Quantum mechanics/molecular mechanics modeling of regioselectivity of drug metabolism in cytochrome P450 2C9

Abstract

Cytochrome P450 enzymes (P450s) are important in drug metabolism and have been linked to adverse drug reactions. P450s display broad substrate reactivity, and prediction of metabolites is complex. QM/MM studies of P450 reactivity have provided insight into important details of the reaction mechanisms and have the potential to make predictions of metabolite formation. Here we present a comprehensive study of the oxidation of three widely used pharmaceutical compounds (S-ibuprofen, diclofenac, and S-warfarin) by one of the major drug-metabolizing P450 isoforms, CYP2C9. The reaction barriers to substrate oxidation by the iron-oxo species (Compound I) have been calculated at the B3LYP-D/CHARMM27 level for different possible metabolism sites for each drug, on multiple pathways. In the cases of ibuprofen and warfarin, the process with the lowest activation energy is consistent with the experimentally preferred metabolite. For diclofenac, the pathway leading to the experimentally observed metabolite is not the one with the lowest activation energy. This apparent inconsistency with experiment might be explained by the two very different binding modes involved in oxidation at the two competing positions. The carboxylate of diclofenac interacts strongly with the CYP2C9 Arg108 side chain in the transition state for formation of the observed metabolite-but not in that for the competing pathway. We compare reaction barriers calculated both in the presence and in the absence of the protein and observe a marked improvement in selectivity prediction ability upon inclusion of the protein for all of the substrates studied. The barriers calculated with the protein are generally higher than those calculated in the gas phase. This suggests that active-site residues surrounding the substrate play an important role in controlling selectivity in CYP2C9. The results show that inclusion of sampling (particularly) and dispersion effects is important in making accurate predictions of drug metabolism selectivity of P450s using QM/MM methods.

Figure 1

Chemical structures of the drug molecules studied in this work: diclofenac (1), S-ibuprofen (2), and S-warfarin (3). Atoms numbered according to convention used in the current work. Diclofenac and ibuprofen were modeled in the QM/MM calculations here as the carboxylate (negatively charged) forms.

Figure 2

Chemical structures of the metabolites formed during oxidation of S-ibuprofen by CYP2C9: major metabolites S,R-3-hydroxyibuprofen (4) and S,S-3-hydroxyibuprofen (5), and minor metabolite S-2-hydroxyibuprofen (6).

Figure 3

Chemical structures of the metabolites formed during oxidation of diclofenac. 4′-Hydroxydiclofenac (7) is the major metabolite formed during oxidation by CYP2C9. 5-Hydroxydiclofenac (8) is the major metabolite formed during oxidation by CYP3A4.

Figure 4

Chemical structure of S-7-hydroxywarfarin (9) and S-6-hydroxywarfarin (10), the major and minor metabolites formed during oxidation of S-warfarin by CYP2C9. These structures correspond to the open form of S-warfarin.

Figure 5

Criteria used for selection of MD structures for QM/MM modeling of (a) aliphatic hydroxylation and (b) aromatic hydroxylation by CYP2C9.

Figure 6

(a) QM representation of Cpd I used in QM and QM/MM calculations. (b) Separation of S-warfarin into QM and MM regions in QM/MM calculations.

Figure 7

Rebound mechanism for aliphatic hydroxylation catalyzed by P450s. The first step is hydrogen abstraction, resulting in the formation of a substrate radical which then rebounds with the Fe-bound OH group to form the hydroxylated product.

Figure 8

Reactant complex structures for hydrogen abstraction from (a) C3 and (b) C2 of S-ibuprofen, calculated at the B3LYP(BSI)/CHARMM27 level of theory (corresponding to doublet profiles 2-3 and 3-1 in Table 1).

Figure 9

Transition-state structures for hydrogen abstraction from C3 and C2 of S-ibuprofen, calculated at the B3LYP-D(BSI)/CHARMM27 level of theory (corresponding to doublet profiles 2-3 and 3-1 in Table 1). The carbon atoms corresponding to C2 and C3 hydroxylation are in green and orange, respectively.

