11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1, encoded by Hsd11b1) is an enzyme that predominantly converts inactive glucocorticoids (cortisone in human and most mammals, 11dehydro-corticosterone in mice and rats) into their active forms (cortisol and corticosterone, respectively). Thus 11β-HSD1 amplifies intracellular levels of glucocorticoids. Studies in globally 11β-HSD1 deficient mice have revealed changes in glucocorticoid-regulated physiological and pathological processes, including metabolism, aging, arthritis and angiogenesis. The function of macrophages, which play an important role in inflammation, is also altered. For example, 11β-HSD1 deficiency in macrophages causes a delay in their acquisition of phagocytic capacity. To dissect the role of macrophage 11β-HSD1 in angiogenesis, arthritis and obesity, both in vitro macrophage stimulation and in vivo functional assays in macrophage-specific 11β-HSD1 knockout mice, were conducted. Thioglycollate-elicited peritoneal macrophages from globally 11β-HSD1 deficient and control C57BL/6 mice were used for in vitro studies. In M1/M2 macrophage polarisation experiments, 11β-HSD1 deficient macrophages showed increased expression of mRNAs encoding pro-inflammatory factors upon lipopolysaccharide and interferon-ϒ treatment and decreased expression of pro-resolution genes with interleukin-4 stimulation. However, at cytokine or protein levels, there was little difference between the genotypes except for decrease IL12 p40 levels in 11β-HSD1 deficient macrophages. Hypoxic stress failed to show differences between genotypes in hypoxia-regulated gene expression. These data do not support a strong role for macrophage 11β-HSD1 in inflammation regulation, nor in response to hypoxia, at least when measured in vitro. The discrepancy between transcriptional and translational responses is currently unexplained, but may reflect altered posttranscriptional activity. To investigate the role of macrophage 11β-HSD1 in vivo, macrophage-specific Hsd11b1 knockout mice, LysM-Cre Hsd11b1 flox/flox (MKO) mice and Hsd11b1flox/flox littermate controls were generated. In MKO mice, 11β-HSD1 protein levels and enzyme activity were reduced by >80% in resident peritoneal macrophages. However, 11β-HSD1 protein and enzyme activity levels were unchanged or only modestly reduced in thioglycocollate-elicited peritoneal neutrophils, monocytes/macrophages, or in bone marrow-derived macrophages, despite >80% decrease in Hsd11b1 mRNA levels in these cells. A relatively long half-life of 11β-HSD1 protein compared to that of circulating myeloid cells may underlie this mismatch between transcriptional and translational expression. Furthermore, following 12 days of inflammatory arthritis induced by K/BxN serum transfer, the reduction in 11β-HSD1 protein levels in circulating neutrophils of MKO mice is consistently around 50%, which corroborates the above explanation. MKO mice and littermate controls were subjected to inflammatory models which may involve resident macrophages. First, to address the role of 11β-HSD1 in macrophages in angiogenesis, sponge implants were inserted subcutaneously into the flanks of adult male mice and harvested after 21 days. Chalkley counting on hematoxylin and eosin stained sponge sections showed significantly increased angiogenesis in MKO mice (scores: 5.2±1.0 versus 4.3±0.7; p<0.05, n=9-11). Cdh5 expression (encoding VE-cadherin, a marker of endothelial cells) was higher in sponges from MKO mice (relative expression: 1.5±0.9 versus 0.8±0.6; p<0.05), as was Il1b (encoding IL-1 beta, a marker of inflammation, relative expression: 6.5±6.4 versus 1.5±0.9; p<0.05). Vegfa mRNA (encoding vascular endothelial growth factor alpha) was unchanged, with a trend for higher Angpt1 (encoding angiopoietin 1, p=0.09) expression levels in the MKO group. These results suggest that lack of 11β- HSD1 in resident macrophages increases their pro-angiogenic activity, independently of VEGF-. The K/BxN serum transfer model of arthritis was used to investigate the role of macrophage 11β-HSD1 in arthritis. Adult male MKO and control mice received a single i.p. injection of 125μl K/BxN serum per mouse, followed by 21 days of clinical scoring to assess joint inflammation. The onset of inflammation (d1-8) was similar between MKO and control mice, but MKO mice exhibited greater clinical inflammation scores in the resolution phase of arthritis (d13-21; area-under-the-curve: 86.6±14.7 versus 60.1±13.4; p<0.005), indistinguishable from globally 11β-HSD1- deficient mice. Hematoxylin and eosin staining revealed pronounced fibroplasia predominantly in the supporting mesenchyme associated with the tenosynovium, with new bone and blood vessel formation. These results suggest that macrophage 11β-HSD1 deficiency is fully accountable for the worse arthritis resolution phenotype in the globally 11β-HSD1 deficient mice, but not the earlier onset of inflammation with global 11β-HSD1 deficiency. Macrophage activation states are closely linked with adipose insulin sensitivity. Globally 11β-HSD1 deficient mice are protected from high fat diet induced insulin resistance and adipose tissue hypoxia and fibrosis. To study the effect of macrophage 11β-HSD1 deficiency on insulin sensitivity, adult male MKO and control mice were given a 14 week high fat diet, which typically causes insulin resistance in control but not globally 11β-HSD1 KO mice. The level of fibrosis in subcutaneous adipose tissues was reduced as indicated by quantification of picrosirius red staining of collagen, though GTT data so far does not support protection from insulin resistance in MKO mice. In summary, in vitro macrophage polarisation experiments do not support a strong role of 11β-HSD in M1/M2 macrophage polarisations or response to hypoxia. However, MKO mice reveal, for the first time, an important in vivo role of macrophage 11β-HSD1 to promote angiogenesis and facilitate resolution of K/BxN serum transfer induced arthritis. Modulation of fibrosis is context dependent. Reduced adipose fibrosis may be one of the mechanisms that improve insulin sensitivity. Meanwhile, these findings suggest caution regarding the potential side effects of 11β-HSD1 inhibitors in treating metabolic disease in patients with inflammation-related co-morbidities, such as rheumatoid arthritis.

