How Do Food Compounds Interact with Their Protein Targets? Alternative Modes for Protein Binding

A large body of research is oriented towards the determination of the mechanism of action of food compounds at the molecular level, in order to rationalize the important role of these molecules in health, and as source of new drugs. In this work we perform a systematic analysis of all the food-protein complexes at atomic resolution present in the Protein Data Bank. We analyze both the interaction types used in their binding, as well as the functional groups involved in these. For the analysis, food compounds are divided into a set of molecules derived from fatty lipids (FoodFL, which includes glycerolipids, glycerophospholipids, and fatty acyls), and the rest of the molecules (FoodnoFL), since these correspond to highly dissimilar chemical spaces. As a control, a set of drugs is used. From this analysis it is found that the three compound sets provide very different solutions to the protein-ligand binding problem. FoodnoFL compounds stabilize their binding mainly through hydrogen bonds, salt bridges, cation-π interactions, and metal coordination, while FoodFL do so through hydrophobics and close contacts. In turn, drugs prefer the use of hydrophobic, π-π, cation-π, and halogen bonds interactions. These differences result from differences in the types and relative abundances of functional groups, and/or differential usage of interaction types by the same functional groups. This knowledge can be exploited in the design new drugs inspired in food compounds.