Fast Methods for Prediction of Aldehyde Oxidase-Mediated Site-of-Metabolism

Abstract

Aldehyde Oxidase (AO) is an enzyme involved in the metabolism of aldehydes and N-containing heterocyclic compounds. Many drug compounds contain heterocyclic moieties, and AO metabolism has lead to failure of several late-stage drug candidates. Therefore, it is important to take AO-mediated metabolism into account early in the drug discovery process, and thus, to have fast and reliable models to predict the site of metabolism (SOM). We have collected a dataset of 78 substrates of human AO with a total of 89 SOMs and 347 non-SOMs and determined atomic descriptors for each compound. The descriptors comprise NMR shielding and ESP charges from density functional theory (DFT), NMR chemical shift from ChemBioDraw, and Gasteiger charges from RDKit. Additionally, atomic accessibility was considered using 2D-SASA and relative span descriptors from SMARTCyp. Finally, stability of the product, the metabolite, was determined with DFT and also used as a descriptor. All descriptors have AUC larger than 0.75. In particular, descriptors related to the chemical shielding and chemical shift (AUC = 0.96) and ESP charges (AUC = 0.96) proved to be good descriptors. We recommend two simple methods to identify the SOM for a given molecule: 1) use ChemBioDraw to calculate the chemical shift or 2) calculate ESP charges or chemical shift using DFT. The first approach is fast but somewhat difficult to automate, while the second is more time-consuming, but can easily be automated. The two methods predict correctly 93% and 91%, respectively, of the 89 experimentally observed SOMs.

Revisiting Aldehyde Oxidase Mediated Metabolism in Drug-like Molecules: An Improved Computational Model

“…For the electronic properties like E LUMO , electronegativity, and hardness, there were no significant difference between substrates and nonsubstrates (Figure B,D). As for the ESP properties, several studies demonstrated that the partial charges such as MK charge or ESP partial charge of the metabolic sites, namely, the electron-deficient carbon adjacent to the nitrogen, were more positive than non-SOMs. ,, Because the ESP partial charge of the atom was the integral of electrostatic potential surface on the atom, and for the nitrogen-containing heterocycle rings, the more positive charge of the SOMs might implicate the higher value of the ESP energy. Thus, we speculated that the substrate might have higher maximum values of the ESP energies (Max_ElPot) on the electrostatic potential maps or the nonsubstrates might have lower minimum values of the ESP energies (MIN_ElPot) on the electrostatic potential maps.…”

“…Decision tree model DTN AOX (A) for predicting hAOX-mediated SOMs and its performance on (B) the test data set and (C) the external data set, compared with our previous DT AOX model and three descriptors proposed by Montefiori et al …”

“…For this external data set, the DTN AOX model demonstrated decent performance (Figure C) and significantly exceeded DT AOX , especially in terms of SE score. Prediction results of the methods proposed by Montefiori et al were also benchmarked here for reference.…”

“…Cruciani et al developed a probability function for hAOX-metabolized oxidation with 88% prediction accuracy based on a large set of homogeneous experimental data, in which both exposure and electrophilic effects were taken into account . Furthermore, Montefiori et al collected a data set of 78 substrates with known SOMs information and constructed models to predict the AOX-mediated SOMs, in which the importance of the chemical shift and ESP charge of the deficient C of SOMs were highlighted …”

“…17 Furthermore, Montefiori et al collected a data set of 78 substrates with known SOMs information and constructed models to predict the AOXmediated SOMs, in which the importance of the chemical shift and ESP charge of the deficient C of SOMs were highlighted. 18 In our previous work, the reaction energy barrier (ΔE inter ) and steric hindrance (Steric) were adopted to predict the potential SOMs of the molecules toward hAOX-mediated metabolism. As the results illuminated, the DT AOX model exhibited good predictive performance and gave a clear explanation of the relationship between the exposure, reaction energy barrier effects, and the hAOX-mediated metabolism.…”