The Metabolic Rainbow: Deep Learning Phase I Metabolism in Five Colors

Effect of sublethal concentrations of the bioinsecticide spinosyn treatment of Trichoplusia ni eggs on the caterpillar and its parasitoid, Trichogramma brassicae

BACKGROUND

Integrated Pest Management (IPM) seeks to combine multiple management strategies for optimal pest control. One method that is successfully employed in IPM is the use of beneficial organisms. However, in severe circumstances when pest insects exceed threshold limits, insecticides may still need to be implemented. Thus, understanding the effects of insecticides on biocontrol agents, such as parasitoid wasps, is paramount to ensure sustainable agroecosystems. Sublethal effects of the bioinsecticide spinosyn, a mixture of the bacterial Saccharopolyspora spinosa (Mertz and Yao) fermentation products spinosyn A and D, on eggs of Trichoplusia ni (Hübner), a cruciferous crop pest, and its egg parasitoid Trichogramma brassicae (Bezdenko) was investigated.

RESULTS

The LC50 for spinosyn A and D (dissolved in ethanol) on T. ni eggs is 54 ng mL-1. Transcriptomics on caterpillars (1st and 3rd instars) that hatched from eggs treated with sublethal concentrations of spinosyn identified the upregulation of several genes encoding proteins that may be involved in insecticide resistance including detoxification enzymes, such as cytochrome P450s, glutathione S-transferases and esterases. Sublethal T. ni egg treatments did not affect parasitoid emergence, however, there was a marked increase in the size of T. brassicae hind tibia and wings that emerged from spinosyn-treated eggs.

CONCLUSIONS

For the caterpillar, treatment of eggs with sublethal concentrations of spinosyn may induce insecticide resistance mechanisms. For the parasitoids, their increased size when reared in spinosyn-treated eggs suggests that the emerged wasps may have higher performance. © 2024 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

DOCK-PET: database of CNS kinetic parameters in the healthy human brain for existing PET tracers

PURPOSE

Information about developed positron emission tomography (PET) tracers and obtained clinical PET images is publicly available in a database. However, findings regarding the kinetic parameters of PET tracers are yet to be summarized. Therefore, in this study, we created an open-access database of central nervous system (CNS) kinetic parameters in the healthy human brain for existing PET tracers (DOCK-PET).

METHODS

Our database includes information on the kinetic parameters and compounds of existing CNS-PET tracers. The kinetic parameter dataset comprises the analysis methods, VT, BPND, K parameters, relevant literature, and study details. The list of PET tracers and kinetic parameter information was compiled through keyword-based searches of PubMed and the Molecular Imaging and Contrast Agent Database (MICAD). The kinetic parameters obtained, including VT, BPND, and K parameters, were reorganized based on the defined brain anatomical regions. All data were rigorously double-checked before being summarized in Microsoft Excel and JavaScript Object Notation (JSON) formats.

RESULTS

Of the 247 PET tracers identified through searches using the PubMed and MICAD websites, the kinetic parameters of 120 PET tracers were available. Among the 120 PET tracers, compound structures with chemical and physical properties were obtained from the PubChem website or the ChemDraw software. Furthermore, the affinity information of the 104 PET tracers was gathered from PubChem or extensive literature surveys of the 120 PET tracers.

CONCLUSIONS

We developed a comprehensive open-access database, DOCK-PET, that includes both kinetic parameters of healthy humans and compound information for existing CNS-PET tracers.

Exploration of morpholine-thiophene hybrid thiosemicarbazones for the treatment of ureolytic bacterial infections via targeting urease enzyme: Synthesis, biochemical screening and computational analysis

