Explorer 11-Hydroxysteroid dehydrogenase type 1 contributes to the balance between 7-keto-and 7-hydroxy-oxysterols in vivo

11b-Hydroxysteroid dehydrogenase 1 (11bHSD1; EC 1.1.1.146) generates active glucocorticoids from inert 11-keto metabolites. However, it can also metabolize alternative substrates, including 7bhydroxyand 7-keto-cholesterol (7bOHC, 7KC). This has been demonstrated in vitro but its consequences in vivo are uncertain. We used genetically modified mice to investigate the contribution of 11bHSD1 to the balance of circulating levels of 7KC and 7bOHC in vivo, and dissected in vitro the kinetics of the interactions between oxysterols and glucocorticoids for metabolism by the mouse enzyme. Circulating levels of 7KC and 7bOHC in mice were 91.3 22.3 and 22.6 5.7 nM respectively, increasing to 1240 220 and 406 39 nM in ApoE / mice receiving atherogenic western diet. Disruption of 11bHSD1 in mice increased (p < 0.05) the 7KC/7bOHC ratio in plasma (by 20%) and also in isolated microsomes (2 fold). The 7KC/7bOHC ratio was similarly increased when NADPH generation was restricted by disruption of hexose-6-phosphate dehydrogenase. Reduction and oxidation of 7-oxysterols by murine 11bHSD1 proceeded more slowly and substrate affinity was lower than for glucocorticoids. in vitro 7bOHC was a competitive inhibitor of oxidation of corticosterone (Ki = 0.9 mM), whereas 7KC only weakly inhibited reduction of 11-dehydrocorticosterone. However, supplementation of 7-oxysterols in cultured cells, secondary to cholesterol loading, preferentially slowed reduction of glucocorticoids, rather than oxidation. Thus, in mouse, 11bHSD1 influenced the abundance and balance of circulating and tissue levels of 7bOHC and 7KC, promoting reduction of 7KC. In health, 7-oxysterols are unlikely to regulate glucocorticoid metabolism. However, in hyperlipidaemia, 7-oxysterols may inhibit glucocorticoid metabolism and modulate signaling through corticosteroid receptors. 2013 Elsevier Inc. All rights reserved. * Corresponding author. Tel.: +44 131 242 6763; fax: +44 131 242 6779. E-mail addresses: Tijana.mitic@bristol.ac.uk (T. Mitić), s.shave@ed.ac.uk (S. Shave), nina_sem@hotmail.com (N. Semjonous), Iain.McNae@ed.ac.uk (I. McNae), d.f.cobice@sms.ed.ac.uk (D.F. Cobice), g.g.lavery@bham.ac.uk (G.G. Lavery), scott.webster@ed.ac.uk (S.P. Webster), patrick.hadoke@ed.ac.uk (Patrick W.F. Hadoke), b.walker@ed.ac.uk (B.R. Walker), r.andrew@ed.ac.uk (R. Andrew).