Glucocorticoids profoundly influence the immune system and pharmacological doses exert potent anti-inflammatory actions. During inflammation, glucocorticoids limit oedema and influence cell trafficking, differentiation programmes and gene transcription in glucocorticoid-sensitive leukocytes. Within cells, glucocorticoid action is modulated by a pre-receptor mechanism; glucocorticoid metabolism by the enzyme 11β- hydroxysteroid dehydrogenase (11β-HSD). Two 11β-HSD isozymes exist: 11β-HSD1, which catalyses amplification of glucocorticoid levels in intact cells by oxo-reduction of intrinsically inert cortisone (11-dehydrocorticosterone in rodents) into active cortisol (corticosterone in rodents) and 11β-HSD2, which performs the opposite reaction. Thus, amplification of intracellular glucocorticoid levels by 11β-HSD1 may represent an endogenous anti-inflammatory mechanism. This hypothesis has been tested in Hsd11b1-/- mice (homozygous for a targeted disruption in the Hsd11b1 gene, encoding 11β-HSD1), using carageenan-induced pleurisy and experimental model of arthritis induced by injection of arthritogenic antibodies. In both models, Hsd11b1-/- mice showed more severe acute inflammation than control mice. During carrageenan-induced pleurisy, Hsd11b1-/- mice recruited more inflammatory cells to the pleural cavity than congenic controls, with a greater proportion of viable cells, at the onset and peak of pleurisy, suggesting a worse inflammatory response. Histological examination suggested impaired resolution of inflammation in Hsd11b1-/- mice with persistence of inflammation in the visceral pleura, activation of lymphoid aggregates, and uniquely in Hsd11b1-/- mice, formation of fibrous adhesions between lung lobes 48h after initiation of pleurisy. During experimental arthritis induced by injection of serum from arthritic K/BxN mice, clinical signs of inflammation occurred earlier in Hsd11b1-/- mice and were slower to resolve than in control mice. Histological assessment of the acute phase (2d) of arthritis showed no difference in joint pathology between genotypes, despite greater oedema and higher clinical scores in the Hsd11b1-/- mice. However, when the inflammation had resolved (21d following injection of serum), compared to control mice, Hsd11b1-/- mice showed more severe exostosis, intense periarticular inflammation, more collagen deposition and uniquely, ganglion cyst formation. At 21d, whereas basal (morning) plasma corticosterone levels were normal in control mice, they remained elevated in Hsd11b1-/- mice, suggesting ongoing inflammation and persistent activation of the hypothalamic-pituitary-adrenal axis. Mast cells are critical in the initiation of an inflammatory response and are essential in this model of arthritis. Mast cells expressed 11β-HSD1 (but not 11β-HSD2) mRNA and activity. Although mast cell number did not differ in joints or peritoneum of Hsd11b1-/- mice, 11-HSD1-deficient mast cells had a lower threshold for degranulation induced by K/BxN arthritogenic serum. As well as implicating a role for mast cell 11β-HSD1 in limiting initial inflammation in arthritis, these findings also have implications for infection, allergy and tolerance. Collectively, these data suggest that 11β-HSD1 deficiency worsens acute inflammation and results in slower resolution. Therefore, amplification of intracellular glucocorticoids levels, by 11β-HSD1, may represent an important mechanism to limit the acute inflammatory response and programme its subsequent resolution. Increasing leukocyte 11β-HSD1 or local delivery of substrate affords a novel approach for anti-inflammatory therapy.

Diabetes Mellitus is characterized by high blood sugar and is caused by resistance to (type 2) or insufficiency of (type 1) the pancreatic β-cell hormone insulin. Most commonly, type 2 diabetes is associated with obesity whereas type 1 diabetes is largely a result of immune-mediated destruction of the β-cell. One rare but significant cause of type 2 diabetes is excess blood glucocorticoid levels (Cushing’s syndrome). High circulating glucocorticoids potently induce metabolic disorders including peripheral insulin resistance in key metabolic tissues (muscle, liver and fat) as well as directly suppressing β-cell function and can precipitate type 2 diabetes. However, in common forms of metabolic syndrome (visceral obesity, type 2 diabetes, increased cardiovascular disease risk) it appears that amplification of local tissue glucocorticoid action by increased levels of the intracellular enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), particularly in adipose tissue, is a key driver of the adverse metabolic phenotype rather than altered circulating glucocorticoid levels. 11β-HSD1 is also elevated in pancreatic islets from obese rodents. This thesis aimed to determine the role of 11β-HSD1 in pancreatic islets (β-cells) under normal conditions and its potential pathogenic role in the development of diabetes. We first determined that 11β-HSD1 acted primarily as a reductase (amplifying glucocorticoid action) in pancreatic islets. We then determined that islet 11β-HSD1 transcription is under the control of the promoters that express in other tissues like liver. Islet 11β-HSD1 is significantly regulated by factors relevant to the diabetic state; high glucose and insulin suppressed whereas fatty acids and TNFα increased 11β-HSD1 activity. To test whether the high islet 11β-HSD1 found in obese rodents was directly diabetogenic, we generated transgenic mice specifically overexpressing β-cell 11β-HSD1 under the mouse insulin promoter (MIP-HSD1 mice) in a mouse strain prone to develop β-cell failure when subjected to diabetic challenge (eg. chronic high fat feeding). Unexpectedly, MIP-HSD1tg/+ mice (expressing ~2 fold elevated 11β-HSD1 activity) exhibited markedly improved β-cell insulin secretory responses, whereas MIP-HSD1tg/tg mice had partially impaired β-cell insulin secretory function in vivo and in vitro. Moreover, MIP-HSD1tg/+ mice completely resisted the mild hyperglycaemia induced by multiple-low doses of the β-cell toxin streptozotocin (40mg/kg i.p. for 5 days) and partially resisted the profound hyperglycaemia induced by a single high dose of streptozotocin (180mg/kg). Notably, MIP-HSD1tg/+ mice exhibited lower macrophage infiltration (MAC-2) and higher T-regulatory cell (Foxp3) infiltration after these challenges with evidence of increased insulin-positive cells and maintenance of normal levels of proliferation-competent β-cells. Overall, MIP-HSD1tg/tg exhibited a partial protection from the streptozotocin challenge. Modestly increased 11β-HSD1 expression in β-cells unexpectedly supports compensatory insulin hypersecretion preventing type 2 diabetes and protects β-cells from inflammatory mediated damage in the setting of type 1 diabetes. Above a protective threshold, elevated β-cell 11β-HSD1 may result in β-cell dysfunction and diabetes. These findings have important implications for the currently advocated therapeutic strategies to inhibit 11β-HSD1 in the context of obesity and diabetes.

