Community Research and Development Information Service - CORDIS

FP6

INSILICOTOX Report Summary

Project ID: 42328
Funded under: FP6-MOBILITY
Country: United Kingdom

Final Activity Report Summary - INSILICOTOX (In Silico Tools to Reduce the Use of Animals in Toxicity Testing for Bioreactive Chemicals)

An increasing part of developing alternatives for the toxicological assessment of chemicals, including in chemico (experimental measurement of reactivity) and in silico (computational) approaches, involves the understanding of biochemical reactivity in toxicology. These techniques have come to the fore recently due to legislation such as European Union's REACH and Cosmetics regulations. The aim of the INSILICOTOX project was to develop in silico models for reactive toxicity, also based on the compilation of an in chemico database. Reactive acting chemicals are defined as chemicals which interact covalently with biomolecules, possibly after bio-activation. The electrophilic reaction can elicit a number of toxic effects, called reactive toxicity, dependent on species, route of exposure, and dose. Interactions involving covalent binding differ from those involving reversible receptor-binding and those resulting in depletion of certain defence molecules e.g. GSH (glutathione). Through their reactivity, reactive acting chemicals may lead, e.g., to genotoxic effects (i.e. DNA reactive). DNA binding triggers other cellular events as protein-binding. Reactive acting chemicals may also lead to other effects, e.g., skin sensitisation and irritation.

A review of the mechanisms of activity associated with reactive toxicity has been performed. This has identified organic chemistry mechanisms as being the most important and therefore more appropriate and essential for analysis (i.e., Michael-type addition, SNAr, SN2, Schiff base formation, and alkylation). These mechanisms are especially important for mutagenicity, skin sensitisation and acute excess toxicity. To support mechanistically based category formation (or the grouping of chemicals) to predict toxicity, a chemical reactivity database has been compiled, which allows for the combination of structural, reactivity, and toxicological data. Reactivity data taken from the compiled database were also used to develop correlations between different endpoints and reactivity data. The outcome will help in the decision making, whether a compound acts via a specific endpoint or not. This procedure can be very powerful to fill data gaps in the field of integrated testing strategies (ITS).

Quantitative structure-activity relationships (QSARs) provide a useful tool for defining a mathematical relationship between chemical structure and toxicity, and for applying such statistically derived models for predicting the toxicity of untested chemicals. The prediction of the toxicity of reactive chemicals requires the use of chemical descriptors that encapsulate how covalently reactive a given chemical is. Quantum chemical descriptors are related to the electronic structure of reactive chemicals and to the chemical mechanisms that are involved in covalent bond formation between biological nucleophiles and electrophilic chemicals. A number of computational methods are available for the calculation of theoretical descriptors. For example, we have introduced a model to predict quantitatively reactivity and toxicity of alpha, beta-unsaturated compounds, including various compound classes involved in Michael addition. The models have good statistical performance and are mechanistically interpretable, based on easily computed ground state properties.

The validity of this reactivity model can be confirmed by transition state calculations, where the mechanism is examined in detail. Factors affecting rate constants have been elucidated, especially solvent effects and the influence of steric hindrance. This knowledge might help in the reactivity or toxicity profiling of electrophilic compounds. It can support mechanistic grouping to order compounds according to reactivity and also confirm, that a compound is reactive.

Reported by

LIVERPOOL JOHN MOORES UNIVERSITY
LIVERPOOL
United Kingdom
See on map
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top