On the recognition of mammalian microsomal cytochrome P450 substrates and their characteristics: towards the prediction of human p450 substrate specificity and metabolism

Multidisciplinary Insights into the Structure-Function Relationship of the CYP2B6 Active Site

Cytochrome P450 2B6 (CYP2B6) is a highly polymorphic human enzyme involved in the metabolism of many clinically relevant drugs, environmental toxins, and endogenous molecules with disparate structures. Over the last 20-plus years, in silico and in vitro studies of CYP2B6 using various ligands have provided foundational information regarding the substrate specificity and structure-function relationship of this enzyme. Approaches such as homology modeling, X-ray crystallography, molecular docking, and kinetic activity assays coupled with CYP2B6 mutagenesis have done much to characterize this originally neglected monooxygenase. However, a complete understanding of the structural details that make new chemical entities substrates of CYP2B6 is still lacking. Surprisingly little in vitro data has been obtained about the structure-function relationship of amino acids identified to be in the CYP2B6 active site. Since much attention has already been devoted to elucidating the function of CYP2B6 allelic variants, here we review the salient findings of in silico and in vitro studies of the CYP2B6 structure-function relationship with a deliberate focus on the active site. In addition to summarizing these complementary approaches to studying structure-function relationships, we note gaps/challenges in existing data such as the need for more CYP2B6 crystal structures, molecular docking results with various ligands, and data coupling CYP2B6 active site mutagenesis with kinetic parameter measurement under standard expression conditions. Harnessing in silico and in vitro techniques in tandem to understand the CYP2B6 structure-function relationship will likely offer further insights into CYP2B6-mediated metabolism. SIGNIFICANCE STATEMENT: The apparent importance of cytochrome P450 2B6 (CYP2B6) in the metabolism of various xenobiotics and endogenous molecules has grown since its discovery with many in silico and in vitro studies offering a partial description of its structure-function relationship. Determining the structure-function relationship of CYP2B6 is difficult but may be aided by thorough biochemical investigations of the CYP2B6 active site that provide a more complete pharmacological understanding of this important enzyme.

Strategies for Assessing Acceptable Intakes for Novel N-Nitrosamines Derived from Active Pharmaceutical Ingredients

The detection of N-nitrosamines, derived from solvents and reagents and, on occasion, the active pharmaceutical ingredient (API) at higher than acceptable levels in drug products, has led regulators to request a detailed review for their presence in all medicinal products. In the absence of rodent carcinogenicity data for novel N-nitrosamines derived from amine-containing APIs, a conservative class limit of 18 ng/day (based on the most carcinogenic N-nitrosamines) or the derivation of acceptable intakes (AIs) using structurally related surrogates with robust rodent carcinogenicity data is recommended. The guidance has implications for the pharmaceutical industry given the vast number of marketed amine-containing drugs. In this perspective, the rate-limiting step in N-nitrosamine carcinogenicity, involving cytochrome P450-mediated α-carbon hydroxylation to yield DNA-reactive diazonium or carbonium ion intermediates, is discussed with reference to the selection of read-across analogs to derive AIs. Risk-mitigation strategies for managing putative N-nitrosamines in the preclinical discovery setting are also presented.

The Central Role of Cytochrome P450 in Xenobiotic Metabolism-A Brief Review on a Fascinating Enzyme Family

Human Cytochrome P450 (CYP) enzymes constitute a superfamily of membrane-bound hemoproteins that are responsible for the metabolism of a wide variety of clinically, physiologically, and toxicologically important compounds. These heme-thiolate monooxygenases play a pivotal role in the detoxification of xenobiotics, participating in the metabolism of many structurally diverge compounds. This short-review is intended to provide a summary on the major roles of CYPs in Phase I xenobiotic metabolism. The manuscript is focused on eight main topics that include the most relevant aspects of past and current CYP research. Initially, (I) a general overview of the main aspects of absorption, distribution, metabolism, and excretion (ADME) of xenobiotics are presented. This is followed by (II) a background overview on major achievements in the past of the CYP research field. (III) Classification and nomenclature of CYPs is briefly reviewed, followed by (IV) a summary description on CYP’s location and function in mammals. Subsequently, (V) the physiological relevance of CYP as the cornerstone of Phase I xenobiotic metabolism is highlighted, followed by (VI) reviewing both genetic determinants and (VI) nongenetic factors in CYP function and activity. The last topic of the review (VIII) is focused on the current challenges of the CYP research field.

Designing QSARs for Parameters of High-Throughput Toxicokinetic Models Using Open-Source Descriptors

The intrinsic metabolic clearance rate (Cl_int) and the fraction of the chemical unbound in plasma (f_up) serve as important parameters for high-throughput toxicokinetic (TK) models, but experimental data are limited for many chemicals. Open-source quantitative structure-activity relationship (QSAR) models for both parameters were developed to offer reliable in silico predictions for a diverse set of chemicals regulated under the U.S. law, including pharmaceuticals, pesticides, and industrial chemicals. As a case study to demonstrate their utility, model predictions served as inputs to the TK component of a risk-based prioritization approach based on bioactivity/exposure ratios (BERs), in which a BER < 1 indicates that exposures are predicted to exceed a biological activity threshold. When applied to a subset of the Tox21 screening library (6484 chemicals), we found that the proportion of chemicals with BER <1 was similar using either in silico (1133/6484; 17.5%) or in vitro (148/848; 17.5%) parameters. Further, when considering only the chemicals in the Tox21 set with in vitro data, there was a high concordance of chemicals classified with either BER <1 or >1 using either in silico or in vitro parameters (767/848, 90.4%). Thus, the presented QSARs may be suitable for prioritizing the risk posed by many chemicals for which measured in vitro TK data are lacking.

