They metabolize vast array of xenobiotic compounds including drugs, environmental pollutants, and endogenous compounds such as steroids and prostaglandins. Structure-activity relationships of drug metabolizing enzymes and their substrates comprise an important area of research that impacts pharmacology, toxicology and basic enzymology.
There are increasing efforts to incorporate metabolism studies early in the drug discovery process because poor pharmacokinetics account for a substantial proportion of clinical failures. To accelerate these efforts, Life Technologies provides recombinant drug metabolizing enzymes and assay methods that enable scientists to screen large numbers of diverse compounds for their drug metabolism profiles.
Risperidone Risperdal , tramadol Ultram. In target-based methods, the structure of the enzyme is the starting point for model generation. Combining structure- and ligand-based approaches to improve site of metabolism prediction in CYP2C9 substrates. Many other tissues contain also xenobiotic metabolizing enzymes. Clin Pharmacol Ther ; The great diversity of reactions catalyzed by cytochromes P Development of a combined protein and pharmacophore model for cytochrome P 2C9.
Note: You clicked on an external link, which has been disabled in order to keep your shopping session open. Search Thermo Fisher Scientific. The important field of in silico tools for predicting general ADMET properties is extensively covered in recent reviews Cronin and Madden, ; Pelkonen et al. Several different types of in silico methods have been developed; the simplest way to classify them is to distinguish physics-based and empirical models Figure 1.
Physics-based methods include for example molecular dynamics and the prediction of binding affinity by methods such as free energy perturbation and quantum chemical QC calculations. Empirical methods, based on existing experimental data without knowledge of the physics of the system, may be divided to ligand-based and target-based approaches. In ligand-based methods, structures of known active and inactive compounds are modeled to derive quantitative structure-activity relationships QSARs and other properties such as sites of metabolism SOM , i.
Also various rule-based expert systems belong to this category. In target-based methods, the structure of the enzyme is the starting point for model generation. Models integrating both ligands and enzymes are known as combined or mechanism-based methods. Types of in silico models.
Numerous specific methods exist in each category. In CoMFA, ligand-receptor interactions are represented by standard potential energy fields such as steric and electrostatic interactions. Differences in these interaction field intensities in a set of molecules are related to differences in their biological response. Calculation of steric and electrostatic fields is carried out by placing aligned molecules from a dataset into a cubic lattice in which probe atoms surround the molecules.
CoMFA uses a partial least-squares PLS method in the analysis to predict activity from energy values at the grid points. The results of the PLS analysis are often presented as a 3D coefficient contour map which show favorable and unfavorable steric and electrostatic regions.
CoMSIA uses a Gaussian function to calculate similarity indices for a data set of pre-aligned molecules at regularly spaced grid points. Expert systems mimic human reasoning and formalize existing knowledge. These are programs in which a computer solves problems by applying rules from a knowledge base. Such rules may be a combination of factual and heuristic types, and are usually non-numerical. In most cases, 3D structures of compounds are not required.
Metabolic pathways are sometimes very different even in closely related mammalian species, thus some expert systems allow filtering of specific subsets of the data to a specific species Kirchmair et al. Expert systems exploit the extensive databases of experimentally derived metabolic pathways. Of the target-based methods, docking analysis mimics the binding of a ligand to a biological macromolecule, usually a protein. Typically, in docking simulation the conformational space of the ligand is sampled within the ligand binding cavity of the target protein to identify the most likely binding conformation s for the ligand.
The binding affinity or fitness of the ligand is estimated rapidly for all sampled conformations with a scoring function.
In principle, docking predicts energetically favorable conformations of ligands and also reveals key groups or atoms for binding. With crystal structures available for the major human CYPs, protein-ligand docking methods are suitable for the analysis and prediction of CYP—ligand interactions. However, docking accounts poorly for substrate reactivity Kirchmair et al. Presentation of details of all the in silico methods is outside the scope of this review; we will provide here a general view of the state-of-art.
Several recent reviews cover the technical aspects extensively Shaik et al.
Today in silico methods used to evaluate CYP—ligand interactions typically combine techniques from physics-based and empirical models. With appropriate combinations, the strengths of individual in silico methods complement each other. The crystal structures of all these CYP enzymes have been elucidated. Ligands to these CYP enzymes are either substrates that are metabolized or inhibitors that decrease substrate turnover.
Elucidation of the binding cavities of individual CYPs to their ligand profiles has revealed that the size and shape of the binding cavity are critical for selective ligand binding. A good ligand is able to complement the binding cavity in size, shape, and electrostatic interactions. Numerous in silico models on these nine CYPs have been published since the s using various approaches.
Together, these studies have yielded a fairly detailed picture on the main features of ligand-enzyme interactions.
