Cytochroms P450, Aldehide oxidase and Flavin containing monooxygenase 3
Consensus SoM prediction models
The consensus SoM prediction models perform an averaging process on the results coming from the independent predictions performed with the enzymes that are most relevant for the specific organ in the human body.
The current version of the software implements the following consensus models:
- LIVER (CYP2C9, CYP2D6, CYP3A4)
- SKIN (CYP1A1, CYP2E1, CYP3A5)
- BRAIN (CYP1A1, CYP2D6, CYP3A4)
The consensus SoM prediction relies on how frequently the same site of a fixed substrate structure is selected among the individually predicted most probable sites; it doesn’t therefore directly depend on the Psm numerical values. The consensus is obtained by assigning the same relative weights to each CYPs, although we know that some CYP isoforms are more abundant than others depending on the particular organ. The relative weights will be modified only when we have a clear indication of the percentage of CYP involved in the metabolism of each molecule.
Human AOX enzyme
tudies of the metabolism mediated by aldehyde oxidase (AOX) have assumed a prominent role as demonstrated by the numerous papers on this subject and by the various experimental strategies developed to determine the propensity of a compound to be an AOX substrate. AOX is a cytosolic enzyme that catalyzes oxidation by a nucleophile mechanism using water as a source of oxygen that is incorporated in the metabolite. AOX also shows significant ability to hydrolyze the amide bonds that are generally metabolically stable and hardly modified by CYP enzymes. The AOX enzyme is composed of two identical subunits of 150 kDa and is prevalently expressed in the liver but also in other human tissues.
The structure of the human AOX enzyme has recently been solved by x-ray crystallography (code 4UHX). This protein structure is missing two important polar amino acids (Asp881, Glu882) situated in the loop between the Leu880 and Ser883 residues that are only at 4.5 Å distance from the substrate phthalazine. It is clear that they carry out a significant role in molecular interaction with potential substrates. Furthermore, also residues Phe655, Phe656, Thr657, Glu658, Ala659 and Glu660, located in the lower part of the binding site, are missing. These residues, mainly hydrophobic, belong to the gate 1 region, while the previous two are part of the gate 2. Both gate 1 and gate 2 are dynamic loops involved in regulating the ligands entrance into the pocket.
To avoid false results from these structural gaps, and from other loops not resolved in the original x-ray structure, we built a model of the entire protein using the 4UHX structure as template and adding the missing loops. The resulting AOX catalytic cavity includes 30 amino acids, which, notwithstanding their different physical-chemical natures and different polarity, have generated a large pathway of hydrophobic interaction with fewer areas of H-bonding interaction with H-bond interaction donors being almost absent.
Calculations made using water-Flap have shown the presence of a water molecule in proximity to the molybdenum cofactor. This water is mobile but localized in a well-defined region, where the enzyme-substrate nucleophile attack can take place, and the oxygen of the water molecule is transferred to the AOX substrate.
It is clear that the entire protein and, in particular, the binding site displays a significant degree of flexibility, likely fundamental for the protein to perform its action. To better investigate the protein intrinsic dynamic we performed plain Molecular Dynamics (MD) studies starting from the protein model previously mentioned, which was minimized, equilibrated and submitted to 200 ns long MD simulations. The obtained trajectory was clustered according to the variation of the Molecular Interaction Fields (MIFs) of the pocket, as defined by the first mentioned thirty residues. The analyses confirmed a high flexibility of the 655-663 and of the 881-885 segments, supporting their role in regulating the pocket accessibility. This intrinsic dynamic triggers a significant variation of the pocket MIFs and volume, and of the chemical and energetic properties of the catalytic site, but also a dramatic change in the pocket dimension and accessibility. These features could reasonably be responsible of the enzyme capability of accepting and metabolizing a number of different and variable chemical entities.
It was possible to demonstrate that the metabolism of azaheterocycle compounds is a complex function of several factors that range from the atomic charge at the most positive C-H, presence and effects of substituents (EDG or EWG) in the azaheterocycle ring, and protonation of the azaheterocycle for compounds with basic centers having a pKa greater than 7.0. Furthermore, hindered and hydrophobic substituents are often able to dramatically change the substrate exposure, thus enhancing the metabolic stability. This is often caused by interfering with the water network inside or in the proximity of the catalytic center. Regarding amides, the same effects are present but they are more relevant in the amine counterpart, which is more sensitive to hAOX susceptibility being established.
3.8. FMO3 enzyme
Flavin containing monooxygenase 3 (FMO3) is an important enzyme which metabolize drugs oxidizing them through nucleophilic addition reactions. FMO3 uses NADPH and oxygen to convert molecules containing primarily nucleophilic centers, in particular nitrogen, sulfur, phosphorous and selenium atoms, all of which have a free lone pair of electrons, into the corresponding oxides. FMOs are responsible for about 6% of all the phase I metabolic reactions. Since FMO3 is the isoform of FMO which is most dominant in the liver of human adults, almost all the reactions of pharmaceutical interest are mediated by this enzyme. FMO3 is the major representative of the FMO family just as CYP3A4 is the major representative of the cytochrome family.
It is not simple to understand whether a xenobiotic is a substrate of the FMOs or not, especially for FMO3. It is also difficult to predict the site of oxidation and any potential competition with CYP3A4. MetaSite can be extremely useful in helping the researcher formulate hypothesis on FMO3-drug interactions and explain/interpret experiments of metabolite identification driven by FMO3.
As with the cytochromes, even in FMO3 case the oxidation reaction happens by a combination of substrate reactivity (described here by its nucleophilicity) and by the productive spatial interaction with the enzyme (favorable enzyme recognition). Once the enzyme-accessibility and nucleophilic-reactivity components are calculated, the site of metabolism is described by a probability function PSoM (probability for the site of metabolism), which is equivalent to the function used in the case of CYPs enzymes. (see reference for explanations)