Application of cytochrome P450 reactivity on the characterization of chemical compounds and its association with repeated-dose toxicity

Repeated-dose toxicity (RDT) studies are one of the critical studies to assess chemical safety. There have been some studies attempting to predict RDT endpoints based on chemical substructures, but it remains very difficult to establish such a method, and a more detailed characterization of chemical compounds seems necessary. Cytochrome P450s (P450s) comprise multiple forms with different substrate specificities and play important roles in both the detoxification and metabolic activation of xenobiotics. In this study, we investigated possible use of P450 reactivity of chemical compounds to classify the compounds. A total of 148 compounds with available rat RDT test data were used as test compounds and subjected to inhibition assays against 18 human and rat P450s. Among the tested compounds, 82 compounds inhibited at least one P450 form. Hierarchical clustering analyses using the P450 inhibitory profiles divided the 82 compounds into nine groups, some of which showed characteristic chemical and biological properties. Principal component analyses of the P450 inhibition data in combination with the calculated chemical descriptors demonstrated that P450 inhibition data were plotted differently than most chemical descriptors in the loading plots. Finally, association analyses between P450 inhibition and RDT endpoints showed that some endpoints related to the liver, kidney and hematology were significantly associated with the inhibition of some P450s. Our present results suggest that the P450 reactivity profiles can be used as novel descriptors for characterizing chemical compounds for the investigation of the toxicity mechanism and/or the establishment of a toxicity prediction model.

Concepts of Artificial Intelligence for Computer-Assisted Drug Discovery

Artificial intelligence (AI), and, in particular, deep learning as a subcategory of AI, provides opportunities for the discovery and development of innovative drugs. Various machine learning approaches have recently (re)emerged, some of which may be considered instances of domain-specific AI which have been successfully employed for drug discovery and design. This review provides a comprehensive portrayal of these machine learning techniques and of their applications in medicinal chemistry. After introducing the basic principles, alongside some application notes, of the various machine learning algorithms, the current state-of-the art of AI-assisted pharmaceutical discovery is discussed, including applications in structure- and ligand-based virtual screening, de novo drug design, physicochemical and pharmacokinetic property prediction, drug repurposing, and related aspects. Finally, several challenges and limitations of the current methods are summarized, with a view to potential future directions for AI-assisted drug discovery and design.

Predicting Mouse Liver Microsomal Stability with "Pruned" Machine Learning Models and Public Data

PURPOSE

Mouse efficacy studies are a critical hurdle to advance translational research of potential therapeutic compounds for many diseases. Although mouse liver microsomal (MLM) stability studies are not a perfect surrogate for in vivo studies of metabolic clearance, they are the initial model system used to assess metabolic stability. Consequently, we explored the development of machine learning models that can enhance the probability of identifying compounds possessing MLM stability.

METHODS

Published assays on MLM half-life values were identified in PubChem, reformatted, and curated to create a training set with 894 unique small molecules. These data were used to construct machine learning models assessed with internal cross-validation, external tests with a published set of antitubercular compounds, and independent validation with an additional diverse set of 571 compounds (PubChem data on percent metabolism).

RESULTS

"Pruning" out the moderately unstable / moderately stable compounds from the training set produced models with superior predictive power. Bayesian models displayed the best predictive power for identifying compounds with a half-life ≥1 h.

CONCLUSIONS

Our results suggest the pruning strategy may be of general benefit to improve test set enrichment and provide machine learning models with enhanced predictive value for the MLM stability of small organic molecules. This study represents the most exhaustive study to date of using machine learning approaches with MLM data from public sources.

A Mechanism-Based Model for the Prediction of the Metabolic Sites of Steroids Mediated by Cytochrome P450 3A4

Early prediction of xenobiotic metabolism is essential for drug discovery and development. As the most important human drug-metabolizing enzyme, cytochrome P450 3A4 has a large active cavity and metabolizes a broad spectrum of substrates. The poor substrate specificity of CYP3A4 makes it a huge challenge to predict the metabolic site(s) on its substrates. This study aimed to develop a mechanism-based prediction model based on two key parameters, including the binding conformation and the reaction activity of ligands, which could reveal the process of real metabolic reaction(s) and the site(s) of modification. The newly established model was applied to predict the metabolic site(s) of steroids; a class of CYP3A4-preferred substrates. 38 steroids and 12 non-steroids were randomly divided into training and test sets. Two major metabolic reactions, including aliphatic hydroxylation and N-dealkylation, were involved in this study. At least one of the top three predicted metabolic sites was validated by the experimental data. The overall accuracy for the training and test were 82.14% and 86.36%, respectively. In summary, a mechanism-based prediction model was established for the first time, which could be used to predict the metabolic site(s) of CYP3A4 on steroids with high predictive accuracy.