An important component of the pathogenicity of potentially pathogenic bacteria in humans is the urease enzyme. In order to avoid the detrimental impact of ureolytic bacterial infections, the inhibition of urease enzyme appears to be an appealing approach. Therefore, in the current study, morpholine-thiophene hybrid thiosemicarbazone derivatives (5a-i) were designed, synthesized and characterized through FTIR, 1H NMR, 13C NMR spectroscopy and mass spectrometry. A range of substituents including electron-rich, electron-deficient and inductively electron-withdrawing groups on the thiophene ring was successfully tolerated. The synthesized derivatives were evaluated in vitro for their potential to inhibit urease enzyme using the indophenol method. The majority of compounds were noticeably more potent than the conventional inhibitor, thiourea. The lead inhibitor, 2-(1-(5-chlorothiophen-2-yl)ethylidene)-N-(2-morpholinoethyl)hydrazinecarbothioamide (5g) inhibited the urease in an uncompetitive manner with an IC50 value of 3.80 ± 1.9 µM. The findings of the docking studies demonstrated that compound 5g has a strong affinity for the urease active site. Significant docking scores and efficient binding free energies were displayed by the lead inhibitor. Finally, the ADME properties of lead inhibitor (5g) suggested the druglikeness behavior with zero violation.

Polymorphisms in drug-metabolizing genes and urinary bladder cancer susceptibility and prognosis: Possible impacts and future management

Epidemiological studies have shown the association of genetic variants with risks of occupational and environmentally induced cancers, including bladder (BC). The current review summarizes the effects of variants in genes encoding phase I and II enzymes in well-designed studies to highlight their contribution to BC susceptibility and prognosis. Polymorphisms in genes codifying drug-metabolizing proteins are of particular interest because of their involvement in the metabolism of exogenous genotoxic compounds, such as tobacco and agrochemicals. The prognosis between muscle-invasive and non-muscle-invasive diseases is very different, and it is difficult to predict which will progress worse. Web of Science, PubMed, and Medline were searched to identify studies published between January 1, 2010, and February 2023. We included 73 eligible studies, more than 300 polymorphisms, and 46 genes/loci. The most studied candidate genes/loci of phase I metabolism were CYP1B1, CYP1A1, CYP1A2, CYP3A4, CYP2D6, CYP2A6, CYP3E1, and ALDH2, and those in phase II were GSTM1, GSTT1, NAT2, GSTP1, GSTA1, GSTO1, and UGT1A1. We used the 46 genes to construct a network of proteins and to evaluate their biological functions based on the Reactome and KEGG databases. Lastly, we assessed their expression in different tissues, including normal bladder and BC samples. The drug-metabolizing pathway plays a relevant role in BC, and our review discusses a list of genes that could provide clues for further exploration of susceptibility and prognostic biomarkers.

D-CyPre: a machine learning-based tool for accurate prediction of human CYP450 enzyme metabolic sites

The advancement of graph neural networks (GNNs) has made it possible to accurately predict metabolic sites. Despite the combination of GNNs with XGBOOST showing impressive performance, this technology has not yet been applied in the realm of metabolic site prediction. Previous metabolic site prediction tools focused on bonds and atoms, regardless of the overall molecular skeleton. This study introduces a novel tool, named D-CyPre, that amalgamates atom, bond, and molecular skeleton information via two directed message-passing neural networks (D-MPNN) to predict the metabolic sites of the nine cytochrome P450 enzymes using XGBOOST. In D-CyPre Precision Mode, the model produces fewer, but more accurate results (Jaccard score: 0.497, F1: 0.660, and precision: 0.737 in the test set). In D-CyPre Recall Mode, the model produces less accurate, but more comprehensive results (Jaccard score: 0.506, F1: 0.669, and recall: 0.720 in the test set). In the test set of 68 reactants, D-CyPre outperformed BioTransformer on all isoenzymes and CyProduct on most isoenzymes (5/9). For the subtypes where D-CyPre outperformed CyProducts, the Jaccard score and F1 scores increased by 24% and 16% in Precision Mode (4/9) and 19% and 12% in Recall Mode (5/9), respectively, relative to the second-best CyProduct. Overall, D-CyPre provides more accurate prediction results for human CYP450 enzyme metabolic sites.

