Inquiry of Literature Evidence for Induced Fit of Cytochrome P450 2C9 for Warfarin, Phenprocoumon, Flurbiprofen and Clopidogrel: A Critical Review
1Faculty of Chemical Sciences, Benemérita Universidad Autónoma de Puebla, CP 72570 Puebla, State Puebla, Mexico
2Center for Biosystem Analysis ZBSA, University of Freiburg, Freiburg, Germany
*Correspondence to: Thomas Scior, E-mail: firstname.lastname@example.org
Citation: Scior T, Quiroga- Montes I, Kammerer B (2018) Inquiry of Literature Evidence for Induced Fit of Cytochrome P450 2C9 for Warfarin, Phenprocoumon, Flurbiprofen and Clopidogrel: A Critical Review. SCIOL Biotechnol 2018;1:30-48.
Reports on substrate hydroxylation by Cytochrome P450 enzymes (CYP) were gathered to review data on induced fit and regioselectivities. Molecular models thereof were generated based on available crystal structures. To this end, PDB database entries were retrieved to assemble pairs of liganded and unliganded complexes of the same CYP. Then, their structures were superimposed and geometrical differences correlated to document known cases of induced fit. For the bibliographic revision of regioselectivities the scope was narrowed to focus on hydroxylation patterns of 4-hydroxycoumarin-like warfarin and similar drugs like phenprocoumon by CYP2C9. This combined analysis of literature and related structural information lent detailed insight into not only structural variation between all those CYPs which have been reported as responsible for substrate hydroxylation, but also variability in the sites of metabolism (SoM). All of which reflects substrate recognition and regioselectivities at an atom scale. In conclusion, it was found that induced fit and regioselectivities have been frequently cited in the extant literature body but poorly documented.
Vitamin K, Acenocoumarol, Ritonavir CYP2C19, CYP3A
List of Abbreviations
PPC: Phenprocoumon; SRM: Selected-Reaction-Monitoring (chromatogram); TS: Transition State; WFN: Warfarin
Cytochrome P450 oxygenases (CYP for short) are liver microsomal enzymes with a prosthetic hem(e) group that catalyze various types of hepatic biotransformation reactions including aromatic and aliphatic oxidation, N- and O-dealkylation, S- and N-oxidation, sulfoxide/sulfone formation, oxidative deamination, desulfuration, and dehalogenation [1,2]. Almost 90% of commonly used drugs are metabolized by CYP1A2, CYP2C9, CYP2C19, CYP2D6 and CYP3A4 . They metabolize a remarkably diverse group of endogenous substrates as well as (exogenous) xenobiotics, like nutrients, plant ingredients, microbial toxins, or synthetic drugs. They are sometimes referred to as isoenzymes - a fairly inappropriate terminology because they do not always share the same catalytic reaction, i.e. neither mechanisms nor substrates. Nevertheless, its name P450 was properly attributed to these heme proteins alluding to the Protein's visible 450 nm spectral absorption line observed after chemical reduction and exposure to carbon monoxide gas for analytical characterization . Here we restricted the bibliographic search on cases related to drug biotransformation by hydroxylation of warfarin, phenprocoumon, flurbiprofen, ritonavir and clopidogrel.
Bibliographic Evidence for Hydroxylation Reaction Mechanism
Reviewing the literature on biochemical and mechanistic aspects
Earlier studies about drug metabolism have focuses on the hydroxylation processes of aliphatic as well as aryl groups [2,4], for example the crystal structures of diclofenac-liganded rabbit CYP2C5 aromatic hydroxylation (PDB codes: 1NR6) . For details on the biochemical and mechanistic aspects further reading is recommended (Table 1).
Table 1: Listing of suggested publications to further the understanding of the biochemical and mechanistic aspects. View Table 1
A major obstruction to the determination of drug metabolites is time and costs. To this regard efforts to predict them computationally have become a complementary tool [2,34-38]. To date, few publications can be related to in silico methods of modeling the CYP2C9 metabolic pathways (Table 2). Mechanistic insight of the CYP2C9 oxidation for diclofenac, ibuprofen and warfarin was gained by a combination of conformational sampling by molecular dynamics and DFT-based QM and QM/MM modeling (CHARMM force field) .
Especially modeling the potential CYP-mediated regioselectivity for oxidative biotransformations is a daunting task since they belong to a group of proteins with structural and functional flexibilities like the nuclear receptors [60,61]. They may adopt different backbone and side chain conformations when bound to different ligands (induced fit by substrates, hormones or co-factors) [5,27,44,62-64]. Induced fit reflects changes in the molecular geometry upon ligand binding. Such conformational changes can be regarded as an efficient selection mechanism called "conformational selection" [65,66]. As a proof of concept the following concerning CYP2B4 should also hold in the case of CYP2C9 and therefore is literally cited here: "The CYP2B4 active site is able to accommodate small ligands by moving only a small number of side chains, suggesting that ligand reorientation is energetically favored over protein conformational changes for binding of these similarly sized molecules. Adjusting both protein conformation and ligand orientation in the active site gives CYP2B4 the flexibility to bind to the widest range of molecules, while also being energetically favorable" (cited from ). To reflect such protein flexibilities three approaches have been widely accepted: soft docking (deliberately reducing repulsion forces), managing (a few) side chain rotations (rotamer libraries), or using several structures with different conformations of the same protein or related enzymes .
In today's gold standard for evaluating (benchmarking tests) of regiospecificity prediction, the outcome should find the experimentally observed sites of metabolism (SoM) in the ligand structures among the first or second group of solution ranks (scored clusters) [2,54]. Some published limitations of the established computational approaches, however, "highlight the potential difficulties in creating broadly applicable metabolite prediction models" (text cited from ). In a seminal work Andreas Bender and colleagues wrote a synopsis in the field of CYP activity predictions, the implied tenets, computational tools thereof and their limitations . In 2017 a complementary review focused on QSAR tools to predict CYP - ligand interactions .
Reports on molecular mechanism of substrate hydroxylation
Monohydroxylation is a possible biotransformation mechanism on aromatic or even aliphatic scaffolds with an inert carbon - hydrogen bond. The mechanism via an epoxide intermediate was outlined in the specialized literature (Figure 1) [30,41,68-72].
