From Structural Alerts to Signature Fragment Alerts: A Case Study on Pyrrolizidine Alkaloids.

Even though modeling is considered a valid alternative to mutagenicity testing for substances with known structures, it can be applied for mixtures only if all of the single chemical structures are identified. Within the present work, we investigate a new avenue to exploit computational toxicology for mixtures, such as plant-based food ingredients. Indeed, considering that in the absence of toxicological information, an important early consideration is whether any substance may be genotoxic through the mutagenic mechanism of action, we tried to establish a correspondence between genotoxic structural alerts (SAs) and so-called signature fragment alerts (SFAs). Once this correspondence is established, chromatograms could be screened for chemical features associated with genotoxic alerts. Pyrrolizidine alkaloids (PAs), a large group of natural toxins (several of them known as genotoxic) were used as a case study because their early identification would bring significant benefits. The method was built using 56 PA pure standards, resulting in the characterization of signature fragment alerts. Finally, the approach was verified in real plant-based samples such as herbal tea and alfalfa, where the screening of signature fragment alerts allowed highlighting quickly the presence of genotoxic PAs in plant-based mixtures. Therefore, the SFA analysis can be used for risk prioritization of newly identified PAs and for their identification in mixtures, contributing to the unnecessary use of animal experimentation for genotoxicity testing.

Mutagenic potential and structural alerts of phytotoxins.

Toxic plant-produced chemicals, so-called phytotoxins, constitute a category of natural compounds belonging to a diversity of chemical classes. Some of them (e.g., alkaloids, terpenes, saponins) are associated with high toxic potency, while for many of others no toxicological data is available. In this study, the mutagenic potential of 1586 phytotoxins, as obtained from a publicly available database, was investigated applying different in silico approaches. (Q)SAR models (including statistical-based and rule-based systems) were used for the prediction of bacterial in vitro mutagenicity (Ames test) and the results from multiple tools were combined to assign consensus predicted values (i.e., positive, negative, inconclusive). The overall consensus outcome was then employed to investigate relationships between structural features of classes of phytotoxins and potential mutagenicity, allowing the identification of structural alerts raising a specific concern. The results highlighted that about 10% of the screened compounds were predicted to have mutagenic potential and the critical classes of concern, such as alkaloids, were further investigated in terms of subclasses (e.g., indole alkaloids, isoquinoline alkaloids), getting a deeper insight into the mutagenic potential of possible naturally occurring chemicals in plant materials and their structural alerts.

Evaluating Confidence in Toxicity Assessments Based on Experimental Data and In Silico Predictions.

Understanding the reliability and relevance of a toxicological assessment is important for gauging the overall confidence and communicating the degree of uncertainty related to it. The process involved in assessing reliability and relevance is well defined for experimental data. Similar criteria need to be established for in silico predictions, as they become increasingly more important to fill data gaps and need to be reasonably integrated as additional lines of evidence. Thus, in silico assessments could be communicated with greater confidence and in a more harmonized manner. The current work expands on previous definitions of reliability, relevance, and confidence and establishes a conceptional framework to apply those to in silico data. The approach is used in two case studies: 1) phthalic anhydride, where experimental data are readily available and 2) 4-hydroxy-3-propoxybenzaldehyde, a data poor case which relies predominantly on in silico methods, showing that reliability, relevance, and confidence of in silico assessments can be effectively communicated within Integrated approaches to testing and assessment (IATA).

In Silico Approaches In Carcinogenicity Hazard Assessment: Current Status and Future Needs.

Historically, identifying carcinogens has relied primarily on tumor studies in rodents, which require enormous resources in both money and time. In silico models have been developed for predicting rodent carcinogens but have not yet found general regulatory acceptance, in part due to the lack of a generally accepted protocol for performing such an assessment as well as limitations in predictive performance and scope. There remains a need for additional, improved in silico carcinogenicity models, especially ones that are more human-relevant, for use in research and regulatory decision-making. As part of an international effort to develop in silico toxicological protocols, a consortium of toxicologists, computational scientists, and regulatory scientists across several industries and governmental agencies evaluated the extent to which in silico models exist for each of the recently defined 10 key characteristics (KCs) of carcinogens. This position paper summarizes the current status of in silico tools for the assessment of each KC and identifies the data gaps that need to be addressed before a comprehensive in silico carcinogenicity protocol can be developed for regulatory use.