7-Oxysterols constitute the major component (40%) of oxidized low-density lipoprotein (oxLDL). They arise in the body via auto-oxidation of cholesterol and are known to induce endothelial dysfunction, oxidative stress and apoptosis in the vascular wall, prior to development of atherosclerosis. A novel pathway has been described for hepatic inter-conversion of 7-ketocholesterol (7-KC) and 7β -hydroxycholesterol (7β OHC) by the enzyme 11β-hydroxysteroid dehydrogenase type-1 (11β HSD1), better known for metabolizing glucocorticoids. Inhibition of 11βHSD1 is atheroprotective and the potential underlying mechanism for this may involve altered metabolism and actions of glucocorticoids. However, alterations in the metabolism of 7-oxysterols may also play an important role in this atheroprotective effect. The work described here addresses the hypotheses that (i) 7-oxysterols are substrates for murine 11βHSD1; (ii) inhibition of 11β HSD1 may abolish cellular metabolism of 7-oxysterols; (iii) this route of metabolism may modulate the actions of 7-oxysterols and glucocorticoids on murine vascular physiology. Murine 11β HSD1 inter-converted 7-oxysterols (Km=327.6±98ìM, Vmax=0.01±0.001pmol/ìg/min) but the regulation of reaction direction is different from that for glucocorticoids. Predominant dehydrogenation of 7β OHC to 7-KC was quantified in several models (recombinant protein, cultured cells stably transfected with 11β HSD1), in which predominant reduction of glucocorticoids was measured. Furthermore, in murine hepatic microsomes, dehydrogenation of 7β OHC occurred exclusively. In aortic rings in culture, however, both reduction and dehydrogenation of 7-oxysterols were evident. 7-Oxysterols and glucocorticoid substrates competed for metabolism by 11β HSD1, with 7β OHC inhibiting dehydrogenation of glucocorticoids (Ki=908±53nM). The circulating concentrations of 7-oxysterols in the plasma of C57Bl6 and 11β HSD1-/- mice were in the ìM range (0.02 – 0.13ìM). The disruption of 11β HSD1 has resulted in increased ratios of 7-KC and 7β OHC over total plasma cholesterol levels (*p<0.05). This finding suggested that 11β HSD1 is involved in metabolizing and determining the plasma levels of 7-KC and 7β OHC. To assess the consequences of these alterations for vascular function, studies were undertaken in aortic rings. Prolonged incubation with 7-oxysterols (20-25 ìM) showed a tendency to attenuate noradrenaline-mediated contractions of C57Bl6 aortae, but had no effect on contractions in response to 5-hydroxytryptamine or KCl. Similarly, endothelium-dependent and -independent relaxations of murine aortae were unaltered after exposure to 7-oxysterols. Thus in the mouse, 11β HSD1 may influence the balance of circulating and cellular 7-oxysterols which may have consequential effects on glucocorticoid action. Although this work suggests that concentrations present in murine tissues are unlikely to cause vascular dysfunction, they may influence further cellular events as yet undescribed. Under pathological conditions where high concentrations of 7-oxysterols occur, 11β HSD1 may influence the extracellular-transport and delivery of 7-KC and 7β OHC to the plaque. This work therefore proposes that inhibition of metabolism of 7-oxysterols by 11β HSD1 inhibitors, may contribute to the atheroprotective effects of these drugs.

Alterations in glucocorticoid (GC) biosynthesis and metabolism are associated with a variety of pathophysiological disorders including cholestasis, diabetes and other metabolic disorders. Bile acids (BA) are also important modulators of metabolic functions and regulate cholesterol, triglyceride and glucose homeostasis as well as being critical for dietary fat digestion, enterohepatic function, and postprandial thermogenesis. In intact cells and in vivo, the 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) enzyme converts inactive GC precursors (cortisone in humans, and 11-dehydrocorticosterone in mice and rats) into their active forms (cortisol and corticosterone, respectively) thereby amplifying local intracellular GC levels. Interconversion by 11β-HSD1 of other sterols has also been described. These include conversions of 7keto-cholesterol to 7β-hydroxycholesterol, 7-oxodehydroepiandrosterone (7-oxo-DHEA) to 7α-hydroxy- and 7β-hydroxy DHEA, 7- oxo-lithocholic acid (LCA, a bile acid; BA) to chenodeoxycholic acid (CDCA, a 7α- hydroxylated BA) and ursodeoxycholic acid (UDCA, a 7β-hydroxylated BA) in human liver microsomes. In the liver, BA inhibit 11β-HSD1 but whether 11β-HSD1 regulates BA homeostasis is unclear. Evidence of molecular regulation of the enterohepatic recycling of bile acids by liver glucocorticoid receptor (GR) in mice does suggest a role for 11β-HSD1. It was therefore hypothesised that disruption of 11β-HSD1 expression in mice would impair BA recycling and might affect the relative concentrations of BA within the enterohepatic circuit. The primary objective of the current work was to investigate the impact of altered 11β-HSD1 on BA homeostasis. This was achieved using genetically modified mouse models with altered 11β-HSD1 expression, either globally or restricted to hepatocytes. BA are stored in the gall bladder and are released postprandially, to aid digestion. It was hypothesised that 11β-HSD1 deficiency might the affect the process of postprandial gall bladder emptying/refilling. Mice with global 11β-HSD1 knockout (Hsd11b1-/-) and age-matched control mice (C57Bl/6) were either fasted for 4h and culled or fasted for 4h and re-fed for another 4h before culling. Their response to fasting and re-feeding was assessed with specific focus on organs associated with BA recycling in the enterohepatic circuit (liver, gall bladder, serum and small intestine). Gall bladders of fasted Hsd11b1-/- and C57Bl/6 mice had similar volumes of bile but in fasted Hsd11b1-/- mice, BA concentrations were higher in serum and liver. As expected, re-feeding caused gall bladder emptying in C57Bl/6 mice with consequent increased serum and liver bile acid concentrations. In Hsd11b1-/- mice, the gall bladder did not empty and serum and liver BA concentrations were similar to the fasted state. To explore possible reasons for this, levels of mRNA encoding proteins known to be involved in hepatic BA transport were quantified using real-time