Figure 1: Schematic representation of the two possible chemical reaction types for PPC or WFN. Aromatic hydroxylation (to the left) is shown with an epoxide intermediate state (see step 7a in Figure 2). Aliphatic hydroxylation occurs on the propyl or butyl side chains of their common scaffold [70,72]. View Figure 1
The enzyme's crystal structures show a central metal ion (Fe III) that is in a square coordination environment with four pyrrole moieties on the cyclic porphyrin system. An additional anionic thiolate group (cysteine) constitutes the fifth coordination site to form a square pyramidal geometry. The remaining sixth open coordination site can bind a dioxygen species which in turn is in proximity to the SoM for hydroxylation or epoxidation of substrate (Figure 1). The resulting substrate-enzyme complex is coupled with an oxygen transferring process (Figure 2). The heme iron is essentially electrophilic . During the central reaction step of CYP hydroxylation, an electrophilic oxygen moiety is interacting with the hydro-carbon fragment of the ligand to form a carbon-oxygen bond. Bond forming is swiftly followed by a transient stage to allow the oxygen to connect the adjacent carbon atom. The rearrangement (shift) is assisted by the hydrogen atom in presence of the lone pair electrons of the oxygen atom ending with the OH group insertion at the adjacent C- atom [69,74]. As in the case of other prosthetic groups the heme is noncovalently bound to its apoprotein (Figure 2). Iron (Fe III) is the metal center of its porphyrin ring system which functions as a cofactor for catalysis. Cationic Fe III is reduced to Fe II prior to its reception of molecular oxygen (steps 1, 2 and 3 in Figure 2). It is assumed that O2 is converted into a radical bioxygen species in exchange iron regains its initial oxidation state as Fe III for the sake of electronic bookkeeping (steps 4 and 5 in Figure 2). The substrate oxygenation, mostly epoxidation and hydroxylation does not take place until an activated monooxgen (or oxo) species is coordinated to Fe IV forming a cationic radical heme group (step 5 in Figure 2). The highly instable complex reacts swiftly with inert carbon atoms of alkyl chains or aryl rings of liganded substrates because the complex frees larger quantities of potential energy, providing (tunneling) the Arrhenius activation energy reaching the transition state on a defined hyperdimensional coordination path. Although the exact mechanism is still under substantial debate, it is safe to utter that those double bonds on alkenes or aromatic rings can readily undergo oxygen transfer (oxygenation) reactions (Figure 2). The aromatic hydroxylation is best explained by a radical pathway. Korzekwa, Swinney and Trager showed in isotope metabolic studies that a direct epoxidation of a aryl ring is most unlikely . Instead it can be assumed that an early tetrahedral intermediate (step 6 in Figure 2) leads swiftly to either an instable aryl epoxide (step 7a in Figure 2) or arene-on (step 7b in Figure 2). Both converge into a final product, a monohydroxylated metabolite (step 8 in Figure 2). Studies at the National Institute of Health, USA, with deuterium demonstrated that the deuterium-labeled hydrogen shifts to the adjacent "ortho" position of the ring. The "leaving group" upon rearrangement is either the original hydrogen atom in "ortho" position (H*) under basic reaction conditions, or, the deuterium-labeled hydrogen in acidic milieu. Of note, the so-called NIH shift also works fine with small organic substituents other than hydrogen (alias deuterium) .
Figure 2: Established view on the biochemical mechanism of aromatic hydroxylation by heme-containing cytochrome P450 oxygenases. Step 7 is branched passing through either an expoxide (7a) or a ketone (7b) intermediate of aromatic substrate (S). The hydrogen atom of the NIH shift (cyclic arrows in steps 7a or 6 leading to 7b) was labeled "D" (deuterium isotope experiments ). The aromatic oxygenation requires the assistance of flavin-coenzymes and NADPH cofactors. The active site heme is represented by its protoporphyrin ring system. Its central iron cation is labeled by its oxidation states (II, III, or IV). The product (8) encompasses hydroxylated phenprocoumon or warfarin metabolites among other drugs with a common aryl substructure (5). View Figure 2
Substrate selectivities and sites of hydroxylation metabolism
Given the chiral nature of the amino acids (L-isomers for higher plants and animals) it is not surprising that substrate oxygenation by (oxygenases = oxygen transfer, not oxidases = electron transfer) results in stereoselectivity (one over the other) if not stereospecificity (only one). Given the wealth of studies on drug biotransformations we focused warfarin (WFN) and Phenprocoumon (PPC) which prevent blood coagulation. Intriguingly, WFN constitutes an enantiomeric drug (R- and S- forms). In more general terms racemic drugs are stereoselectively metabolized, i.e. the isomers interact with CYP enzymes with different affinities, binding modes or reaction rates, all of which leads to distinct metabolic pathways to produce various metabolites in different amounts. Regioselectivity arises here as a consequence because multiple binding modes result in a structural pattern of distinct oxygenation sites (cf. site of metabolism, SoM for short). Phenprocoumon (PPC) is structurally and pharmacologically related to warfarin (WFN). Both belong to the 4-hydroxycoumarin class of oral anticoagulant drugs (see Figure 3). Mechanistically, phenprocoumon and warfarin inhibit the vitamin K-dependent carboxylation of glutamate to form an unusual posttranslational amino acid derivative which is needed in many blood clothing factors to modulate calcium binding. In their molecular pathway 4-hydroxycoumarins target an epoxide reductase which regains vitamin K through reduction of its epoxide form (see Figure 3) [77,78]. Intriguingly, the bacterial and human epoxide reductases function without a prosthetic heme group . The latter can be replaced (heme group imitation, mimics) under in vitro conditions in organic synthesis by heavy metal ions like iron (3+) which were found useful for epoxide catalysis [80,81].
WFN and PPC are chiral, synthetic compounds with a single asymmetric carbon atom C9. Both drugs are commercially used as racemates although it is clinically known that the S- form always acts stronger than the corresponding R-isomer when tested as anticoagulant agents [13,41,77]. In theory either pharmacodynamic (better affinity to the protein targeting the coagulation mechanism) or pharmacokinetic (ADME) reasons (better bioavailability and distribution or lower rate of metabolism and elimination) - or even both together - could cause the stronger activity for the respective (S)-forms. Commercial dosage forms deliver racemic drugs, albeit academic research has elucidated that the (R)-enantiomers showed invariably higher hepatic biotransformation rates over their (S)-isomers. Despite their chemical similarity metabolic fate of PPC and WFN diverges [41,82] (Table 3).
The unique structural difference between PPC and WFN is the side chain decoration of the 4-hydroxycoumarin scaffold. WFN possesses an acetonyl side chain attached to chiral carbon atom C9 whereas PPC shows a shorter ethyl side chain instead. In the literature, we found three mayor tenets concerning the substrate recognition: (1) Changes in the tautomeric behavior reflect the CYP2C9-based metabolic differences between PPC and WFN (see Figure 2 and Figure 3 in  as well as Figure 1 in ) . (2) Changes in the C9 side chain of PPC compared to WFN account for the divergent metabolic fate between PPC and WFN. Owing to the larger and branched side chain attached to C9 WFN cannot access the narrow cleft at the distal side of the heme group (see Figure 1 in ). Molecular dynamics studies were conducted to demonstrate on theoretical grounds the existence of an induced fit mechanism, i.e. main and side chain rearrangements of a liganded CYP compared to its unliganded (apo-form) . (3) Drugs with different size and shape and yet, were found complexed to the same CYP (e.g. CYP2C5, PDB codes: 1NR6 , 1N6B , 1DT6 ).
Of note, WFN possesses more tautomeric states than PPC. The differing electronic behaviors impacts in substrate recognition in a way how (long and strongly) tautomers bind or interact with CYP2C9 residues. Methoxyl derivatives of PPC were synthesized to "freeze" it into certain tautomeric forms and were also recognized as hydroxylation substrates by the CYP enzymes .
Bibliographic Evidence for Specific Hydroxylation Patterns
The reported hydroxylation metabolites of phenprocoumon
On a molecular level it was unraveled that CYP2C9 constitutes the main enzyme for PPC hydroxylation in hepatic tissue [83,91]. PPC hydroxylation rates were significantly correlated with liver enzymes CYP2C9 and CYP3A4 protein activities or gene expression. The influence of CYP2C9 genotypes decreases from 7-OH-PPC > 6-OH-PPC > 4´-OH-PPC = 2'-OH-PPC > "remaining"-OH-PPC. But for 4' hydroxylation there is already a notable contribution of CYP3A4 and CYP2C8 (see Figure 4).