In vitro effect of important herbal active constituents on human cytochrome P450 1A2 (CYP1A2) activity

This study was designed to investigate eight herbal active constituents (andrographolide, asiaticoside, asiatic acid, madecassic acid, eupatorin, sinensetin, caffeic acid, and rosmarinic acid) on their potential inhibitory effects on human cytochrome P450 1A2 (CYP1A2) activity. A fluorescence-based enzyme assay was performed by co-incubating human cDNA-expressed CYP1A2 with its selective probe substrate, 3-cyano-7-ethoxycoumarin (CEC), in the absence or presence of various concentrations of herbal active constituents. The metabolite (cyano-hydroxycoumarin) formed was subsequently measured in order to obtain IC50 values. The results indicated that only eupatorin and sinensetin moderately inhibited CYP1A2 with IC50 values of 50.8 and 40.2 μM, while the other active compounds did not significantly affect CYP1A2 activity with IC50 values more than 100 μM. Ki values further determined for eupatorin and sinensetin were 46.4 and 35.2 μM, respectively. Our data indicated that most of the investigated herbal constituents have negligible CYP1A2 inhibitory effect. In vivo studies however may be warranted to ascertain the inhibitory effect of eupatorin and sinensetin on CYP1A2 activity in clinical situations.

Precursors for cytochrome P450 profiling breath tests from an in silico screening approach

The family of cytochrome P450 enzymes (CYPs) is a major player in the metabolism of drugs and xenobiotics. Genetic polymorphisms and transcriptional regulation give a complex patient-individual CYP activity profile for each human being. Therefore, personalized medicine demands easy and non-invasive measurement of the CYP phenotype. Breath tests detect volatile organic compounds (VOCs) in the patients' exhaled air after administration of a precursor molecule. CYP breath tests established for individual CYP isoforms are based on the detection of (13)CO2 or (14)CO2 originating from CYP-catalyzed oxidative degradation reactions of isotopically labeled precursors.We present an in silico work-flow aiming at the identification of novel precursor molecules, likely to result in VOCs other than CO2 upon oxidative degradation as we aim at label-free precursor molecules. The ligand-based work-flow comprises five parts: (1) CYP profiling was encoded as a decision tree based on 2D molecular descriptors derived from established models in the literature and validated against publicly available data extracted from the DrugBank. (2) Likely sites of metabolism were identified by reactivity and accessibility estimation for abstractable hydrogen radical. (3) Oxidative degradation reactions (O- and N-dealkylations) were found to be most promising in the release of VOCs. Thus, the CYP-catalyzed oxidative degradation reaction was encoded as SMIRKS (a programming language style to implement reactions based on the SMARTS description) to enumerate possible reaction products. (4) A quantitative structure property relation (QSPR) model aiming to predict the Henry constant H was derived from data for 488 organic compounds and identifies potentially VOCs amongst CYP reaction products. (5) A blacklist of naturally occurring breath components was implemented to identify marker molecules allowing straightforward detection within the exhaled air.Evident oxidative degradation reactions served as test case for the screening approach. Comparisons to metabolism data from literature support the results' plausibility. Thus, a large scale screening for potential novel breath test precursor using the presented five stage work-flow is promising.

The in vitro metabolism of phospho-sulindac amide, a novel potential anticancer agent

Phospho-sulindac amide (PSA) is a novel potential anti-cancer and anti-inflammatory agent. Here we report the metabolism of PSA in vitro. PSA was rapidly hydroxylated at its butane-phosphate moiety to form two di-hydroxyl-PSA and four mono-hydroxyl-PSA metabolites in mouse and human liver microsomes. PSA also can be oxidized or reduced at its sulindac moiety to form PSA sulfone and PSA sulfide, respectively. PSA was mono-hydroxylated and cleared more rapidly in mouse liver microsomes than in human liver microsomes. Of eight major human cytochrome P450s (CYPs), CYP3A4 and CYP2D6 exclusively catalyzed the hydroxylation and sulfoxidation reactions of PSA, respectively. We also examined the metabolism of PSA by three major human flavin monooxygenases (FMOs). FMO1, FMO3 and FMO5 were all capable of catalyzing the sulfoxidation (but not hydroxylation) of PSA, with FMO1 being by far the most active isoform. PSA was predominantly sulfoxidized in human kidney microsomes because FMO1 is the dominant isoform in human kidney. PSA (versus sulindac) is a preferred substrate of both CYPs and FMOs, likely because of its greater lipophilicity and masked-COOH group. Ketoconazole (a CYP3A4 inhibitor) and alkaline pH strongly inhibited the hydroxylation of PSA, but moderately suppressed its sulfoxidation in liver microsomes. Together, our results establish the metabolic pathways of PSA, identify the major enzymes mediating its biotransformations and reveal significant inter-species and inter-tissue differences in its metabolism.

Mechanistically-based QSARs to describe metabolic constants in mammals

Biotransformation is one of the processes which influence the bioaccumulation of chemicals. The enzymatic action of metabolism involves two processes, i.e. the binding of the substrate to the enzyme followed by a catalytic reaction, which are described by the Michaelis-Menten constant (Km) and the maximum rate (Vmax). Here, we developed Quantitative Structure-Activity Relationships (QSARs) for Log(1/Km) and LogVmax for substrates of four enzyme classes. We focused on oxidations catalysed by alcohol dehydrogenase (ADH), aldehyde dehydrogenase (ALDH), flavin-containing monooxygenase (FMO) and cytochrome P450 (CYP) in mammals. The chemicals investigated were xenobiotics, including alcohols, aldehydes, pesticides and drugs. We applied general linear models for this purpose, employing descriptors related to partitioning, geometric characteristics, and electronic properties of the substrates, which can be interpreted mechanistically. The explained variance of the QSARs varied between 20% and 70%, and it was larger for Log(1/Km) than for LogVmax. The increase of 1/Km with compound logP and size suggests that weak interactions are important, e.g. by substrate binding via desolvation processes. The importance of electronic factors for 1/Km was described in relation to the catalytic mechanism of the enzymes. Vmax was particularly influenced by electronic properties, such as dipole moment and energy of the lowest unoccupied molecular orbital. This can be explained by the nature of the catalysis, characterised by the cleavage and formation of covalent or ionic bonds (strong interactions). The present study may be helpful to understand the underlying principles of the chemical specific activity of four important oxidising enzymes.