Trichoplusia ni Transcriptomic Responses to the Phytosaponin Aglycone Hederagenin: Sex-Related Differences

Many plant species, particularly legumes, protect themselves with saponins. Previously, a correlation was observed between levels of oleanolic acid-derived saponins, such as hederagenin-derived compounds, in the legume Medicago truncatula and caterpillar deterrence. Using concentrations that reflect the foliar levels of hederagenin-type saponins, the sapogenin hederagenin was not toxic to 4th instar caterpillars of the cabbage looper Trichoplusia ni nor did it act as a feeding deterrent. Female caterpillars consumed more diet than males, presumably to obtain the additional nutrients required for oogenesis, and are, thus, exposed to higher hederagenin levels. When fed the hederagenin diet, male caterpillars expressed genes encoding trypsin-like proteins (LOC113500509, LOC113501951, LOC113501953, LOC113501966, LOC113501965, LOC113499659, LOC113501950, LOC113501948, LOC113501957, LOC113501962, LOC113497819, LOC113501946, LOC113503910) as well as stress-responsive (LOC113503484, LOC113505107) proteins and cytochrome P450 6B2-like (LOC113493761) at higher levels than females. In comparison, female caterpillars expressed higher levels of cytochrome P450 6B7-like (LOC113492289). Bioinformatic tools predict that cytochrome P450s could catalyze the oxygenation of hederagenin which would increase the hydrophilicity of the compound. Expression of a Major Facilitator Subfamily (MFS) transporter (LOC113492899) showed a hederagenin dose-dependent increase in gene expression suggesting that this transporter may be involved in sapogenin efflux. These sex-related differences in feeding and detoxification should be taken into consideration in insecticide evaluations to minimize pesticide resistance.

In vitro metabolism of the emerging contaminant 6PPD-quinone in human and rat liver microsomes: Kinetics, pathways, and mechanism

N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine-quinone (6PPD-Q) is an ozonation product of the rubber antioxidant N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine (6PPD). 6PPD-Q has recently been detected in various environmental media, which may enter the human body via inhalation and skin contact pathways. However, the human metabolism of 6PPD-Q has remained unknown. This study investigated the in vitro Cytochrome P450-mediated metabolism of 6PPD-Q in human and rat liver microsomes (HLMs and RLMs). 6PPD-Q was significantly metabolized at lower concentrations but slowed at high concentrations. The intrinsic clearance (CLint) of 6PPD-Q was 21.10 and 18.58 μL min-1 mg-1 protein of HLMs and RLMs, respectively, suggesting low metabolic ability compared with other reported pollutants. Seven metabolites and one intermediate were identified, and metabolites were predicted immunotoxic or mutagenic toxicity. Mono- and di-oxygenation reactions were the main phase I in vitro metabolic pathways. Enzyme inhibition experiments and molecular docking techniques were further used to reveal the metabolic mechanism. CYP1A2, 3A4, and 2C19, especially CYP1A2, play critical roles in 6PPD-Q metabolism in HLMs, whereas 6PPD-Q is extensively metabolized in RLMs. Our study is the first to demonstrate the in vitro metabolic profile of 6PPD-Q in HLMs and RLMs. The results will significantly contribute to future human health management targeting the emerging pollutant 6PPD-Q.

Vanadium Complexes with Thioanilide Derivatives of Amino Acids: Inhibition of Human Phosphatases and Specificity in Various Cell Models of Metabolic Disturbances

In the text, the synthesis and characteristics of the novel ONS-type vanadium (V) complexes with thioanilide derivatives of amino acids are described. They showed the inhibition of human protein tyrosine phosphatases (PTP1B, LAR, SHP1, and SHP2) in the submicromolar range, as well as the inhibition of non-tyrosine phosphatases (CDC25A and PPA2) similar to bis(maltolato)oxidovanadium(IV) (BMOV). The ONS complexes increased [14C]-deoxy-D-glucose transport into C2C12 myocytes, and one of them, VC070, also enhanced this transport in 3T3-L1 adipocytes. These complexes inhibited gluconeogenesis in hepatocytes HepG2, but none of them decreased lipid accumulation in the non-alcoholic fatty liver disease model using the same cells. Compared to the tested ONO-type vanadium complexes with 5-bromosalicylaldehyde and substituted benzhydrazides as Schiff base ligand components, the ONS complexes revealed stronger inhibition of protein tyrosine phosphatases, but the ONO complexes showed greater activity in the cell models in general. Moreover, the majority of the active complexes from both groups showed better effects than VOSO4 and BMOV. Complexes from both groups activated AKT and ERK signaling pathways in hepatocytes to a comparable extent. One of the ONO complexes, VC068, showed activity in all of the above models, including also glucose utilizatiand ONO Complexes are Inhibitors ofon in the myocytes and glucose transport in insulin-resistant hepatocytes. The discussion section explicates the results within the wider scope of the knowledge about vanadium complexes.