Figure 4: Localization (arrows) of the experimentally known sites of metabolism (monohydroxylation) on each enantiomer of PPC (S-Left; R-Right). The arrow labels indicate the metabolizing cytochrome P450 enzyme, thick arrows symbolize higher rates [41,83]. The pharmacokinetics tool box ADMET predictor assessed the following hydroxylation sites: PPC: 6, 7, 8 and 4' by CYP2C9; 6, 7, 9 and 4' by CYP3A4. The position 9 was predicted, too, in contrast to the extant literature. View Figure 4
Concerning the stereoselectivity CYP2C9 and CYP3A4 constitute the major catalysts of (S) - and (R)-PPC hydroxylation, whereas CYP2C8 partly catalyzed the (S)-C4'- hydroxylation. Moreover CYP2C9 acts as a major catalyst of the C6- and C7-hydroxylation for both enantiomers . However, CYP2C8 was equally important regarding the (S)-C4'- hydroxylation. Two or more biocatalysts are involved in the C4'-, C6- and C7-hydroxylation of both enantiomers of PPC . In the liver CYP2C9 also mediates the biotransformation of the oral anticoagulants S-warfarin and its nitro derivative R- and S-acenocoumarol . The role of CYP2C9 has previously been found to be of greatest importance for the metabolism of the more potent (S)-enantiomer of warfarin . CYP2C9 activity for drugs falls in the order: warfarin > acenocoumarol > phenprocoumon . Clinical studies proofed that the anticoagulant potency of (R)-PPC is much lower than that of its (S)-enantiomer . In 2005 Kammerer and Ufer, et al.  demonstrated that (R)-PPC was detectable in higher concentrations than its (S)-enantiomer in blood and liver samples . Apparently (R)-PPC is less metabolized. In good keeping, Ufer and coworkers revealed in kinetic experiments that the C7 hydroxylation rate of (S)-PPC was extremely high [83,92].
Besides 8-OH-PPC  which is not present in the blood stream in measurable amounts, five monohydroxylated metabolites were detected and their amounts quantified in a chromatogram (decreasing AUC: 2´-OH-PPC > 7-OH-PPC >>> 4´-OH-PPC = 6-OH-PPC > "?"-OH-PPC (see Figure 2 in ). Intriguingly, CYP2C19 is the main enzyme for C8 hydroxylation of WFN (Km = 0.3 mM)  compared to 1A2 and 1A1 with lower Km values (1.4 mM and 1.2 mM, respectively) . Until recently, no further metabolite has been detected for the phase-I metabolism of PPC .
The reported regioselective metabolites of warfarin
In 1997 the stereospecific metabolic pathways for the R- and S-enantiomers of WFN became known and involved CYP3A4 and other isozymes (Figure 5) [2,82]. Especially, CYP2C9 mainly metabolizes the more potent S-enantiomer of WFN . To look for differences the published data on hydroxylation patterns of WFN was compared to those of PPC (Figure 4 and Figure 5) .
Figure 5: Localization (arrows) of the experimentally known sites of metabolism (monohydroxylation) on each enantiomer of WFN (S-Left; R-Right). The arrow labels indicate the metabolizing cytochrome P450 enzyme, thick arrows symbolize higher rates . The pharmacokinetics tool box ADMET predictor assessed the following hydroxylation sites: WFN: 6, 7 and 4' by CYP2C9; 10 by CYP3A4. The position 9 was predicted, too, in contrast to extant literature. View Figure 5
Comparison between 4-hydroxycoumarins
In addition to the hydroxylation of PPC and WFN, CYP2C9 hydroxylates acenocoumarol which is a closely related anticoagulant: A nitro derivative with the same 4-hydroxycoumarin scaffold (Figure 3) . CYP2C9 activity for drugs falls in the order: warfarin > acenocoumarol > phenprocoumon . Intriguingly, pharmacokinetics studies showed drug clearance decreases in the order WFN > acenocoumarol > PPC . It was also found that mutational changes (genetic polymorphisms of patients) in the primary sequence of CYP2C9 do influence the PPC elimination rate to a lesser extent than can be expected. To explanations are possible, either CYP3A4 assists the catalysis into hydroxylated metabolites or unchanged PPC is eliminated in higher amounts . The hepatic PPC metabolism was quantitatively characterized and its metabolite (S)-7-OH-PPC was found at highest rates while monoclonal antibody studies revealed among the CYP 2C subfamily (with 2C8, 2C9, 2C18 and 2C19) that CYP2C9 was the most active "hydroxylator" for positions 6- and 7 of (R)- as well as (S)- enantiomers .
Bibliographic Evidence for CYP Regioselectivity
Structural knowledge of CYP2C9 enzyme
After many years of experimental studies, the structure-dependent functions of CYP2C9 activities have been summarized in a seminal work [49,94]. The earlier findings about substrate recognition by CYP2C9 still hold in sight of today's under-standing. It became clear that an intramolecular combination of anionic and hydrophobic features forged the enzyme's preferences and substrate specificities [19,95]. With the advent of crystal structures molecular modeling has been applied in the field of cytochrome P450 enzymes. In 1987 Poulos published the crystal structure of a microbial cytochrome P450. It was co-crystallized with camphor as a substrate attacked by a heme-bound oxygen species for oxidation . Camphor elucidated the way how even a hydrogen - carbon bond on an alkyl chain could be attacked by a heme-coordinated oxygen species [4,6]. Further release of microbial crystal structures paved the way toward a deeper understanding of the mechanistic aspects between heme iron, reactive oxygen and substrate (PDB codes: 1NOO , 4C9N ). A few years after Poulos' seminal work of 1987 a first computational model dealt with CYP2C9 substrate recognition. The model established a general view on the binding features and geometries: (1) A preferentially anionic substrate which lies (2) At a distance of about 0.4 nm away from (3) A postulated cationic binding site. The third landmark constitutes the hydroxylation site which forms (4) An almost orthogonal angle with the anionic and cationic sites. (5) The distance between the anionic and hydroxylation sites was found to range somewhere between 0.7 to 1.0 nm .
In 1999 the regioselective 4-hydroxycoumarin hydroxylation by CYP2C9 was reported: carbon atoms with id numbers 6 and 7 on (S)-WFN and on (R)-WFN but to a very small amount. The two enantiomeric forms of PPC were found hydroxylated in positions 6, 7, 8 and 4' .
Site directed mutagenesis studies pinpointed binding-relevant amino acids and more crystal structures were inspected to infer general rules . Recently, in 2013, QSAR studies reflected that "CYP2C9 showed the selectivity towards slightly acidic compounds, large molecular mass, large polar surface area, and larger number of hydrogen-bond acceptor atoms" (cited from ). Recently, over 40 computational methods or tools were reviewed and listed which have an applicability domain for the biochemical field of CYP metabolism . After collecting the literature data we carried out sequence alignments and structure superpositions and compared the results to the literature descriptions for substrate - enzyme complexes. The differential amino acids of the CYP sequences were identified by Multiple Sequence Alignment (MSA) studies (Figure 6) .