q-PCR. Levels of mRNA encoding NTCP/ SCL10A1/ SCL10A1, the transporter responsible for most hepatocyte BA uptake, were increased in livers of fasted Hsd11b1-/- mice whereas levels of Slc51b mRNA, encoding the OST- transporter that facilitates BA removal from liver to the systemic circulation, and levels of Mrp2 and Atp8b1/FIC1 mRNAs (both encoding proteins which transport BA from liver into gall bladder) were decreased. This suggests that in fasted Hsd11b1-/- mice, BA transporter expression is altered to increase BA influx into hepatocytes and decrease efflux, to compensate for reduced levels of liver BA. These data together imply that bile acid recycling is controlled by 11β-HSD1 activity which regulates gall bladder emptying, hepatic BA concentration and BA transporter activity to ensure continuity of BA recycling within the enterohepatic circuit compartments. These changes may also affect digestion of lipids and fat-soluble micronutrients. Because 11β-HSD1 can directly metabolise secondary BA, it was predicted that 11β-HSD1 deficiency would lead to changes in the BA profile. Profiling of BA in the gall bladder was performed using mass spectrophotometry. In Hsd11b1-/- mice, 7α-hydroxylated BA predominated (cholic acid [CA]>α-muricholic acid [α- MCA]>CDCA>others), in contrast to C57Bl/6 mice in which 7β-hydroxylated BA predominated (ω-MCA>β-MCA>UDCA>others). The ratio of 7α:7β acids was therefore >100-fold greater in Hsd11b1-/- mice. This suggests that 11β-HSD1 either directly or indirectly controls the epimerisation of 7α- to 7β- hydroxylated BAs. Measurement of mRNAs encoding proteins important for hepatic BA biosynthesis in livers of fasted Hsd11b1-/- mice showed decreased expression of Scarb1/SR-B1, Cyp39a1 and Cyp27a1 (though with no change in levels of CDCA, the product of CYP27A1, in liver or bile fluid), compared to fasted control mice. Hepatic levels of Gpbar1/TGR5/GPBAR1 and Cyp3a11 mRNAs, encoding proteins important in BA detoxification, were increased and decreased, respectively. This suggests that Gpbar1/TGR5/GPBAR1, encoding G-protein coupled bile acid receptor (also called TGR5/GPBAR1) and an FXR target, could be induced to detoxify 7α-hydroxylated BA whereas expression of Cyp3a11, which catalyses the conversion of LCA to hyodeoxycholic acid (HDCA) is decreased; bile fluid of Hsd11b1-/- mice contained lower levels of LCA and little to no HDCA, though LCA and HDCA levels in liver were unaltered. Currently, the functional differences between 7α- and 7β- hydroxylated BA are not clear. However, these findings could have significant implications for bile acid-mediated transcription which, in turn, might affect lipid and sterol metabolism. Also, alterations in BA composition may have other physiological consequences via other pathways. Because cholesterol is the precursor of BA synthesis, it was hypothesised that western diet (WD) (containing cholesterol) would exacerbate and/or alter the phenotype of Hsd11b1-/- mice. Gall bladder weights of fasted Hsd11b1-/- and control C57Bl/6 mice did not change with western diet compared to chow diet. In control C57Bl/6 mice, the total BA concentration in the gall bladder increased in response to WD in comparison to chow diet. In contrast, Hsd11b1-/- mice showed no change in total BA concentration when fed on WD in comparison to chow. These data indicate that 11β-HSD1 is required by mice for the normal increase in total BA concentration in bile in response to dietary cholesterol. BA profiling of bile from control mice fed on WD showed no difference in the relative amounts of 7β-hydroxylated BA and 7α-hydroxylated BA to littermates fed on chow diet with the exception of β–MCA which increased, and α–MCA which decreased. Like chow-fed Hsd11b1-/- mice, BA profiling of bile from WD-fed Hsd11b1-/- mice showed a significant decrease in relative levels of 7β-hydroxylated BA (UDCA < β-MCA < others) and an increase in percentage of 7α-hydroxylated BAs (CA>α-MCA>CDCA>others) compared to C57Bl/6 controls. These data show that Hsd11b1-/- mice fail to show the normal increase in 7β-hydroxylated BA and decrease in 7α-hydroxylated BA observed in control mice in response to a cholesterol containing diet, suggesting 11β-HSD1 deficiency blunts the influence of cholesterol on BA composition. Measurement of hepatic mRNAs encoding BA transporters suggest that hepatocyte uptake of BA is decreased in C57Bl/6 on WD compared to those mice on chow diet, whereas this was not the case in Hsd11b1-/- mice where hepatic expression did not change with diet. Thus, Hsd11b1-/- mice failed to increase expression of Ntcp/ Scl10a1/ Scl10a1 appropriately, suggesting impaired hepatic BA uptake, while Slc51b (encoding OST-β) expression was increased, compared to control mice, possibly to reduce hepatic BA concentration by transporting BA out of hepatocytes into the systemic circulation. Therefore, Hsd11b1-/- mice may adapt to a cholesterol-induced increase in hepatic BA by blunting hepatic BA uptake via NTCP/ SCL10A1/ SCL10A1 and increasing hepatic efflux via OST-β. The effects of 11β-HSD1 deficiency upon BA recycling and BA profile and concentration within the enterohepatic circuit, could reflect 11β-HSD1 action within the liver or could be due to actions in other tissues. / To investigate the role of hepatic 11β-HSD1 specifically, 11β-HSD1 liver-specific knockout (Hsd11b1LKO), 11β- HSD1 liver-specific over-expressors (Hsd11b1LOE) and control mice with exon 3 of the Hsd11b1 gene “floxed” (Hsd11b1F) were studied. Findings from this study indicate a role for 11β-HSD1 in adaption to dietary cholesterol and suggest that hepatic 11β-HSD1 (as opposed to 11β-HSD1 in extra-hepatic tissues) is the main factor regulating BA metabolism. Also, work from this thesis demonstrates 11β-HSD1 is an important regulator of gall bladder emptying and filling, an important component of enterohepatic bile acid recycling. Based on these findings it is anticipated that therapeutic use of 11β-HSD1 inhibitors will result in BA imbalances within the enterohepatic circuit and therefore BA homeostasis. Care must therefore be observed when implementing therapeutic use of 11β-HSD1 inhibitors, with particular focus on patients with cholestasis, Addison’s disease and critically ill patients who already have known BA imbalances in their enterohepatic system.

