| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Department of Endocrinology, Diabetes, and Nutrition, Klinikum Benjamin Franklin, Freie Universität Berlin (S.D., E.E., P.B., M.Q., C.B.-V., M.R., W.O., V.B.), and Research Laboratories of Schering AG (D.S., P. E.), 12200 Berlin, Germany
Address all correspondence and requests for reprints to: Dr. Sven Diederich, Department of Endocrinology, Diabetes, and Nutrition, Klinikum Benjamin Franklin, Freie Universität Berlin, Hindenburgdamm 30, 12200 Berlin, Germany. E-mail: diederich{at}ukbf.fu-berlin.de.
Abstract
The 11ß-hydroxysteroid dehydrogenase (11ß-HSD) system plays a pivotal role in glucocorticoid (GC) and mineralocorticoid (MC) action. Although 11ß-HSD activities are important determinants for the efficacy of synthetic MCs and GCs, corresponding pharmacokinetic data are scanty. Therefore, we characterized 11ß-HSD profiles for a wide range of steroids often used in clinical practice. 11ß-HSD1 and 11ß-HSD2 were selectively examined in 1) human liver and kidney cortex microsomes, and 2) Chinese hamster ovarian cells stably transfected with 11ß-HSD1 or 11ß-HSD2 expression vectors. Both systems produced concordant evidence for the following conclusions.
Oxidation of steroids by 11ß-HSD2 is diminished if they are fluorinated in position 6
or 9 (e.g. in dexamethasone) or methylated at 2 or 6 (in methylprednisolone) or 16 or 16ß, by a methylene group at 16 (in prednylidene), methyloxazoline at 16, 17 (in deflazacort), or a 2-chlor configuration. Whereas the methyl groups also decrease reductase activity (steric effects), fluorination increases reductase activity (negative inductive effect), leading to a shift to reductase activity. This may explain the strong MC activity of 9-fluorocortisol and should be considered in GC therapy directed to 11ß-HSD2-expressing tissues (kidney, colon, and placentofetal unit). 11ß-HSD2 oxidation of prednisolone is more effective than that of cortisol, explaining the reduced MC activity of prednisolone compared with cortisol.
Reduction by 11ß-HSD1 is diminished by 16-methyl, 16ß-methyl, 2-methyl, and 2-chlor substitution, whereas it is increased by the 
1-dehydro configuration in prednisone, resulting in higher hepatic first pass activation of prednisone compared with cortisone.
To characterize a GC or a MC as substrate for the different 11ßHSDs may be essential for an optimized steroid therapy.
THE PRESENCE OF an 11ß-hydroxyl group (Fig. 1
) is essential for the antiinflammatory and immunosuppressive effects of glucocorticoids (GCs) (1, 2) and for the sodium-retaining effects of mineralocorticoids (MCs) (3). Therefore, the interconversion of the 11ß-hydroxyl into the corresponding 11ß-keto group and vice versa plays a pivotal role for the efficacy of these steroids (4). These reactions are catalyzed by the two enzymes 11ß-hydroxysteroid dehydrogenase type 1 (11ß-HSD1) and type 2 (11ß-HSD2). 11ß-HSD1 is expressed in a wide range of tissues (5). Whereas in vitro 11ß-HSD1 functions as a bidirectional enzyme (6), in vivo it works mainly as a reducing and activating enzyme (7). The postulated function of 11ß-HSD1 seems to be autocrine or paracrine modulation of the GC status in nearly all GC target tissues (8).