Figure 6: The sequence of CYP2C9 is aligned and compared to other CYPs which are relevant in 4-hydroxycoumarins metabolism. Lower case letters stand for one letter codes of amino acids (aa). Upper case letters symbolize the aa of the catalytic site. Underlined upper case letters represent the aa in the catalytic site interacting with the PPC metabolites.
Line 1: Cytochrome P450 family 2 subfamily C polypeptide 19 [Homo sapiens]. NCBI Reference Sequence: NP_000760.1.
Line 2: Cytochrome P450 family 2 subfamily C polypeptide 8 [Homo sapiens]. NCBI Reference Sequence: NP_000761.3
Line 3: Cytochrome P450 family 2 subfamily D polypeptide 6 [Homo sapiens]. NCBI Reference Sequence: NP_000097.3
Line 4: Cytochrome P450 family 1 subfamily A polypeptide 2 [Homo sapiens]. NCBI Reference Sequence: NP_000752.2
Line 5: Cytochrome P450 family 3 subfamily A polypeptide 4 [Homo sapiens]. NCBI Reference Sequence: NP_059488.2
Line 6: Cytochrome P450 family 2 subfamily C polypeptide 9 [Homo sapiens]. NCBI Reference Sequence: NP_000762.2.
Line 7: Secondary structure of CYP2C9, alpha helices A, beta sheets B, loops L.
Line 8: Identity of secondary structure of CYP2C9.
Line 9: Homology symbols: "*" identical residue; ":" conserved residue; "." weakly similar residue. View Figure 6
Important for the regioselectivity is N297 the role of which was elucidated for the oxidation of coumarin at the outspokenly hydrophobic site of human CYP2A6 (PDB code: 1Z10 ). In addition, flurbiprofen and other ligand complexes (PDB codes: 1R9O , 1Z10 , 1Z11 ) or a rabbit protein (CYP2B4) in complex with (S)-clopidogrel (PDB code: 3ME6 ) were also studied. The binding position of the chlorophenyl ring and orientation of clopidogrel were inspected to compare the start positions for phenyl ring hydroxylation simulations. The structure alignments helped observing the conformational flexibilities necessary for molecular recognition between substrates and protein according to the established tenet concerning "conformational selections" .
Bibliographic Evidence for Induced Fit with CYP
Structural knowledge about induced fit for CYP enzymes
To find examples of induced fit scenarios we gathered crystal structures of other CYP proteins with ligand interacting residues. We found crystal complexes of CYP3A4 with ligands structurally related to WFN and PPC in terms of overall molecular volume, shape and aromatic rings (Table 4).
We collected pairs of crystal structures of liganded CYP complexes and their respective apo-forms (unliganded) to observe the induced fit phenomena upon ligand binding to the active site (Table 4). To our surprise not many suited database entries were found despite the plethora of known CYP complexes. Only one such pair (CYP 3A4 with ritonavir, PDB Codes: 1TQN , 3NXU ) gave a clear evidence of induced fit. After superposing the liganded form onto the apo-form (by Vega ZZ ), the ligand clearly overlapped with side chains of the apo-form (Table 5).
Table 5: Structures comparison between unliganded and liganded complexes for different pairs of CYPs. If side chain atoms of the apo (unliganded) forms clash into the ligands, this can be interpreted as the necessity for conformational rearrangements to avoid such clashes. In other words, backbone and side chain rearrangements in the liganded complex when compared to the apo form, can be interpreted in terms of an induced fit phenomenon. View Table 5
Revision of induced fit and regioselectivity for CYP2C9 and CYP3A4
CYP3A4 hydroxylates WFN at C10 . The same position on PPC was studied to evaluate its possible hydroxylation by CYP2C9. While WFN should dock into CYP3A4 in a pose allowing for C10 hydroxylation, CYP2C9 should not. The superposition of both P450 enzymes revealed how the differential side chains favor or disfavor the C10- hydroxylation on the scaffold of WFN but not of PPC (see Figures 5, Figure 7 and Figure 8) . The related CYP3A4 complexes were also studied (Table 4).
Figure 7: Superposition of CYP3A4 (silver) and CYP2C9 (gold). The red backbone represents the segments of CYP2C9 with a reduced catalytic site; the green backbone of CYP3A4 displays all the segments which correspond to the reduced catalytic site space of CYP2C9. The heme-bound (R) - warfarin (yellow) in the position for C10- hydroxylation can only exist in case of CYP3A4. The steric hindrance of CYP2C9 can be observed by the red line in close contact to the yellow ligand warfarin. The heme group (grey colors) is hardly visible (upper left corner). View Figure 7
Figure 8: Superposition of CYP3A4 (silver) and CYP2C9 (gold). The red sequences are the segments of CYP2C9 that reduce the space at the active site; in green are the sequences of CYP3A4 allowing a wider cavity. Yellow is Warfarin. The figure displays a rotated view by 90 degrees in x and z axes. L366 of CYP2C9 in red is the amino acid responsible of the main reduction of space while its analog E374 in CYP3A4, in green color, together with its shifted backbone is turned outwards allowing the widening of the catalytic site cavity. The heme group is hardly visible (central background). View Figure 8
In the multiple sequence alignment (MSA) studies the interacting (differential) amino acids at the catalytic site are marked in upper case letters (Figure 6). Enzymes CYP2C9 and CYP3A4 differ in the side chains responsible for ligand binding (lines 5 and 6 in Figure 6). In particular, S365, L366, P367, H368 are reducing the catalytic site volume by 4 × 15 Å3 through backbone rotations towards the Fe-heme group. In addition, main chain rotation together with residue changes in P211, I213, Q214, V215, and P220 reduce the space by forming a flap into the core region of the catalytic site. Finally, F100, P101, L102 cause a similar main chain shift and spatial changes. Apparently, the proline residues are not conserved throughout the isozymes' superfamily members and therefore account for the individual backbone kinks (Figure 9).
Figure 9: Display of the (S)-enantiomer of clopidogrel - heme complex which served as a 3D template for orientating the ligand and positioning the equivalent phenyl rings of PPC (and WFN) during manual docking (PDB code: 3ME6 ). Color code: O red, N blue, C grey, Fe orange, S yellow and Cl green. View Figure 9
Comparing the sequences of CYP2C9 to CYP3A4 and visualizing their 3D models revealed their fairly low sequence similarity which comes as a surprise considering that they recognize similar substrates and perform the hydroxylation at positions C6-, C7- and C4'- carbon atoms of a common scaffold. In contrast to the far higher similarity of their primary sequences, CYP2C9 and CYP2C8 do not share the same substrates, e.g. while CYP2C9 metabolizes PPC, CYP2C8 does not (Figure 4 and Figure 6) . Human recombinant CYP2C9 catalyzes the formation of PPC to four different monohydroxylated metabolites in the C4'-, C6-, C7- and C8- positions on (S)-PPC. Earlier protein kinetic studies revealed (S)-7- hydroxylation of PPC as quantitatively most important [41,83]. This is in good keeping with diclofenac a similar case of high regioselectivity of aromatic hydroxylation by CYP2C9. Intriguingly, like CYP2C9, another enzyme, here CYP2C5, acts with high regioselectivity of aromatic hydroxylation to produce the same para-hydroxy-dichlorophenyl moiety . In addition CYP2C9 constitutes a major catalyst of the C6- and C7-hydroxylation for both enantiomers (Figure 10 and Figure 11) . In the case of PPC C2' docked solutions reflected conformations other than the available crystal structures all of which indicated induced fit because other protein conformations were needed than the crystal ones .