Introduction
Intracellular generation of active glucocorticoids (cortisol in humans, corticosterone in mice) is catalyzed by 11b-hydroxysteroid dehydrogenase (11bHSD) type 1 (EC 1.1.1.146). The potential for 11bHSD1 to regulate fuel metabolism has been demonstrated in murine models, in which disruption of the enzyme protects from metabolic dyshomeostasis [1,2] and, more recently, in humans in whom specific 11bHSD1 inhibitors improve hyperglycaemia [3]. In murine models, inhibition of 11bHSD1 also offers atheroprotection [4][5][6]. Therefore inhibition of the reductase activity of 11bHSD1 is a tractable target for drug development, but to fully understand the spectrum of actions and side-effects of such drugs, effects on other substrates of 11bHSD1 must be considered. This is, as yet, unexplored in vivo, either in genetically modified mice or following selective pharmacological manipulation.
In addition to metabolizing glucocorticoids, 11bHSD1 can catalyze the inter-conversion of 7-keto-and 7b-hydroxy-sterols and steroids (Fig. 1a) (e.g. 7-oxygenated metabolites of dehydroepiandrosterone [7] and highly cytotoxic cholesterol metabolites, the 7-oxysterols [8,9]). 7-Oxysterols are formed from cholesterol both enzymatically and by auto-oxidation [10]. They accumulate in atherosclerotic plaques, a site of 11bHSD1 expression [11], with 7-ketocholesterol (7KC) being the most abundant, closely followed 11b-Hydroxysteroid dehydrogenase 1 (11bHSD1; EC 1.1.1.146) generates active glucocorticoids from inert 11-keto metabolites. However, it can also metabolize alternative substrates, including 7bhydroxy-and 7-keto-cholesterol (7bOHC, 7KC). This has been demonstrated in vitro but its consequences in vivo are uncertain. We used genetically modified mice to investigate the contribution of 11bHSD1 to the balance of circulating levels of 7KC and 7bOHC in vivo, and dissected in vitro the kinetics of the interactions between oxysterols and glucocorticoids for metabolism by the mouse enzyme.
Circulating levels of 7KC and 7bOHC in mice were 91.3 AE 22.3 and 22.6 AE 5.7 nM respectively, increasing to 1240 AE 220 and 406 AE 39 nM in ApoE À/À mice receiving atherogenic western diet. Disruption of 11bHSD1 in mice increased (p < 0.05) the 7KC/7bOHC ratio in plasma (by 20%) and also in isolated microsomes (2 fold). The 7KC/7bOHC ratio was similarly increased when NADPH generation was restricted by disruption of hexose-6-phosphate dehydrogenase.
Reduction and oxidation of 7-oxysterols by murine 11bHSD1 proceeded more slowly and substrate affinity was lower than for glucocorticoids. in vitro 7bOHC was a competitive inhibitor of oxidation of corticosterone (K i = 0.9 mM), whereas 7KC only weakly inhibited reduction of 11-dehydrocorticosterone.
However, supplementation of 7-oxysterols in cultured cells, secondary to cholesterol loading, preferentially slowed reduction of glucocorticoids, rather than oxidation.
Thus, in mouse, 11bHSD1 influenced the abundance and balance of circulating and tissue levels of 7bOHC and 7KC, promoting reduction of 7KC. In health, 7-oxysterols are unlikely to regulate glucocorticoid metabolism. However, in hyperlipidaemia, 7-oxysterols may inhibit glucocorticoid metabolism and modulate signaling through corticosteroid receptors.
by 7b-hydroxycholesterol (7bOHC) [12]. Early reports [13,14] revealed that 11bHSD1 converted 7bOHC to 7KC in hepatic microsomes from all vertebrates tested (human, guinea-pig, rat, hamster and chicken) and that rat hepatic 11bHSD1 also reduced 7KC to 7bOHC. However, this has not been studied in other species and it remains unclear whether enzymes other than 11bHSD1 also catalyze interconversion of 7bOHC and 7KC. 11bHSD1 is a bi-directional enzyme (Fig. 1a) and both dehydrogenase (inactivating glucocorticoids) and reductase (regenerating glucocorticoids) activities can be measured in tissues [15,16]. The prevalent direction of 11bHSD1, with respect to metabolism of glucocorticoids, is reduction and is dependent on the availability of endogenous co-factor (NADPH), which is generated by hexose-6-phosphate dehydrogenase (H6PDH) within the endoplasmic reticulum (ER) [17]. Mice lacking H6PDH are unable to regenerate glucocorticoids by 11bHSD1 [18] but it is unclear if NADPH supply physiologically regulates the balance between reductase and dehydrogenase activities and the contribution of H6PDH in vivo has not been investigated for 7-oxysterols. Pharmacological inhibition of 11bHSD1 in rats caused hepatic accumulation of 7KC [9] suggesting that, as with glucocorticoids, the predominant direction of metabolism of 7-oxysterols by 11bHSD1 in vivo is reduction. Tissue-specific differences in the equilibrium position of metabolism of glucocorticoids by 11bHSD1 may indeed be due to the presence of competitive substrates, as some reports have suggested that 7-oxygenated compounds inhibit metabolism of glucocorticoids by 11bHSD1 [19]. For example, 7KC and 7bOHC inhibit 11bHSD1 activity in mouse adipocyte (3T3-L1 and 3T3-F442) cell lines [20] and in differentiated human THP-1 macrophages [21], modulating the downstream actions of glucocorticoids.
We hypothesized that 11bHSD1 is a key determinant of the balance of 7bOHC and 7KC in vivo. Depending on their levels in the circulation and tissues, 7KC and 7bOHC may differentially inhibit either reduction or dehydrogenation of glucocorticoids, respectively. Since these oxysterols accumulate in tissues that express 11bHSD1 [10] (e.g. macrophages, foam cells, adipose, atherosclerotic plaques [11]), the relative proportion of 7KC to 7bOHC may influence the amount of active glucocorticoid within cells. To address this hypothesis we investigated the balance of 7KC and 7bOHC in mice with transgenic disruption of 11bHSD1 and H6PDH, and abilities of these 7-oxysterols to influence the equilibrium between the dehydrogenase and reductase activities of glucocorticoid metabolism by murine 11bHSD1.