Value and limitation of structure-based profilers to characterize developmental and reproductive toxicity potential.

The uncertainty regarding the safety of chemicals leaching from food packaging triggers attention. In silico models provide solutions for screening of these chemicals, since many are toxicologically uncharacterized. For hazard assessment, information on developmental and reproductive toxicity (DART) is needed. The possibility to apply in silico toxicology to identify and quantify DART alerts was investigated. Open-source models and profilers were applied to 195 packaging chemicals and analogues. An approach based on DART and estrogen receptor (ER) binding profilers and molecular docking was able to identify all except for one chemical with documented DART properties. Twenty percent of the chemicals in the database known to be negative in experimental studies were classified as positive. The scheme was then applied to 121 untested chemicals. Alerts were identified for sixteen of them, five being packaging substances, the others structural analogues. Read-across was then developed to translate alerts into quantitative toxicological values. They can be used to calculate margins of exposure (MoE), the size of which reflects safety concern. The application of this approach appears valuable for hazard characterization of toxicologically untested packaging migrants. It is an alternative to the use of default uncertainty factor (UF) applied to animal chronic toxicity value to handle absence of DART data in hazard characterization.

Bisphenol A binding promiscuity: A virtual journey through the universe of proteins.

Significant efforts are currently being made to move toxicity testing from animal experimentation to human relevant, mechanism-based approaches. In this context, the identification of molecular target(s) responsible for mechanisms of action is an essential step. Inspired by the recent concept of polypharmacology (the ability of drugs to interact with multiple targets) we argue that whole proteome virtual screening might become a breakthrough tool in toxicology reflecting the real complexity of chemical-biological interactions. Therefore, we investigated the value of performing ligand-protein binding prediction screening across the full proteome to identify new mechanisms of action for food chemicals. We applied the new approach to make a broader comparison between bisphenol A (BPA) (food-packaging chemical) and the endogenous estrogen, 17ß-estradiol (EST). Applying a novel high-throughput ligand-protein binding prediction tool (BioGPS) by the Amazon Web Services (AWS) cloud (to speed-up the calculation), we investigated the value of performing in silico screening across the full proteome (all human and rodent x-ray protein structures available in the Protein Data Bank). The strong correlation between in silico predictions and available in vitro data demonstrates the high predictive power of the method used. The most striking results obtained was that BPA was predicted to bind to many more proteins than the ones already known, most of which were common to EST. Our findings provide a new and unprecedented insight on the complexity of chemical-protein interactions, highlighting the binding promiscuity of BPA and its broader similarity compared to the female sex hormone, EST.

Using bisphenol A and its analogs to address the feasibility and usefulness of the CALUX-PPARγ assay to identify chemicals with obesogenic potential.

Environmental chemical exposures have been implicated in the obesity epidemic as potential mis-regulators of a variety of metabolic pathways. As agonism of the peroxisome proliferator-activated nuclear hormone receptor γ (PPARγ) is one of the suspected mechanisms involved, a PPARγ screening assay may have relevance for the biodetection of such effects of environmental chemicals. To test this hypothesis, we established the PPARγ2-CALUX® assay in-house and tested it against a number of known and suspected PPARγ modulators. Furthermore, we added a rat liver S9 metabolizing system to the protocol to introduce metabolic competence to the assay. Our results confirmed the responsiveness of the cell line to the known PPARγ agonists and antagonists: rosiglitazone, tributyltin, 15-deoxy-Δ12,14-prostaglandin J2, GW9662 and diclofenac. These data are in agreement with previous studies in various models. Seven bisphenol analogs tested induced little to no agonist activity, but all demonstrated antagonistic properties. These findings were contrary to both our assumptions and literature reports. Addition of the S9-metabolizing system to each of these tests did not alter any of the measured activities. Taken together, it seems probable that there are additional obesogenic effects of these chemicals which would not be detected by this assay.