The following text and Table 1 summarize the main characteristics of these CYPs: typical substrates and inhibitors, common features of the ligands, and main characteristics of the enzyme binding cavities and active sites, i. Coumarin fits excellently in the narrow binding cavity of CYP2A6. There are basic residue s in the active site as acidic compounds are oxidized efficiently or inhibit the enzyme, for example the acyl glucuronides of gemfibrozil and clopidogrel Ogilvie et al.
The crystal structure of CYP2C9 shows that Arg plays a significant role in the binding of acidic substrates such as flurbiprofen Mo et al. The tertiary structures of 2C19 and CYP2C8 are highly similar, although their binding cavities differ greatly due to amino acid differences that directly alter the topography and the hydrophobic and polar landscapes of the cavities Niwa and Yamazaki, ; Reynald et al. This structural knowledge has helped in understanding why CYP2E1 generally catalyzes small molecular substrates, such as acetaminophen and halothane Porubsky et al.
Numerous in silico models have given important insights into the nature of interactions between individual CYP forms and their ligands substrates and inhibitors. To be of practical use, the main parameters to be predicted for CYP-mediated metabolism are: 1 reactions catalyzed, and SOMs and K m and V max values for the reaction, and 2 inhibition of CYP-specific reactions and the inhibition mechanism and key constants e.
Ability to predict and identify metabolites of candidate drug molecules is essential to modern drug discovery, because it is crucial to know if the metabolites are active or inactive or possibly reactive and thus toxic. Unfavorable metabolic pathways may exclude a drug candidate from further development, as they may cause toxicity in later, more costly phases of development.
The same information is also critical when elucidating the possible effects of any xenobiotic in the body. Many bioactivation pathways to reactive metabolites are known; therefore specific structural alerts are scrutinized especially in drug candidates Kalgutkar et al. Oxidation of substrates by CYPs is a multistep process. The rate-determining step involves hydrogen or electron abstraction from the substrate followed by oxygen rebound or a concerted oxygenation via formation of a complementary interaction between the substrate and amino acid residues in the active site near the oxygen coordinated to heme iron.
Drug Metabolizing Enzymes: Cytochrome P and Other Enzymes in Drug Discovery and Development - CRC Press Book. Drug Metabolizing Enzymes: Cytochrome P and Other Enzymes in Drug Discovery and Development [Jae Lee, R. Scott Obach, Michael B. Fisher] on.
Thus, hydrogen abstraction energy is an important determinant for SOM of a substrate. However, the most reactive site of a substrate may not be the predicted SOM, because different sizes, shapes, and electrostatic forces of complementary interaction in the active sites of various CYPs determine the orientation of substrate toward to the activated oxygen coordinated to heme. Active site differences make the regioselectivity of oxygenation reactions CYP form-specific. It is thus necessity to consider substrate-enzyme recognition in predicting SOMs Stjernschantz et al.
Various ligand-based and target-based as well as combined methods have been used for SOM prediction. Examples of these methods are given in Table 2. Ligand-based methods concentrate on finding common trends and patterns of size, shape, and atomic or physicochemical environment of the substrates and their relation to SOM.
Methods used include pharmacophore and QSAR models, fragment analysis and atomic environment fingerprints, often in combination with rule systems. Target-based methods focus on discovering active conformations for substrates in the CYP active site. Target-based SOM prediction utilizes docking, homology modeling, molecular dynamics simulations and fingerprints. Two basic assumptions are made when docking is used to find active conformations for substrates. Second, the binding energy of the substrate should be low. These basics are taken into account when the success of the method is validated.
Further requirements are that there are no other atoms between the heme iron and the target oxidized atom. Docking is always CYP form-specific, since it relies on the 3D structure of the binding cavity of the particular enzyme. However, as docking predicts only the active conformation of a substrate, performing it alone is not enough to predict specific SOMs.
The flexibility of CYP enzymes needs to be taken into account in modeling. However, the fastest and most basic way is to use rigid CYP crystal structures, which often leave little space of freedom for active conformations. The physical space of rigid structures is often specific for the cocrystallized ligand due to the induced fit effect, and this may mask the true active conformation of the enzyme for another substrate.
Induced fit effects have been considered by docking substrates to multiple structures crystallized with varying ligands, flexible structures, or ensembles of a CYP enzyme from molecular dynamics Hritz et al. Ensemble docking is a time-consuming process and thus it is rational to use only a few target protein structures.
Although computationally expensive, molecular dynamics on enzyme-substrate complexes is also a valuable tool for confirming active conformations and flexibility of CYP binding cavities Park and Harris, A long radius from the iron leads to a wide accepted area above the heme plane in many CYP structures, leaving space for other substrate atoms besides a SOM.
Having multiple substrate atoms in the accepted space leaves the method very error-prone if one wants to define the primary SOM at atomic precision.