Selective androgen receptor modulator use and related adverse events including drug-induced liver injury: Analysis of suspected cases

PURPOSE

Selective androgen receptor modulators (SARMs) have demonstrated agonist activity on the androgen receptor in various tissues, stimulating muscle mass growth and improving bone reconstruction. Despite being in clinical trials, none has been approved by the Food and Drug Administration (FDA) or European Medicines Agency for pharmacotherapy. Still, SARMs are very popular as performance-enhancing drugs. The FDA has issued warnings about the health risks associated with SARMs, but the long-term exposure and possible adverse events still need to be fully understood. This review aims to evaluate the adverse events associated with using SARMs by humans.

METHODS

PubMed database was searched from September 16, 2022, to October 2, 2023. In total, 20 records were included in the final review. Data from preclinical and clinical studies supported the review.

RESULTS

Since 2020, 20 reports of adverse events, most described as drug-induced liver injury associated with the use of SARM agonists, have been published. The main symptoms mentioned were cholestatic or hepatocellular liver injury and jaundice. Limited data are related to the dosages and purity of SARM supplements.

CONCLUSION

Promoting SARMs as an anabolic agent in combination with other performance-enhancing drugs poses a risk to users not only due to doping controls but also to health safety. The lack of quality control of consumed supplements makes it very difficult to assess the direct impact of SARMs on the liver and their potential hepatotoxic effects. Therefore, more detailed analyses are needed to determine the safety of using SARMs.

Dual Inhibition of Mycobacterium tuberculosis and the Host TGFBR1 by an Anilinoquinazoline

Tuberculosis (TB) control is complicated by the emergence of drug resistance. Promising strategies to prevent drug resistance are the targeting of nonreplicating, drug-tolerant bacterial populations and targeting of the host, but inhibitors and targets for either are still rare. In a cell-based screen of ATP-competitive inhibitors, we identified compounds with in vitro activity against replicating Mycobacterium tuberculosis (Mtb), and an anilinoquinazoline (AQA) that also had potent activity against nonreplicating and persistent Mtb. AQA was originally developed to inhibit human transforming growth factor receptor 1 (TGFBR1), a host kinase that is predicted to have host-adverse effects during Mtb infection. The structure-activity relationship of this dually active compound identified the pyridyl-6-methyl group as being required for potent Mtb inhibition but a liability for P450 metabolism. Pyrrolopyrimidine (43) emerged as the optimal compound that balanced micromolar inhibition of nonreplicating Mtb and TGFBR1 while also demonstrating improved metabolic stability and pharmacokinetic profiles.

Message Passing Neural Networks Improve Prediction of Metabolite Authenticity

Cytochrome P450 enzymes aid in the elimination of a preponderance of small molecule drugs, but can generate reactive metabolites that may adversely react with protein and DNA and prompt drug candidate attrition or market withdrawal. Previously developed models help understand how these enzymes modify molecule structure by predicting sites of metabolism or characterizing formation of metabolite-biomolecule adducts. However, the majority of reactive metabolites are formed by multiple metabolic steps, and understanding the progenitor molecule's network-level behavior necessitates an integrative approach that blends multiple site of metabolism and structure inference models. Our previously developed tool, XenoNet 1.0, generates metabolic networks, where nodes are molecules and weighted edges are metabolic transformations. We extend XenoNet with a bidirectional message passing neural network that integrates edge feature information and local network structure using edge-conditioned graph convolutions and jumping knowledge to predict the authenticity of inferred Phase I metabolite structures. Our model significantly outperformed prior work and algorithmic baselines on a data set of 311 networks and 6606 intermediates annotated using a chemically diverse set of 20 736 individual in vitro and in vivo reaction records accounting for 92.3% of all human Phase I metabolism in the Accelrys Metabolite Database. Cross-validated predictions resulted in area under the receiver operating characteristic curves of 88.5% and 87.6% for separating experimentally observed and unobserved metabolites at global and network levels, respectively. Further analysis verified robustness to networks of varying depth and breadth, accurate detection of metabolites, such as d,l-methamphetamine, that are experimentally observed or unobserved in different network contexts, extraction of important metabolic subnetworks, and identification of known bioactivation pathways, such as for nimesulide and terbinafine. By exploiting network structures, our approach accurately suggests unreported metabolites for experimental study and may rationalize modifications for avoiding deleterious pathways antecedent to reactive metabolite formation.