Figure 10: FlexX docking  of PPC into the cavity of human CYP2C9 after removing of the co-crystalized ligand flurbiprofen (PDB code: 1R9O ) [84,92]. The accessibility of the 6- and 7- positions to the heme-based oxygen atom for aryl ring hydroxylation becomes evident. The final pose of PPC is displayed for its (R)-enantiomeric, dominating tautomeric and undissociated forms. The cavity surface is color coded: blue (hydrophilic), over green (neutral) to brownish (lipophilic). Atom color code: white C; blue N; red O; light blue H (only shown on PPC) . View Figure 10
Figure 11: FlexX docking  of undissociated (R)-PPC in its dominating tautomeric form into the cavity of human CYP2C9 after removing of the co-crystalized ligand flurbiprofen (PDB code: 1R9O ) . The accessibility of the meta-and para-phenyl positions (C3' and C4') to the heme-based oxygen atom for the phenyl ring hydroxylation becomes evident. Color code in Figure 10 . View Figure 11
Specialized literature was studied and revised along with computational modeling to obtain molecular insight into the P450 biotransformation of substrates. We focused on warfarin and phenprocoumon, and to a lesser extent on flurbiprofen, ritonavir and clopidogrel to reflect the documented regioselectivities of metabolite hydroxylation for some cytochrome P450 enzymes, mainly CYP2C9 and CYP3A4. In addition, many reports on flexible protein conformations upon ligand binding (induced fit) have been found but very few articles contained structural evidences. Induced fit was made visible upon super-positioning of liganded and unliganded crystal geometries of the same CYP enzyme.
In more general terms the reviewed literature about regioselectivities and induced fit reflect the difficulty to report on nature's evolutionary changes: Variable environment, variable habitats and food intake has ever since favored sequence changes in nonfunctional parts over functional parts (i.e. entrance vestibules vs. oxidation sites). Some mutations have modified the cavity topology - and with it - the steric requirements for xenobiotics to be recognized as substrates for oxidation and subsequent hydrophilic (urinal) elimination. In need of adapting or improving activity range of living species (for vertebrates) their habitats have been changed. In contrast, the catalytic site and with it the electronic ground work of the multi-step oxygenation process has remained fundamentally the same (conserved amino acids) . Here it was possible to lay the knowledge bases for our recently published findings by applying ligand docking and molecular dynamics in 2018 [120, 121]. Yet, in the near future, publications with new data might present their preselective role in the living cells in addition to the factors of molecular flexibility, shape and size dependencies that we modeled here. Limiting factors were also met and concerned the need of dedicated input data (more CYP2C9 crystal structures). Precisely, to assess the contributions to the substrate recognition by induced fit mechanisms only little insight has been reported due to missing pairs of liganded/unliganded CYP complexes for the mechanism under scrutiny.
Conflict of Interest
The authors report no conflict of interest or having received any kind of financial contributions of funding for the study.
Graduate student Israel Quiroga is grateful for both CONACyT grants (graduate studies of Master in Science and PhD degree) during 2013 to 2018. We feel very much beholden to CA-120 and VIEP at BUAP for support, and Prof Dr Enrique Meléndez, University of Puerto Rico, Department of Chemistry, PO Box 9019, Mayagüez, PR 00681, USA, for discussion. The authors thankfully acknowledge computer resources, technical expertise and support provided by Laboratorio Nacional de Supercómputo del Sureste de México, which is a member of the CONACyT network of national laboratories.
- D Korolev, KV Balakin, Y Nikolsky, et al. Modeling of human cytochrome P450-mediated drug metabolism using unsupervised machine learning approach. J Med Chem 2003;46:3631-43.
- J Zaretzki, C Bergeron, P Rydberg, TW Huang, KP Bennett, CM Breneman. RS-Predictor: A new tool for predicting sites of cytochrome P450-mediated metabolism applied to CYP 3A4. J Chem Inf Model 2011;51:1667-89.
- J Hodgson. ADMET-turning chemicals into drugs. Nat Biotechnol 2001;19:722-6.
- PR Ortiz de Montellano. Hydrocarbon hydroxylation by cytochrome P450 enzymes. Chem Rev 2010;110:932-48.
- MR Wester, EF Johnson, C Marques-Soares, et al. Structure of mammalian cytochrome P450 2C5 complexed with diclofenac at 2.1 Å resolution: Evidence for an induced fit model of substrate binding. Biochemistry 2003;42:9335-45.
- TL Poulos, BC Finzel, AJ Howard. High-resolution crystal structure of cytochrome P450cam. J Mol Biol 1987;195:687-700.
- R Raag, TL Poulos. Crystal structure of the carbon monoxide-substrate-cytochrome P-450CAM ternary complex. Biochemistry 1989;28:7586-92.
- H Li, S Narasimhulu, LM Havran, JD Winkler, TL Poulos. Crystal structure of cytochrome P450cam complexed with its catalytic product, 5-exo-hydroxycamphor. Journal of the American Chemical Society 1995;117:6297-9.
- C Bathelt, RD Schmid, J Pleiss. Regioselectivity of CYP2B6: Homology modeling, molecular dynamics simulation, docking. J Mol Model 2002;8:327-35.
- PR Ortiz de Montellano, JJ De Voss. Oxidizing species in the mechanism of cytochrome P450. Nat Prod Rep 2002;19:477-93.
- M Azevedo, VC Martins, DM Prazeres, V Vojinović, JM Cabral, LP Fonseca. Horseradish peroxidase: A valuable tool in biotechnology. Biotechnol Annu Rev 2003;9:199-247.
- X He, PR Ortiz de Montellano. Radical rebound mechanism in cytochrome P-450-catalyzed hydroxylation of the multifaceted radical clocks α-and β-thujone. J Biol Chem 2004;279:39479-84.
- M Ufer. Comparative pharmacokinetics of vitamin K antagonists. Clin Pharmacokinet 2005;44:1227-46.
- M Fischer, M Knoll, D Sirim, F Wagner, S Funke, J Pleiss. The Cytochrome P450 Engineering Database: A navigation and prediction tool for the cytochrome P450 protein family. Bioinformatics 2007;23:2015-7.
- SB. Nabuurs, M Wagener, J De Vlieg. A flexible approach to induced fit docking. J Med Chem 2007;50:6507-18.
- DF Lewis, Y Ito. Cytochrome P450 structure and function: An evolutionary perspective. In: Cytochromes P450, RSC, 2008;3-45.
- A Seifert, J Pleiss. Identification of selectivity?determining residues in cytochrome P450 monooxygenases: A systematic analysis of the substrate recognition site 5. Proteins 2009;74:1028-35.
- JA Pavon. Mechanism of oxygen activation and hydroxylation by the aromatic amino acid hydroxylases. Doctoral dissertation, Texas A&M University, 2009.
