Mast Cells Express 11 beta-hydroxysteroid Dehydrogenase Type 1: A Role in Restraining Mast Cell Degranulation

Mast cells are key initiators of allergic, anaphylactic and inflammatory reactions, producing mediators that affect vascular permeability, angiogenesis and fibrosis. Glucocorticoid pharmacotherapy reduces mast cell number, maturation and activation but effects at physiological levels are unknown. Within cells, glucocorticoid concentration is modulated by the 11 b -hydroxysteroid dehydrogenases (11 b -HSDs). Here we show expression and activity of 11 b -HSD1, but not 11 b -HSD2, in mouse mast cells with 11 b -HSD activity only in the keto-reductase direction, regenerating active glucocorticoids (cortisol, corticosterone) from inert substrates (cortisone, 11-dehydrocorticosterone). Mast cells from 11 b -HSD1-deficient mice show ultrastructural evidence of increased activation, including piecemeal degranulation and have a reduced threshold for IgG immune complex-induced mast cell degranulation. Consistent with reduced intracellular glucocorticoid action in mast cells, levels of carboxypeptidase A3 mRNA, a glucocorticoid-inducible mast cell-specific transcript, are lower in peritoneal cells from 11 b -HSD1-deficient than control mice. These findings suggest that 11 b -HSD1-generated glucocorticoids may tonically restrain mast cell degranulation, potentially influencing allergic, anaphylactic and inflammatory responses.


Introduction
Mast cells play a central (typically initial) role in inflammatory and allergic reactions.They guard against bacterial pathogens and participate in tissue repair by producing mediators that promote vascular permeability, angiogenesis and fibrosis.Mast cells accumulate in chronically inflamed tissues in humans and in mice, and have consistently been observed in high numbers in human rheumatoid arthritis (reviewed, [1]), in Crohn's disease (reviewed, [2]) and in the bronchus of asthmatic patients [3,4].Mast cells contain preformed tumour necrosis factor (TNF)-a in granules and can rapidly produce large amounts of both TNF-a and interleukin (IL)-1 [5] as well as other mediators, including histamine, eicosanoids (particularly prostaglandin D 2 ) and vascular endothelial growth factor, which contribute to oedema, inflammation, hyperplasia and neovascularisation.
Glucocorticoids reduce mast cell number, maturation and activation [6,7,8,9], contributing to the potent anti-allergic and anti-inflammatory effects of these steroids.Blood glucocorticoid levels depend upon activity of the hypothalamic-pituitary-adrenal axis.However, intracellular glucocorticoid concentrations can differ greatly from blood levels due to the action of 11bhydroxysteroid dehydrogenase (11b-HSD), an enzyme that interconverts active glucocorticoids (cortisol in humans, corticosterone in rodents) and intrinsically inert 11-keto metabolites (cortisone, 11-dehydrocorticosterone). Two isozymes exist; 11b-HSD1 and 11b-HSD2.Whereas 11b-HSD2 inactivates glucocorticoids and is largely restricted to mineralocorticoid target tissues in the adult, 11b-HSD1 catalyses the opposite reaction in vivo, predominantly reactivating glucocorticoids by converting inert 11keto-glucocorticoids into active forms and shows a more widespread distribution [10].Mice homozygous for a targeted disruption of the Hsd11b1 gene that encodes 11b-HSD1 (Hsd11b1 2/2 mice) have normal blood glucocorticoid levels on the C57BL/6J strain background [11] yet have a phenotype consistent with intracellular glucocorticoid deficiency (reviewed, [12]).Thus, they exhibit more severe acute inflammation in models of myocardial infarction, arthritis, sterile peritonitis and carageenan-induced pleurisy [13,14]}, suggesting 11b-HSD1 normally exerts a restraining influence upon the early inflammatory response.In vivo, 11b-HSD1 expression is rapidly and markedly increased at sites of inflammation, including in peritoneal immune cells during sterile peritonitis [15], in colitis [16,17] and in the arthritic joint [18].11b-HSD1 is expressed in macrophages [15,19] where it performs an anti-inflammatory function, accelerating acquisition of macrophage phagocytic competence [15]).Expression has also been shown in dendritic cells [20], neutrophils [21] and lymphocytes [22], though its role in these cells remains uncharacterised.Here, we describe expression and activity of 11b-HSD1 in mast cells, classical glucocorticoidtargets in allergic and anaphylactic reactions, and demonstrate a restraining influence of 11b-HSD1 on mast cell degranulation.