|
In contrast to endogenous glucocorticoids, 9-fluorinated GCs show weak oxidase but strong reductase activity with 11ß-HSD2 (11, 12). This 9-fluor induced shift to reductase activity is an explanation for the strong MC activity of 9-fluorocortisol (13) and may be useful for targeted renal immunosuppression (14). These examples show that consideration of the metabolism of synthetic GCs and MCs by 11ß-HSDs may be very important for optimizing systemic GC and MC therapy. Therefore, we expand our previous examinations (15) on a large number of widely used synthetic steroids and characterized their substrate specificity for human 11ß-HSD1 and 11ß-HSD2. As human liver (11ß-HSD1) and kidney cortex (11ß-HSD2) selectively express these enzymes, incubations with microsome preparations from these organs are a valid and reproducible system for studying 11ß-HSDs (14). On the other hand, this system is somehow artificial, because the 11ß-HSD1 in isolated liver microsomes is not able to reduce cortisone to cortisol, which is the main and most important reaction in vivo and is also observed in tissue slices and whole cells. Therefore, we complemented our experiments by using intact Chinese hamster ovarian (CHO) cells selectively transfected with human 11ß-HSD1 or 11ß-HSD2 (16).
Materials and Methods
Synthetic steroids
The following steroids were used as 11ß-HSD substrates. Cortisol (Fig. 1a
), 2-methyl-cortisol, prednisolone (1,4-pregnadien-11ß,17,21-triol-3,20-dione; Fig. 1b), dexamethasone (9-fluoro-16-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione; Fig. 1c), betamethasone (9-fluoro-16ß-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione), budesonide (16,17-butilidene-bis[oxy]-1,4-pregnadien-11ß,21-dihydroxy-3,20-dione), and 9-fluoro- cortisol (9-fluoro-11ß,17,21-trihydroxy-4-pregnen-3,20-dione) were acquired from Sigma (St. Louis, MO). 6-Methyl-prednisolone (6-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione), 16-methyl-prednisolone (16-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione), 16ß-methyl-prednisolone (16ß-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione), 6-fluoro-cortisol (6-flouoro-11ß,17,21-trihydroxy-4-pregnen-3,20-dione), fluocortolone (6-flouoro-16-methyl-1,4-pregnadien-11ß,21-dihydroxy-3,20-dione; Fig. 1e), diflucortolone (6,9-diflouoro-16-methyl-1,4-pregnadien-11ß,21-dihydroxy-3,20-dione), and 2-chloro-fluocortolone (2-chloro-6-flouoro-16-methyl-1,4-pregnadien-11ß,21-dihydroxy3,20-dione) were all purchased from Schering AG (Berlin, Germany). Isoflupredone (9-fluoro-1,4-pregnadien-11ß,17,21-triol-3,20-dione), 2-methyl-9-fluoro-cortisone (2-methyl-9-fluoro-17,21-dihydroxy-4-pregnen-3,11,20-trione), desoxymetasone (9-fluoro-16-methyl-1,4-pregnadien-11ß,21-diol-3,20-dione), and flumethasone (6,9-fluoro-16-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione) were obtained from Paesel and Lorei (Hanau, Germany). 2-Methyl-9-fluoro-cortisone was a special synthesis ordered by our group. As we mainly needed it for testing reductase activity (17), we have ordered the synthesis of the 11-oxo form.
Beclomethasone (9-chloro-16ß-methyl-1,4-pregnadien-11ß,17,21-triol-3,20-dione) was purchased from Glaxo Wellcome GmbH (Bad Oldesloe, Germany), prednylidene (16-methylene-1,4-pregnadien-11ß,17,21-triol-3,20-dione) was obtained from Merck \|[amp ]\| Co., Inc. (Darmstadt, Germany), flunisolide (6-flouoro-11,16,17,21-tetrahydroxy-1,4-pregnadien-3,20-dione-16,17-acetonid) was obtained from Boehringer Ingelheim GmbH (Ingelheim, Germany), and deacetyl-deflazacort (16,17-methyloxazoline-1,4-pregnadien-11ß,17,21-triol-3,20-dione; Fig. 1d was purchased from Hoechst Marion Roussel, Inc. (Bad Soden, Germany). Deflazacort is deacetylated in a first pass metabolism, so that deacetyl-deflazacort is the systemically available metabolite (18). If not available, the corresponding 11-oxo-steroids were synthesized by oxidation with chromium VI oxide (11). All steroids were purified by HPLC before use.