- D Sirim, M Widmann, F Wagner, J Pleiss. Prediction and analysis of the modular structure of cytochrome P450 monooxygenases. BMC Struct Biol 2010;10:34-46.
- L Zhang. Heme Biology: The secret life of heme in regulating diverse biological processes. World Scientific Publishing, Singapore, 2011.
- P Vasanthanathan, J Lastdrager, C Oostenbrink, et al. Identification of CYP1A2 ligands by structure-based and ligand-based virtual screening. Med Chem Comm 2011;2:853-9.
- A Cecil, C Rikanovic, K Ohlsen, et al. Modeling antibiotic and cytotoxic effects of the dimeric isoquinoline IQ-143 on metabolism and its regulation in Staphylococcus aureus, Staphylococcus epidermidis and human cells. Genome Biol 2011;12:24.
- L Quintieri, P Palatini, MO Stefano, M Floreani. Inhibition of cytochrome P450 2C8-mediated drug metabolism by the flavonoid diosmetin. Drug Metab Pharmacokinet 2011;26:559-68.
- E Meléndez. Metallocenes as target specific drugs for cancer treatment. Inorganica Chim Acta 2012;393:36-52.
- KM Kadish, KM Smith, R Guilard. The porphyrin handbook. World Scientific Publishing, Singapore, 2012.
- VB Urlacher, M Girhard. Cytochrome P450 monooxygenases: An update on perspectives for synthetic application. Trends Biotechnol 2012;30:26-36.
- A Seifert, J Pleiss. Identification of selectivity determinants in CYP monooxygenases by modelling and systematic analysis of sequence and structure. Curr Drug Metab 2012;13:197-202.
- J Kirchmair, MJ Williamson, JD Tyzack, et al. Computational prediction of metabolism: Sites, products, SAR, P450 enzyme dynamics, and mechanisms. J Chem Inf Model 2012;52:617-48.
- D Batabyal, TL Poulos. Crystal structures and functional characterization of wild-type CYP101D1 and its active site mutants. Biochemistry 2013;52:8898-906.
- KH Tiong, NAM Yunus, BC Yiap, EL Tan, R Ismail, CE Ong. Inhibitory potency of 8-methoxypsoralen on cytochrome P450 2A6 (CYP2A6) allelic variants CYP2A6* 15, CYP2A6* 16, CYP2A6* 21 and CYP2A6* 22: Differential susceptibility due to different sequence locations of the mutations. PLoS One 2014;9:e86230.
- Ł Gricman, C Vogel, J Pleiss. Conservation analysis of class?specific positions in cytochrome P450 monooxygenases: functional and structural relevance. Proteins 2014;82:491-504.
- VV Shumyantseva, AA Makhova, TV Bulko, et al. Electrocatalytic cycle of P450 cytochromes: the protective and stimulating roles of antioxidants. RSC Adv 2015;5:71306-13.
- N Wattanachai, S Kaewmoongkun, W Tassaneeyakul. Development of a pharmacogenetics-based warfarin dosing algorithm for Thai patients. Drug Metabolism and Pharmacokinetics 2017;32.
- A Vedani, M Dobler, M Smieško. VirtualToxLab-A platform for estimating the toxic potential of drugs, chemicals and natural products. Toxicol Appl Pharmacol 2012;261:142-53.
- EJ Mendieta-Wejebe, J Correa-Basurto, EM Garcia-Segovia, G Ceballos-Cancino, MC Rosales-Hernandez. Molecular modeling used to evaluate CYP2C9-dependent metabolism: homology modeling, molecular dynamics and docking simulations. Curr Drug Metab 2011;12:533-48.
- MC Rosales-Hernández, JE Mendieta-Wejebe, JG Trujillo-Ferrara, J Correa-Basurto. Homology modeling and molecular dynamics of CYP1A1 and CYP2B1 to explore the metabolism of aryl derivatives by docking and experimental assays. European Journal of Medicinal Chemistry 2010;45:4845-55.
- JE Mendieta-Wejebe, MC Rosales-Hernández, H Rios, et al. Comparing the electronic properties and docking calculations of heme derivatives on CYP2B4. J Mol Model 2008;14:537-45.
- M Bello, JE Mendieta-Wejebe, J Correa-Basurto. Structural and energetic analysis to provide insight residues of CYP2C9, 2C11 and 2E1 involved in valproic acid dehydrogenation selectivity. Biochem Pharmacol 2014;90:145-58.
- A Mancy, P Broto, S Dijols, PM Dansette, D Mansuy. The substrate binding site of human liver cytochrome P450 2C9: an approach using designed tienilic acid derivatives and molecular modeling. Biochemistry 1995;34:10365-75.
- VA Payne, YT Chang, GH Loew. Homology modeling and substrate binding study of human CYP2C9 enzyme. Proteins 1999;37:176-90.
- M He, KR Korzekwa, JP Jones, AE Rettie, WF Trager. Structural forms of phenprocoumon and warfarin that are metabolized at the active site of CYP2C9. Arch Biochem Biophys 1999;372:16-28.
- DF Lewis, M Dickins, BG Lake, PS Goldfarb. Investigation of enzyme selectivity in the human CYP2C subfamily: homology modelling of CYP2C8, CYP2C9 and CYP2C19 from the CYP2C5 crystallographic template. Drug Metabol Drug Interact 2003;19:257-85.
- L Afzelius. Computational modelling of structures and ligands of CYP2C9. Uppsala, 2004.
- A Seifert, S Tatzel, RD Schmid, J Pleiss. Multiple molecular dynamics simulations of human p450 monooxygenase CYP2C9: the molecular basis of substrate binding and regioselectivity toward warfarin. Proteins 2006;64:147-55.
- RR Ayscue, Primera. Computer modeling of dapsone-mediated heteroactivation of flurbiprofen metabolism by CYP2C9. ProQuest, Morgantown, 2008.
- A Belic, M Temesvari, K Kohalmy, et al. Investigation of the CYP2C9 induction profile in human hepatocytes by combining experimental and modelling approaches. Curr Drug Metab 2009;10:1066-74.
- A Vedani, M Smiesko. In silico toxicology in drug discovery-concepts based on three-dimensional models. Altern Lab Anim 2009;37:477-96.
- MW Linder, MB Homme, KK Reynolds, et al. Interactive modeling for ongoing utility of pharmacogenetic diagnostic testing: application for warfarin therapy. Clin Chem 2009;55:1861-8.
- SF Zhou, ZW Zhou, LP Yang, JP Cai. Substrates, inducers, inhibitors and structure-activity relationships of human Cytochrome P450 2C9 and implications in drug development. Curr Med Chem 2009;16:3480-675.
- G Rossato, B Ernst, M Smiesko, M Spreafico, A Vedani. Probing small?molecule binding to cytochrome P450 2D6 and 2C9: An in silico protocol for generating toxicity alerts. ChemMedChem 2010;5:2088-101.
- á Tarcsay, GM Keserű. In silico site of metabolism prediction of cytochrome P450-mediated biotransformations. Expert Opin Drug Metab Toxicol 2011;7:299-312.
- RC Braga, VM Alves, CA Fraga, EJ Barreiro, V de Oliveira, CH Andrade. Combination of docking, molecular dynamics and quantum mechanical calculations for metabolism prediction of 3,4-methylenedioxybenzoyl-2-thienylhydrazone. J Mol Model 2012;18:2065-78.
