The prediction of hydrogen-bond (H-bond) acceptor ability is crucial in drug design. This important property is quantified in a large pKBHX database of consistently measured values. We aim to expand the chemical diversity of the studied H-bond acceptors and to increase the range of H-bond strength considered. Two quantum chemical descriptors are contrasted, called ΔE(H) (the change in the energy of a topological hydrogen atom upon complexation) and Vmin (the local minimum in the electrostatic potential on the H-bond accepting site). We performed a systematic analysis of the correlations between pKBHX and Vmin for an initial set of 106 compounds (including O- and N-containing subsets, as well as compounds including sulfur, chlorine, and π-bases). Correlations improve for family dependent subsets, and after outlier treatment, a set of 90 compounds was used to set up a linear equation to predict pKBHX from Vmin. This equation and a previously published equation [Green and Popelier J. Chem. Inf.2014, 54 (2), 553-561], to predict pKBHX from ΔE(H), were used to predict the pKBHX values for 22 potentially biologically active heteroaromatic ring compounds, [Pitt et al. J. Med. Chem. 2009, 52 (9), 2952-2963], among which several structures still remain experimentally unavailable. H-Bond basicity of sp(2) nitrogen sites were consistently predicted with both descriptors. A worse agreement was found with carbonyl acceptor sites, with the stronger deviations observed for the lactam groups. It was shown that secondary interactions involving the neighboring NH group were influencing the results. Substitution of the NH group with an NMe group resulted in an improved consistency from both Vmin and ΔE(H) predictions, the latter being more prominently affected by the methyl substitution. Both approaches appear as efficient procedures for the H-bond ability prediction of novel heteroaromatic rings. Nevertheless, the ΔE(H) parameter presents slight chemical structure limitations and requires more detailed H-bond structure investigations.

作者：Jér?me, Graton；Jean-Yves, Le Questel；Peter, Maxwell；Paul, Popelier

来源：Journal of chemical information and modeling 2016 年 56卷 2期