11β-Hydroxysteroid Dehydrogenase Type 1: An Explanation of the Metabolic Syndrome - Cushing’s Disease Paradox?

J. Seckl

Endocrinology Unit, University of Edinburgh, Western General Hospital, Edinburgh, U.K.

Conventional wisdom states that glucocorticoid action is determined by the levels of steroid in the circulation and the density of receptors in target cells. Recently however, a novel level of control of glucocorticoid action upon target tissues has been elucidated, cellular metabolism by intracellular enzymes, notably 11beta-hydroxysteroid dehydrogenases (11β-HSDs). There are 2 isozymes, the products of distinct and only distantly related genes. The function of 11β-HSD-2, a potent dehydrogenase, is to protect receptors from active glucocorticoids (cortisol in humans, corticosterone in rats and mice) by rapidly inactivating them to inert 11-keto forms (cortisone, 11-dehydrocorticosterone). Thus 11β-HSD-2 prevents glucocorticoid access to intrinsically non-selective mineralocorticoid receptors in the distal nephron, producing aldosterone selectivity in vivo. In contrast, the function of the ‘cinderella’ type 1 isozyme (11β-HSD-1), though the first to be isolated and cloned, was obscure until recent studies showed that in intact cells and organs it catalyses the reverse or reductase reaction. This suggests that 11β-HSD-1 regenerates active glucocorticoids from inert keto-steroids, notably in glucocorticoid receptor rich tissues such as liver, adipose and the brain.

A key example of the importance of 11β-HSD-1 has come from recent studies of the prevalent Metabolic Syndrome of visceral obesity, insulin resistance, hyperglycaemia, dyslipidaemia and hypertension. The phenotype closely resembles the rare Cushing’s syndrome of cortisol excess, but in simple obesity plasma cortisol levels are not elevated. Recent evidence has shown an adipose-selective 2 - 3-fold increase in 11β-HSD-1 activity and mRNA levels in rodent models (Zucker rats, ob/ob mice) and human populations with the metabolic syndrome or abdominal obesity. We and others hypothesised that, by locally increasing glucocorticoid action, this might produce a localised ‘Cushing’s syndrome of adipose tissue’. To address causation transgenic mice with ∼ 2.5 fold overexpression of 11β-HSD-1 only in adipose tissue (under the aP2 promoter) were generated. These mice have elevated adipose corticosterone levels, whilst systemic plasma levels of the glucocorticoid remain unaltered. The transgenic mice show 3-fold visceral obesity, hyperglycaemia, insulin resistance, dyslipidaemia and hypertension. The latter is apparently driven by angiotensinogen over-production (4-fold) selectively in adipose tissue and is normalised by low doses of an angiotensin 2 receptor antagonist. This mouse also has elevated portal blood levels of corticosterone, implicating excess liver glucocorticoid exposure in its phenotype. To dissect this, we recently made liver-specific 11β-HSD-1 transgenic mice lines driven by the apoE promoter. These mice have intrahepatic fat accumulation and hyperlipidaemia, modest insulin resistance and hypertension, but have normal body weight and fat distribution and normal glucose tolerance. Thus the effects of adipose overexpression alone can model the metabolic syndrome fully.

To explore the physiological importance and therapeutic potential of 11β-HSD-1 inhibition we made 11β-HSD-1 knock-out mice on both obesity-prone (C57BL/6J) and obesity-resistant (MF-1) strain backgrounds. The null mice have improved glucose tolerance and lower plasma triglycerides driven, in part, by reduced hepatic gluconeogenesis and increased liver beta-lipid oxidation and insulin sensitivity. Adipose tissue is also important in the 11β-HSD-1^-/- phenotype. Thus, 11β-HSD-1^-/- mice have 67 % lower intra-adipose corticosterone levels. Both strains of 11β-HSD-1^-/- mice have reduced fat pad weights. Intriguingly, with chronic high fat (HF) feeding, both strains exhibit reduced visceral fat accumulation, redistributing fat to ‘metabolically safer’ peripheral depots (Fig. [1]). On the C57BL/6J obesity-prone background 11β-HSD-1^-/- mice had 15 % lower weight gain despite consuming 12 % more calories than wild types. The favourably altered fat distribution was associated with improved glucose tolerance and insulin sensitivity in vivo. The null mice also had reduced free fatty acids and a ‘cardioprotective’ serum lipid profile (low LDL, high HDL-cholesterol). Isolated adipocytes from 11β-HSD-1^-/- mice exhibited higher basal and insulin-stimulated glucose uptake indicating insulin sensitisation. The biology appears conserved, at least in part, since administration of the 11β-HSD inhibitor carbenoxolone to humans also increases insulin sensitivity.

Fig. 1 Adipose depots in 11β-HSD1-/- mice with high fat (HF) diet for 18 weeks. With HF the wild type mice gain adipose mass in the metabolically disadvantageous visceral mesenteric depot. In contrast, 11β-HSD1-/- mice gain adipose tissue in the “safer” peripheral (e.g. epididymal) depots. Morton NM et al, submitted.

Unexpectedly, wild type mice down-regulated 11β-HSD-1 in white and brown fat in response to HF feeding. Moreover, obesity-resistant A/J mice had both lower basal levels and down-regulated adipose 11β-HSD-1 more markedly than obesity-prone C57BL/6J mice. Down-regulation of adipose 11β-HSD-1 in response to dietary fat represents a novel adaptive mechanism, more pronounced in obesity-resistant animals, that counteracts central adiposity and its metabolic consequences.

11β-HSD-1 is also highly expressed in the brain, notably in hippocampus and cerebellum. 11β-HSD-1 null mice show modestly elevated plasma corticosterone levels at the diurnal nadir compatible with blunted feedback upon the hypothalamic-pituitary-adrenal axis. Despite this, 11β-HSD-1-/- mice have lower intrahippocampal corticosterone levels and resist glucocorticoid-associated impairments of cognitive (hippocampal) function with ageing. Again the biology may be conserved as 11β-HSD-1 is expressed in the human hippocampus and carbenoxolone improves aspects of cognitive performance in healthy elderly humans. Thus 11β-HSD-1 represents a novel control of intracellular glucocorticoid action prior to glucocorticoid receptor binding and may be a useful therapeutic target in the Metabolic Syndrome and age-related cognitive disorders.