Using graph neural networks for site-of-metabolism prediction and its applications to ranking promiscuous enzymatic products

MOTIVATION

While traditionally utilized for identifying site-specific metabolic activity within a compound to alter its interaction with a metabolizing enzyme, predicting the site-of-metabolism (SOM) is essential in analyzing the promiscuity of enzymes on substrates. The successful prediction of SOMs and the relevant promiscuous products has a wide range of applications that include creating extended metabolic models (EMMs) that account for enzyme promiscuity and the construction of novel heterologous synthesis pathways. There is therefore a need to develop generalized methods that can predict molecular SOMs for a wide range of metabolizing enzymes.

RESULTS

This article develops a Graph Neural Network (GNN) model for the classification of an atom (or a bond) being an SOM. Our model, GNN-SOM, is trained on enzymatic interactions, available in the KEGG database, that span all enzyme commission numbers. We demonstrate that GNN-SOM consistently outperforms baseline machine learning models, when trained on all enzymes, on Cytochrome P450 (CYP) enzymes, or on non-CYP enzymes. We showcase the utility of GNN-SOM in prioritizing predicted enzymatic products due to enzyme promiscuity for two biological applications: the construction of EMMs and the construction of synthesis pathways.

AVAILABILITY AND IMPLEMENTATION

A python implementation of the trained SOM predictor model can be found at https://github.com/HassounLab/GNN-SOM.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Active Compounds in Zingiber officinale as Possible Redox Inhibitors of 5-Lipoxygenase Using an In Silico Approach

5-Lipoxygenase (5-LOX) converts arachidonic acid to lipidic inflammatory mediators such as leukotrienes (LTs). In diseases such as asthma, LTs contribute to a physiopathology that could be reverted by blocking 5-LOX. Natural products with anti-inflammatory potential such as ginger have been used as nutraceuticals since ancient times. 6-Gingerol and 6-shogaol are the most abundant compounds in the ginger rhizome; they possess anti-inflammatory, antioxidant, and chemopreventive properties. In the present study, 6-gingerol and 6-shogaol structures were analyzed and compared with two commercial 5-LOX inhibitors (zileuton and atreleuton) and with other inhibitor candidates (3f, NDGA, CP 209, caffeic acid, and caffeic acid phenethyl ester (CAPE)). The pharmacokinetics and toxicological properties of 6-gingerol, 6-shogaol, and the other compounds were evaluated. Targeted molecular coupling was performed to identify the optimal catalytic pocket for 5-LOX inhibition. The results showed that 6-gingerol and 6-shogaol follow all of the recommended pharmacokinetic parameters. These compounds could be inhibitors of 5-LOX because they present specific interactions with the residues involved in molecular inhibition. The current study demonstrated the potential of 6-gingerol and 6-shogaol as anti-inflammatory agents that inhibit 5-LOX, as they present a high level of performance in the toxicological analysis and could be catabolized by the cytochrome p450 enzymatic complex; however, 6-gingerol was superior in safety compared to 6-shogaol.