A large comparison of integrated SAR/QSAR models of the Ames test for mutagenicity$.

Results from the Ames test are the first outcome considered to assess the possible mutagenicity of substances. Many QSAR models and structural alerts are available to predict this endpoint. From a regulatory point of view, the recommendation from international authorities is to consider the predictions of more than one model and to combine results in order to develop conclusions about the mutagenicity risk posed by chemicals. However, the results of those models are often conflicting, and the existing inconsistency in the predictions requires intelligent strategies to integrate them. In our study, we evaluated different strategies for combining results of models for Ames mutagenicity, starting from a set of 10 diverse individual models, each built on a dataset of around 6000 compounds. The novelty of our study is that we collected a much larger set of about 18,000 compounds and used the new data to build a family of integrated models. These integrations used probabilistic approaches, decision theory, machine learning, and voting strategies in the integration scheme. Results are discussed considering balanced or conservative perspectives, regarding the possible uses for different purposes, including screening of large collection of substances for prioritization.

Modeling Chronic Toxicity: A Comparison of Experimental Variability With (Q)SAR/Read-Across Predictions.

This study compares the accuracy of (Q)SAR/read-across predictions with the experimental variability of chronic lowest-observed-adverse-effect levels (LOAELs) from in vivo experiments. We could demonstrate that predictions of the lazy structure-activity relationships (lazar) algorithm within the applicability domain of the training data have the same variability as the experimental training data. Predictions with a lower similarity threshold (i.e., a larger distance from the applicability domain) are also significantly better than random guessing, but the errors to be expected are higher and a manual inspection of prediction results is highly recommended.

Integrated strategy for mutagenicity prediction applied to food contact chemicals.

Food contamination due to unintentional leakage of chemicals from food contact materials (FCM) is a source of increasing concern. Since for many of these substances, only limited or no toxicological data are available, the development of alternative methodologies to establish rapidly and cost-efficiently level of safety concern is critical to ensure adequate consumer protection. Computational toxicology methods are considered the most promising solutions to cope with this data gap. In particular, mutagenicity assessment has a particular relevance and is a mandatory requirement for all substances released from plastic FCM, regardless how low migration and exposure are. In the present work, a strategy integrating a number of (Quantitative) Structure Activity Relationship ((Q)SAR) models for Ames mutagenicity predictions is proposed. A list of chemicals representing likely migrating moieties from FCM was selected to test the value of the newly defined strategy and the possibility to combine predictions given by the different algorithms was evaluated. In particular, a scheme to integrate mutagenicity estimations into a single final assessment was developed resulting in an increased domain of applicability. In most cases, a deeper analysis of experimental data, where available, allowed fixing misclassification errors, highlighting the importance of data curation in the development, validation and application of in silico methods. The high accuracy of the strategy provided the rationales for its application for toxicologically uncharacterized chemicals. Finally, the overall strategy of integration will be automated through its implementation into a freely available software application.

Hazard assessment through hybrid in vitro / in silico approach: The case of zearalenone.

Within the framework of reduction, refinement and replacement of animal experiments, new approaches for identification and characterization of chemical hazards have been developed. Grouping and read across has been promoted as a most promising alternative approach. It uses existing toxicological information on a group of chemicals to make predictions on the toxicity of uncharacterized ones. In the present work, the feasibility of applying in vitro and in silico techniques to group chemicals for read across was studied using the food mycotoxin zearalenone (ZEN) and metabolites as a case study. ZEN and its reduced metabolites are known to act through activation of the estrogen receptor α (ERα). The ranking of their estrogenic potencies appeared highly conserved across test systems including binding, in vitro and in vivo assays. This data suggests that activation of ERα may play a role in the molecular initiating event (MIE) and be predictive of adverse effects and provides the rationale to model receptor-binding for hazard identification. The investigation of receptor-ligand interactions through docking simulation proved to accurately rank estrogenic potencies of ZEN and reduced metabolites, showing the suitability of the model to address estrogenic potency for this group of compounds. Therefore, the model was further applied to biologically uncharacterized, commercially unavailable, oxidized ZEN metabolites (6α-, 6ß-, 8α-, 8ß-, 13- and 15-OH-ZEN). Except for 15-OH-ZEN, the data indicate that in general, the oxidized metabolites would be considered a lower estrogenic concern than ZEN and reduced metabolites.