Microsomes
Human liver and kidney cortex microsomes were prepared as previously described (14). Protein was quantified before every incubation using Bradford analysis. Substrate concentrations were chosen in the region of maximum velocity [kidney cortex (11ß-HSD2), 10-6 mol/liter, liver (11ß-HSD1), 10–5 mol/liter]. The cosubstrate concentration (kidney: NAD for oxidation, NADH for reduction; liver: NADP for oxidation, NADPH for reduction) was 10-3 mol/liter. The pH for oxidation was 8.5, and that for reduction was 6.0. For each steroid and each reaction tested, pilot studies for time and protein kinetics were performed so that initial velocities could be measured in the linear range.
Transfected CHO cells
The vector p11ßHSD1 was created using pSKHSD1, which was a gift from Dr. A. K. Agarwal (19). The insert coding for the 11ßHSD1 was excised from pSKHSD1 with HindIII and XbaI and ligated into the poly cloning site (PCS) of the mammalian expression vector pcDNA3.1 (Invitrogen, San Diego, CA) to produce p11ßHSD1. Before ligation the PCS was treated with HindIII and XbaI, and the small cleavage product was removed. For creation of p11ßHSD2, the expression plasmid pcDNA1.1HSD2 coding for the 11ß-HSD2 (20) from Dr. A. K. Agarwal was cut with BamHI and NotI. The cleavage was ligated into the PCS of pcDNA3.1 using the cleavage sites for BamHI and NotI.
CHO-K1 cells (American Type Culture Collection, Manassas, VA) were cultured in Ham’s F-12 medium (with stable glutamine derivative; Biochrome Berlin, Germany) supplemented with 10% fetal bovine serum, penicillin (5000 U/ml), streptomycin (5000 µg/ml), and amphotericin B (2.5 µg/ml). Cells were grown in six-well tissue culture plates to 60–70% confluence (37 C, 5% CO2). Transfection was performed using Lipofectamine (Life Technologies, Inc., Karlsruhe, Germany): 1 µg DNA (p11ßHSD1 or p11ßHSD2) was diluted in 100 µl OptiMEM reduced serum medium (Life Technologies, Inc.), and 4 µl Lipofectamine were diluted in 100 µl Earle’s MEM (EMEM) with 2 mM proline, without serum and antibiotics. Both solutions were combined and mixed gently. After 30 min, 800 µl EMEM with 2 mM proline without serum and antibiotics were added to the lipid-DNA complexes, and this suspension was added to the washed cells. After 5-h incubation (37 C, 5% CO2), 1 ml EMEM containing 10% fetal bovine serum and 2 mM proline was added. After 72 h Ham’s F-12 medium with the additives described above and an additional 400 µg/ml geniticin (Life Technologies, Inc.) was added to select transfected cells. Selected clones were propagated in the same medium, but with reduced geniticin (200 µg/ml).
Incubations were performed in flasks with a 75-cm2 culture area (Falcon 3084, BD Biosciences, Heidelberg, Germany). For each reaction tested, pilot studies were performed to determine optimal conditions. For 11ß-HSD2 oxidation, the incubation volume was 13 ml with 2.7 x 106 cells, the incubation time was 12 h, and the substrate concentration was 10-6 mol/liter. For each experiment, cortisol served as the control steroid (percent conversion, 23.7–46.5%; n = 27; mean, 34.2%; SD, 5.7%). For 11ß-HSD2 reduction, the incubation volume was 13 ml with 27 x 106 cells, the incubation time was 12 h, and the substrate concentration was 10-6 mol/liter. For 11ß-HSD1 reduction, the incubation volume was 13 ml with 2.7 x 106 cells, the incubation time was 24 h, and the substrate concentration was 10-5 mol/liter. In each experiment cortisone served as the control steroid (percent conversion, 12.5–28.0%; n = 19; mean, 18.9%; SD, 4.6%).