Discovery of Novel Reductive Elimination Pathway for 10-Hydroxywarfarin

Coumadin (R/S-warfarin) anticoagulant therapy is highly efficacious in preventing the formation of blood clots; however, significant inter-individual variations in response risks over or under dosing resulting in adverse bleeding events or ineffective therapy, respectively. Levels of pharmacologically active forms of the drug and metabolites depend on a diversity of metabolic pathways. Cytochromes P450 play a major role in oxidizing R- and S-warfarin to 6-, 7-, 8-, 10-, and 4′-hydroxywarfarin, and warfarin alcohols form through a minor metabolic pathway involving reduction at the C11 position. We hypothesized that due to structural similarities with warfarin, hydroxywarfarins undergo reduction, possibly impacting their pharmacological activity and elimination. We modeled reduction reactions and carried out experimental steady-state reactions with human liver cytosol for conversion of rac-6-, 7-, 8-, 4′-hydroxywarfarin and 10-hydroxywarfarin isomers to the corresponding alcohols. The modeling correctly predicted the more efficient reduction of 10-hydroxywarfarin over warfarin but not the order of the remaining hydroxywarfarins. Experimental studies did not indicate any clear trends in the reduction for rac-hydroxywarfarins or 10-hydroxywarfarin into alcohol 1 and 2. The collective findings indicated the location of the hydroxyl group significantly impacted reduction selectivity among the hydroxywarfarins, as well as the specificity for the resulting metabolites. Based on studies with R- and S-7-hydroxywarfarin, we predicted that all hydroxywarfarin reductions are enantioselective toward R substrates and enantiospecific for S alcohol metabolites. CBR1 and to a lesser extent AKR1C3 reductases are responsible for those reactions. Due to the inefficiency of reactions, only reduction of 10-hydroxywarfarin is likely to be important in clearance of the metabolite. This pathway for 10-hydroxywarfarin may have clinical relevance as well given its anticoagulant activity and capacity to inhibit S-warfarin metabolism.

Recent advances in computational metabolite structure predictions and altered metabolic pathways assessment to inform drug development processes

Many drug candidates fail during preclinical and clinical trials due to variable or unexpected metabolism which may lead to variability in drug efficacy or adverse drug reactions. The drug metabolism field aims to address this important issue from many angles which range from the study of drug-drug interactions, pharmacogenomics, computational metabolic modeling, and others. This manuscript aims to provide brief but comprehensive manuscript summaries highlighting the conclusions and scientific importance of seven exceptional manuscripts published in recent years within the field of drug metabolism. Two main topics within the field are reviewed: novel computational metabolic modeling approaches which provide complex outputs beyond site of metabolism predictions, and experimental approaches designed to discern the impacts of interindividual variability and species differences on drug metabolism. The computational approaches discussed provide novel outputs in metabolite structure and formation likelihood and/or extend beyond the saturated field of drug phase I metabolism, while the experimental metabolic pathways assessments aim to highlight the impacts of genetic polymorphisms and clinical animal model metabolic differences on human metabolism and subsequent health outcomes.

Cannabis and Cannabis Edibles: A Review

Cannabis is an excellent natural source of fiber and various bioactive cannabinoids. So far, at least 120 cannabinoids have been identified, and more novel cannabinoids are gradually being unveiled by detailed cannabis studies. However, cannabinoids in both natural and isolated forms are especially vulnerable to oxygen, heat, and light. Therefore, a diversity of cannabinoids is associated with their chemical instability to a large extent. The research status of structural conversion of cannabinoids is introduced. On the other hand, the use of drug-type cannabis and the phytocannabinoids thereof has been rapidly popularized and plays an indispensable role in both medical therapy and daily recreation. The recent legalization of edible cannabis further extends its application into the food industry. The varieties of legal edible cannabis products in the current commercial market are relatively monotonous due to rigorous restrictions under the framework of Cannabis Regulations and infancy of novel developments. Meanwhile, patents/studies related to the safety and quality assurance systems of cannabis edibles are still rare and need to be developed. Furthermore, along with cannabinoids, many phytochemicals such as flavonoids, lignans, terpenoids, and polysaccharides exist in the cannabis matrix, and these may exhibit prebiotic/probiotic properties and improve the composition of the gut microbiome. During metabolism and excretion, the bioactive phytochemicals of cannabis, mostly the cannabinoids, may be structurally modified during enterohepatic detoxification and gut fermentation. However, the potential adverse effects of both acute and chronic exposure to cannabinoids and their vulnerable groups have been clearly recognized. Therefore, a comprehensive understanding of the chemistry, metabolism, toxicity, commercialization, and regulations regarding cannabinoid edibles is reviewed and updated in this contribution.

