Metabolomics and In Silico Docking-Directed Discovery of Small-Molecule Enzyme Targets

The identification of target proteins for small molecules is of great importance in drug discovery and for understanding the cellular mode of action (MOA) of toxicants. Herein, a bottom-up oriented target finding strategy is proposed based on the principle that the targeted enzymes can be inferred according to their phenotypic changes at the metabolome level. Meanwhile, computer-assisted in silico molecular docking analysis was performed to evaluate the binding affinities between the chemicals and the target enzymes to further rank the possible targets. In this study, triphenyl phosphate (TPhP) was used as an example to illustrate the workflow. After a comprehensive metabolome and lipidome analysis, 51 related metabolic enzymes were selected for ranking binding energies, wherein 25 proteins exhibited a higher affinity for TPhP than for their endogenous substrates. Two proteins, hydroxyacyl-coenzyme A dehydrogenase (HADH) and 3-hydroxyacyl-CoA dehydrogenase type-2 (HSD17B10), were further confirmed by surface phasma resonance (SPR) and isothermal titration calorimetry (ITC) analysis, displayed K-d values at low micromolar levels for TPhP. Overall, the proposed strategy has provided a feasible means for discovering enzymatic targets for the large-scale small-molecule sets, with the advantages of closely associating with the phenotype change, reducing the cost of groping, and improving the accuracy of target prediction.

Automated summary

The study introduces a bottom-up target finding strategy based on metabolome level, where targeted enzymes are inferred from phenotypic changes. Computer-assisted molecular docking analysis is used to evaluate binding affinities and rank possible targets, with the example of triphenyl phosphate (TPhP). Two proteins, HADH and HSD17B10, were confirmed to have high affinity for TPhP through SPR and ITC analysis.