A simple litmus test for aldehyde oxidase metabolism of heteroarenes

The bioavailability of aromatic azaheterocyclic drugs can be affected by the activity of aldehyde oxidase (AO). Susceptibility to AO metabolism is difficult to predict computationally and can be complicated in vivo by differences between species. Here we report the use of bis(((difluoromethyl)sulfinyl)oxy)zinc (DFMS) as a source of CF₂H radical for a rapid and inexpensive chemical “litmus test” for the early identification of heteroaromatic drug candidates that have a high probability of metabolism by AO.

PBTK modelling platforms and parameter estimation tools to enable animal-free risk assessment: recommendations from a joint EPAA--EURL ECVAM ADME workshop

Information on toxicokinetics is critical for animal-free human risk assessment. Human external exposure must be translated into human tissue doses and compared with in vitro actual cell exposure associated to effects (in vitro-in vivo comparison). Data on absorption, distribution, metabolism and excretion in humans (ADME) could be generated using in vitro and QSAR tools. Physiologically-based toxicokinetic (PBTK) computer modelling could serve to integrate disparate in vitro and in silico findings. However, there are only few freely-available PBTK platforms currently available. And although some ADME parameters can be reasonably estimated in vitro or in silico, important gaps exist. Examples include unknown or limited applicability domains and lack of (high-throughput) tools to measure penetration of barriers, partitioning between blood and tissues and metabolic clearance. This paper is based on a joint EPAA--EURL ECVAM expert meeting. It provides a state-of-the-art overview of the availability of PBTK platforms as well as the in vitro and in silico methods to parameterise basic (Tier 1) PBTK models. Five high-priority issues are presented that provide the prerequisites for wider use of non-animal based PBTK modelling for animal-free chemical risk assessment.

In silico physicochemical parameter predictions

Drug discovery is a complex process with the aim of discovering efficacious molecules where their potency and selectivity are balanced against ADMET properties to set the appropriate dose and dosing interval. The link between physicochemical properties and molecular structure are well established. The subsequent connections between physicochemical properties and a drug's biological behavior provide an indirect link back to structure, facilitating the prediction of a biological property as a consequence of a particular molecular manipulation. Due to this understanding, during early drug discovery in vitro physicochemical property assays are commonly performed to eliminate compounds with properties commensurate with high attrition risks. However, the goal is to accurately predict physicochemical properties to prevent the synthesis of high risk compounds and hence minimize wasted drug discovery efforts. This paper will review the relevance to ADMET behaviors of key physicochemical properties, such as ionization, aqueous solubility, hydrogen bonding strength and hydrophobicity, and the in silico methodology for predicting them.

Regioselective oxidation of phospho-NSAIDs by human cytochrome P450 and flavin monooxygenase isoforms: implications for their pharmacokinetic properties and safety

BACKGROUND AND PURPOSE

Phospho-ibuprofen (MDC-917) and phospho-sulindac (OXT-328) are highly effective in cancer and arthritis treatment in preclinical models. Here, we investigated their metabolism by major human cytochrome P450s (CYPs) and flavin monooxygenases (FMOs).

EXPERIMENTAL APPROACH

The CYP/FMO-catalysed metabolism of phospho-ibuprofen and phospho-sulindac was studied by using in silico prediction modelling and a direct experimental approach.

KEY RESULTS

The CYP isoforms catalyse the oxidation of non-steroidal anti-inflammatory drugs (NSAIDs) and phospho-NSAIDs, with distinct activity and regioselectivity. CYP1A2, 2C19, 2D6 and 3A4 oxidize phospho-ibuprofen, but not ibuprofen; whereas CYP2C9 oxidizes ibuprofen, but not phospho-ibuprofen. All CYPs tested oxidize phospho-sulindac, but not sulindac. Among the five CYPs evaluated, CYP3A4 and 2D6 are the most active in the oxidation of phospho-ibuprofen and phospho-sulindac respectively. FMOs oxidized phospho-sulindac and sulindac, but not phospho-ibuprofen or ibuprofen. FMOs were more active towards phospho-sulindac than sulindac, indicating that phospho-sulindac is a preferred substrate of FMOs. The susceptibility of phospho-NSAIDs to CYP/FMO-mediated metabolism was also reflected in their rapid oxidation by human and mouse liver microsomes, which contain a full complement of CYPs and FMOs. Compared with conventional NSAIDs, the higher activity of CYPs towards phospho-ibuprofen and phospho-sulindac may be due to their greater lipophilicity, a key parameter for CYP binding.

CONCLUSIONS AND IMPLICATIONS

CYPs and FMOs play an important role in the metabolism of phospho-NSAIDs, resulting in differential pharmacokinetic profiles between phospho-NSAIDs and NSAIDs in vivo. The consequently more rapid detoxification of phospho-NSAIDs is likely to contribute to their greater safety.

In vitro and in vivo metabolic studies of phospho-aspirin (MDC-22)

PURPOSE

To investigate the metabolism of phospho-aspirin (PA, MDC-22), a novel anti-cancer and anti-inflammatory agent.

METHODS

The metabolism of PA was studied in the liver and intestinal microsomes from mouse, rat and human.