- S Ekins, G Bravi, JH Wikel, SA Wrighton. Three-dimensional-quantitative structure activity relationship analysis of cytochrome P-450 3A4 substrates. J Pharmacol Exp Ther 1999;291:424-33.
- JD Tyzack, HY Mussa, MJ Williamson, J Kirchmair, RC Glen. Cytochrome P450 site of metabolism prediction from 2D topological fingerprints using GPU accelerated probabilistic classifiers. J Cheminform 2014;6:29.
- KA Ford, G Ryslik, J Sodhi, et al. Computational predictions of the site of metabolism of cytochrome P450 2D6 substrates: comparative analysis, molecular docking, bioactivation and toxicological implications. Drug Metab Rev 2015;47:291-319.
- YP Chang, CC Huang, CC Shen, KC Tsai, YF Ueng. Differential inhibition of CYP1-catalyzed regioselective hydroxylation of estradiol by berberine and its oxidative metabolites. Drug Metab Pharmacokinet 2017;30:374-83.
- T Niwa, N Murayama, Y Imagawa, H Yamazaki. Comparison of catalytic and inhibitory properties in human drug-metabolizing cytochrome P450 3A4 and 3A5 for various compounds including endogenous steroid hormones and azole antifungals by molecular docking simulation. Drug Metabolism and Pharmacokinetics 2017;32.
- H Kato, N Yamaotsu, N Iwazaki, S Okamura, T Kume, S Hirono. Precise prediction of activators for the human constitutive androstane receptor using structure-based three-dimensional quantitative structure-activity relationship methods. Drug Metabolism and Pharmacokinetics 2017;32:179-88.
- R Lonsdale, K Houghton, J Żurek, et al. Quantum mechanics/molecular mechanics modeling of regioselectivity of drug metabolism in cytochrome P450 2C9. J Am Chem Soc 2013;135:8001-15.
- L Perić-Hassler, E Stjernschantz, C Oostenbrink, DP Geerke. CYP 2D6 binding affinity predictions using multiple ligand and protein conformations. Int J Mol Sci 2013;14:24514-30.
- T Scior, M Verhoff, I Gutierrez-Aztatzi, HP Ammon, S Laufer, O Werz. Interference of boswellic acids with the ligand binding domain of the glucocorticoid receptor. J Chem Inf Model 2014;54:978-86.
- J Skopalík, P Anzenbacher, M Otyepka. Flexibility of human cytochromes P450: molecular dynamics reveals differences between CYPs 3A4, 2C9, and 2A6, which correlate with their substrate preferences. J Phys Chem B 2008;112:8165-73.
- B Sotriffer. Accounting for induced-fit effects in docking: what is possible and what is not? Curr Top Med Chem 2011;11:179-91.
- X Yu, V Cojocaru, RC Wade. Conformational diversity and ligand tunnels of mammalian cytochrome P450s. Biotechnol Appl Biochem 2013;60:134-45.
- P Chakraborty, E Di Cera. Induced fit is a special case of conformational selection. Biochemistry 2017;56:2853-9.
- HN Motlagh, JO Wrabl, J Li, VJ Hilser. The ensemble nature of allostery. Nature 2014;508:331-9.
- SC Gay, AG Roberts, K Maekawa, et al. Structures of cytochrome P450 2B4 complexed with the antiplatelet drugs ticlopidine and clopidogrel. Biochemistry 2010;49:8709-20.
- M Remko, R Broer, A Remková. A comparative study of the molecular structure, lipophilicity, solubility, acidity, absorption and polar surface area of coumarinic anticoagulants and direct thrombin inhibitors. RSC Advances 2014;4:8072-84.
- FP Guengerich. Cytochrome P450s and other enzymes in drug metabolism and toxicity. AAPS J 2006;8:E101-11.
- A Crocoll. Biosynthesis of the phenolic monoterpenes, thymol and carvacrol, by terpene synthases and cytochrome P450s in oregano and thyme. der Biologisch-Pharmazeutischen Fakultät der Friedrich-Schiller-Universität, Jena, 2011.
- K Czarnecki, S Nimri, Z Gross, LM Proniewicz, JR Kincaid. Direct resonance raman evidence for a trans influence on the ferryl fragment in models of compound i intermediates of heme enzymes. J Am Chem Soc 1996;118:2929-35.
- AE Shilov, GB Shul'pin. Activation of CH bonds by metal complexes. Chem Rev 1997;97:2879-932.
- H Becker. Organikum: organisch-chemisches Grundpraktikum. Deutscher Verlag der Wissenschaften, Berlin, 1967.
- JW Daly, DM Jerina, B Witkop. Arene oxides and the NIH shift: the metabolism, toxicity and carcinogenicity of aromatic compounds. Experientia 1972;28:1129-49.
- KR Korzekwa, DC Swinney, WF Trager. Isotopically labeled chlorobenzenes as probes for the mechanism of cytochrome P-450 catalyzed aromatic hydroxylation. Biochemistry 1989;28:9019-27.
- G Guroff, J Daly, D Jerina, J Renson, B Witkop, S Udenfriend. Hydroxylation-induced migration: the NIH shift. Recent experiments reveal an unexpected and general result of enzymatic hydroxylation of aromatic compounds. Science 1967;157:1524-30.
- H Auterhoff, J Knabe, HD Höltje. Lehrbuch der pharmazeutischen Chemie. Wissenschaftliche verlagsgesellschaft, Stuttgart, 1991.
- I Sugiura, B Furie, CT Walsh, BC Furie. Propeptide and glutamate-containing substrates bound to the vitamin K-dependent carboxylase convert its vitamin K epoxidase function from an inactive to an active state. Proc Natl Acad Sci USA 1997;94:9069-74.
- R Wallin, SM Hutson, D Cain, A Sweatt, DC Sane. A molecular mechanism for genetic warfarin resistance in the rat. FASEB J 2001;15:2542-4.
- B Plietker, Primera. Iron catalysis in organic chemistry: Reactions and applications. Stuttgart: John Wiley & Sons, 2008.
- W Li, S Schulman, RJ Dutton, D Boyd, J Beckwith, TA Rapoport. Structure of a bacterial homologue of vitamin K epoxide reductase. Nature 2010;463:507-12.
- LS Kaminsky, ZY Zhang. Human P450 metabolism of warfarin. Pharmacol Ther 1997;73:67-74.
- M Ufer, JO Svensson, KW Krausz, HV Gelboin, A Rane, G Tybring. Identification of cytochromes P 450 2C9 and 3A4 as the major catalysts of phenprocoumon hydroxylation in vitro. Eur J Clin Pharmacol 2004;60:173-82.
- B Kammerer, R Kahlich, M Ufer, A Schenkel, S Laufer, CH Gleiter. Stereospecific pharmacokinetic characterisation of phenprocoumon metabolites, and mass-spectrometric identification of two novel metabolites in human plasma and liver microsomes. Anal Bioanal Chem 2005;383:909-17.
- TH Sullivan-Klose, BI Ghanayem, DA Bell, et al. The role of the CFP2C9-Leu 359 allelic variant in the tolbutamide polymorphism. Pharmacogenetics 1996;6:341-9.
