (Biosoft, Cambridge, UK); the results (GBA in human serum) are expressed in nanomolar cortisol equivalents. Differences in mean values were tested by paired and unpaired Student’s t tests, when appropriate; and, for simultaneous comparisons of three groups, ANOVA followed by Scheffé’s post hoc analysis was employed. Pearson’s correlation coefficient was calculated between two related variables to investigate their relationship. The effects of inhalation treatment on serum GBA were investigated by first calculating the relationship between endogenous cortisol and GBA levels before the onset of medications. A regression equation representing this relationship (y = 0.404 + x·0.148; R2 = 0.81, see Fig. 2
) was then used to calculate the expected GBA from the measured serum cortisol concentrations during asthma treatment. The difference between the measured and the expected GBA is referred to as the excess GBA. The deviations of mean excess GBA from zero were tested with one-sample t tests. Results are given as mean ±SEM. Statistical significance was accepted for P < 0.05.
|
Bioassay parameters
We transfected COS-1 cells transiently with plasmids encoding hGR together with a steroid receptor coactivator, ARIP3, and corresponding reporter plasmid encoding luciferase enzyme under the control of liganded GR. ARIP3 was included in the assay, because it is the strongest coactivator of the proteins that we have tested on the minimal GR-dependent promoters (Ref. 21 and unpublished observations). In the present study, ARIP3 increased the maximal response 6.5-fold without rendering the assay inaccurate or affecting the lowest tested concentrations by which glucocorticoids activated the reporter. A small aliquot of FCS, with different concentrations of cortisol (standard curve) or human serum (unknown GBA) diluted 1:10, was added to the cell culture medium after transfection; and, on the following day, relative luciferase activities were measured and calculated (luciferase activity/ß-galactosidase activity). Dose-response curves from six different transfections are shown (see Fig. 1A). The highest standard used in the bioassay (1000 nM cortisol in FCS) induced an average of approximately 90-fold increase in the relative luciferase activity.
|
-methylprednisolone = prednisolone > cortisol (Fig. 1B
). Cortisone did not activate the bioassay (Fig. 1B), indicating the lack of 11ß-hydroxysteroid dehydrogenase type I activity in the recombinant cells. Not shown in Fig. 1B, 4 µM of 11-deoxycortisol in culture medium exhibited activity corresponding to 
10 nM cortisol in culture medium (relative glucocorticoid activity 0.25% of that of cortisol). Relative glucocorticoid activities displayed by progesterone or 17-hydroxyprogesterone were 0.05% of that of cortisol. The reporter gene was not activated by the highest DHT concentration tested (10 nM DHT in culture medium). Mifepristone suppressed glucocorticoid-dependent reporter gene activity in a dose-dependent fashion; in the presence of 100 nM mifepristone in the culture medium, the mean relative luciferase activity elicited by human serum pool containing 260 nM cortisol (26 nM in culture medium) was suppressed to approximately one seventh (from 632 ± 21 to 97 ± 9; P < 0.001). The within-assay CV for human serum was less than 8%, and the between-assay CV (determined from 11 consecutive assay runs) was 10%. Serum GBA in children
The mean (±SEM) serum GBA at baseline was 37.1 ± 2.9 nM cortisol equivalents (cortisol equivalent refers to the concentration of cortisol in FCS that induces a relative reporter activity equal to that elicited by a human serum sample with unknown GBA), which was clearly lower than the mean serum cortisol level (249 ± 17.7 nM) measured by RIA. However, there was a strong positive correlation between serum GBA (bioassay) and cortisol (RIA) levels (r = 0.90, P < 0.0001, Fig. 2). Serum GBA did not correlate with the age of the children [r = 0.02, P = NS (not significant)], and the mean GBA did not differ significantly between the boys and the girls (33.4 ± 2.0 vs. 40.8 ± 5.4 nM cortisol equivalents, respectively; P = NS), or between the treatment groups (32.5 ± 3.1, 44.0 ± 5.9, and 31.9 ± 2.5 nM cortisol equivalents in the BUD, fluticasone propionate, and CROM groups, respectively). The mean serum transcortin level was 1.15 ± 0.05 µM. Pretreatment transcortin levels were higher in the CROM group than in the BUD group (1.42 ± 0.2 vs. 1.02 ± 0.06 µM; P < 0.05) but did not differ between the BUD and fluticasone propionate groups (ANOVA; P = NS). Serum GBA and transcortin levels did not correlate (r = 0.09, P = NS). The BUD, fluticasone propionate, and CROM groups did not differ, with respect to age (9.6 ± 0.7, 9.9 ± 0.7, and 9.2 ± 1.4 yr, respectively) or baseline serum cortisol levels (216 ± 22, 282 ± 34, and 246 ± 25 nM, respectively).