Automated and reproducible read-across like models for predicting carcinogenic potency.

Several qualitative (hazard-based) models for chronic toxicity prediction are available through commercial and freely available software, but in the context of risk assessment a quantitative value is mandatory in order to be able to apply a Margin of Exposure (predicted toxicity/exposure estimate) approach to interpret the data. Recently quantitative models for the prediction of the carcinogenic potency have been developed, opening some hopes in this area, but this promising approach is currently limited by the fact that the proposed programs are neither publically nor commercially available. In this article we describe how two models (one for mouse and one for rat) for the carcinogenic potency (TD50) prediction have been developed, using lazar (Lazy Structure Activity Relationships), a procedure similar to read-across, but automated and reproducible. The models obtained have been compared with the recently published ones, resulting in a similar performance. Our aim is also to make the models freely available in the near future thought a user friendly internet web site.

Establishing the level of safety concern for chemicals in food without the need for toxicity testing.

There is demand for methodologies to establish levels of safety concern associated with dietary exposures to chemicals for which no toxicological data are available. In such situations, the application of in silico methods appears promising. To make safety statement requires quantitative predictions of toxicological reference points such as no observed adverse effect level and carcinogenic potency for DNA-reacting chemicals. A decision tree (DT) has been developed to aid integrating exposure information and predicted toxicological reference points obtained with quantitative structure activity relationship ((Q)SAR) software and read across techniques. The predicted toxicological values are compared with exposure to obtain margins of exposure (MoE). The size of the MoE defines the level of safety concern and should account for a number of uncertainties such as the classical interspecies and inter-individual variability as well as others determined on a case by case basis. An analysis of the uncertainties of in silico approaches together with results from case studies suggest that establishing safety concern based on application of the DT is unlikely to be significantly more uncertain than based on experimental data. The DT makes a full use of all data available, ensuring an adequate degree of conservatism. It can be used when fast decision making is required.

Minimum information about a bioactive entity (MIABE).

Bioactive molecules such as drugs, pesticides and food additives are produced in large numbers by many commercial and academic groups around the world. Enormous quantities of data are generated on the biological properties and quality of these molecules. Access to such data - both on licensed and commercially available compounds, and also on those that fail during development - is crucial for understanding how improved molecules could be developed. For example, computational analysis of aggregated data on molecules that are investigated in drug discovery programmes has led to a greater understanding of the properties of successful drugs. However, the information required to perform these analyses is rarely published, and when it is made available it is often missing crucial data or is in a format that is inappropriate for efficient data-mining. Here, we propose a solution: the definition of reporting guidelines for bioactive entities - the Minimum Information About a Bioactive Entity (MIABE) - which has been developed by representatives of pharmaceutical companies, data resource providers and academic groups.

Use of computational tools in the field of food safety.

In this article we give an overview of how computational methods are currently used in the field of food safety by national regulatory bodies, international advisory organisations and the food industry. Our results show that currently the majority of stakeholders in the field of food safety do not apply computational methods on a routine basis, mainly because of a lack of in-house expertise. Some organisations, however, are very experienced in their use and have developed specialised in-house approaches. Despite this variable situation, computational tools are widely perceived to be a useful tool to support regulatory assessments and decision making in the field of food safety. Recognized, however, is a widespread need to develop guidance documents and software tools that will promote and harmonise the use of computational methods, together with appropriate training.

Flavonoids for controlling starch digestion: structural requirements for inhibiting human alpha-amylase.

