Regulation of Cholesterol Sulfotransferase SULT2B1b by Hepatocyte Nuclear Factor 4α Constitutes a Negative Feedback Control of Hepatic Gluconeogenesis.

The cholesterol sulfotransferase SULT2B1b converts cholesterol to cholesterol sulfate (CS). We previously reported that SULT2B1b inhibits hepatic gluconeogenesis by antagonizing the gluconeogenic activity of hepatocyte nuclear factor 4α (HNF4α). In this study, we showed that the SULT2B1b gene is a transcriptional target of HNF4α, which led to our hypothesis that the induction of SULT2B1b by HNF4α represents a negative feedback to limit the gluconeogenic activity of HNF4α. Indeed, downregulation of Sult2B1b enhanced the gluconeogenic activity of HNF4α, which may have been accounted for by the increased acetylation of HNF4α as a result of decreased expression of the HNF4α deacetylase sirtuin 1 (Sirt1). The expression of Sult2B1b was also induced by HNF4α upon fasting, and the Sult2B1b null (Sult2B1b-/-) mice showed increased gluconeogenic gene expression and an elevated fasting glucose level, suggesting that SULT2B1b also plays a restrictive role in HNF4α-mediated fasting-responsive gluconeogenesis. We also developed thiocholesterol, a hydrolysis-resistant derivative of CS, which showed superior activity to that of the native CS in inhibiting gluconeogenesis and improving insulin sensitivity in high-fat-diet-induced diabetic mice. We conclude that the HNF4α-SULT2B1b-CS axis represents a key endogenous mechanism to prevent uncontrolled gluconeogenesis. Thiocholesterol may be used as a therapeutic agent to manage hyperglycemia.

Hepatocyte nuclear factor-4α interacts with other hepatocyte nuclear factors in regulating transthyretin gene expression.

Transthyretin is a negative acute phase protein whose serum level decreases during the acute phase response. Transthyretin gene expression in the liver is regulated at the transcriptional level, and is controlled by hepatocyte nuclear factor (HNF)-4α and other HNFs. The site-directed mutagenesis of HNF-4, HNF-1, HNF-3 and HNF-6 binding sites in the transthyretin proximal promoter dramatically decreases transthyretin promoter activity. Interestingly, the mutation of the HNF-4 binding site not only abolishes the response to HNF-4α, but also reduces significantly the response to other HNFs. However, mutation of the HNF-4 binding site merely affects the specific binding of HNF-4α, but not other HNFs, suggesting that an intact HNF-4 binding site not only provides a platform for specific interaction with HNF-4α, but also facilitates the interaction of HNF-4α with other HNFs. In a cytokine-induced acute phase response cell culture model, we observed a significant reduction in the binding of HNF-4α, HNF-1α, HNF-3ß and HNF-6α to the transthyretin promoter, which correlates with a decrease in transthyretin expression after injury. These findings provide new insights into the mechanism of the negative transcriptional regulation of the transthyretin gene after injury caused by a decrease in the binding of HNFs and a modulation in their coordinated interactions.

Stability of fatty acyl-coenzyme A thioester ligands of hepatocyte nuclear factor-4alpha and peroxisome proliferator-activated receptor-alpha.

Although long-chain fatty acyl-coenzyme A (LCFA-CoA) thioesters are specific high-affinity ligands for hepatocyte nuclear factor-4alpha (HNF-4alpha) and peroxisome proliferator-activated receptor-alpha (PPARalpha), X-ray crystals of the respective purified recombinant ligand-binding domains (LBD) do not contain LCFA-CoA, but instead exhibit bound LCFA or have lost all ligands during the purification process, respectively. As shown herein: (i) The acyl chain composition of LCFA bound to recombinant HNF-4alpha reflected that of the bacterial LCFA-CoA pool, rather than the bacterial LCFA pool. (ii) Bacteria used to produce the respective HNF-4alpha and PPARalpha contained nearly 100-fold less LCFA-CoA than LCFA. (iii) Under conditions used to crystallize LBD (at least 3 wk at room temperature in aqueous buffer), 16:1-CoA was very unstable in buffer alone. (iv) In the presence of the respective nuclear receptor (i.e., HNF-4alpha and PPARalpha), LBD 70-75% of 16:1-CoA was degraded after 1 d at room temperature in the crystallization buffer, whereas as much as 94-97% of 16:1-CoA was degraded by 3 wk. (v) Cytoplasmic LCFA-CoA binding proteins such as acyl-CoA binding protein, sterol carrier protein-2, and liver-FA binding protein slowed the process of 16:1-CoA degradation proportional to their respective affinities for this ligand. Taken together, these data for the first time indicated that the absence of LCFA-CoA in the crystallized HNF-4alpha and PPARalpha was due to the paucity of LCFA-CoA in bacteria as well as to the instability of LCFA-CoA in aqueous buffers and the conditions used for LBD crystallization. Furthermore, instead of protecting bound LCFA-CoA from autohydrolysis like several cytoplasmic LCFA-CoA binding proteins, these nuclear receptors facilitated LCFA-CoA degradation.