Novel in silico methods for predicting metabolite formation by Cytochrome P450 enzymes

Samenvatting

Drug metabolism plays a major role in possible toxicity and failure of drugs or drug candidates. Therefore it is of uttermost importance to be able to predict into which metabolites drugs and drug-like compounds are converted. Cytochrome P450 enzymes (CYPs, P450s) metabolize 75% of the currently marketed drugs. However, predicting substrate binding modes and selectivity for CYPs (needed to predict P450 metabolite formation) is until now difficult, because of the large protein flexibility or plasticity of CYPs. The current proposal aims on developing novel in silico methods for accurate P450 metabolite prediction, by efficiently including plasticity effects. For this purpose, plasticity models will be developed for human CYPs involved in drug metabolism, and for bacterial P450 BM3 mutants that are used as biocatalysts to selectively produce, e.g., drug metabolites for drug discovery and development purposes.

First, plasticity models for metabolite prediction will be developed by carefully selecting conformationally functional P450 structures from molecular dynamics simulations, for use in docking-based virtual screening. Second, for specific combinations of CYP isoforms and substrates, atomic reactivities will be evaluated for sites-of-metabolism as identified using the plasticity models, to obtain optimal accuracy in metabolite prediction. Third, methods for CYP-binding selectivity prediction will be developed using novel, time-efficient free-energy methods. The selectivity models are needed for metabolite prediction, because different P450s may form different metabolites. Finally, an automated workflow for using the plasticity and selectivity models is foreseen, thereby addressing the issue of automatically assigning force-field parameters.

With the novel plasticity and selectivity models, accurate and efficient prediction of P450 metabolites and of stability of drugs and drug-like compounds becomes available, which is very relevant to pharmaceutical scientists and molecular toxicologists. Recent market research demonstrated utilization potential of the plasticity and selectivity models, which will be highly increased by their anticipated accuracy, efficiency and automated implementation.

Producten

Wetenschappelijk artikel

Kenmerken

Projectnummer

723.012.105

Hoofdaanvrager

Dr. D.P. Geerke

Verbonden aan

Vrije Universiteit Amsterdam, Faculteit der BĂ©tawetenschappen, Afdeling Scheikunde & Farmaceutische Wetenschappen

Uitvoerders

Dr. M. van Dijk, Dr. M. van Dijk, Dr. D.P. Geerke, R.A. Luirink, K.M. Visscher, C.R. Vosmeer MSc

Looptijd

01/11/2013 tot 01/02/2019