Animals
Male mice (10-16 weeks, n = 6-8/group [2,23]) with disruption of 11bHSD1 (Hsd11b1 À/À ) or H6PDH (H6pdh À/À ) or both (Hsd11b1 À/À /H6pdh À/À ) and their wild-type littermate controls (15 weeks) were maintained on chow diet and tap water ad libitum, under a 16 h/8 h light/dark cycle at 21-24 8C. Male ApoE À/ À mice (in-house colony, 8 weeks; n = 6) were maintained on a western Diet (D12079B, Research Diets, USA) for 14 weeks. All licensed procedures were performed under accepted standards of humane animal care, as outlined in the UK Home Office Ethical Guidelines. Animals were culled by cervical dislocation at 10:00 h. Tissues and fluids were snap-frozen and stored at À80 8C. (a) Interconversion of glucocorticoids and 7-oxysterols catalyzed by 11bhydroxysteroid dehydrogenase type 1 (11bHSD1). The equilibrium of interconversion of inert 11-keto and active 11b-hydroxy forms of glucocorticoids (shown here as 11-dehydrocorticosterone and corticosterone, the principle rodent glucocorticoids) favors predominant reduction. 11bHSD1 can also interconvert 7-keto and 7b-hydroxycholesterol but the favored equilibrium position between the two reactions is not understood. (b)-(e) In Silico modeling of interactions between 7-oxysterols and residues in the active site of murine 11bhydroxysteroid dehydrogenase 1 (m11bHSD1). 2D Modeling of the active site of m11bHSD1 (retrieved from PDB 1Y5 M) using LigPlot. Hydrogen bond lengths of interactions between (b) 7-ketocholesterol and (c) 7b-hydroxycholesterol and the critical residues of catalytic tetrad are shorter than those for (d) 7ahydroxycholesterol (7aOHC). (e) 3D modeling of interactions between active site residues Serine 170 (S170) and Tyrosine 183 (Y183) of m11bHSD1 and the 7oxygenated moieties using PyMOL. Positioning of 7bOHC (pink) or 7KC (yellow) into the active site demonstrated their more favorable orientation over 7aOHC (turquoise), for hydrogen bonding with key amino acids of m11bHSD1 active site.

Quantitation of 11bHSD1 enzyme kinetics
Inter-conversion of substrates and products was quantified under conditions of first order kinetics. Three forms of murine enzyme (n = 6/group) were used: (1) a truncated form of recombinant m11bHSD1 protein (N23D, gift from Dr Webster), (2) enzyme contained within murine hepatic microsomes and (3) a full-length m11bHSD1 protein expressed in stably transfected HEK293 cells [24].

Metabolism by recombinant murine 11bHSD1 expressed in stably transfected cells
HEK293 cells, stably transfected to produce m11bHSD1, were seeded onto a 5 cm dish and incubated overnight with 7KC, 7bOHC or 7aOHC (1 mM), or with steroid (30 nM) for 45 min.

Supplementation of cholesterol in stably transfected cells
To enrich cellular cholesterol and 7-oxysterol content, HEK293 cells stably expressing m11bHSD1 were incubated (37 8C, 30 min) with cholesterol-loaded methyl-b-cyclodextrin (1:6, 10 mM in DMEM) [26]) and kinetic experiments performed within 24 h. Following manipulation, cells were washed with DMEM (37 8C) followed by phosphate buffered saline, and then incubated ( (1 mg) were added and oxysterols and cholesterol were immediately extracted into methanol:hexane (2:5, 7 mL, 50 mg/mL BHT, 2 mM EDTA). The dried organic extract was dissolved in chloroform: methanol (2:1) and processed for quantitation by GCMS. All final measurements were expressed as a ratio of the total protein content in the cells.