RESULTS

PA is rapidly deacetylated to phospho-salicylic acid (PSA), which undergoes regioselective oxidation to generate 3-OH-PSA and 5-OH-PSA. PSA also can be hydrolyzed to give salicylic acid (SA), which can be further glucuronidated. PA is far more stable in human liver or intestinal microsomes compared to those from mouse or rat due to its slowest deacetylation in human microsomes. Of the five major human cytochrome P450 (CYP) isoforms, CYP2C19 and 2D6 are the most active towards PSA. In contrast to PSA, conventional SA is not appreciably oxidized by the CYPs and liver microsomes, indicating that PSA is a preferred substrate of CYPs. Similarly, PA, in contrast to PSA, cannot be directly oxidized by CYPs and liver microsomes, indicating that the acetyl group of PA abrogates its oxidation by CYPs.

CONCLUSIONS

Our findings establish the metabolism of PA, reveal significant inter-species differences in its metabolic transformations, and provide an insight into the role of CYPs in these processes.

Quantitative Property-Property Relationship for Screening-Level Prediction of Intrinsic Clearance of Volatile Organic Chemicals in Rats and Its Integration within PBPK Models to Predict Inhalation Pharmacokinetics in Humans

The objectives of this study were (i) to develop a screening-level Quantitative property-property relationship (QPPR) for intrinsic clearance (CL_int) obtained from in vivo animal studies and (ii) to incorporate it with human physiology in a PBPK model for predicting the inhalation pharmacokinetics of VOCs. CL_int, calculated as the ratio of the in vivo V_max (μmol/h/kg bw rat) to the K_m (μM), was obtained for 26 VOCs from the literature. The QPPR model resulting from stepwise linear regression analysis passed the validation step (R² = 0.8; leave-one-out cross-validation Q² = 0.75) for CL_int normalized to the phospholipid (PL) affinity of the VOCs. The QPPR facilitated the calculation of CL_int (L PL/h/kg bw rat) from the input data on log P_ow, log blood: water PC and ionization potential. The predictions of the QPPR as lower and upper bounds of the 95% mean confidence intervals (LMCI and UMCI, resp.) were then integrated within a human PBPK model. The ratio of the maximum (using LMCI for CL_int) to minimum (using UMCI for CL_int) AUC predicted by the QPPR-PBPK model was 1.36 ± 0.4 and ranged from 1.06 (1,1-dichloroethylene) to 2.8 (isoprene). Overall, the integrated QPPR-PBPK modeling method developed in this study is a pragmatic way of characterizing the impact of the lack of knowledge of CL_int in predicting human pharmacokinetics of VOCs, as well as the impact of prediction uncertainty of CL_int on human pharmacokinetics of VOCs.

Selecting Relevant Descriptors for Classification by Bayesian Estimates: A Comparison with Decision Trees and Support Vector Machines Approaches for Disparate Data Sets

Classification algorithms suffer from the curse of dimensionality, which leads to overfitting, particularly if the problem is over-determined. Therefore it is of particular interest to identify the most relevant descriptors to reduce the complexity. We applied Bayesian estimates to model the probability distribution of descriptors values used for binary classification using n-fold cross-validation. As a measure for the discriminative power of the classifiers, the symmetric form of the Kullback-Leibler divergence of their probability distributions was computed. We found that the most relevant descriptors possess a Gaussian-like distribution of their values, show the largest divergences, and therefore appear most often in the cross-validation scenario. The results were compared to those of the LASSO feature selection method applied to multiple decision trees and support vector machine approaches for data sets of substrates and nonsubstrates of three Cytochrome P450 isoenzymes, which comprise strongly unbalanced compound distributions. In contrast to decision trees and support vector machines, the performance of Bayesian estimates is less affected by unbalanced data sets. This strategy reveals those descriptors that allow a simple linear separation of the classes, whereas the superior accuracy of decision trees and support vector machines can be attributed to nonlinear separation, which are in turn more prone to overfitting.

Novel natural inhibitors of CYP1A2 identified by in silico and in vitro screening

Inhibition of cytochrome P450 (CYP) is a major cause of herb–drug interactions. The CYP1A2 enzyme plays a major role in the metabolism of drugs in humans. Its broad substrate specificity, as well as its inhibition by a vast array of structurally diverse herbal active ingredients, has indicated the possibility of metabolic herb–drug interactions. Therefore nowadays searching inhibitors for CYP1A2 from herbal medicines are drawing much more attention by biological, chemical and pharmological scientists. In our work, a pharmacophore model as well as the docking technology is proposed to screen inhibitors from herbal ingredients data. Firstly different pharmaphore models were constructed and then validated and modified by 202 herbal ingredients. Secondly the best pharmaphore model was chosen to virtually screen the herbal data (a curated database of 989 herbal compounds). Then the hits (147 herbal compounds) were continued to be filtered by a docking process, and were tested in vitro successively. Finally, five of eighteen candidate compounds (272, 284, 300, 616 and 817) were found to have inhibition of CYP1A2 activity. The model developed in our study is efficient for in silico screening of large herbal databases in the identification of CYP1A2 inhibitors. It will play an important role to prevent the risk of herb–drug interactions at an early stage of the drug development process.