Chronic glucocorticoid (GC) excess (Cushing’s syndrome, pharmacotherapy) causes metabolic and cardiovascular disease. This might be predicted from the known metabolic (dyslipidaemia, insulin resistance/hyperglycaemia) and hypertensive effects of chronically elevated GC levels. Intracellular GC levels within target tissues are controlled by 11β-hydroxysteroid dehydrogenases. 11β-Hydroxysteroid dehydrogenase type 1 (11β-HSD1, encoded by Hsd11b1) is an enzyme that, in intact cells and in vivo, converts inert GCs (cortisone in humans, and 11- dehydrocorticosterone in mice and rats) into their active forms (cortisol and corticosterone, respectively). Consequently, 11β-HSD1 amplifies intracellular GC levels. Additionally, 11β-HSD1 is also involved in the metabolism of 7-oxysterols; it catalyses the reduction of 7-ketocholesterol (7-KC) to 7β-hydroxycholesterol (7β- HC). 7-KC may inhibit cholesterol biosynthesis through its ability to inhibit cleavage/processing of sterol regulatory element binding protein-2 (SREBP-2), the key regulator of cholesterol synthesis. Alteration of cholesterol homeostasis is a major risk factor for cardiovascular disease. Improvement of metabolic syndrome and attenuation of atherosclerosis are observed in susceptible rodent models with 11β- HSD1 deficiency or inhibition. Conversely, pilot data showed decreased levels of 7- KC as well as increased levels of cleaved SREBP-2 protein (the transcriptionally active form) in liver of mice with hepatic 11β-HSD1 overexpression (LOE mice), suggesting increased cholesterol biosynthesis. It was hypothesised that hepatic 11β- HSD1 promotes cholesterol biosynthesis through hepatic induction of SREBP-2 target genes in the cholesterol biosynthetic pathway. The hypothesis was tested in adult, male LOE and wild-type C57BL/6 mice. In mice fed a standard chow diet, hepatic levels of mRNA encoding hydroxymethylglutarylcoenzyme A (HMG-CoA) reductase and HMG-CoA synthase, SREBP-2 targets in the cholesterol biosynthetic pathway, did not differ between genotypes. Compared to chow, a cholesterol-rich ‘Western’ diet (WD) decreased hepatic levels of mRNA encoding SREBP-2, HMG-CoA reductase and HMG-CoA synthase in wild-type as well as in LOE mice. These data imply that LOE mice show a normal physiological response with respect to cholesterol synthesis when challenged with cholesterol-rich diet, and, contrary to the hypothesis, liver 11β-HSD1 does not increase cholesterol biosynthesis via elevated expression of mRNAs encoding hepatic cholesterol biosynthetic enzymes. The liver X receptors (LXR) are well-known as sensors of oxysterols and regulators of genes involved in processes that decrease total body cholesterol levels i.e. reverse cholesterol transport and cholesterol excretion into bile. Cholesterol is the precursor to oxysterol LXR ligands. It was predicted that liver overexpression of 11β-HSD1 leads to activation of LXRα (the isoform with dominant roles in reverse cholesterol transport and whole-body cholesterol homeostasis) and its downstream targets involved in cholesterol efflux and excretion, in response to increased intracellular cholesterol levels. Indeed, levels of Lxrα mRNA were increased in livers of WD-fed LOE mice compared to wild-type mice on the same diet. There was no evidence for increased cholesterol clearance through bile acid synthesis in LOE mice as indicated by unchanged levels of hepatic Cyp7a1 mRNA between LOE and wild-type mice. However, consistent with being direct targets of LXRα, increased Abcg5 and Abcg8 mRNA levels were observed in livers of WD-fed LOE mice compared to WD-fed wild-type mice. These results corroborate findings in chow-fed LOE mice. Moreover, these data suggest that LOE mice ‘sense’ intracellular cholesterol excess and respond to it by increasing cholesterol efflux into the biliary lumen for excretion, thereby supporting a role for hepatic 11β-HSD1 in promoting biliary cholesterol secretion. To assess the effect(s) of hepatic 11β-HSD1 deficiency on cholesterol homeostasis as well as evaluate the importance of liver 11β-HSD1 in metabolic syndrome, liver-specific 11β-HSD1 knockout (LKO) mice were generated by crossing “floxed” Hsd11b1 mice with Alb-Cre transgenic mice in which Cre expression is restricted to hepatocytes. In liver of LKO mice, 11β-HSD1 mRNA, protein and enzyme activity were reduced by >80%, with no differences in 11β-HSD1 protein levels in kidney, adipose tissue or muscle between LKO and floxed Hsd11b1 littermate controls. These results indicate liver-specificity of Hsd11b1 knockdown in these mice. Body weight and weights of liver, adipose tissue, adrenal, muscle, kidney and brain were unaltered by liver-specific 11β-HSD1 deficiency on a standard chow diet. These mice were subject to a 14-week high fat (HF) diet, which typically causes metabolic syndrome in control but not globally 11β-HSD1 deficient mice. In HF-fed LKO mice, weights of the subcutaneous and epididymal fat depots were decreased compared to HF-fed control mice, resulting in an overall decrease in total white adipose tissue weight. Although no differences were observed in subcutaneous adipocyte hypertrophy between HF-fed LKO and control mice in a small number of samples tested, the above finding suggests that liver 11β-HSD1 influences adiposity and that liver-specific deficiency of 11β-HSD1 may reduce diet-induced adiposity. In terms of cholesterol homeostasis, no differences were observed in hepatic levels of mRNAs encoding cholesterol biosynthetic enzymes as well as those encoding enzymes/transporters for cholesterol catabolism/excretion between LKO and control mice, on either chow or HF diet. In summary, these data do not support a role for hepatic 11β-HSD1 in cholesterol synthesis. However, my evidence suggests that increased hepatic 11β-HSD1 promotes hepatobiliary cholesterol secretion. Finally, knockdown of liver 11β-HSD1, combined with HF feeding, reduces adiposity, suggesting that hepatic 11β-HSD1 may play a key role in adipose tissue lipogenesis/lipolysis and/or lipid storage, and that liver-specific 11β-HSD1 inhibition (or deficiency) may be advantageous in diet-induced obesity. Data presented in this thesis contribute to the understanding of the role of hepatic 11β-HSD1 in cholesterol homeostasis and metabolic syndrome.