Modeling the Bioactivation and Subsequent Reactivity of Drugs

Electrophilically reactive drug metabolites are implicated in many adverse drug reactions. In this mechanism-termed bioactivation-metabolic enzymes convert drugs into reactive metabolites that often conjugate to nucleophilic sites within biological macromolecules like proteins. Toxic metabolite-product adducts induce severe immune responses that can cause sometimes fatal disorders, most commonly in the form of liver injury, blood dyscrasia, or the dermatologic conditions toxic epidermal necrolysis and Stevens-Johnson syndrome. This study models four of the most common metabolic transformations that result in bioactivation: quinone formation, epoxidation, thiophene sulfur-oxidation, and nitroaromatic reduction, by synthesizing models of metabolism and reactivity. First, the metabolism models predict the formation probabilities of all possible metabolites among the pathways studied. Second, the exact structures of these metabolites are enumerated. Third, using these structures, the reactivity model predicts the reactivity of each metabolite. Finally, a feedfoward neural network converts the metabolism and reactivity predictions to a bioactivation prediction for each possible metabolite. These bioactivation predictions represent the joint probability that a metabolite forms and that this metabolite subsequently conjugates to protein or glutathione. Among molecules bioactivated by these pathways, we predicted the correct pathway with an AUC accuracy of 89.98%. Furthermore, the model predicts whether molecules will be bioactivated, distinguishing bioactivated and nonbioactivated molecules with 81.06% AUC. We applied this algorithm to withdrawn drugs. The known bioactivation pathways of alclofenac and benzbromarone were identified by the algorithm, and high probability bioactivation pathways not yet confirmed were identified for safrazine, zimelidine, and astemizole. This bioactivation model-the first of its kind that jointly considers both metabolism and reactivity-enables drug candidates to be quickly evaluated for a toxicity risk that often evades detection during preclinical trials. The XenoSite bioactivation model is available at http://swami.wustl.edu/xenosite/p/bioactivation.

Human Biotransformation Pathway of Temephos Using an In Silico Approach

Temephos is an organophosphorothioate (OPT) larvicide used for controlling vectors of diseases such as dengue, chikungunya, and Zika. OPTs require a metabolic activation mediated by cytochrome P540 (CYP) to cause toxic effects, such as acetylcholinesterase (AChE) activity inhibition. There is no information about temephos biotransformation in humans, and it is considered to have low toxicity in mammals. Recent studies have reported that temephos-oxidized derivatives cause AChE inhibition. The aim of this study was to propose the human biotransformation pathway of temephos using in silico tools. The metabolic pathway was proposed using the MetaUltra program of MultiCase software as well as the Way2Drug and Xenosite web servers. The results show the following three essential reactions of phase I metabolism: (1) S-oxidation, (2) oxidative desulfurization, and (3) dephosphorylation, as well as the formation of 19 possible intermediary metabolites. Temephos dephosphorylation is the most likely reaction, and it enables phase II metabolism for glucuronidation to be excreted. However, the CYP-dependent metabolism showed that temephos oxon can be formed, which could lead to toxic effects in mammals. CYP2B6, 2C9, and 2C19 are the main isoforms involved in temephos metabolism, and CYP3A4 and 2D6 have minor contributions. According to computational predictions, the highest probability of temephos metabolism is dephosphorylation and phase II reactions that do not produce cholinergic toxic effects; nonetheless, the participation of CYPs is highly possible if the primary reaction is depleted.