Analysis of 11ß-HSD activity
Steroid extraction, separation, and quantitation were performed as described previously (21). As the steroids tested were only available as nonradioactive substrates, quantification of the corresponding 11- hydroxy and 11-oxoderivatives was performed using UV detection after extraction by Sep-Pak C18 cartridges (Waters Corp., Milford, MA) and separation by reverse phase HPLC (22). In all incubations (microsomes and CHO cells), we were unable to detect any other metabolites than the 11-oxo- or 11-hydroxysteroids. Calculation of the percent conversion rates allowed determination of the initial velocities.
Statistics
Statistical calculations were performed using the SPSS program from SPSS, Inc. (Chicago, IL). An unpaired t test was used.
Results
11ß-HSD2: steroid metabolism in human kidney cortex microsomes (Table 1) and by transfected CHO cells (Figs. 2 and 3)
Whereas oxidation by 11ß-HSD2 was the sole reaction for unfluorinated steroids, fluorinated steroids were metabolized bidirectionally, with a strong preference for reductase activity (Table 1).
|
|
|
Oxidation by 11ß-HSD2 was diminished by 6-methyl, 16-methyl, 16ß-methyl, and 6-fluor or 9-fluor substitution (Table 1, footnotes a and b). In addition, a 2-methyl group (cortisol vs. 2-methyl-cortisol), a 16,17-methyloxazoline configuration (prednisolone vs. deacetyl-deflazacort, Fig. 1d), a 16-methylene substitution (prednisolone vs. prednylidene), and a 2-chlor substitution (fluocortolone vs. 2-chlor-fluocortolone) also decreased 11ß-HSD2 oxidation (Fig. 2). Both topically used GCs tested showed little oxidation by 11ß-HSD2 (budesonide and flunisolide, Fig. 2). Multiple substitutions in these steroids did not allow systematic comparison with the other steroids tested.
11ß-HSD2 oxidation was increased by a 1-dehydro configuration (prednisolone vs. cortisol; Table 1, footnote b, and Fig. 2). In addition, fluocortolone (Fig. 1e; 17-desoxy configuration and 6-fluor substitution) showed surprisingly high oxidative activity in the group of fluorinated steroids (Table 1, footnote f, and Fig. 2).
11ß-HSD2 reduction
Reduction by 11ß-HSD2 was decreased by 2-methyl (Fig. 3), 16-methyl, and 16ß-methyl substitution (dexamethasone and betamethasone vs. isoflupredone, Table 1).
Whereas 6- or 9-fluorination decreased 11ß-oxidation, it dramatically increased 11ß-reduction, leading to a shift of the 11ß-HSD2 redox equilibrium to the active 11-hydroxy side (Table 1). 6- and 9-fluorination had additive effects in augmenting reductase activity (flumethasone vs. dexamethasone, diflucortolone vs. desoxymetasone; Table 1). In concordance with the effects on 11ß-oxidation, the 1-dehydro configuration also increased 11ß-reduction (isoflupredone vs. 9-fluoro-cortisol, Table 1).
The effects of the 17-desoxy configuration seem to be influenced by other modifications in the structure of the steroid. In double fluorinated steroids such as flumethasone, it increased reductase activity and decreased oxidase activity, leading to a shift to the active 11-hydroxy side (flumethasone vs. diflucortolone, Table 1). In monofluorinated steroids in position 9, it had no significant effects (dexamethasone vs. desoxymetasone, Table 1), whereas monofluorination in position 6, as in fluocortolone (Fig. 1e), combined with the 17-desoxy configuration led to an impressing increase in oxidase activity (Table 1, footnote f) and a strong decrease in reductase activity (Table 1, footnote h). Therefore, fluocortolone is unique in the group of fluorinated steroids, because it has its redox equilibrium on the inactive 11-oxo side, as do the unfluorinated steroids (Table 1).