- ZY Zhang, MJ Fasco, L Huang, FP Guengerich, LS Kaminsky. Characterization of purified human recombinant cytochrome P4501A1-Ile462 and-Val462: assessment of a role for the rare allele in carcinogenesis. Cancer Res 1996;56:3926-33.
- LC Wienkers, CJ Wurden, E Storch, KL Kunze, AE Rettie, WF Trager. Formation of (R)-8-hydroxywarfarin in human liver microsomes. A new metabolic marker for the (S)-mephenytoin hydroxylase, P4502C19. Drug Metab Dispos1996;24:610-4.
- AE Rettie, KR Korzekwa, KL Kunze, et al. Hydroxylation of warfarin by human cDNA-expressed cytochrome P-450: a role for P-4502C9 in the etiology of (S)-warfarin-drug interactions. Chem Res Toxicol 1992;5:54-9.
- A Cairns. Essentials of pharmaceutical chemistry. Pharmaceutical Press, Padstow, 2012.
- PA Williams, J Cosme, V Sridhar, EF Johnson, DE McRee. Mammalian microsomal cytochrome P450 monooxygenase: structural adaptations for membrane binding and functional diversity. Mol Cell 2000;5:121-31.
- J Kirchheiner, M Ufer, EC Walter, et al. Effects of CYP2C9 polymorphisms on the pharmacokinetics of R-and S-phenprocoumon in healthy volunteers. Pharmacogenetics 2004;14:19-26.
- M Ufer, B Kammerer, R Kahlich, et al. Genetic polymorphisms of cytochrome P450 2C9 causing reduced phenprocoumon (S)-7-hydroxylation in vitro and in vivo. Xenobiotica 2004;34:847-59.
- Y van Leeuwen. Towards Improvement. Primera, Enschede: Gildeprint, 2009.
- D Van Booven, S Marsh, H McLeod, et al. Cytochrome P450 2C9-CYP2C9. Pharmacogenet Genomics 2010;20:277-81.
- RL Reynald, S Sansen, CD Stout, EF Johnson. Structural characterization of human cytochrome P450 2C19 active site differences between P450s 2C8, 2C9, AND 2C19. J Biol Chem 2012;287:44581-91.
- SL Mo, ZW Zhou, LP Yang, MQ Wei, SF Zhou. New insights into the structural features and functional relevance of human cytochrome P450 2C9. Curr Drug Metab 2009;10:1127-50.
- I Sridhar, J Liu, M Foroozesh, CLK Stevens. Insights on cytochrome p450 enzymes and inhibitors obtained through QSAR studies. Molecules 2012;17:9283-305.
- TA Hall. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 1999;41:95-8.
- JK Yano, MH Hsu, KJ Griffin, CD Stout, EF Johnson. Structures of human microsomal cytochrome P450 2A6 complexed with coumarin and methoxsalen. Nat Struct Mol Biol 2005;12:822-3.
- IR Wester, JK Yano, GA Schoch, et al. The structure of human cytochrome P450 2C9 complexed with flurbiprofen at 2.0-Å resolution. J Biol Chem 2004;279:35630-7.
- FC Bernstein, TF Koetzle, GJ Williams, et al. The protein data bank. Eur J Biochem 1977;80:319-24.
- PA Williams, J Cosme, DM Vinković, et al. Crystal structures of human cytochrome P450 3A4 bound to metyrapone and progesterone. Science 2004;305:683-6.
- JK Yano, MR Wester, GA Schoch, KJ Griffin, CD Stout, EF Johnson. The structure of human microsomal cytochrome P450 3A4 determined by X-ray crystallography to 2.05-Å resolution. J Biol Chem 2004;279:38091-4.
- I Ekroos, T Sjögren. Structural basis for ligand promiscuity in cytochrome P450 3A4. Proc Natl Acad Sci U S A 2006;103:13682-7.
- IF Sevrioukova, TL Poulos. Structure and mechanism of the complex between cytochrome P4503A4 and ritonavir. Proc Natl Acad Sci U S A 2010;107:18422-7.
- IF Sevrioukova, TL Poulos. Interaction of human cytochrome P4503A4 with ritonavir analogs. Arch Biochem Biophys 2012;520:108-16.
- IF Sevrioukova, TL Poulos. Pyridine-substituted desoxyritonavir is a more potent inhibitor of cytochrome P450 3A4 than ritonavir. J Med Chem 2013;56:3733-41.
- IF Sevrioukova, TL Poulos. Dissecting cytochrome P450 3A4-ligand interactions using ritonavir analogues. Biochemistry 2013;52:4474-81.
- IF Sevrioukova, TL Poulos. Structural and mechanistic insights into the interaction of cytochrome P4503A4 with bromoergocryptine, a type I ligand. J Biol Chem 2012;287:3510-7.
- A Pedretti, L Villa, G Vistoli. VEGA: a versatile program to convert, handle and visualize molecular structure on Windows-based PCs. J Mol Graph Model 2002;21:47-9.
- S Sansen, MH Hsu, CD Stout, EF Johnson. Structural insight into the altered substrate specificity of human cytochrome P450 2A6 mutants. Arch Biochem Biophys 2007;464:197-206.
- IK Yano, TT Denton, MA Cerny, X Zhang, EF Johnson, JR Cashman. Synthetic inhibitors of cytochrome P-450 2A6: inhibitory activity, difference spectra, mechanism of inhibition, and protein cocrystallization. J Med Chem 2006;49:6987-7001.
- AA Schoch, JK Yano, MR Wester, KJ Griffin, CD Stout, EF Johnson. Structure of human microsomal cytochrome P450 2C8 Evidence for a peripheral fatty acid binding site. J Biol Chem 2004;279:9497-503.
- AA Schoch, JK Yano, S Sansen, PM Dansette, CD Stout, EF Johnson. Determinants of cytochrome P450 2C8 substrate binding structures of complexes with montelukast, troglitazone, felodipine, and 9-cis-retinoic acid. J Biol Chem 2008;283:17227-37.
- GM Morris, R Huey, W Lindstrom, et al. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J Comput Chem 2009;30:2785-91.
- Tripos, Sybyl X, St Louis MO 63144-2319 USA: Tripos Inc, 2010.
- M Clark, RD Cramer, N Van Opdenbosch. Validation of the general purpose Tripos 5.2 force field. Journal of Computational Chemistry 1989;10:982-1012.
- M Rarey, B Kramer, T Lengauer, G Klebe. A fast flexible docking method using an incremental construction algorithm. Journal of Molecular Biology 1996;261:470-89.
- ED Brodie, BJ Ridenhour, ED Brodie. The evolutionary response of predators to dangerous prey: hotspots and coldspots in the geographic mosaic of coevolution between garter snakes and newts. Evolution 2002;56:2067-82.
- I Quiroga, FJ Melendez, K Salvador-Atonal, B Kammerer, T Scior. Identification of a new site of metabolism for phenprocoumon by modeling it's CYP2C9 hydroxylation pattern. SAJ Pharm Pharmacol 2018;5:101.
- I Quiroga, T Scior. Structure-function analysis of the cytochromes P450, Responsible for phenprocoumon metabolism. Journal of the Mexican Chemical Society 2017;61:349-60.