Endocrine Abstracts (2019) 65 P171 | DOI: 10.1530/endoabs.65.P171

5[beta]-reductase (AKR1D1) isoforms differentially regulate natural and synthetic glucocorticoid clearance and glucocorticoid receptor activation in vitro

Nathan Appanna1, Anastasia Arvaniti1, Elena Gangitano1, Karen Morris2, Sherly George2, Brian Keevil2, Jeremy Tomlinson1 & Nikolaos Nikolaou1


1University of Oxford, Oxford, UK; 2University of Manchester, Manchester, UK


Metabolic syndrome and its hepatic manifestation, non-alcoholic fatty liver disease (NAFLD), are increasing in prevalence. Steroid hormones are established regulators of metabolic phenotype. 5β-reductase (AKR1D1) is highly expressed in human liver, inactivating steroid hormones, including glucocorticoids and androgens. The human AKR1D1 gene contains 9 exons; six splice variants have been identified and three lead to functional protein isoforms (AKR1D1-001, -002, and -006). The AKR1D1-002 isoform is the most well-characterised and we have shown that it is able to modulate hepatic glucocorticoid availability and glucocorticoid receptor (GR) activation. However, the potential of the other AKR1D1 isoforms to regulate steroid hormone availability is unknown. AKR1D1 splice variants were over-expressed in HEK293 cells, and incubated with cortisol, dexamethasone or prednisolone. AKR1D1-002 over-expression resulted in rapid cortisol clearance (as measured by liquid chromatography-mass spectrometry). However, the clearance of the synthetic glucocorticoids, dexamethasone and prednisolone, was much more limited. Consistent with these data, GR activation, as measured by a luciferase-reported assay, was decreased in cortisol-treated AKR1D1-002 over-expressing cells, in comparison with empty vector controls. However, no differences in GR activation were observed in AKR1D1-002 over-expressing cells, when treated with dexamethasone and prednisolone. In contrast to AKR1D1-002, over-expression of either AKR1D1-001 or AKR1D1-006 had no effect on cortisol clearance. However, AKR1D1-001 significantly decreased GR activation following dexamethasone and prednisolone treatment, suggestive of increased clearance of synthetic glucocorticoids. AKR1D1-006 failed to regulate GR activation, following either natural or synthetic glucocorticoid treatment. Through genetic manipulation of expression of AKR1D1 splice variants, we have demonstrated their differential ability to regulate natural and synthetic glucocorticoid metabolism. Our data suggest that, while AKR1D1-006 may be functionally inactive, the AKR1D1-001 isoform may have an important role in regulating exogenous glucocorticoid availability. This might have implications for patients being treated with synthetic glucocorticoids and their risk of developing adverse metabolic side effects.