To investigate the effect of exogenous steroids on circulating GBA, the difference between the measured (Fig. 3A) and the expected GBA value was calculated; the difference was termed: excess GBA. This parameter did not differ significantly from zero in any group at baseline or after 4 months of treatment with CROMs (Fig. 3B). Furthermore, there were no differences in excess GBA at baseline in the three treatment groups (ANOVA, P = NS, Fig. 3B). In the BUD group, however, serum excess GBA increased significantly between 0 and 2 months of treatment, from 0.1 ± 1.6 to 12.1 ± 2.3 nM cortisol equivalents (P < 0.001), and differed from zero at the end of the study period (P < 0.001), indicating the presence of excess GBA in circulation (Fig. 3B). In the fluticasone propionate group, this parameter did not increase significantly during the course of the treatment; at the end of the study, the presence of serum excess GBA was at the borderline of statistical significance (P < 0.08, Fig. 3B). During the treatment, the mean serum transcortin concentration increased significantly in the BUD group (from 1.02 ± 0.06 µM to 1.16 ± 0.06 µM; P < 0.001) but not in the other two groups studied.
|
Drugs acting through the GR are widely used in the clinical setting, but the evaluation of their effects at the target cell level is impossible with antibody-based methods, receptor-binding assays, or methods based on the use of mass spectrometry, because of the different transcriptional properties of hGR agonists and antagonists. We have, therefore, developed a bioassay that employs responses of living mammalian cells to any permeable ligand that activates hGR. Although the bioactivities of glucocorticoids have been previously evaluated using transactivation assays (22, 23) the current bioassay is, to the best of our knowledge, the first to quantify GBA directly in human serum.
The property of the assay to differentiate between glucocorticoid biopotencies suggests that it enables investigation of circulating glucocorticoid milieu also in patients on long-term systemic glucocorticoid therapy, who usually display suppressed endogenous cortisol levels. The order of biopotencies of selected glucocorticoids measured by the assay is in agreement with that observed previously using A549 cells (22). Glucocorticoid-dependent reporter gene activity was blocked by mifepristone, a compound with both antiprogestin and antiglucocorticoid activity (24). This drug acts through the GR, and it has been used to treat Cushing’s syndrome in both children and adults (25, 26, 27, 28, 29). However, an optimal therapeutic mifepristone dose has been difficult to establish because of the lack of a suitable biomarker of GR activity (29). A bioassay such as the one described in this work should be beneficial in evaluating serum GBA also in patients receiving drugs with antiglucocorticoid activity.
We investigated serum GBA levels in children before and during treatment with inhaled steroids. Overall, serum GBA levels were lower than cortisol levels measured with RIA. The reason for this finding is unclear. However, approximately 90% of endogenous cortisol is tightly bound to a plasma carrier protein, transcortin (30); and, in the course of the assay, sera were diluted 10-fold with cell culture medium. This is expected to dissociate the weakest protein-steroid complexes, and steroids that are not bound to transcortin should be available for entering the cells. This notion is supported by the finding that denaturation of transcortin by heating of human serum increases GBA levels measured with the bioassay (Ref. 31 and unpublished observation). However, the effect of transcortin on the results obtained by the bioassay requires further investigation, especially because serum transcortin and GBA levels did not correlate negatively.