Quantitation of circulating and tissue levels of 7-oxysterols
7-Oxysterols were quantified in plasma (0.4-1 mL) prepared from trunk blood collected (pooled if necessary) in EDTA-coated tubes from mice (n = 8/group). Plasma was prepared from blood collected in EDTA-coated (1.6 mg/mL) vials. The effects of disruption of Hsd1b1, H6pdh or both were explored in hepatic microsomes and cytosol (0.05-0.5 mg/mL protein) from mice homozygous for the disrupted allele (n = 6/group) versus their littermate controls. All samples were flushed under argon prior to extraction and BHT (45 mM, in ethanol) added before 7-oxysterols were extracted and converted to their trimethylsilyl derivatives [28] prior to analysis by GCMS [29].
2.8. In silico modeling of interactions between 7-oxysterols and residues in the active site of murine 11b-hydroxysteroid dehydrogenase 1 (m11bHSD1) 3D Macromolecular structural information about m11bHSD1 was obtained from the Research Laboratory for Structural Bioinformatics Protein Data Bank. 1Y5 M represented a dimeric m11bHSD1 bound with NADP+ and 1Y5R represented m11bHSD1 bound with NADP+ and corticosterone [30]. The structure of 7ahydroxysteroid dehydrogenase (EC1. 1.1.159, 7aHSD, PDBID 1FMC) in complex with 7-oxoglycochenodeoxycholic acid [31] was a template for modeling the steric orientation of 7aOHC, allowing alignment of 7a-and 7b-hydroxyl and 7-keto groups into the active site, when 7aHSD and 11bHSD1 were subsequently superimposed. Energy maps for all ligand atoms around the active site were generated using the virtual screening program LIDAEUS (Ligand Discovery At Edinburgh University). Energy minimization routines were used to aid the positioning of substrate within the active site of 11bHSD1. 2D Representations of protein-ligand complexes from modeled structures were created using LigPlot (Cambridge, UK), the output of which was then augmented by 2D representations of substrates generated by MARVINVIEW 1 (ChemAxon, Budapest, Hungary) to distinguish between the steric orientation of 7a-and 7bOHC. Visualization of 3D structures was performed using PyMOL (open source, DeLano Scientific LLC).   [29]. Levels increased more than 10 fold (1240 AE 22 (7KC) and 406 AE 39 (7bOHC) nM) in ApoE À/À mice on an atherogenic, western diet. Following disruption of Hsd11b1, there was a trend (p = 0.08) for an increase in concentrations of 7KC (133.8 AE 16.8 nM) but not 7bOHC (23.6 AE 2.2 nM) [29]. However, the ratio of 7KC/7bOHC in plasma significantly increased in the Hsd11b1 À/À (5.4 AE 0.5) vs. control mice (4.1 AE 0.4, n = 9, p < 0.05).
Both 7KC and 7bOHC were detected in microsomes from control mice. Disruption of Hsd11b1 caused a profound reduction in hepatic microsomal concentrations of both oxysterols (Table 1), with an increase in the 7KC/7bOHC ratio. In the cytosols from control murine liver, only 7KC (25.3 AE 13.4 ng/mg protein) was detected, but following disruption of 11bHSD1, levels of 7KC became undetectable. Disruption of H6pdh, or both H6pdh and Hsd1b1 also lowered the levels of 7bOHC and 7KC in the hepatic microsomes compared with littermate controls ( Table 1). The 7KC/7bOHC ratio increased with disruption of H6pdh and disruption of both H6pdh and Hsd11b1 did not have any further effect over lack of 11bHSD1 alone ( Table 1).

Oxysterols Inhibit oxidation and/or reduction of glucocorticoids
Competition between 7-oxysterols and glucocorticoids for metabolism by 11bHSD1 across physiological and pathophysiological concentration ranges was investigated using three preparations of murine enzyme. In all preparations, 7aOHC was not accepted as a substrate and not generated upon reduction of 7KC (not shown).