Ethics Statement
All animal experimentation was conducted in strict accord with accepted standards of humane animal care under the auspices of the Animal (Scientific Procedures) Act UK 1986 following prior approval by the local University of Edinburgh ethical committee.

Animals
Mice homozygous for a targeted disruption of the Hsd11b1 gene on a C57BL/6J background (.8 backcrosses) have been described [15].Control age-matched C57BL/6J (Hsd11b1 +/+ ) mice were bred in-house.Mice were housed in groups of 2-5 per cage under controlled conditions (12 h-light/dark cycle, 21uC) with ad libitum access to water and standard rodent chow.

Generation of Anti-glucose 6-phosphate Isomerase IgG Immune Complexes
Arthritogenic K/BxN serum containing anti-glucose 6-phosphate isomerase (GPI) IgG immune complexes was generated in house from arthritic K/BxN mice (expressing both the KRN T cell receptor transgene and the MHC class II molecule A g7 ) as described [14].

Analysis of Peritoneal Mast Cells
For analysis of peritoneal mast cells, 8-12 week old male and female mice were used.Following peritoneal lavage, enriched peritoneal mast cells were prepared from freshly isolated peritoneal cells, positively selected for CD117 + mast cells using magnetic beads (Miltenyi Biotec, Surrey UK) according to the manufacturer's instructions.Generally, around 50% pure mast cell populations were achieved based on histochemical staining of cytocentrifuge preparations or flow cytometry analysis.Mast cell degranulation was determined by measuring the release of bhexosaminidase as described [25].Briefly, peritoneal cells were incubated in triplicate for 15 min at 37uC in Tyrode's Buffer and were untreated or treated with 10 mM ionomycin or diluted K/ BxN serum.Following incubation, cells were collected by centrifugation.Aliquots (in triplicate) of the supernatant were transferred to a 96-well plate.The remaining supernatant was carefully removed and cell pellets solubilized in Tyrode's buffer with the addition of 0.5% Triton X-100.Aliquots (in triplicate) of the solubilized pellet were also transferred to a 96-well plate.Next, b-hexosaminidase substrate (1 p-nitrophenyl-N-acetyl-b-D-glucosamine) was added to each well and incubated for 40 min at 37uC.Reactions were stopped by the addition of glycine and absorbance was measured at 405 nm.Total mast cell b-hexosaminidase was measured in untreated cells solubilized by the addition of Triton X-100.For each sample, degranulation was calculated as A 405 supernatant/(A 405 supernatant+A 405 pellet).The mean of the untreated samples in each group was arbitrarily set to 100, and all other values expressed relative to this.Net degranulation was calculated as the difference between treated and untreated samples (and is thus % degranulation above levels in untreated cells).In preliminary experiments, no differences in mast cell number or level of degranulation were observed between the sexes of either genotype, therefore total peritoneal cells from both sexes were used.

Transmission Electron Microscopy
Samples were fixed in 3% glutaraldehyde in 0.1M sodium cacodylate buffer, pH 7.3, for 2 h then washed 3 times (each for 10 min) in 0.1M sodium cacodylate.Specimens were post-fixed in 1% osmium tetroxide in 0.1M sodium cacodylate for 45 min, washed again (as above) then dehydrated in 50%, 70%, 90% and 100% acetone (10 min each), then twice further in 100% acetone.Samples were embedded in Araldite resin and 1 mm thick sections cut on a Reichert OMU4 ultramicrotome (Leica Microsystems UK Ltd, Milton Keynes, UK), stained with toluidine blue and viewed in a light microscope to select suitable areas for investigation.Ultrathin (60 nm thick) sections were cut from selected areas, stained in uranyl acetate and lead citrate then viewed in a Phillips CM120 Transmission electron microscope (FEI UK Ltd, Cambridge, UK).Images were captured using a Gatan Orius CCD camera (Gatan UK, Oxon, UK).

Statistics
Student's t-test was used for comparisons between genotypes.Significance was set at p,0.05.Values are means 6 SEM.