Glucocorticoids (GC) have a negative effect on age-associated cognitive decline and the GC metabolising enzyme 11β-hydroxysteroid dehydrogenase Type 1 (11β- HSD1) plays a key role in these effects. Increased glucocorticoids exert detrimental effects on the volume and function of brain regions such as the prefrontal cortex (PFC) and hippocampus that are necessary for cognitive functions such as memory and working memory. Previous research has identified changes in cell populations, metabolite levels and structure within the brain as well as altered levels of inflammation with age, and studies have suggested these biomarkers to be associated with cognitive impairments. Aged mice with a deletion in 11ß-HSD1 (11β-HSD1-/- mice), resulting in lower levels of glucocorticoids within the brain, exhibit attenuated cognitive decline in hippocampal dependent spatial learning and memory with age. However, the mechanisms through which 11β-HSD1 contribute to age-associated cognitive decline remain unknown. However, previous genetic models of 11β-HSD1- /- mice have demonstrated residual 11β-HSD1 activity in the brain which may still exert some effects on cognitive processes. Furthermore, the effect of 11ß-HSD1 on working memory – a more cognitively demanding process essential for everyday decision making - has yet to be determined. This thesis tests the hypothesis that glucocorticoid action mediates age-associated cognitive impairment in spatial learning and memory and spatial working memory through alterations in cell activity, brain metabolite levels and neuroinflammatory processes. Therefore, we aimed to investigate if complete lifelong 11β-HSD1 deficiency would protect against age-associated working memory deficits as well as spatial learning and memory deficits, and its effect on associated neural markers. In particular, we determined changes in hippocampal metabolite levels, cell activity and inflammation as a function of ageing in a longitudinal manner. At 6, 12, 18 and 22 months, male 11β- HSD1-/- and C57BL/6J control mice were cognitively assessed in the Morris Water Maze (MWM) and Radial Arm Water Maze (RAWM) – tests for spatial reference memory and spatial working memory respectively. Magnetic resonance spectroscopy (1H-MRS) was performed to examine the hippocampal metabolite profile in the same mice at 6, 18 and 22 months. Following their final scan, mice were culled and brains dissected for analysis. Results revealed unaltered spatial learning with age in C57BL/6J and 11β-HSD1-/- mice and pointed to a development of alternative strategies for task completion as a result of repeated testing. Spatial memory was more susceptible to age-associated effects with impairments in wild-type mice but not 11β-HSD1-/- mice, though not completely immune from the effects of repeated testing. These impairments were correlated with glutamate/glutamine levels and glial fibrillary acidic protein (GFAP), whilst GFAP was further correlated with 11β-HSD1 protein expression. Working memory was impaired with age in both 11β-HSD1-/- and wild-type mice, suggesting 11β-HSD1 deletion may be detrimental to cognitive processes in the prefrontal cortex. In conclusion, impaired memory with age may be attributed to increased glial reactivity and altered glutamate/glutamine cycling in the hippocampus, and lifelong removal of 11β-HSD1 may alter these processes. However, lifelong removal of 11β-HSD1 may not be as beneficial to working memory processes suggesting that 11β-HSD1 and glucocorticoid action play a key role in working memory processes.