QSARs for PBPK modelling of environmental contaminants

Physiologically-based pharmacokinetic (PBPK) models are increasingly finding use in risk assessment applications of data-rich compounds. However, it is a challenge to determine the chemical-specific parameters for these models, particularly in time- and resource-limiting situations. In this regard, SARs, QSARs and QPPRs are potentially useful for computing the chemical-specific input parameters of PBPK models. Based on the frequency of occurrence of molecular fragments (CH(3), CH(2), CH, C, C=C, H, benzene ring and H in benzene ring structure) and exposure conditions, the available QSAR-PBPK models facilitate the simulation of tissue and blood concentrations for some inhaled volatile organic chemicals. The application domain of existing QSARs for developing PBPK models is limited, due to lack of relevant data for diverse chemicals and mechanisms. Even though this approach is conceptually applicable to non-volatile and high molecular weight organics as well, it is more challenging to predict the other PBPK model parameters required for modelling the kinetics of these chemicals (particularly tissue diffusion coefficients, association constants for binding and oral absorption rates). As the level of our understanding of the mechanistic basis of toxicokinetic processes improves, QSARs to provide a priori predictions of key chemical-specific PBPK parameters can be developed to expedite the internal dose-based health risk assessments in data-poor situations.

Structure, function, regulation and polymorphism and the clinical significance of human cytochrome P450 1A2

Human CYP1A2 is one of the major CYPs in human liver and metabolizes a number of clinical drugs (e.g., clozapine, tacrine, tizanidine, and theophylline; n > 110), a number of procarcinogens (e.g., benzo[a]pyrene and aromatic amines), and several important endogenous compounds (e.g., steroids). CYP1A2 is subject to reversible and/or irreversible inhibition by a number of drugs, natural substances, and other compounds. The CYP1A gene cluster has been mapped on to chromosome 15q24.1, with close link between CYP1A1 and 1A2 sharing a common 5'-flanking region. The human CYP1A2 gene spans almost 7.8 kb comprising seven exons and six introns and codes a 515-residue protein with a molecular mass of 58,294 Da. The recently resolved CYP1A2 structure has a relatively compact, planar active site cavity that is highly adapted for the size and shape of its substrates. The architecture of the active site of 1A2 is characterized by multiple residues on helices F and I that constitutes two parallel substrate binding platforms on either side of the cavity. A large interindividual variability in the expression and activity of CYP1A2 has been observed, which is largely caused by genetic, epigenetic and environmental factors (e.g., smoking). CYP1A2 is primarily regulated by the aromatic hydrocarbon receptor (AhR) and CYP1A2 is induced through AhR-mediated transactivation following ligand binding and nuclear translocation. Induction or inhibition of CYP1A2 may provide partial explanation for some clinical drug interactions. To date, more than 15 variant alleles and a series of subvariants of the CYP1A2 gene have been identified and some of them have been associated with altered drug clearance and response and disease susceptibility. Further studies are warranted to explore the clinical and toxicological significance of altered CYP1A2 expression and activity caused by genetic, epigenetic, and environmental factors.

Health effects from long-range transported contaminants in Arctic top predators: An integrated review based on studies of polar bears and relevant model species

The aim of this review is to provide a thorough overview of the health effects from the complexed biomagnified mixture of long-range transported industrial organochlorines (OCs), polybrominated diphenyl ethers (PBDEs), perfluorinated compounds (PFCs) and mercury (Hg) on polar bear (Ursus maritimus) health. Multiple scientific studies of polar bears indicate negative relationships between exposure to these contaminants and health parameters; however, these are all of a correlative nature and do not represent true cause-and-effects. Therefore, information from controlled studies of farmed Norwegian Arctic foxes (Vulpes lagopus) and housed East and West Greenland sledge dogs (Canis familiaris) were included as supportive weight of evidence in the clarification of contaminant exposure and health effects in polar bears. The review showed that hormone and vitamin concentrations, liver, kidney and thyroid gland morphology as well as reproductive and immune systems of polar bears are likely to be influenced by contaminant exposure. Furthermore, exclusively based on polar bear contaminant studies, bone density reduction and neurochemical disruption and DNA hypomethylation of the brain stem seemed to occur. The range of tissue concentration, at which these alterations were observed in polar bears, were ca. 1-70,000 ng/g lw for OCs (blood plasma concentrations of some PCB metabolites even higher), ca. 1-1000 ng/g lw for PBDEs and for PFCs and Hg 114-3052 ng/g ww and 0.1-50 microg/g ww, respectively. Similar concentrations were found in farmed foxes and housed sledge dogs while the lack of dose response designs did not allow an estimation of threshold levels for oral exposure and accumulated tissue concentrations. Nor was it possible to pinpoint a specific group of contaminants being more important than others nor analyze their interactions. For East Greenland polar bears the corresponding daily SigmaOC and SigmaPBDE oral exposure was estimated to be 35 and 0.34 microg/kg body weight, respectively. Furthermore, PFC concentrations, at which population effect levels could occur, are likely to be reached around year 2012 for the East Greenland polar bear subpopulation if current increasing temporal trends continue. Such proposed reproductive population effects were supported by physiological based pharmacokinetic (PBPK) modelling of critical body residues (CBR) with risk quotients >or=1 for SigmaPCB, dieldrin, SigmaPFC and SigmaOHC (organohalogen contaminant). The estimated daily TEQ for East Greenland polar bears and East Greenland sledge dogs were 32-281-folds above WHO SigmaTEQ guidelines for humans. Compared to human tolerable daily intake (TDI), these were exceeded for PCBs, dieldrin, chlordanes and SigmaHCH in East Greenland polar bears. Comparisons like these should be done with caution, but together with the CBR modelling and T-score estimations, these were the only available tools for polar bear risk evaluation. In conclusion, polar bears seem to be susceptible to contaminant induced stress that may have an overall sub-clinical impact on their health and population status via impacts on their immune and reproductive systems.