11b-HSD1 is Expressed in Mast Cells
With 11-dehydrocorticosterone as substrate, BMD-mast cells show 11b-keto-reductase activity (Figure 1A; 16.961.1 pmol corticosterone/h/10 6 cells).In contrast, BMD-mast cells show negligible 11b-dehydrogenase activity with corticosterone as substrate (Figure 1A) showing the absence of 11b-HSD2 and indicating 11b-HSD1 is a predominant 11b-reductase in mast cells.Consistent with this, 11b-HSD1 mRNA is present and 11b-HSD2 mRNA absent in mast cells (Figure 1B).BMD-mast cells from 11b-HSD1-deficient mice lack 11b-HSD1 activity (Figure 1C), confirming the activity is due to 11b-HSD1.11b-HSD1 mRNA is transcribed from 3 distinct promoters [27,28].Transcription of the Hsd11b1 gene in BMD-mast cells initiates at the upstream P1 promoter, with little or no transcription from the downstream P2 and P3 promoters (Figure 1D).Importantly, this contrasts with BMD-macrophages which exploit the P2 promoter (Figure 1D), suggesting that these P1initiated transcripts are in mast cells and not in contaminating macrophages.CD117 or c-Kit, is a marker for mature mast cells.Flow cytometric staining of lavaged peritoneal cells (of which ,3% are mast cells; see Figure S1 for definition of the CD117 + population) with 11b-HSD1 antibody demonstrates immunoreactive protein in CD117 + cells (Figure 1E), confirming 11b-HSD1 expression in mast cells in vivo as well as in vitro.
To determine whether mast cell contents are affected by 11b-HSD1 deficiency, key glucocorticoid-regulated transcripts were examined.Levels of mRNA encoding carboxypeptidase A3,  generally considered a mast cell-specific product [29,30] that is glucocorticoid-inducible [8], are lower in peritoneal cells from Hsd11b1 2/2 mice, compared to controls despite increased peritoneal mast cell number, consistent with reduced intracellular glucocorticoid action in 11b-HSD1-deficient mast cells (Figure 3A).Moreover, although not mast cell-specific, levels of annexin I mRNA (formerly called lipocortin), a classic glucocorticoidinducible gene [31], are also reduced in peritoneal cells from Hsd11b1 2/2 mice (Figure 3B).
To investigate whether the characteristic granule morphology of mast cells is affected by 11b-HSD1-deficiency, transmission electron microscopy was carried out on an enriched population of peritoneal mast cells (,50% mast cells).Total granule number per cell did not differ between genotypes Hsd11b1 +/+ 55.6614.7 (n = 41) vs Hsd11b1 2/2 61.6619.4(n = 49).However, whereas mast cells from control mice predominantly contain homogeneously electron dense granules (Figure 4A), those from Hsd11b1 2/2 mice are more heterogeneous and contain a greater proportion of lighter granules (Figure 4A, B) as well as enlarged partially filled and empty vesicles, characteristic of piecemeal degranulation [32], a slower form of mast cell degranulation than anaphylactic degranulation, that has been associated with persistent inflammation [32].

11ß-HSD1 Deficiency Reduces the Activation Threshold of Peritoneal Mast Cells
To determine whether 11b-HSD1-deficiency functionally alters mast cells as suggested by the granule morphology, we measured mast cell degranulation by assaying b-hexosaminidase release.Maximal degranulation induced by 10 mM ionomycin was similar in peritoneal cells from Hsd11b1 2/2 and control mice (Figure 5A).A similar maximal level of degranulation was observed following 15 min incubation with a 1:2 dilution of K/BxN serum (containing anti-glucose 6-phosphate isomerase IgG immune complexes, a potent trigger of mast cell degranulation) into culture medium (Figure 5B).However, 15 min incubation with a lower concentration of K/BxN serum (diluted 1:8 in medium) induced significantly more degranulation of peritoneal mast cells from Hsd11b1 2/2 than control mice (Figure 5B).Furthermore, when an enriched population of peritoneal CD117 + cells (,50% mast cells) was incubated for 21 h with K/BxN serum (diluted 1:8 in medium), microscopic examination revealed extensive degranulation of 11b-HSD1-deficient mast cells, whereas controls showed little or no degranulation (Figure 5C).11b-HSD1-deficiency therefore lowers the threshold for degranulation of resident mast cells following IgG immune complex activation.