Metabolic Forest: Predicting the Diverse Structures of Drug Metabolites

Adverse drug metabolism often severely impacts patient morbidity and mortality. Unfortunately, drug metabolism experimental assays are costly, inefficient, and slow. Instead, computational modeling could rapidly flag potentially toxic molecules across thousands of candidates in the early stages of drug development. Most metabolism models focus on predicting sites of metabolism (SOMs): the specific substrate atoms targeted by metabolic enzymes. However, SOMs are merely a proxy for metabolic structures: knowledge of an SOM does not explicitly provide the actual metabolite structure. Without an explicit metabolite structure, computational systems cannot evaluate the new molecule's properties. For example, the metabolite's reactivity cannot be automatically predicted, a crucial limitation because reactive drug metabolites are a key driver of adverse drug reactions (ADRs). Additionally, further metabolic events cannot be forecast, even though the metabolic path of the majority of substrates includes two or more sequential steps. To overcome the myopia of the SOM paradigm, this study constructs a well-defined system-termed the metabolic forest-for generating exact metabolite structures. We validate the metabolic forest with the substrate and product structures from a large, chemically diverse, literature-derived dataset of 20 736 records. The metabolic forest finds a pathway linking each substrate and product for 79.42% of these records. By performing a breadth-first search of depth two or three, we improve performance to 88.43 and 88.77%, respectively. The metabolic forest includes a specialized algorithm for producing accurate quinone structures, the most common type of reactive metabolite. To our knowledge, this quinone structure algorithm is the first of its kind, as the diverse mechanisms of quinone formation are difficult to systematically reproduce. We validate the metabolic forest on a previously published dataset of 576 quinone reactions, predicting their structures with a depth three performance of 91.84%. The metabolic forest accurately enumerates metabolite structures, enabling promising new directions such as joint metabolism and reactivity modeling.

Site-Level Bioactivity of Small-Molecules from Deep-Learned Representations of Quantum Chemistry

Atom- or bond-level chemical properties of interest in medicinal chemistry, such as drug metabolism and electrophilic reactivity, are important to understand and predict across arbitrary new molecules. Deep learning can be used to map molecular structures to their chemical properties, but the data sets for these tasks are relatively small, which can limit accuracy and generalizability. To overcome this limitation, it would be preferable to model these properties on the basis of the underlying quantum chemical characteristics of small molecules. However, it is difficult to learn higher level chemical properties from lower level quantum calculations. To overcome this challenge, we pretrained deep learning models to compute quantum chemical properties and then reused the intermediate representations constructed by the pretrained network. Transfer learning, in this way, substantially outperformed models based on chemical graphs alone or quantum chemical properties alone. This result was robust, observable in five prediction tasks: identifying sites of epoxidation by metabolic enzymes and identifying sites of covalent reactivity with cyanide, glutathione, DNA and protein. We see that this approach may substantially improve the accuracy of deep learning models for specific chemical structures, such as aromatic systems.

XenoNet: Inference and Likelihood of Intermediate Metabolite Formation

Drug metabolism is a common cause of adverse drug reactions. Drug molecules can be metabolized into reactive metabolites, which can conjugate to biomolecules, like protein and DNA, in a process termed bioactivation. To mitigate adverse reactions caused by bioactivation, both experimental and computational screening assays are utilized. Experimental assays for assessing the formation of reactive metabolites are low throughput and expensive to perform, so they are often reserved until later stages of the drug development pipeline when the drug candidate pools are already significantly narrowed. In contrast, computational methods are high throughput and cheap to perform to screen thousands to millions of compounds for potentially toxic molecules during the early stages of the drug development pipeline. Commonly used computational methods focus on detecting and structurally characterizing reactive metabolite-biomolecule adducts or predicting sites on a drug molecule that are liable to form reactive metabolites. However, such methods are often only concerned with the structure of the initial drug molecule or of the adduct formed when a biomolecule conjugates to a reactive metabolite. Thus, these methods are likely to miss intermediate metabolites that may lead to subsequent reactive metabolite formation. To address these shortcomings, we create XenoNet, a metabolic network predictor, that can take a pair of a substrate and a target product as input and (1) enumerate pathways, or sequences of intermediate metabolite structures, between the pair, and (2) compute the likelihood of those pathways and intermediate metabolites. We validate XenoNet on a large, chemically diverse data set of 17 054 metabolic networks built from a literature-derived reaction database. Each metabolic network has a defined substrate molecule that has been experimentally observed to undergo metabolism into a defined product metabolite. XenoNet can predict experimentally observed pathways and intermediate metabolites linking the input substrate and product pair with a recall of 88 and 46%, respectively. Using likelihood scoring, XenoNet also achieves a top-one pathway and intermediate metabolite accuracy of 93.6 and 51.9%, respectively. We further validate XenoNet against prior methods for metabolite prediction. XenoNet significantly outperforms all prior methods across multiple metrics. XenoNet is available at https://swami.wustl.edu/xenonet.