In concordance with effects on 11ß-HSD2 oxidation, a substitution with chlorine in position 2 totally abolished 11ß-HSD2 reductase activity (experiments with 11-dehydro-2-chlor-fluocortolone in 11ß-HSD2-transfected CHO cells; data not shown).
11ß-HSD1: steroid metabolism in human liver microsomes (Table 2) and by transfected CHO cells (Fig. 4)
In agreement with the effects on 11ß-HSD2, the redox equilibrium of 11ß-HSD1 is also shifted to the active 11-hydroxy side by 6- or 9-fluorination (Table 2 and Fig. 4, cortisol vs. 9-fluoro-cortisol). In human liver microsomes, 11ß-reduction could only be demonstrated with fluorinated steroids (Table 2), whereas transfected CHO cells also showed reductase activity with unfluorinated steroids such as cortisone or prednisone (Fig. 4).
|
|
Oxidation by 11ß-HSD1 was decreased by a 16-methyl group (Table 2, footnote a) and was abolished by 6-fluor or 9-fluor substitution (Table 2, all fluorinated steroids). In contrast to 11ß-HSD2 oxidation (Table 1), 6-methyl and 16ß-methyl groups showed no significant effect on 11ß-HSD1 oxidation (Table 2). Oxidation by 11ß-HSD1 was increased by the 1-dehydro configuration (Table 2, footnote a).
11ß-HSD1 reduction
Whereas 16-methyl and 16ß-methyl groups (Table 2, isoflupredone vs. dexamethasone and betamethasone) and 2-methyl and 2-chlor substitutions (Fig. 4, 9-fluoro-cortisol vs. 2-methyl-9-fluoro-cortisol and fluocortolone vs. 2-chlor-fluocortolone) led to a significant decrease in 11ß-HSD1 reduction, 16-methylene and 6-methyl substituents had no effect on 11ß-HSD1 reduction (Fig. 4, prednisolone vs. prednylidene and 6-methyl-prednisolone).
11ß-HSD1 reduction was increased by 6-fluor or 9-fluor substitution (Table 2; all fluorinated steroids) and by the 1-dehydro configuration (Table 2, isoflupredone vs. 9-fluoro-cortisol; Fig. 4, cortisol vs. prednisolone). 9-Chlor substitution instead of 9-fluor substitution significantly augmented the reductase activity of 11ß-HSD1 (beclomethasone vs. betamethasone, Fig. 4). In addition, double fluorination in positions 6 and 9 (flumethasone vs. dexamethasone, Table 2) as well as the 17-desoxy configuration (diflucortolone vs. flumethasone, Table 2) led to increased reductase activity of 11ß-HSD1.
Discussion
Apart from pharmacodynamic parameters like receptor binding and transactivation, the activity of a hormone is determined by pharmacokinetic characteristics. The most important pharmacokinetic systems for GCs and MCs are the 11ß-HSDs, because they regulate the target cell adjustment between the active hydroxy and the inactive oxo form of a steroid (4). In mammalian tissues at least two isoenzymes of 11ß-HSD are responsible for this elegant autocrine fine- tuning. Whereas the widely distributed 11ß-HSD1 acts predominantly as a reductase, facilitating GC hormone action, the selectively expressed 11ß-HSD2 acts as an oxidizing enzyme, thus diminishing the GC and MC activities of a steroid in the corresponding organs. Beside this autocrine function, 11ß-HSD1 has an important pharmacological function, because hepatic 11ßHSD1 activates orally given pro-drugs such as cortisone or prednisone to their active 11-hydroxy derivatives (23, 24).
In this context the knowledge about 11ß-HSD substrate specificity of a GC or a MC seems to be indispensable for an optimized steroid therapy. Moreover, as fluorinated steroids are mainly reduced by 11-ßHSD2, we have developed the idea of targeted renal immunosuppression with a fluorinated 11-dehydrosteroid, which should have high affinity to 11ß-HSD2, but low affinity to 11ß-HSD1 (14). As both isoenzymes share only 14% homology, divergent effects of modifications in the steroid structure can be expected.
