VKORC1 and VKORC1L1 have distinctly different oral anticoagulant dose-response characteristics and binding sites.

Vitamin K reduction is catalyzed by 2 enzymes in vitro: the vitamin K 2,3-epoxide reductase complex subunit 1 (VKORC1) and its isozyme VKORC1-like1 (VKORC1L1). In vivo, VKORC1 reduces vitamin K to sustain γ-carboxylation of vitamin K-dependent proteins, including coagulation factors. Inhibition of VKORC1 by oral anticoagulants (OACs) is clinically used in therapy and in prevention of thrombosis. However, OACs also inhibit VKORC1L1, which was previously shown to play a role in intracellular redox homeostasis in vitro. Here, we report data for the first time on specific inhibition of both VKOR enzymes for various OACs and rodenticides examined in a cell-based assay. Effects on endogenous VKORC1 and VKORC1L1 were independently investigated in genetically engineered HEK 293T cells that were knocked out for the respective genes by CRISPR/Cas9 technology. In general, dose-responses for 4-hydroxycoumarins and 1,3-indandiones were enzyme-dependent, with lower susceptibility for VKORC1L1 compared with VKORC1. In contrast, rodenticides exhibited nearly identical dose-responses for both enzymes. To explain the distinct inhibition pattern, we performed in silico modeling suggesting different warfarin binding sites for VKORC1 and VKORC1L1. We identified arginine residues at positions 38, 42, and 68 in the endoplasmatic reticulum luminal loop of VKORC1L1 responsible for charge-stabilized warfarin binding, resulting in a binding pocket that is diametrically opposite to that of VKORC1. In conclusion, our findings provide insight into structural and molecular drug binding on VKORC1, and especially on VKORC1L1.

Variants in FIX propeptide associated with vitamin K antagonist hypersensitivity: functional analysis and additional data confirming the common founder mutations.

One of the most common and unwanted side effects during oral anticoagulant therapy (OAT) is bleeding complications. In rare cases, vitamin K antagonist (VKA)-related bleeding events are associated with mutations affecting the F9 propeptide at amino acid position 37 due to a substitution of alanine to either valine or threonine. Based on our actual cohort of 18 patients, we update the knowledge on this rare phenotype and its origin. A founder mutation for both variants was reconfirmed by haplotype analysis of intronic and extragenic short tandem repeat (STR) polymorphisms with a higher prevalence in Switzerland than in other regions of Europe. Screening of healthy individuals for the presence of these F9 gene mutations did not identify any of these variants, thus proving the rare occurrence of this genotype. Furthermore, both variants were expressed in vitro and warfarin dose responses were studied. Our warfarin dose response analysis confirmed higher sensitivity of both variants to warfarin with the effect being more apparent for Ala37Thr. Thus, although F9 propeptide mutation-associated hypersensitivity to VKA is a rare phenomenon, awareness towards this bleeding phenotype is important to identify patients at risk.

Warfarin and vitamin K compete for binding to Phe55 in human VKOR.

Vitamin K epoxide reductase (VKOR) catalyzes the reduction of vitamin K quinone and vitamin K 2,3-epoxide, a process essential to sustain γ-carboxylation of vitamin K-dependent proteins. VKOR is also a therapeutic target of warfarin, a treatment for thrombotic disorders. However, the structural and functional basis of vitamin K reduction and the antagonism of warfarin inhibition remain elusive. Here, we identified putative binding sites of both K vitamers and warfarin on human VKOR. The predicted warfarin-binding site was verified by shifted dose-response curves of specified mutated residues. We used CRISPR-Cas9-engineered HEK 293T cells to assess the vitamin K quinone and vitamin K 2,3-epoxide reductase activities of VKOR variants to characterize the vitamin K naphthoquinone head- and isoprenoid side chain-binding regions. Our results challenge the prevailing concept of noncompetitive warfarin inhibition because K vitamers and warfarin share binding sites on VKOR that include Phe55, a key residue binding either the substrate or inhibitor.

Structural Modeling Insights into Human VKORC1 Phenotypes.

Vitamin K 2,3-epoxide reductase complex subunit 1 (VKORC1) catalyses the reduction of vitamin K and its 2,3-epoxide essential to sustain γ-carboxylation of vitamin K-dependent proteins. Two different phenotypes are associated with mutations in human VKORC1. The majority of mutations cause resistance to 4-hydroxycoumarin- and indandione-based vitamin K antagonists (VKA) used in the prevention and therapy of thromboembolism. Patients with these mutations require greater doses of VKA for stable anticoagulation than patients without mutations. The second phenotype, a very rare autosomal-recessive bleeding disorder caused by combined deficiency of vitamin K dependent clotting factors type 2 (VKCFD2) arises from a homozygous Arg98Trp mutation. The bleeding phenotype can be corrected by vitamin K administration. Here, we summarize published experimental data and in silico modeling results in order to rationalize the mechanisms of VKA resistance and VKCFD2.