Die 11beta-HSD1 reguliert intrazellulär die Cortisolkonzentration durch Regeneration von Cortison z.B. aus dem Blutkreislauf, zu Cortisol. Daher stellt diese ein wichtiges Element in der Glucocorticoid-vermittelten Genregulation dar. Die 11beta-HSD1 wird ubiquitär exprimiert, auf hohem Niveau besonders in Leber, Fettgewebe und glatten Muskelzellen. Insbesondere die Bedeutung der 11beta-HSD1 in Leber und Fettgewebe konnte mehrfach nachgewiesen werden. In der Leber führte eine erhöhte Aktivität aufgrund einer Überexpression in Mäusen zu einer verstärkten Gluconeogeneserate. Des Weiteren konnte gezeigt werden, dass eine erhöhte Expression und erhöhte Enzymaktivität der 11beta-HSD1 im subkutanen und viszeralen Fettgewebe assoziiert ist mit Fettleibigkeit, Insulinresistenz und Dyslipidämie. Über die Regulation ist jedoch noch wenig bekannt. Zur Untersuchung der Promotoraktivität wurde der Promotorbereich von -3034 bis +188, vor und nach dem Translations- und Transkriptionsstart, der 11beta-HSD1 kloniert. 8 Promotorfragmente wurden mittels Dual-Luciferase-Assay in humanen HepG2-Zellen sowie undifferenzierten und differenzierten murinen 3T3-L1-Zellen untersucht. Anschließend wurde mittels nicht-radioaktiven EMSA die Bindung des TATA-Binding Proteins (TBP) sowie von CCAAT/Enhancer-Binding-Proteinen (C/EBP) an ausgewählte Promotorregionen analysiert. Nach der Charakterisierung des Promotors wurden spezifische endogene und exogene Regulatoren untersucht. Fettsäuren modifizieren die Entstehung von Adipositas und Insulinresistenz. Ihre Wirkung wird u.a. PPARgamma-abhängig vermittelt und kann durch das Inkretin (Glucose-dependent insulinotropic Peptide) GIP modifiziert werden. So wurden die Effekte von unterschiedlichen Fettsäuren, vom PPARgamma Agonisten Rosiglitazon sowie dem Inkretin GIP auf die Expression und Enzymaktivität der 11beta-HSD1 untersucht. Dies wurde in-vitro-, tierexperimentell und in humanen in-vivo-Studien realisiert. Zuletzt wurden 2 Single Nucleotide Polymorphismen (SNP) im Promotorbereich der 11beta-HSD1 in der Zellkultur im Hinblick auf potentielle Funktionalität analysiert sowie die Assoziation mit Diabetes mellitus Typ 2 und Körpergewicht in der MeSyBePo-Kohorte bei rund 1.800 Personen untersucht. Die Luciferase-Assays zeigten basal eine zell-spezifische Regulation der 11beta-HSD1, wobei in allen 3 untersuchten Zelltypen die Bindung eines Repressors nachgewiesen werden konnte. Zudem konnte eine mögliche Bindung des TBPs sowie von C/EBP-Proteinen an verschiedene Positionen gezeigt werden. Die Transaktivierungsassays mit den C/EBP-Proteinen -alpha, -beta und -delta zeigten eben-falls eine zellspezifische Regulation des 11beta-HSD1-Promotors. Die Aktivität und Expression der 11beta-HSD1 wurde durch die hier untersuchten endogenen und exogenen Faktoren spezifisch modifiziert, was sowohl in-vitro als auch in-vivo in unterschiedlichen Modellsystemen dargestellt werden konnte. Die Charakterisierung der MeSyBePo-Kohorte ergab keine direkten Assoziationen zwischen Polymorphismus und klinischem Phänotyp, jedoch Tendenzen für eine erhöhtes Körper-gewicht und Typ 2 Diabetes mellitus in Abhängigkeit des Genotyps. Der Promotor der 11beta-HSD1 konnte aufgrund der Daten aus den Luciferaseassays sowie den Daten aus den EMSA-Analysen näher charakterisiert werden. Dieser zeigt eine variable und zell-spezifische Regulation. Ein wichtiger Regulator stellen insbesondere in den HepG2-Zellen die C/EBP-Proteine -alpha, -beta und -delta dar. Aus den in-vivo-Studien ergab sich eine Regulation der 11beta-HSD1 durch endogene, exogene und pharmakologische Substanzen, die durch die Zellkulturversuche bestätigt und näher charakterisiert werden konnten. / The enzyme 11beta-HSD1 regulates intracellular the cortisol concentration by regeneration of cortisone to cortisol. Hence, 11beta-HSD1 is an important factor in glucocorticoid-mediated gene expression. It is ubiquitously expressed, but high levels have been specifically described in liver, adipose tissue and smooth muscle cells. A pivotal role for 11beta-HSD1 has been demonstrated with respect to metabolism in liver and adipose tissue. Thus, a liver-specific overexpression results in an elevated gluconeogenesis and hepatic glucose output. Furthermore, a fat-specific overexpression was associated with obesity, insulin resistance and dyslipidemia. Despite these intriguing data, the regulation of the human 11beta-HSD1 gene is still in its infancies. 8 promoter fragments from -3034 to +188 of 11beta-HSD1-gene were cloned to analyze promoter activity. Dual-Luciferase-Assay was used in humane HepG2 cells and in undifferentiated and differentiated 3T3-L1 cells. Furthermore, the region close to the transcription start was studied with a non-radioactive EMSA for binding of TATA-binding protein (TBP) and CCAAT/enhancer-binding-protein (C/EBP). The role of the endogenous and exogenous regulators fatty acids, PPARgamma and the incretin (Glucose-dependent insulinotropic Peptide) GIP was investigated in-vitro and in-vivo. Finally, the functional consequences of 2 Single Nucleotide Polymorphisms (SNP) within the promoter region were studied in cell culture and the MeSyBePo-cohorts for association with diabetes mellitus type 2 and body weight. The Luciferase-assay revealed a cell-specific regulation of 11beta-HSD1 and a repressor, which was active in all 3 cell models. Accordingly, a cell-specific regulation was observed in transactivation-assays with C/EBP-proteins -alpha, -beta and -delta. The 11beta-HSD1 enzyme expression and activity was specifically modified by the here investigated endogenous and exogenous factors, which was demonstrated in-vitro but also in-vivo in various experimental settings. The characterisation of the MeSyBePo-cohorte revealed no association between genotype and clinical phenotype, although a trend for an increased body weight and diabetes mellitus type 2 was detected. This work demonstrated a cell-specific regulation of the 11beta-HSD1 promoter. Furthermore, a binding site for TATA-binding proteins was detected in HepG2 and undifferentiated 3T3-L1 cells. A pivotal role in regulation of 11beta-HSD1 promoter activity was demonstrated for the C/EBP-proteins, especially in liver cells. The in-vivo-Studies revealed a regulation of enzyme expression and activity by endogenous, exogenous and pharmacological substances, which was confirmed and analyzed in more detail in cell culture experiments.