Human kidney cortex and liver microsomes as well as selectively transfected CHO cells are both well established systems for studying 11ß-HSD activity. Although the initial velocities of the different systems cannot be compared because the concentrations of the 11ß-HSDs are not known, comparison between different steroids in the same system is a useful method to determine the effects of specific substitutions on the metabolism by 11ß-HSDs. Moreover, the high agreement of the results in both systems underlines the correctness of our data.
Oxidase activity of 11ß-HSD2 seems to be very important for GC therapy targeted to organs expressing mainly or exclusively 11ß-HSD2 (kidney and colon). High 11ß-HSD2 oxidation of an 11-hydroxy-GC should result in strong inactivation in the corresponding organs, whereas 11-hydroxy GCs with low 11ß-HSD2 oxidation should have significant pharmacokinetic advantages because of their reduced inactivating capacity in the target tissues. As 11ß-HSD2 in the placenta protects the fetus from maternal GCs (25), the best option for intrauterine treatment of the fetus should be steroids with very low oxidation by 11ß-HSD2. If the mother and not the fetus needs GC therapy, steroids with high 11ß-HSD2 oxidation should be the preference.
We have defined many substituents (2-methyl, 2-chlor, 6-methyl, 16-methyl, 16ß-methyl, 16-methylene, and 16,17-methyloxazoline substituents) that lead to steric inhibition of 11ß-HSD2 oxidation and may be favorable for GC therapy directed to 11ß-HSD2-expressing organs. The most important effect of substitution is that of 6- or 9-fluorination, which diminishes 11ß-HSD2 oxidation and dramatically increases 11ß-HSD2 reduction, thus leading to a strong shift to the active 11-hydroxy form of a steroid. Therefore, fluorinated 11-hydroxysteroids show the lowest inactivating capacity by 11ß-HSD2 and should be the best therapeutic option for intrauterine treatment of the fetus or for renal immunosuppression. As double fluorination in positions 6 and 9 augments this shift to the active 11-hydroxy side, flumethasone or diflucortolone seems to be the first choice for these approaches.
The fluor-induced shift to reductase activity can be enhanced by substituting chlorine for fluor in the 9 position (beclomethasone vs. betamethasone). This underlines the thesis that the negative inductive effect of halogen atoms leads to the shift in the redox equilibrium (26).
Both topically used GCs tested (budesonide and flunisolide) show little inactivation by 11ß-HSD2. As the target organs of topical GC therapy (lung, skin, and rectum) also express 11ß-HSD2 (9, 27, 28, 29), a pharmacokinetic profile with little 11ß-HSD2 oxidation seems to be favorable for these substances.
In the group of fluorinated steroids, fluocortolone is an interesting exception, because this steroid, despite a 6-fluor substitution, shows effective oxidation by 11ß-HSD2, which seems to result from combination with the 17-deoxy configuration. In the group of unfluorinated steroids, prednisolone shows the highest activity for 11ß-HSD2 oxidation. Concerning GC therapy, 11-hydroxysteroids with high inactivation by 11ß-HSD2 (e.g. fluocortolone and prednisolone) should be used when the 11ß-HSD2-expressing tissues have to be protected from GC action, e.g. in pregnancy when the mother and not the fetus needs GC treatment.
Apart from the discussed recommendations for GC therapy, our results allow some conclusions concerning the access of 11-hydroxysteroids to the MC receptor, which is the physiolocically most important function of 11ß-HSD2. The increased 11ß-HSD2 oxidation of prednisolone compared with cortisol may lead to enhanced renal inactivation (30), thus giving a good pharmacokinetic explanation for the reduced MC activity of prednisolone compared with cortisol. On the other hand, the diminished 11ß-HSD2 oxidation of 9-fluoro-cortisol is one explanation for its strong MC activity compared with cortisol (13).
