Human Cytochrome P450 8B1 Structure and Function: Supporting Drug Design for NAFLD and Diabetes Mellitus

Human cytochrome P450 8B1 (CYP8B1) performs a key step in human production of bile acids. It 12α-hydroxylates a steroidal substrate to produce the relatively polar primary bile acid cholic acid. The absence of CYP8B1 catalytic function results in the more hydrophobic bile acid chenodeoxycholic acid. Knockouts of CYP8B1 indicated this P450 may be a good drug target for nonalcoholic fatty liver disease due to its alteration of bile acid-mediated fat absorption. Similar studies indicate that the absence of CYP8B1 also has a favorable effect on type 2 diabetes. However, the absence of selective CYP8B1 inhibitors both restricts further study of its physiological roles and potential disease treatment. We recently generated the first CYP8B1 structure, with the nonspecific type II inhibitor (S)-tioconazole, and are now employing a structure/function strategy to understand CYP8B1 interactions with a) its native substrate and b) a wider array of inhibitors. Within the CYP8B1 active site tioconazole is located near the B' helix due to spatial hinderance from W281 and N286 in the I helix on the opposite side of the active site. W281 has previously been suggested as a key residue for CYP8B1 catalysis with its native steroidal substrate. To test this concept, a human CYP8B1 W281F mutant was generated. The W281F enzyme demonstrated both impaired substrate binding and significantly reduced catalytic efficiency. Thus W281 appears to be important in binding both the native substrate and the tioconazole inhibitor. Subsequently a set of additional azole ligands were evaluated to broaden the definition of ligand structures compatible with the CYP8B1 active site and potentially identify even better inhibitors. Changes in visible absorbance spectra were used to determine the binding mode and determine Kd values, while inhibition of steroid 12a-hydroxylation was used to determine IC50 values. Most azole compounds bound CYP8B1, typically via the expected type II imidazole-to-heme-iron interaction like tioconazole, and with Kd values in the nM to µM range. Miconazole and econazole bound with the highest affinity and had the lowest IC50 values, demonstrating improvements over tioconazole. The structural differences between these ligands are a relatively moderate substitution of a thienyl vs. phenyl ring and modified chlorination patterns on that ring, demonstrating CYP8B1 sensitivity to these inhibitor components. Both the CYP8B1 structure and the functional characterization should facilitate the identification increasingly effective CYP8B1 inhibitors. If selectivity can also be incorporated into these compounds, the resulting inhibitors should enable a better understanding of the role of CYP8B1 in normal physiology and facilitate a potential treatment for nonalcoholic fatty liver disease and type 2 diabetes.