QSAR of cytochrome inhibitors

Cytochrome P450 (CYP450) enzymes are predominantly involved in the Phase I metabolism of xenobiotics. Metabolic inhibition and induction can give rise to clinically important drug-drug interactions. Metabolic stability is a prerequisite for sustaining the therapeutically relevant concentrations, and very often drug candidates are sacrificed due to poor metabolic profiles. Computational tools such as quantitative structure-activity relationships are widely used to study different metabolic end points successfully to accelerate the drug discovery process. There are a lot of computational studies on clinically important CYPs already reported in recent years. But other clinically significant families are to yet be explored computationally. Powerfulness of quantitative structure-activity relationship will drive computational chemists to develop new potent and selective inhibitors of different classes of CYPs for the treatment of different diseases with least drug-drug interactions. Furthermore, there is a need to enhance the accuracy, interpretability and confidence in the computational models in accelerating the drug discovery pathways.

Comparative QSAR studies of CYP1A2 inhibitor flavonoids using 2D and 3D descriptors

Comparative Quantitative Structure Activity Relationship (QSAR) analyses have been performed with 21 naturally occurring flavonoids for their inhibitory effects on cytochrome P450 1A2 enzyme using two-dimensional (topological, structural, and thermodynamic) and three-dimensional (spatial) descriptors. The chemometric tools used for the analyses are stepwise multiple linear regression, partial least squares, genetic function approximation, and genetic partial least squares. The data set was divided into a training set (n = 15) and test set (n = 6), based on K-means clustering technique applied on standardized two-dimensional descriptor matrix, and models were developed from the training set compounds. The best model (genetic partial least squares model using two-dimensional descriptors) was selected based on the highest external predictive R(2) (R(2)(pred)) value (0.840) and the lowest root mean square error of prediction value (0.351). The developed QSAR equations suggest the importance of the double bond present at 2 and 3 positions and requirement of absence of hydroxyl substituent or glycosidic linkage at 3 position of the 1,4-benzopyrone nucleus. Furthermore, the phenyl ring present at 2 position of the 1,4-benzopyrone ring should not be substituted with hydroxyl group. Moreover, hydroxyl groups present at 5 and 7 positions of the benzopyran nucleus should not be glycosylated for good cytochrome P450 1A2 enzyme inhibitory activity.

SPORCalc: A development of a database analysis that provides putative metabolic enzyme reactions for ligand-based drug design

Understanding both the enzyme reactions that contribute to intermediate metabolism and the biochemical fate of candidate therapeutic and toxic agents are essential for drug design. Traditional metabolic databases indicate whether reactions have been observed but do not provide the likelihoods of reactions occurring, for example those of mixed function oxygenases and oxidases, during phase I metabolism. The desire for more quantitative predictions motivated the development of the recently introduced Substrate Product Occurrence Ratio Calculator (SPORCalc) that identifies metabolically labile atom positions in candidate compounds. This paper describes a further development and provides a clearer explanation of SPORCalc for the computational pharmacology, medicinal chemistry and drug design communities interested in metabolic prediction of xenobiotics using chemical databases of biotransformations. Examples of reaction centre detection in Metabolite are described followed by a demonstration of almokalant, an anti-arrhythmic agent, undergoing phase I metabolism. In general, occurrence ratio (OR) values are calculated throughout a compound and its transformed metabolites to give propensity (p) values at each atom position. The OR values from substrates and products in the database are essential for addition and elimination reactions. For almokalant, the resulting p values ranged from 10(-1) to 10(-5) and their order of magnitude reflected the known and experimentally observed metabolites. SPORCalc depends entirely on the level of detail from isoform- or species-specific reaction classes in Metabolite. Labile atom positions (sites of metabolism) are identified in both the candidate compound and its metabolites. In general, the likelihood of one enzyme isoform-dependent reaction occurring relative to another and the putative metabolic routes from different isoforms can be investigated. SPORCalc can be developed further to include suitable three-dimensional, structure-activity and physiochemical information.

Hepatic cytochrome P450 activity and pollutant concentrations in paradise shelducks and southern black-backed gulls in the South Island of New Zealand

Cytochrome P450 (CYP) enzymes catalyse the oxidative metabolism of various xenobiotics including environmental pollutants. We investigated liver microsomal CYP marker activities in 60 paradise shelducks (Tadorna variegata; herbivore) and 77 southern black-backed gulls (Larus dominicanus; omnivore) collected at three sites with putatively different levels of pollution in the South Island of New Zealand. Ethoxyresorufin O-deethylase (EROD) activity was high in birds at an urban landfill site compared to those at a relatively pristine and an agricultural site. Analysis of p-nitrophenol hydroxylase and erythromycin demethylase activities indicated the presence of two additional CYP isoforms in shelducks but no additional form in gulls. Total polychlorinated biphenyl (PCB) concentrations (ranges: shelducks, 0.073-6.2; gulls, 8.2-330 ng/g wet weight) were high in landfill samples suggesting a link to EROD induction and, in landfill shelducks, EROD was independently associated with Hg and Pb concentration. PCB congener-specific assessments indicated the metabolism of at least two congeners (#28 and #74) is induced in shelducks. DDE concentrations (ranges: shelducks, 0.85-320; gulls, 44-4800 ng/g) were high in birds at the landfill and agricultural sites. Body weight tended to be lower in landfill birds, but whether this reflects the greater energetic demands of pollutant detoxification remains to be investigated.

Computational prediction of drug binding and rationalisation of selectivity towards cytochromes P450

BACKGROUND

Early in-vitro consideration of metabolism and inhibition of cytochrome P450 has proven its merits over the last 15 years. Simultaneously, many computational drug-design methods have been developed, and are being applied to study the interactions between drug candidates and cytochrome P450 enzymes (P450s).