The most important pharmacological function of 11ß-HSD1 is hepatic first pass activation of orally given inactive 11-dehydrosteroids (23, 24). The increased 11ß-HSD1 reduction of prednisone compared with cortisone may explain why orally given prednisone shows a more effective hepatic activation than cortisone, resulting in higher systemic availability of the active 11-hydroxy form prednisolone (23).
Concerning our idea of renal GC targeting (14), we are interested in substituents that decrease 11ß-HSD1 reduction but do not affect 11ß-HSD-2 reduction. A fluorinated 11-dehydrosteroid with such a modification should pass 11ß-HSD1 in the liver nearly unchanged, resulting in low plasma concentrations of the corresponding 11-hydoxysteroid and thus minimizing systemic side-effects of the steroid. As renal 11ß-HSD2 activity determines local 11-hydroxy/11-dehydro ratios in the kidney (30), the fluorinated 11-dehydrosteroid should be specifically activated in the kidney, leading to a targeted enrichment of the active 11-hydroxysteroid in this organ. Similar to immunosuppression in inflammatory lung and skin diseases, local intrarenal manipulation of the immune response seems to be the most important factor determining renal allograft survival (31, 32). Therefore, our approach may be a promising way for renal immunosuppression with reduced side-effects.
Reduction by 11ß-HSD1 was diminished for 16-methyl-, 16ß-methyl-, and especially 2-methyl- and 2-chlor-substituted steroids. A 2-chlor substitution totally inhibited 11ß-HSD1 reduction, but showed the same effect on oxidation and reduction by 11ß-HSD2. As 2-, 16-, and 16ß-methylations also result in decreased reduction by 11ß-HSD2, these modifications are not optimal for the idea of renal GC targeting. Therefore, further investigations, preferentially with computer-designed substances, have to be performed. An alternative approach for GC targeting to the kidney may be the combination of an available fluorinated 11-dehydrosteroid (e.g. 11-dehydro-dexamethasone or 11-dehydro- diflucortolone) with a selective inhibitor of 11ß-HSD1 (33).
In summary, we have characterized the 11ß-HSD profile of a wide range of clinically often used steroids. Both in vitro systems used were shown to be suitable for this approach: 1) human kidney and liver microsomes (to measure 11ß-HSD1 reductase activity in microsomes, fluorinated steroids have to be used) and 2) stably 11ß-HSD1- and 11ß-HSD2-transfected CHO cells (which have the advantage of permanent availability and relatively high similarity to the in vivo conditions).
Our results allow some pharmacokinetic considerations that have to be proven by clinical investigations. The 1-dehydro configuration (as in prednisolone) increases renal oxidation by 11ß-HSD2 and may be the reason for the diminished MC activity of prednisolone compared with cortisol.
The ideal GC targeted to organs expressing mainly or exclusively 11ß-HSD2 (placentofetal unit, kidney, and colon) should have methyl groups in position 2, 6, or 16 and fluor atoms in position 6 or 9, as these substituents decrease local inactivation by 11ß-oxidation.
Therefore, betamethasone (9-fluor-16ß-methyl-prednisolone) and dexamethasone (9-fluor-16-methyl-prednisolone), which are mainly used in intrauterine GC treatment of the fetus, seem to be suitable for this therapeutic approach. Whether flumethasone, diflucortolone, or 2-chlor-fluocortolone, which show even less inactivation by 11ß-HSD2, may offer clinically relevant benefits seems to be questionable.
As prednisolone, the widely used GC for renal immunosuppression, shows very strong oxidation by 11ßHSD2, this steroid does not seem to be ideal for these patients. Thus, exact clinical studies comparing prednisolone with, for example, 6-methyl-prednisolone, deflazacort, and dexamethasone in kidney transplantation should be very informative.