Murine 11bHSD1 stably transfected into HEK293 cells
Both oxidation and reduction of glucocorticoids were detected in vitro, and reduction was the preferred direction (0.79 AE 0.15 (oxidation) vs. 3.86 AE 0.27 (reduction) pmol/mg/min, respectively, with 30 nM substrate). Both oxidation of 7bOHC and reduction of 7KC, were observed, at similar velocities, which were considerably slower than those measured for glucocorticoids. For example, substrate concentrations of 1 mM were required to achieve rates of oxidation of 7bOHC and reduction of 7KC of 0.90 AE 0.31 vs 0.74 AE 0.04 pmol/mg/min, respectively. Inhibition of metabolism of glucocorticoids by a range of endogenous oxysterols was assessed in both reductase and dehydrogenase directions. 7KC caused the most marked inhibition of reduction of all oxysterols tested, although still only by 40% at the highest concentration used (100 mM; Fig. 2a) and further kinetic analysis was not performed. Of the different oxysterols tested, only 7bOHC inhibited oxidation of corticosterone, with a K i of 1.77 AE 0.09 mM (Fig. 2b).

Murine recombinant 11bHSD1
Although both oxidation and reduction of glucocorticoids were detected using recombinant 11bHSD1, reduction of 11-DHC was the favored reaction (lower K m and higher V max , Table 2). Oxidation of 7bOHC and reduction of 7KC were also detected but proceeded with slower maximal rates and these substrates had poorer affinity (higher K m s; Table 2) than glucocorticoids. 7KC inhibited reduction of 11-DHC (Fig. 2c) with a K i of 7.33 AE 1.76 mM, and 7bOHC inhibited dehydrogenation of corticosterone with a K i of 0.91 AE 0.05 mM. (Fig. 2d). In both cases, the nature of inhibition was competitive, indicated by the regression lines of the Dixon Plots intercepting above the x-axis. Table 1 Effect of disruption of Hsd1b1 or H6pdh on 7-oxysterol concentrations in hepatic microsomes.

11bHSD1 in murine hepatic microsomes
Both oxidation and reduction of glucocorticoids were detected using microsomal 11bHSD1 with reduction being the preferred direction (lower K m and higher V max , Table 2). In contrast, only oxidation of 7bOHC could be measured, forming 7KC at the same rate in the presence of either NAD+ or NADP+ (e.g. 1.25 AE 0.2 vs. 1.35 AE 0.4 pmol/mg/min respectively; 20 mM substrate, n = 3). This reaction was dependent on the presence of 11bHSD1, as 7bOHC was not converted to 7KC by hepatic microsomes from Hsd11b1 À/À mice, with either cofactor. Again, 7-oxysterols demonstrated poorer affinity than glucocorticoids for 11bHSD1. The K m for oxidation of 7bOHC was approximately three orders of magnitude higher than that for glucocorticoids (Table 2), although the maximal velocities achieved were similar for glucocorticoids and 7-oxysterols. Reduction of 7KC could not be demonstrated, even following the addition of the permeabilisation agent, alameticine, or use of NADH as an alternative cofactor [32]. 7KC weakly inhibited reduction of 11-DHC with an IC 50 of 19.4 AE 1.2 mM (Fig. 2e) and further kinetic analysis was not performed. 7bOHC inhibited oxidation with an IC 50 of 2.2 AE 0.4 mM (Fig. 2f).

Supplementation of cellular content of cholesterol and 7oxysterol impedes reduction of glucocorticoids by 11bHSD1
The effect of cholesterol loading was assessed on the equilibrium of 11bHSD1 stably transfected into HEK293 cells. 7KC (19.4 AE 1.08 pmol/mg) and 7bOHC (4.37 AE 1.90 pmol/mg) were present in cells treated with vehicle. Cholesterol loading significantly (p < 0.05) increased the levels of 7KC (39.48 AE 3.01 pmol/mg) and 7bOHC (17.6 AE 2.4 pmol/mg), associated with a slower velocity of reduction of glucocorticoids by 11bHSD1 compared with vehicletreated cells (Fig. 2g).