OBJECTIVE

This review discusses the recent advances of these methods and the implications that are specific for P450s.

METHODS

Mainly focusing on the prediction of binding affinity and ligand selectivity, we outline the applicability of the different methods to answer specific questions. Special emphasis is put on the different levels of theory that are being used in recent computational descriptions of ligand-P450 interactions.

CONCLUSION

P450s offer an additional challenge for computational methods, considering the ambiguities of the catalytic cycle and the significant flexibility of the active site. Different computational methods display different limitations, which is crucial to take into account when choosing the method appropriate to each application.

Variation in CYP1A2 activity and its clinical implications: influence of environmental factors and genetic polymorphisms

CYP1A2 is involved in the metabolism of several widely used drugs and endogenous compounds, and in the activation of procarcinogens. Both genetic and environmental factors influence the activity of this enzyme. The current knowledge regarding factors influencing the activity of CYP1A2 is summarized in this review. Substrates, inhibitors and inducers of CYP1A2 activity, as well as phenotyping probes, are discussed. The functional significance and clinical importance of CYP1A2 gene polymorphisms are reviewed and interethnic differences in the distribution of CYP1A2 variant alleles and haplotypes are summarized. Finally, future perspectives for the possible clinical applications of CYP1A2 genotyping are discussed.

Selectivity of 7-alkoxycoumarins as probe substrates for rat hepatic cytochrome P450 forms is influenced by the substitution pattern on the coumarin nucleus

1. The O-dealkylation of 7-alkoxycoumarins is widely used as an assay to characterize cytochrome P450 (CYP) activity. These substrates can also undergo oxidative attack at additional sites on the coumarin nucleus, which may influence their apparent selectivity for particular CYP forms. 2. Accordingly, the effect of blockade of these additional sites was investigated on the selectivity towards rat hepatic CYP forms, with emphasis on the CYP1A and 2B forms. 3. Blockade of the 3-/4- and 6-positions resulted in substrates for which the CYP1A1/2 selectivity of the unsubstituted 7-alkoxycoumarins was altered to a CYP2B selectivity; this was achieved with little overall change in the molecular dimensions of the substrate. Limited analysis of other inducible CYP forms indicated at most only small effects of structure modification on activity. 4. The findings suggest that the sensitivity of probe substrates for CYP forms may be limited by the occurrence of competing side reactions of the substrate, and that better probes may be derived by blocking the sites of these side reactions.

Homology modelling of CYP3A4 from the CYP2C5 crystallographic template: analysis of typical CYP3A4 substrate interactions

1. The results of homology modelling of cytochrome P4503A4 (CYP3A4), which is a human enzyme of major importance for the Phase 1 metabolism of drug substrates, from the CYP2C5 crystal structure is reported. 2. The overall homology between the two protein sequences was generally good (46%) with 24% of amino acid residues being identical and a 22% similarity between matched pairs in the CYP3A4 and CYP2C5 aligned sequences, thus indicating that CYP2C5 represents a viable template for modelling CYP3A4 by homology. 3. The CYP3A4 model appears to show consistency with the reported findings from the extensive site-directed mutagenesis studies already published. 4. Typical CYP3A4 substrates, such as midazolam, testosterone, nifedipine and verapamil, are shown to fit the putative active site of the enzyme structure in a manner consistent with their known positions of metabolism.

Generation of a set of simple, interpretable ADMET rules of thumb

A set of simple, consistent structure-property guides have been determined from an analysis of a number of key ADMET assays run within GSK: solubility, permeability, bioavailability, volume of distribution, plasma protein binding, CNS penetration, brain tissue binding, P-gp efflux, hERG inhibition, and cytochrome P450 1A2/2C9/2C19/2D6/3A4 inhibition. The rules have been formulated using molecular properties that chemists intuitively know how to alter in a molecule, namely, molecular weight, logP, and ionization state. The rules supplement the more predictive black-box models available to us by clearly illustrating the key underlying trends, which are in line with reports in the literature. It is clear from the analyses reported herein that almost all ADMET parameters deteriorate with either increasing molecular weight, logP, or both, with ionization state playing either a beneficial or detrimental affect depending on the parameter in question. This study re-emphasizes the need to focus on a lower molecular weight and logP area of physicochemical property space to obtain improved ADMET parameters.

Changes in the expression of cytochrome P450 genes in hemin-induced differentiated K562 cells

We have previously reported the expression of CYP genes in human myeloblastic and lymphoid cell lines, and the induction of the CYP3A4 and GSTP1 genes by oxidative stress in the human erythroleukemia cell line, K562. To further elucidate the role of drug metabolizing enzymes in hematogenesis, we have characterized the expression of CYP genes in hemin-induced differentiated K562 cells. After incubation with 50 microM hemin for 3 d, the expression of CYP1A1 and CYP3A4 genes was induced by 2.5- and 3.5-fold, respectively. In contrast, the CYP1B1 and CYP2E1 genes were downregulated in these cells to below 10% of the control levels. Moreover, these changes correlated with the hemin dose and culture time. Metabolism of midazolam, a probe substrate for CYP3A4, in the differentiated K562 cells increased by 2-folds, suggesting that the induction of CYP3A4 activity is consistent with the mRNA level. If these changes in the CYP expression profile in hematopoietic cells occurred, the susceptibility to xenobiotics and/or the therapeutic drugs of the cells might be influenced, and it also affects the metabolism of endogenous substrates, such as steroids and prostaglandins.