Acknowledgments
We thank Dr. A. K. Agarwal for the 11ß-HSD plasmids, and Drs. H. Laurent and H. J. Zentel (Schering AG, Berlin, Germany) for assistance in steroid preparation and separation. We also thank Mrs. G. Kainzbauer for assistance with transfection studies.
Footnotes
This work was supported by a grant from Deutsche Forschungsgesellschaft (DI 741/1-3).
Abbreviations: CHO, Chinese hamster ovarian; EMEM, Earle’s MEM; GC, glucocorticoid; 11ß-HSD, 11ß-hydroxysteroid dehydrogenase; MC, mineralocorticoid; PCS, poly cloning site.
Received June 25, 2002.
Accepted September 6, 2002.
References
-fluorocortisol: an explanation for its mineralocorticoid potency. J Clin Endocrinol Metab 78:928–932[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  K. Lovas, C. G Gjesdal, M. Christensen, A. B Wolff, B. Almas, J. Svartberg, K. J Fougner, U. Syversen, J. Bollerslev, J. A Falch, et al. Glucocorticoid replacement therapy and pharmacogenetics in Addison's disease: effects on bone Eur. J. Endocrinol., June 1, 2009; 160(6): 993 - 1002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Kim, K. X. Kim, N. N. Raveendran, T. Wu, S. R. Pondugula, and D. C. Marcus Regulation of ENaC-mediated sodium transport by glucocorticoids in Reissner's membrane epithelium Am J Physiol Cell Physiol, March 1, 2009; 296(3): C544 - C557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Liu, F. Park, J. L. Pietrusz, G. Jia, R. J. Singh, B. C. Netzel, and M. Liang Suppression of 11{beta}-hydroxysteroid dehydrogenase type 1 with RNA interference substantially attenuates 3T3-L1 adipogenesis Physiol Genomics, February 19, 2008; 32(3): 343 - 351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Tomlinson, M. Sherlock, B. Hughes, S. V. Hughes, F. Kilvington, W. Bartlett, R. Courtney, P. Rejto, W. Carley, and P. M. Stewart Inhibition of 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Activity in Vivo Limits Glucocorticoid Exposure to Human Adipose Tissue and Decreases Lipolysis J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 857 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Jakobsson, J. A. Jorgensen, and A. Jacobsson Differential regulation of fatty acid elongation enzymes in brown adipocytes implies a unique role for Elovl3 during increased fatty acid oxidation Am J Physiol Endocrinol Metab, October 1, 2005; 289(4): E517 - E526. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. O. Leonard, C. Godson, H. R. Brady, and C. T. Taylor Potentiation of Glucocorticoid Activity in Hypoxia through Induction of the Glucocorticoid Receptor J. Immunol., February 15, 2005; 174(4): 2250 - 2257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gulati, K. Gulati, T. Sugiyama, H. Tanaka, T. Taguchi, M. S. Cooper, P. M. Stewart, M. B. Leonard, B. S. Zemel, and V. A. Stallings Glucocorticoids and Bone Mineral Content in the Childhood Nephrotic Syndrome N. Engl. J. Med., December 16, 2004; 351(25): 2655 - 2657. [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Tomlinson, E. A. Walker, I. J. Bujalska, N. Draper, G. G. Lavery, M. S. Cooper, M. Hewison, and P. M. Stewart 11{beta}-Hydroxysteroid Dehydrogenase Type 1: A Tissue-Specific Regulator of Glucocorticoid Response Endocr. Rev., October 1, 2004; 25(5): 831 - 866. [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 |
) of unfluorinated and fluorinated steroids by 11ß-HSD2-transfected CHO cells. Initial velocities (Vo; mean ± SD) are presented. Six replicates were performed for each reaction. The substrate concentration was 10-6 mol/liter; incubation was for 12 h; 2.7 x 106 cells in 13 ml medium were used. n.c., No conversion. **, P < 0.01; ***, P < 0.001. Reductase activity was not measured for these steroids. For steroid structures, see Fig. 1
) of 9