Discussion
Pharmacological levels of glucocorticoids have potent effects on mast cells, but effects of endogenous glucocorticoids at physiological levels are unknown.Mast cells from Hsd11b1 2/2 mice have a phenotype consistent with relative glucocorticoid deficiency, including reduced expression of glucocorticoid-sensitive mast cell-specific carboxypeptidase A3, suggesting 11b-HSD1-mediated glucocorticoid amplification tonically suppresses mast cell responses.
11b-HSD1 is expressed in both BMD-and in peritoneal mast cells.At around 10-20 pmol/h/10 6 cells, the level of ketoreductase activity in BMD-mast cells is considerably higher than that reported for T lymphocytes (,0.1 pmol/h/10 6 cells; [22]) and comparable with levels in immature BMD-dendritic cells (7 pmol/h/10 6 cells; [20]) and BMD-macrophages (12 pmol/h/ 10 6 cells).Thus, the measured activity is likely to be due to mast cells and not to contamination with macrophages or another cell type expressing 11b-HSD1, present at ,2% of the population.This is supported by the use of the P1 promoter in mast cells, which contrasts with most other tissues and cell types, including macrophages, that use the P2 promoter and suggests a distinct regulation in mast cells.
Whilst the maximum response to degranulating stimuli was similar in mast cells from Hsd11b1 2/2 and control mice (suggesting Hsd11b1 2/2 mice have normal levels of b-hexosaminidase in stored granules), 11b-HSD1-deficient cells were sensitive to a low dose of K/BxN serum that was ineffective in control cells, suggesting that it is the threshold for activation that is reduced, rather than the total number of stored granules, which was unchanged in Hsd11b1 2/2 mice.Ultrastructural analysis of peritoneal mast cells supported this ''trigger-happy'' phenotype of Hsd11b1 2/2 mast cells and indicated piecemeal degranulation, a form of non-anaphylactic granule release observed in mast cells in situ, possibly representing up-regulated constitutive secretion [32].This increased sensitivity is likely to reflect a difference in differentiation, maturation or activation of mast cells, as it is preserved in the absence of 11b-HSD1 substrate, ex vivo.Density of FccR, key for IgG-mediated signalling, is normal on 11b-HSD1deficient mast cells, though it is possible that the threshold for FccR activation and signalling is reduced.Mast cell number was increased by ,34% in the peritoneum of Hsd11b1 2/2 mice.Mast cell number is reduced by pharmacological glucocorticoid treatment in vivo, though whether normal physiological glucocorticoid concentrations have a similar effect in vivo has not been reported.However, given that Hsd11b1 2/2 mice have normal plasma corticosterone levels on this genetic background [33] and we have previously shown normal responses of Hsd11b1 2/2 immune cells to exogenous corticosterone [15], the increase in the number of mast cells in the peritoneum of Hsd11b1 2/2 mice suggests that 11b-HSD1-mediated regeneration of endogenous intracellular glucocorticoids limits proliferation of at least some mast cell populations.In this respect, it is interesting that activation products of mast cells are chemoattractants for their progenitors [34,35], suggesting that the basal activation observed in Hsd11b1 2/2 mast cells may underlie the increase in peritoneal mast cell number.
Immune activation is a potent stimulus to the hypothalamicpituitary-adrenal (HPA) axis and removal of endogenous glucocorticoids or blockade of their actions exacerbates immune and inflammatory disease in humans and in animal models (reviewed, [36]).Our data suggest that by providing a ''brake'' to spontaneous mast cell degranulation, 11b-HSD1-mediated amplification of endogenous glucocorticoid action within mast cells may influence allergic and anaphylactic reactions, in which mast cells are central, and for which glucocorticoid phamacotherapy remains a highly effective treatment.Experimental testing of this hypothesis in appropriate in vivo models of anaphylaxis and allergy will be important in the future.These will be important considerations in the clinical introduction of selective 11b-HSD1 inhibitors for treatment of metabolic disease, a side-effect of which could be mast cell hyper-responsiveness.