346
Z. Fu and J. Chen
reaction intermediates; the fast release of the products would facilitate the recovery
of the resting state for the next cyclic process [4]. Taking together, the surrounding
enzymatic matrix can influence the computational results in two main aspects: steric
and electrostatic effects. Thus, these effects should not be neglected in studying P450
catalytic reactions, which demands on simulations with multiscalar techniques, such
as the combined quantum mechanics/molecular mechanics (QM/MM) method.
In 2002, Thiel et al. [22] first performed a DFT(B3LYP)/MM study which delved
into properties of intermediates in the catalytic cycle and mechanisms for Compound
0 protonation and C–H bond hydroxylation by P450 enzymes. Thereafter, Guallar
[23, 24] conducted QM/MM investigations with a restricted open-shell DFT method,
termed DFT(ROB3LYP)/MM, to probe the H-abstraction mechanism by P450cam.
They found that the electrostatic interactions between the negatively charged propionate of the porphyrin side chain and the positively charged surrounding residues
would facilitate C–H hydroxylation. In the past decade, QM/MM investigations on
P450 enzymes have gradually been increasing, with P450cam being the dominant
isoform. For example, to resolve the controversy of a “second oxidant” in P450cam
catalytic cycle, Shaik et al. [9] employed QM/MM calculations to dissect into the
reactivity of three intermediate species, i.e., Compound I, Compound 0, and the ferric hydrogen peroxide Fe
III (O 2 H 2 ). It was shown that the persistence and oxidative
reactivity of Fe
III (O 2 H 2 ) depend primarily on the interplay of camphor and protein.
The presence of camphor in the active site blocks the release of H 2 O 2 (the uncoupling
pathway in Fig. 17.3), which expedites its homolytic O–O bond cleavage to a HO·
radical and Fe
IV -OH species. The HO· radical is further adjusted by hydrogen bonding with adjacent amino acid residues to an appropriate position that facilitates the
H-abstraction from Fe
IV –OH by the HO· radical, leading to Compound I and H 2 O.
Hence in this case, QM/MM simulations serve to better understand the protonation
of intermediates in the catalytic cycle.
In other cases, QM/MM investigations have adequately unveiled some of the
long-standing puzzles in P450 catalytic cycle and contributed to understanding of
the regio- and stereoselectivity in oxidative reactions. By using MD and QM/MM
techniques, Ramanan et al. [25] simulated the binding of P450-BM3 with fatty acids
and revealed that the substrate’s binding with active site residues (e.g., Arg241 and
Pro242) determines the regio- and enantioselective reactivity for fatty acid hydroxylation. These studies further verify that quantum chemical calculations are capable
of predicting SOM and products based on molecular structures (i.e., ab initio). It is
anticipated that QM/MM and MD calculations, along with future improvements in
computational theory and capability, could pave the way for accurate prediction for
metabolic mechanisms of xenobiotics catalyzed by P450 enzymes.
Through half a century’s endeavors, people have acquired in-depth knowledge of
the reaction mechanisms and metabolic profile of P450 enzymes. As noted above, the
truncated cluster model, Compound I, contributed greatly in unveiling the reaction
mechanisms. The geometric structure and electronic orbital occupations for Compound I are depicted in Fig. 17.4. The typical structure for Compound I (Fig. 17.4)
comprises a protoporphyrin with a high-valent iron (IV) atom and –SH representing
cysteine, an Fe=O bond of ca. 1.65 Å and Fe–S bond of ca. 2.30 Å.
Z. Fu and J. Chen
reaction intermediates; the fast release of the products would facilitate the recovery
of the resting state for the next cyclic process [4]. Taking together, the surrounding
enzymatic matrix can influence the computational results in two main aspects: steric
and electrostatic effects. Thus, these effects should not be neglected in studying P450
catalytic reactions, which demands on simulations with multiscalar techniques, such
as the combined quantum mechanics/molecular mechanics (QM/MM) method.
In 2002, Thiel et al. [22] first performed a DFT(B3LYP)/MM study which delved
into properties of intermediates in the catalytic cycle and mechanisms for Compound
0 protonation and C–H bond hydroxylation by P450 enzymes. Thereafter, Guallar
[23, 24] conducted QM/MM investigations with a restricted open-shell DFT method,
termed DFT(ROB3LYP)/MM, to probe the H-abstraction mechanism by P450cam.
They found that the electrostatic interactions between the negatively charged propionate of the porphyrin side chain and the positively charged surrounding residues
would facilitate C–H hydroxylation. In the past decade, QM/MM investigations on
P450 enzymes have gradually been increasing, with P450cam being the dominant
isoform. For example, to resolve the controversy of a “second oxidant” in P450cam
catalytic cycle, Shaik et al. [9] employed QM/MM calculations to dissect into the
reactivity of three intermediate species, i.e., Compound I, Compound 0, and the ferric hydrogen peroxide Fe
III (O 2 H 2 ). It was shown that the persistence and oxidative
reactivity of Fe
III (O 2 H 2 ) depend primarily on the interplay of camphor and protein.
The presence of camphor in the active site blocks the release of H 2 O 2 (the uncoupling
pathway in Fig. 17.3), which expedites its homolytic O–O bond cleavage to a HO·
radical and Fe
IV -OH species. The HO· radical is further adjusted by hydrogen bonding with adjacent amino acid residues to an appropriate position that facilitates the
H-abstraction from Fe
IV –OH by the HO· radical, leading to Compound I and H 2 O.
Hence in this case, QM/MM simulations serve to better understand the protonation
of intermediates in the catalytic cycle.
In other cases, QM/MM investigations have adequately unveiled some of the
long-standing puzzles in P450 catalytic cycle and contributed to understanding of
the regio- and stereoselectivity in oxidative reactions. By using MD and QM/MM
techniques, Ramanan et al. [25] simulated the binding of P450-BM3 with fatty acids
and revealed that the substrate’s binding with active site residues (e.g., Arg241 and
Pro242) determines the regio- and enantioselective reactivity for fatty acid hydroxylation. These studies further verify that quantum chemical calculations are capable
of predicting SOM and products based on molecular structures (i.e., ab initio). It is
anticipated that QM/MM and MD calculations, along with future improvements in
computational theory and capability, could pave the way for accurate prediction for
metabolic mechanisms of xenobiotics catalyzed by P450 enzymes.
Through half a century’s endeavors, people have acquired in-depth knowledge of
the reaction mechanisms and metabolic profile of P450 enzymes. As noted above, the
truncated cluster model, Compound I, contributed greatly in unveiling the reaction
mechanisms. The geometric structure and electronic orbital occupations for Compound I are depicted in Fig. 17.4. The typical structure for Compound I (Fig. 17.4)
comprises a protoporphyrin with a high-valent iron (IV) atom and –SH representing
cysteine, an Fe=O bond of ca. 1.65 Å and Fe–S bond of ca. 2.30 Å.