3D in silico modeling
7-Oxysterols have not been co-crystallised with 11bHSD1. Thus, to establish the spatial orientation of the oxygenated residues at the C7 position, the structure of the closely related 7aHSD in complex with 7-oxoglycochenodeoxycholic acid (1FMC) was used. Tyrosine residues in the active sites of 7aHSD (1FMC) and m11bHSD1 (1Y5R) could be superimposed, allowing the 7ahydroxyl group of 1FMC ligand to overlay the 11b-hydroxyl group of corticosterone docked within 1Y5R. Thus, the 3D structure of 7aOHC was created to resemble that of 7-oxoglycochenodeoxycholic acid, allowing the positions of the 7b-hydroxyl and 7-keto groups of 7bOHC and 7KC respectively to be orientated. 7-Oxysterols were docked into the active site of 1Y5R and 3D representations shown in Fig. 1e. The A-ring of 7-oxysterols (as opposed to the D-ring of glucocorticoids) was orientated toward the interior of the 11bHSD1 active site. Interactions between 7bOHC and Ser170 and Tyr183 of the catalytic tetrad had the shortest bond distances (2.7, 2.8 Å respectively), followed by those of 7KC (3.2, 3.3 Å ) and then 7aOHC (4.5, 4.8 Å ; for comparison corticosterone 3.1, 2.8 Å [30]; 11-DHC 2.8, 2.6 Å respectively) When 7bOHC and 7KC were docked, the Tyr183 residue was 5.1 Å from the nicotinamide C4 and Lys187 was 3.2 Å from the hydroxyl group on the cofactor. When 7aOHC was docked, the Tyr183 residue was 4.20 Å from the nicotinamide C4 and Lys187 was 3.2 Å from the hydroxyl group on the cofactor.

