Re-engineering of CYP2C9 to Probe Acid-Base Substrate Selectivity

Abstract

ABSTRACT:A common feature of many CYP2C9 ligands is their weak acidity. As revealed by crystallography, the structural basis for this behavior involves a charge-pairing interaction between an anionic moiety on the substrate and an active site R108 residue. In the present study we attempted to re-engineer CYP2C9 to better accept basic ligands by charge reversal at this key residue. We expressed and purified the R108E and R108E/D293N mutants and compared their ability with that of native CYP2C9 to interact with (S)-warfarin, diclofenac, pyrene, propranolol, and ibuprofen amine. As expected, the R108E mutant maintained all the native enzyme's pyrene 1-hydroxylation activity, but catalytic activity toward diclofenac and (S)-warfarin was abrogated. In contrast, the double mutant displayed much less selectivity in its behavior toward these control ligands. Neither of the mutants displayed significant enhancement of propranolol metabolism, and all three preparations exhibited a type II (inhibitor) rather than type I (substrate) spectrum with ibuprofen amine, although binding became progressively weaker with the single and double mutants. Collectively, these data underscore the importance of the amino acid at position 108 in the acid substrate selectivity of CYP2C9, highlight the accommodating nature of the CYP2C9 active site, and provide a cautionary note regarding facile re-engineering of these complex cytochrome P450 active sites.CYP2C9 is a major form of human liver cytochrome P450 (P450) involved in the metabolism of ϳ15% of all the therapeutic agents that are cleared by oxidative metabolic processes (Williams et al., 2004). Many of these drugs, including (S)-warfarin, phenytoin, and a wide variety of nonsteroidal anti-inflammatory drugs (NSAIDs), are hydrophobic and weakly acidic (Miners and Birkett, 1998;Rettie and Jones, 2005). Early site-directed mutagenesis studies suggested that CYP2C9 residues F114, V113, and F476 were involved in hydrophobic interactions with the substrates (S)-warfarin, diclofenac, and the competitive inhibitor sulfaphenazole (Haining et al., 1999;Melet et al., 2003). In addition, the I helix residues S286 and N289 have been implicated as important determinants of CYP2C9-dependent NSAID metabolism and high affinity sulfaphenazole binding (Jung et al., 1998;Klose et al., 1998). Mutagenesis of the charged residues at R97, R108, D293, and D360 was also found to alter CYP2C9 substrate binding and metabolism (Ridderström et al., 2000;Flanagan et al., 2003;Dickmann et al., 2004). In particular, the R108A mutant displayed a 100-fold lower activity toward diclofenac compared with wild-type (Ridderström et al., 2000), and previous studies from this laboratory showed that the R108F mutant exhibited diminished catalytic activity toward diclofenac and (S)-warfarin while maintaining most of the wild-type activity toward the uncharged molecule pyrene (Dickmann et al., 2004). This latter finding of ligand-dependent activity changes showed that activity alterations were not a global phenomenon and indicat...

Quantum Mechanics/Molecular Mechanics Modeling of Regioselectivity of Drug Metabolism in Cytochrome P450 2C9

“…The R108E mutant CYP2C9 was found to exhibit greatly reduced hydroxylation of diclofenac at the 4′-position; however, no corresponding increase in 5-hydroxylation was reported. 86 …”

“…In the C7 simulation, the average distance between the groups is larger than in the C6 simulation (average values of the Arg108 HH22–warfarin O2 distance 2.30 and 3.29 Å for the C6 and C7 simulations, respectively). Mutagenesis studies of CYP2C9 with S -warfarin revealed a loss of formation of 7-hydroxywarfarin for R108F and R108E mutations. , The third hydrogen-bonding interaction between the enzyme and substrate that differs between the two simulations is that between the aliphatic ketone oxygen of warfarin (O4 in Figure ) and the carboxamide NH 2 group of Asn204. The average distance between the closest amide proton of Asn204 and the O4 of warfarin is 3.03 and 2.40 Å in the C6 and C7 MD simulations, respectively.…”

Pharmacogenomics of CYP2C9: Functional and Clinical Considerations

“…Site-directed mutagenesis data are consistent with a pivotal role for Arg108 in the binding of acidic substrates. Substitution of Arg108 with Ala, Glu or Phe greatly reduced or abolished the metabolism of diclofenac and S -warfarin, whereas the Arg108Glu mutation had a negligible effect on the 1-hydroxylation of the unsubstituted polycyclic aromatic hydrocarbon pyrene and the 4-hydroxlation and dealkylation of propranolol (a basic compound) [ 30 , 31 , 32 ]. Confirmation of the roles of Val113 and Phe114 in stabilizing substrate binding via π–π interactions was demonstrated by reduced or complete loss of S -warfarin and diclofenac metabolism following substitution with Leu or Ile [ 33 , 34 ].…”

Mechanistic Studies of the Cationic Binding Pocket of CYP2C9 In Vitro and In Silico: Metabolism of Nonionizable Analogs of Tienilic Acid

“…This model shows that they are oriented in a similar fashion with the thiophene ring (R1; Table 1) located proximally over the heme and the carboxylate moiety (R2) positioned distal to the heme and coordinated with a cationic binding pocket in a proposed ionic salt bridge with R108 (Poli-Scaife et al, 1997). Many other studies have confirmed that there is an important role for the arginine 108 (R108) in the substrate recognition and orientation of other weakly acidic ligands (Ridderström et al, 2000;Dickmann et al, 2004;Tai et al, 2008). Perhaps the strongest evidence of this interaction comes from the crystal structure of CYP2C9 with flurbiprofen bound showing the carboxylate interaction with R108 (Wester et al, 2004).…”

“…Despite the evidence showing the importance of ionic interactions in the metabolism of weakly acidic ligands by CYP2C9, the substrate recognition and orientation are also directed by interactions with aromatic residues in the active site such as phenylalanine 114 and 476 (F114 and F476) Dickmann et al, 2004;Mosher et al, 2008;Tai et al, 2008). Other CYP2C9 substrates, including ketamine (Hijazi and Boulieu, 2002), amitryptiline (Venkatakrishnan et al, 1998), tamoxifen (Crewe et al, 2002), and terbinafine (Vickers et al, 1999), do not fall in the anionic class of CYP2C9 substrates.…”