In this study we investigated the structural requirements for inhibition of human salivary alpha-amylase by flavonoids. Four flavonols and three flavones, out of the 19 flavonoids tested, exhibited IC50 values less than 100 microM against human salivary alpha-amylase activity. Structure-activity relationships of these inhibitors by computational ligand docking showed that the inhibitory activity of flavonols and flavones depends on (i) hydrogen bonds between the hydroxyl groups of the polyphenol ligands and the catalytic residues of the binding site and (ii) formation of a conjugated pi-system that stabilizes the interaction with the active site. Our findings show that certain naturally occurring flavonoids act as inhibitors of human alpha-amylase, which makes them promising candidates for controlling the digestion of starch and postprandial glycemia.

Virtual screening for aryl hydrocarbon receptor binding prediction.

The overall goal of this study has been to validate computational models for predicting aryl hydrocarbon receptor (AhR) binding. Due to the unavailability of the AhR X-ray crystal structure we have decided to use QSARs models for the binding prediction virtual screening. We have built up CoMFA, Volsurf, and HQSAR models using as a training set 84 AhR ligands. Additionally, we have built a hybrid model combining two of the final selected models in order to give a single operational system. The results show that CoMFA, VolSurf, HQSAR, and the hybrid models gives good results (R(2) equal to 0.91, 0.79, 0.85, and 0.82 and q(2) 0.62, 0.58, 0.62, and 0.70, respectively). Since the techniques analyzed show a good correlation and good prediction also for an external test set, particularly the HQSAR and the hybrid model, we can conclude that these models can be used for predicting AhR binding in virtual screening.

Preliminary analysis of toxicity of benzoxazinones and their metabolites for folsomia Candida.

The overall objective of this study was to explore the toxicity of benzoxazinone allelochemicals and their metabolites to Folsomia candida (Collembola: Isotomidae) (Willem, 1902). Experimental tests showed transformation products to have more pronounced toxicity than parent compounds. The underlying relationship between the chemical structure and toxicity was then studied using three-dimensional QSAR approaches, and results highlighted the role of the steric contribution.

QSAR models for Daphnia magna toxicity prediction of benzoxazinone allelochemicals and their transformation products.

The overall objective of this study is the ecotoxicological characterization of the benzoxazinone 2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one (DIMBOA), the benzoxazolinones benzoxazolin-2-one (BOA) and 6-methoxybenzoxazolin-2-one (MBOA), and their transformation products: phenoxazinones 2-acetylamino-7-methoxy-3H-phenoxazin-3-one (AAMPO), 2-acetylamino-3H-phenoxazin-3-one (AAPO), 2-amino-7-methoxy-3H-phenoxazin-3-one (AMPO), and 2-amino-3H-phenoxazin-3-one (APO); aminophenol 2-aminophenol AP); acetamide N-(2-hydroxyphenyl)acetamide (HPAA); and malonamic acid amide N-(2-hydroxyphenyl)malonamic acid (HPMA). A comparison between empirical results and theoretical ones using rules-based prediction of toxicity was done, and it can be concluded that only the degradation metabolites exhibited significant ecotoxic effect. Using synthetic pesticides knowledge, several QSAR models were trained with various approaches and descriptors. The models generated exhibited good internal predictive ability (R(cv)2 > 0.6) and were used to predict the toxicity of the natural compounds studied.

QSAR model for predicting pesticide aquatic toxicity.

A hierarchical QSAR approach was applied for the prediction of acute aquatic toxicity. Chemical structures were encoded into molecular descriptors by an automated, seamless procedure available within the OpenMolGRID system. Finally, various linear and nonlinear regression techniques were used to obtain stable and thoroughly validated QSARs. The final model was developed by a counterpropagation neural network coupled with genetic algorithms for variable selection. The proposed QSAR is consistent with McFarland's principle for biological activity and makes use of seven molecular descriptors, namely HACA-2, HOMO-LUMO energy gap, Kier and Hall index, HA dependent HDSA-1, BETA polarizability, FHBCA fractional HBSA, and LogP. The model was extensively tested by the test set (R2= 0.79), the y-scrambling test, and sensitivity/stability tests.