Figure 10

Transition-state structures for hydrogen abstraction from C3 and C2 of S-ibuprofen, calculated in vacuo at the B3LYP-D/BSIII level of theory. The carbon atoms corresponding to C2 and C3 hydroxylation are in green and orange, respectively.

Figure 11

(a) Effect of angle of approach of substrate (Sub) to Cpd I on barrier to aliphatic and aromatic hydroxylation. Optimal orbital overlap leads to low barrier when angle of approach is approximately 130° (green). At larger angles of approach, non-optimal orbital overlap occurs, leading to higher barriers (red). (b) Orbital energy diagram for the first electron transfer step in the oxidation of substrate by Cpd I. The electron may transfer to either of the two singly occupied Fe–O π* orbitals.

Figure 12

Addition/rearrangement mechanism for aromatic hydroxylation catalyzed by P450s.^, The orbital energy levels are indicated schematically. The first step is C–O bond formation, resulting in the formation of a σ-adduct with either (a) cationic or (b) radical character. (c) The second step is rearrangement to form the hydroxylated product.

Figure 13

Reactant complex structures for hydrogen abstraction from (a) C6 and (b) C7 of S-warfarin, calculated at the B3LYP-D(BSI)/CHARMM27 level of theory (corresponding to doublet profiles 6-8 and 7-9 in Table 3).

Figure 14

Transition-state structures for C–O bond formation between C6 and C7 of S-warfarin and the Cpd I ferryl oxygen of CYP2C9, calculated at the B3LYP-D(BSI)/CHARMM27 level of theory (corresponding to doublet profiles 6-8 and 7-9 in Table 3). The carbon atoms corresponding to C6 and C7 hydroxylation are displayed in green and orange, respectively.

Figure 15

Transition-state structures for hydrogen abstraction from C6 and C7 of a truncated model of S-warfarin, calculated in vacuo at the B3LYP-D/BSIII level of theory. The carbon atoms corresponding to C6 and C7 hydroxylation are in purple and yellow, respectively.

Figure 16

Reactant complex structures for hydrogen abstraction from (a) C4′ and (b) C5 of diclofenac, calculated at the B3LYP-D(BSI)/CHARMM27 level of theory (corresponding to doublet profiles 4′-2 and 5-3 in Table 5).

Figure 17

QM/MM transition-state structures for C–O bond formation between C4′ and C5 of diclofenac and the Cpd I ferryl oxygen of CYP2C9, calculated at the B3LYP-D-6-31G(d)/CHARMM27 level of theory (corresponding to doublet profiles 4′-2 and 5-3 in Table 5). The carbon atoms corresponding to C4′ and C5 hydroxylation are displayed in green and orange, respectively.

Figure 18

Gas-phase transition state structures for C–O bond formation between C4′ and C5 of diclofenac and the Cpd I ferryl oxygen of CYP2C9, calculated at the B3LYP-D/6-31G(d) level of theory. The carbon atoms corresponding to C4′ and C5 hydroxylation are displayed in green and orange, respectively.

Figure 19

Putative energy level diagram for CYP2C9 oxidation of diclofenac. 2C9+diclofenac refers to the free energy of the separate diclofenac molecule and CYP2C9 enzyme. 5-TS/5-RC and 4′-TS/4′-RC correspond to the free energies of the transition states/reactant complexes to 5- and 4′-hydroxylation, respectively. In this hypothesis, the energy barrier to 4′-hydroxylation (Δ^⧧G_4′) is larger than that of 5-hydroxylation (Δ^⧧G₅). However, the free energy of binding is greater for 4′-hydroxylation, and the relative energy of the transition state to 4′-hydroxylation is lower than that for 5-hydroxylation; hence 4′-hydroxylation would be the preferred pathway.