It has been recently shown that fluticasone propionate is a more potent transactivator of hGR than BUD (23). On the other hand, BUD has better oral bioavailability than fluticasone propionate (12, 13), which may bear clinical significance, because the majority of inhaled steroid dose becomes swallowed after deposition to the oropharynx (32). The lungs are another important route for systemic delivery of inhaled steroids (33). Regardless of the way of steroid entry to circulation, plasma concentrations of BUD and fluticasone propionate that are reached after inhalation are suggested to be high enough to transactivate hGR (23). The children in the present study used these drugs for 2 months. By using the bioassay, we found that they had (especially those receiving BUD) excess GBA in circulation that was not attributable to endogenous cortisol. A systemic glucocorticoid effect in the BUD group is also inferred from our previous study showing that long-term use of BUD interfered more clearly than fluticasone propionate with the growth of the children studied in this work (11).
There are some limitations in an assay such as the present one. The bioassay measures circulating GBA only at a given time, and the results do not reflect directly the long-term effects of steroid therapy on glucocorticoid target cells. Moreover, the assay is conducted in a single cell line and with one reporter gene only; and, therefore, it cannot mimic all glucocorticoid functions in target tissues (34). The strength of the assay is, however, that it is able to measure glucocorticoid or antiglucocorticoid action of a given compound in circulation, independent of the structure of the chemical. In addition, the bioassay is easy to perform because: the patient serum is ready for the assay immediately after filtering, transfection protocol employs commercial reagents, the plasmids are delivered to the cells as a single mastermix, and less than 10 h of hands-on time is required to run over 60 samples. The small sample volume is also a clear benefit, and it should enable the use of the bioassay to investigate GBA in neonates.
Taken together, we have developed a precise transactivation assay based on the use of recombinant mammalian cells, which enables the measurement of GBA from a small aliquot of human serum. By using the assay to investigate children with asthma, we found that inhaled glucocorticoid treatment affected circulating GBA levels. We expect that this bioassay will have further clinical applications in investigating glucocorticoid milieu in patients receiving drugs acting through the GR.
Acknowledgments
We thank Johanna Iso-Oja for excellent technical assistance.
Footnotes
This work was supported by grants from the Medical Research Council (Academy of Finland), the National Technology Agency (TEKES), the Helsinki University Central Hospital, Research Foundation of Orion Corporation, Finnish Cultural Foundation, Research Foundation of Instrumentarium Corporation, the Foundation for Pediatric Research, and Kuopio University Hospital.
Abbreviations: BUD, Budesonide; CROM, Cromone; CV, coefficient of variation; DHT, 5-dihydrotestosterone; FCS, fetal calf serum; GBA, glucocorticoid bioactivity; GR, glucocorticoid receptor; hGR, human GR; NS, not significant.
Received February 4, 2002.
Accepted April 25, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  P. Nykanen, T. Raivio, K. Heinonen, O. A Janne, and R. Voutilainen Circulating glucocorticoid bioactivity and serum cortisol concentrations in premature infants: the influence of exogenous glucocorticoids and clinical factors Eur. J. Endocrinol., May 1, 2007; 156(5): 577 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Schmidt, J. Rainer, S. Riml, C. Ploner, S. Jesacher, C. Achmuller, E. Presul, S. Skvortsov, R. Crazzolara, M. Fiegl, et al. Identification of glucocorticoid-response genes in children with acute lymphoblastic leukemia Blood, March 1, 2006; 107(5): 2061 - 2069. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Leminen, T. Raivio, S. Ranta, J. Oehler, H. von Hertzen, O. A Janne, and O. Heikinheimo Late follicular phase administration of mifepristone suppresses circulating leptin and FSH - mechanism(s) of action in emergency contraception? Eur. J. Endocrinol., March 1, 2005; 152(3): 411 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kajantie, T. Raivio, O. a. Janne, P. Hovi, L. Dunkel, and S. Andersson Circulating Glucocorticoid Bioactivity in the Preterm Newborn after Antenatal Betamethasone Treatment J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3999 - 4003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Raivio, L. Dunkel, S. Wickman, and O. A. Janne Serum Androgen Bioactivity in Adolescence: A Longitudinal Study of Boys with Constitutional Delay of Puberty J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1188 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Heikinheimo, T. Raivio, H. Honkanen, S. Ranta, and O. A. Janne Termination of Pregnancy with Mifepristone and Prostaglandin Suppresses Transiently Circulating Glucocorticoid Bioactivity J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 323 - 326. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
