endocrine-abstracts.org/ea/0028/ea0028p333A.htm
Regulation of de novo lipogenesis in human liver by 5α-reductase
Maryam Nasiri, Laura Gathercole, Paul Stewart & Jeremy Tomlinson
Clinical and Experimental Medicine, University of Birmingham, Birmingham, United Kingdom.
The potent effects of glucocorticoids (GCs) upon carbohydrate metabolism are well described. However, their actions upon lipid metabolism are poorly characterized. Patients with GC excess (Cushing’s syndrome) develop central obesity, insulin resistance and hepatic steatosis in up to 20% of cases. The A-ring reductases (5α-reductase type 1 [5αR1] and type 2 [5αR2]), inactivate cortisol as well as generate dihydrotestosterone (DHT) from testosterone (T) and are highly expressed in human liver. We propose that 5αR expression is a critical regulator of GC action. Using the human hepatoma derived C3A cell line which expresses 5αR1, but not 5αR2, the effects of GCs and 5αR over-expression upon de novo lipogenesis (DNL) were measured by 1-[14C] acetate incorporation in triglyceride. Expression levels were quantified by real-time PCR, and activity (both cortisol clearance and DHT generation) measured using a combination of gas and liquid chromatography / mass spectroscopy. The impact of 5αR over expression upon DNL measured in the presence and absence of cortisol. Cortisol decreased lipogenesis in a dose-dependent manner (85.6 6.6% [100 nM], 73.5 7.9% [250 nM], 55.0 5.6% [1000 nM], p<0.05). 5αR2 over-expression was confirmed by real-time PCR (>1000-fold) and functional activity demonstrated through increased DHT generation following incubation with T (50 nM, 24 h) (T:DHT ratio; 514.6 [vector only] vs. 1.5 [5αR2]) and increased cortisol clearance after incubation with cortisol (200 nM, 24 h) (media cortisol concentrations; 77±3 nM vs.130±9 nM). In the absence of cortisol, 5αR2 transfection did not alter rates of DNL. However, in the presence of cortisol, 5αR2 completely restored rates of lipogenesis to those of untreated controls (e.g. 61.9±7.6% [1000 nM cortisol] vs. 103.8±8.8% [5αR2+1000 nM cortisol], P<0.05, untreated control=100%). This study defines the impact of GCs upon DNL in human liver, but in addition, clearly demonstrates the potent impact that modulation of 5αR2 can have upon GC action and raises the possibility that this might reflect a novel metabolic therapeutic target.