Probe Substrate Dependencies in CYP3A4 Allosteric Inhibition: A Novel Molecular Mechanism Involving F–F′ Loop Perturbations

Abstract

The biochemical basis for substrate dependences in apparent inhibition constant values (K i ) remains unknown. Our study aims to elucidate plausible structural determinants underpinning these observations. In vitro steady-state inhibition assays conducted using human recombinant CYP3A4 enzyme and testosterone substrate revealed that fibroblast growth factor receptor (FGFR) inhibitors erdafitinib and pemigatinib noncompetitively inhibited CYP3A4 with apparent K i values of 10.2 ± 1.1 and 3.3 ± 0.9 μM, respectively. However, when rivaroxaban was adopted as the probe substrate, there were 2.0-and 3.2-fold decreases in its apparent K i values. To glean mechanistic insights into this phenomenon, erdafitinib and pemigatinib were docked to allosteric sites in CYP3A4. Subsequently, molecular dynamics (MD) simulations of apo-and holo-CYP3A4 were conducted to investigate the structural changes induced. Comparative structural analyses of representative MD frames extracted by hierarchical clustering revealed that the allosteric inhibition of CYP3A4 by erdafitinib and pemigatinib did not substantially modulate its active site characteristics. In contrast, we discovered that allosteric binding of the FGFR inhibitors reduces the structural flexibility of the F−F′ loop region, an important gating mechanism to regulate access of the substrate to the catalytic heme. We surmised that the increased rigidity of the F-F′ loop engenders a more constrained entrance to the CYP3A4 active site, which in turn impedes access to the larger rivaroxaban molecule to a greater extent than testosterone and culminates in more potent inhibition of its CYP3A4-mediated metabolism. Our findings suggest a potential mechanism to rationalize probe substrate dependencies in K i arising from the allosteric noncompetitive inhibition of CYP3A4.