Two enzymes catalyze vitamin K 2,3-epoxide reductase activity in mouse: VKORC1 is highly expressed in exocrine tissues while VKORC1L1 is highly expressed in brain.

VKORC1 and VKORC1L1 are enzymes that both catalyze the reduction of vitamin K2,3-epoxide via vitamin K quinone to vitamin K hydroquinone. VKORC1 is the key enzyme of the classical vitamin K cycle by which vitamin K-dependent (VKD) proteins are γ-carboxylated by the hepatic γ-glutamyl carboxylase (GGCX). In contrast, the VKORC1 paralog enzyme, VKORC1L1, is chiefly responsible for antioxidative function by reduction of vitamin K to prevent damage by intracellular reactive oxygen species. To investigate tissue-specific vitamin K 2,3-epoxide reductase (VKOR) function of both enzymes, we quantified mRNA levels for VKORC1, VKORC1L1, GGCX, and NQO1 and measured VKOR enzymatic activities in 29 different mouse tissues. VKORC1 and GGCX are highly expressed in liver, lung and exocrine tissues including mammary gland, salivary gland and prostate suggesting important extrahepatic roles for the vitamin K cycle. Interestingly, VKORC1L1 showed highest transcription levels in brain. Due to the absence of detectable NQO1 transcription in liver, we assume this enzyme has no bypass function with respect to activation of VKD coagulation proteins. Our data strongly suggest diverse functions for the vitamin K cycle in extrahepatic biological pathways.

The Arg98Trp mutation in human VKORC1 causing VKCFD2 disrupts a di-arginine-based ER retention motif.

Vitamin K 2,3-epoxide reductase complex subunit 1 (VKORC1) is an enzyme localized to the endoplasmic reticulum (ER) membrane. VKORC1 catalyzes the reduction of vitamin K 2,3-epoxide to vitamin K and to vitamin K hydroquinone, the latter required by the enzyme γ-carboxylase for γ-carboxylation of all vitamin K-dependent (VKD) proteins. Until now, only 1 human VKORC1 mutation, p.Arg98Trp, is known to cause combined deficiency of VKD clotting factors type 2 (VKCFD2), a disease phenotype reported in 3 unrelated families. VKCFD2 patients suffer from spontaneous bleeding episodes because of decreased levels of γ-carboxylated VKD clotting factors. Daily supraphysiological vitamin K supplementation restores clotting for VKCFD2 patients and results in high serum levels of vitamin K 2,3-epoxide, suggesting that supplemented vitamin K is reduced in vivo. Although the p.Arg98Trp mutation results in reduced vitamin K 2,3-epoxide reductase activity, the molecular mechanism underlying this pathophysiology is unknown. Using a combination of in silico analysis and confocal microscopy, we demonstrate for the first time that VKORC1:p.Arg98Trp disrupts a di-arginine ER retention motif resulting in 20% ER colocalization only. As a consequence, VKORC1 exits the ER membrane by cellular quality control systems and results in the observed VKCFD2 phenotype.

Human VKORC1 mutations cause variable degrees of 4-hydroxycoumarin resistance and affect putative warfarin binding interfaces.

Since the discovery of warfarin-sensitive vitamin K 2,3-epoxide reductase complex subunit 1 (VKORC1), 26 human VKORC1 (hVKORC1) missense mutations have been associated with oral anticoagulant resistance (OACR). Assessment of warfarin resistance using the "classical" dithiothreitol-driven vitamin K 2,3-epoxide reductase (VKOR) assay has not reflected clinical resistance phenotypes for most mutations. Here, we present half maximal inhibitory concentrations (IC50) results for 21 further hVKORC1 mutations obtained using a recently validated cell-based assay (J Thromb Haemost 11(5):872). In contrast to results from the dithiothreitol-driven VKOR assay, all mutations exhibited basal VKOR activity and warfarin IC50 values that correspond well to patient OACR phenotypes. Thus, the present assay is useful for functional investigations of VKORC1 and oral anticoagulant inhibition of the vitamin K cycle. Additionally, we modeled hVKORC1 on the previously solved structure of a homologous bacterial enzyme and performed in silico docking of warfarin on this model. We identified one binding site delineated by 3 putative binding interfaces. These interfaces comprise linear sequences of the endoplasmic reticulum-lumenal loop (Ser52-Phe55) and the first (Leu22-Lys30) and fourth (Phe131-Thr137) transmembrane helices. All known OACR-associated hVKORC1 mutations are located in or around these putative interfaces, supporting our model.