Atherosclerosis and post-angioplasty neointimal proliferation, which are leading causes of cardiovascular morbidity and mortality, develop as a result of chronic or acute vascular injury producing inflammatory and proliferative responses in the vessel wall. Glucocorticoids, the stress hormones produced by the adrenal cortex, have anti-inflammatory and anti-proliferative characteristics and can also influence systemic cardiovascular risk factors. The systemic levels of these hormones are controlled by the hypothalamic pituitary adrenal axis. However, there is also a tissue-specific pre-receptor regulation of these hormones by the two isozymes of 11β-hydroxysteroid dehydrogenase (11β-HSD); type 1 regenerates active glucocorticoids within the cells and type 2 inactivates glucocorticoids. Whilst it has been shown that the inhibition of 11β-HSD1 has favourable effect on cardiovascular risk factors and the inhibition of 11β-HSD2 results in hypertension; the effect of these enzymes on vascular lesion development is not known. The work described in this thesis tested the hypothesis that 11β-HSD1 inhibition reduces vascular lesion development due to improvement in cardiovascular risk factors, whereas 11β-HSD2 inhibition leads to adverse vascular remodelling. Apolipoprotein-E deficient (ApoE-/-) mice fed on western diet were used to study atherosclerosis, whereas neointimal proliferation was investigated using a well-established mouse model of wire-angioplasty. Vascular lesions were assessed using novel imaging and standard histological techniques. 11β-HSD1 inhibition reduced the size of atherosclerotic lesions and improved markers of plaque stability with a reduction in lipid content and increase in collagen content of the plaques. This was associated with a reduction in weight gain and blood pressure but without any effect on lipid profile. 11β-HSD1 inhibition did not produce any significant effect on neointimal proliferation in C57Bl/6J mice. However in ApoE-/- mice, 11β-HSD1 inhibition reduced neointimal proliferation with corresponding increase in size of patent lumen and with an associated reduction in macrophage content of neointimal lesions. 11β-HSD2 deletion produced an outward remodelling in un-injured vessels but there was no effect on neointimal proliferation after wire-angioplasty. Administration of a selective mineralocorticoid antagonist, eplerenone, reduced neointimal lesions significantly but to a similar degree in both C57Bl/6J and 11β-HSD2-/- mice, associated with a significant reduction in macrophage content of lesions but without any effect on blood pressure. Data in this thesis highlight the potential therapeutic application of 11β-HSD1 inhibition in reducing the size and vulnerability of atherosclerotic plaques and also reduction in neointimal proliferation (and hence post-angioplasty restenosis) in high risk patients with „metabolic syndrome‟ phenotype. The results also indicate that 11β-HSD2 has a limited, if any, role to play in the development of neointimal lesions.

Non-alcoholic fatty liver disease (NAFLD) is a worldwide health problem which includes steatosis (triglyceride accumulation alone), non-alcoholic steatohepatitis (NASH, with liver inflammation), fibrosis, cirrhosis and hepatocellular carcinoma. Liver fibrosis, which is a reversible response, is the final phase of most chronic liver disease and is characterized by accumulation of extracellular matrix (ECM) from activated hepatic stellate cells (HSCs). Glucocorticoids (GCs) regulate many aspects of metabolism involved in NAFLD. Also, GCs limit HSC activation in vitro. Tissue GC levels are regulated by 11β- hydroxysteroid dehydrogenase-1 (11β-HSD1) which converts inactive 11- dehydrocorticosterone (DHC) into active corticosterone. Previous studies demonstrate that 11β-HSD1 deficiency improves fatty liver in obesity models, but the role of 11β-HSD1 in mechanisms involved in the progression and/or resolution of hepatic injury is largely unknown. I hypothesized that 11β-HSD1 modulates fibrotic and inflammatory responses during hepatic injury and/or the resolution phase. First I sought to address if the levels of 11β-HSD1 during different models of liver injury are dysregulated. In mice, 11β-HSD1 was down-regulated in choline deficient diet (CDD) induced steatosis, methionine and choline deficient diet (MCDD) induced NASH, carbon tetrachloride (CCL4) induced liver fibrosis and thioacetamide (TAA) induced liver fibrosis. In CCL4 injured livers, the down regulation of 11β- HSD1 was observed around the scar area. To test if 11β-HSD1 plays a key role in modulating liver inflammation and fibrosis responses in NAFLD and liver fibrosis I used initially11β-HSD1 knockout (KO) mice. 11β-HSD1 KO showed higher HSC activation only in the High fat feeding model but not in CDD and MCDD models. In the CCL4 injury model, despite reduced hepatocellular injury, 11β-HSD1 KO mice showed enhanced collagen deposition during peak injury and increased fibrotic gene expression during the early resolution phase although unaltered inflammatory markers during both peak injury and resolution. To further dissect cell-specificity on the effect of 11β-HSD1, I repeated the CCL4-injury model using the hepatocyte-specific 11β-HSD1 KO (Alb-HSD1). Alb-HSD1 mice did not show increased susceptibility to fibrosis compared to control littermates suggesting that the 11β- HSD1 possibly modulates fibrotic response by affecting HSC function. To mechanistically address how GCs inhibit HSC activation in vitro I studied the effects of 11β-HSD1 on HSC in vitro. 11β-HSD1 expression was down-regulated during ‘spontaneous’ HSC activation, and 11β-HSD1 deficiency enhanced susceptibility to activation. The GC (11-DHC)’s inhibitory effect on HSC activation was reversed by 11β-HSD1 inhibition. Finally, to address the clinical relevance of 11β-HSD1 in hepatic injury and/or resolution a selective 11β-HSD1 inhibitor, UE2316, was used. UE2316 induced a pro-fibrotic phenotype in ob/ob mice and CCL4-treated C57BL/6 mice, but had no effect when administered only during injury resolution. In conclusion, 11β-HSD1 deficiency causes increased activation of HSCs following diet and chemical injury and promotes liver fibrosis. Effects of 11β-HSD1 inhibitors, which are a potential treatment for metabolic syndrome, are perhaps offset by adverse outcomes in liver.