A switch in hepatic cortisol metabolism across the spectrum of non alcoholic fatty liver disease

Abstract

Context: Non alcoholic fatty liver disease (NAFLD) is the hepatic manifestation of the metabolic syndrome. NAFLD represents a spectrum of liver disease ranging from reversible hepatic steatosis, to non alcoholic steato-hepatitis (NASH) and cirrhosis. The potential role of glucocorticoids (GC) in the pathogenesis of NAFLD is highlighted in patients with GC excess, Cushing's syndrome, who develop central adiposity, insulin resistance and in 20% of cases, NAFLD. Although in most cases of NAFLD, circulating cortisol levels are normal, hepatic cortisol availability is controlled by enzymes that regenerate cortisol (F) from inactive cortisone (E) (11β-hydroxysteroid dehydrogenase type 1, 11β-HSD1), or inactivate cortisol through A-ring metabolism (5α- and 5β-reductase, 5αR and 5βR).

Objective and methods: In vitro studies defined 11β-HSD1 expression in normal and NASH liver samples. We then characterised hepatic cortisol metabolism in 16 patients with histologically proven NAFLD compared to 32 obese controls using gas chromatographic analysis of 24 hour urine collection and plasma cortisol generation profile following oral cortisone.

Results: In patients with steatosis 5αR activity was increased, with a decrease in hepatic 11β-HSD1 activity. Total cortisol metabolites were increased in this group consistent with increased GC production rate. In contrast, in patients with NASH, 11β-HSD1 activity was increased both in comparison to patients with steatosis, and controls. Endorsing these findings, 11β-HSD1 mRNA and immunostaining was markedly increased in NASH patients in peri septal hepatocytes and within CD68 positive macrophages within inflamed cirrhotic septa.

Conclusion: Patients with hepatic steatosis have increased clearance and decreased hepatic regeneration of cortisol and we propose that this may represent a protective mechanism to decrease local GC availability to preserve hepatic metabolic phenotype. With progression to NASH, increased 11β-HSD1 activity and consequent cortisol regeneration may serve to limit hepatic inflammation.

Figure 1. 24 hour urine steroid metabolite analysis from patients with steatosis and steatohepatitis compared with obese controls.

(A): 5α-reductase activity as depicted by the urinary 5αTHF/THF ratio (mean ± SEM). (B): total 24 Urine 5α-reduced metabolites (mean ± SEM) (Andros: androsterone). (C): total 24 hr Urine F metabolites (mean ± SEM).

Figure 2. 11β-HSD1 activity assessed by:

(A) 24 hr urine cortols/cortolones and 5αTHF+THF/THE ratios (mean ± SEM) in patients with steatosis and steatohepatitis compared with controls. (B) Hepatic cortisol generation measured by cortisol generation profiles (mean AUC ± SEM) in patients with steatosis and steatohepatitis compared with controls.

Figure 3. Real time PCR mRNA expression data on whole liver samples from 5 normal patients and 5 NASH patients (expressed as arbitrary units ± SEM) for (A)HSD11B1 (11β-HSD 1), (B)SRD5A2 (5α-reductase 2), (C)GRα.

** p<0.01 NASH vs controls; * p<0.05 NASH vs controls.

Figure 4. Hepatic 11β-HSD 1 immunoreactivity in patients with severe NASH compared to normal controls.

There was generally increased staining for 11β-HSD1 throughout the liver parenchyma in (A) NASH samples compared with (B) Normal liver ×20. (C) and (D) Increased staining at the limiting plate in peri-septal areas and strongly staining specific cells within the inflammatory infiltrate in NASH ×10(C) and ×20(D) (E) Confocal microscopy on severe NASH cryosections. Green - 11β-HSD1, red – CD68 IgG macrophage marker, yellow – colocalisation of 11β-HSD1 and CD68 positive macrophages. (F) Western blot analysis of human liver microsomes from normal and NASH livers.

Figure 5. Schematic: Hepatic glucocorticoid metabolism and its modulation in response to disease progression in NAFLD.