Discussion
These data demonstrate that reduction of 7KC to 7bOHC is the preferred direction of metabolism of 7-oxtserols by 11bHSD1 in vivo in mouse. Metabolism of 7-oxysterols (at least dehydrogenation) was not detected in microsomes of 11bHSD1 null mice, supporting the notion that it is the only enzyme catalyzing this reaction. While 7-oxysterols were competitive inhibitors of metabolism of glucocorticoids by 11bHSD1, it is unlikely that in health [34] they will be sufficiently potent to exert this effect. Inhibition may become important in hyperlipidaemia [10], or at sites where oxysterols accumulate, such as in adipose and atherosclerotic lesions.
Structural modeling of the murine protein confirmed the potential for interactions of 7-oxysterols with the catalytic tetrad of the enzymatic active site. 7-Oxygenated substrates, in contrast to steroids, interact with 11bHSD1 with their A-ring orientated toward the interior of the binding pocket, in agreement with models in other species [21,30,35]. The higher K m values describing metabolism of 7oxysterols compared with glucocorticoids, however, indicated they were poorer affinity substrates. Circulating concentrations of 7oxysterols in the mouse were comparable in magnitude to those in other species [34] and increased in hyperlipidaemia [10]. However, it is likely metabolism by 11bHSD1 would not proceed at maximal velocity in the presence of the endogenous concentrations reported here or by others [34,36].
While disruption of 11bHSD1 only tended to alter circulating 7oxysterol levels subtly [29], it substantially lowered the levels in hepatic sub-cellular fractions. Oxysterols can be synthesized from spontaneous oxidation of cholesterol and are derived in large part from dietary sources [10]. Therefore the reduction in absolute levels may arise because Hsd11b1 À/À mice have an improved metabolic profile with lower circulating cholesterol concentrations [37], and thus less precursor for auto-oxidation. The specific contribution of 11bHSD1 to the proportions of 7-keto and hydroxy oxysterols was revealed in the increase in the ratio of 7KC/7bOHC ratio in plasma and microsomes, following targeted disruption of Hsd11b1, suggesting that 11bHSD1 catalyses reduction of 7KC in vivo. This corroborates previous studies in rats in which hepatic 7KC accumulated following administration of the non-specific 11bHSD inhibitor carbenoxolone [9]. Lack of NADPH supply due to genetic disruption of H6pdh again increased the 7KC/7bOHC ratio, confirming in vitro findings [38] that H6PDH promotes catalysis of 7KC to form 7bOHC in vivo, similarly to glucocorticoids. Indeed, H6PDH appeared to be the only source of co-factor, as double knockout of H6pdh and Hsd11b1, yielded the same ratio of 7oxysterols, as with disruption of H6pdh alone. 11bHSD1 may therefore play a similar role in regulating actions of 7-oxysterols in vivo as it does glucocorticoids. The importance of metabolism of glucocorticoids by 11bHSD1 is readily apparent since the 11-keto steroid is inert and the hydroxy form is active. However, distinct biological roles for 7KC and 7bOHC are not established and a target receptor has not been defined, although there are a number of reports of subtle differences in their actions (e.g. 7bOHC has a greater ability than 7KC to induce apoptosis in human umbilical vein endothelial cells [39]). However 7-oxysterols can be subject to further metabolism and recent reports suggest that the 25-and 27-hydroxy metabolites of 7a-and 7bOHC play potential roles in regulating the immune response via the novel Gprotein coupled receptor, EB12 [40,41]. Interestingly there is one report showing that 7KC but not 7bOHC limits SCAP exit from the ER within cells [42], which further prevents excess synthesis of cholesterol. Hence, it follows that the increased proportion of 7KC to 7bOHC upon inhibition of 11bHSD1 in vivo may exert a brake on cholesterol synthesis. Other oxysterols modulate nuclear hormone signaling pathways, but the possibility of activation of LXR, at least, by 7-oxysterols has largely dismissed [20].
Work with cells stably transfected to express human 11bHSD1 or with adipocytes [20] has shown that 7-oxysterols (in keeping with other 7-hydroxylated substrates [7]) may compete differentially with glucocorticoids for metabolism by 11bHSD1 and thus modulate glucocorticoid action. Inhibition appears cell-type specific, potentially explained by differential metabolism, accumulation or export of oxysterols [10]; adipocytes and macrophages sequester oxysterols readily [10] whereas macrophages export 7KC and other oxysterols via the ABCG1 transporter [43]. Balá zs et al. did not detect any inhibition of human 11bHSD1 reductase activity in lysates or HEK293 cells by 7KC or 7bOHC, but showed an inhibition of 11bHSD1-reductase activity by 7KC (IC 50 8.1 AE 0.9 mM) in differentiated THP-1 macrophages [21]. Inhibition of glucocorticoid metabolism by co-incubation with 7-oxysterols was investigated here using three models of murine 11bHSD1, in all of which reduction of glucocorticoids was favored. Our data concur with the proposal that 7-oxysterols compete with glucocorticoids for metabolism, with 7KC being consistently less effective at inhibiting 11-DHC reduction by isolated enzyme in vitro, than 7bOHC was at preventing oxidation. Taking into account the IC 50 values, inhibition Table 2 Kinetic parameters describing metabolism of 7-oxysterols and glucocorticoids by murine 11b-hydroxysteroid dehydrogenase 1 (11bHSD1 Velocities of metabolism of substrates by murine recombinant or microsomal 11bHSD1 were assessed and kinetic parameters (K m , V max and V max /K m , true or apparent) assigned following Lineweaver-Burke transformation of data fitted to Michaelis-Menten kinetics. The velocities were quantified; for reduction of 11-dehydrocorticosterone or 7-ketocholesterol in the presence of NADPH or oxidation of corticosterone or 7b-hydroxycholesterol in the presence of NADP + . Data are mean AE SEM, obtained from at least three independent experiments. V max expressed as pmol/mg/min. V max /K m expressed as L/mg/min Â 10 À6 .
of glucocorticoid metabolism is unlikely to be important in health. However, at concentrations in the low micromolar range, as seen in atherosclerosis [44,45], 7bOHC or 7KC may compete for oxidation preventing glucocorticoid inactivation or reduction, respectively. 7bOHC is highly abundant in fatty streaks in developing lesions [46,47] and the 7bOH/7KC ratio is increased. If 7bOHC dominates to inhibit glucocorticoid oxidation, the cells in the lesion and adjacent normal intima may become exposed to increased local glucocorticoid levels, with adverse consequences [48]. However, when endogenous 7-oxysterols were enriched secondary to cholesterol loading in cultured cells, the predominant effect was to suppress reduction of glucocorticoids, suggesting protection from excess glucocorticoid. These findings concur with reduction of 7KC being the major route of metabolism of 7-oxysterols in vivo.
In conclusion, 7KC and 7bOHC are poor affinity substrates for murine 11bHSD1 and are interconverted at a slower rate than glucocorticoids. While differences exist in the patterns of in vitro and in vivo metabolism, reduction of 7KC to 7bOHC appears the predominant reaction in vivo. Although it seems unlikely that the competition with oxysterols will determine predominant direction for glucocorticoid metabolism by 11bHSD1 in health, it may play a role in hyperlipidaemia and atherosclerosis. A greater knowledge of the actions of these 7-oxysterols is required to fully understand the consequences of inhibition or over-activity of 11bHSD1 pathway.