A strategy for screening novel umami dipeptides based on common feature pharmacophore and molecular docking

Abstract

To shorten the complex and tedious process of traditional umami peptide identification, a novel model based on common feature pharmacophore (HipHop, a ligand molecule-based screening method) and molecular docking (a receptor protein-based screening method) was established for umami peptide screening. In this study, HipHop was used to perform a preliminary screening of peptides. Dipeptides with potential umami activity were docked into the umami taste receptor T1R1/T1R3 for a second screening. Twenty previously unreported umami dipeptides identified through virtual screening were validated using sensory evaluation and electronic tongue analysis. All 20 dipeptides (HE, HD, KE, EH, ET, EQ, DH, DR, DQ, DN, DY, DM, DI, DV, QE, QD, NE, ND, CE, and SE) had umami taste with umami threshold values ranging from 0.094 to 1.517 mmol/L. Therefore, when we increased the screening criteria for docking energy to -60 kcal/mol, the virtual screening results had 100% accuracy. The T1R1-peptide complexes of the four dipeptides with the lowest umami threshold values were subjected to molecular dynamics (MD) simulations for 100 ns, and the results showed that the four umami dipeptides remained in the starting active cavity. Overall, this screening strategy could be applied to the rapid screening of umami peptides in food products.

Keywords: Common feature pharmacophore model; Molecular docking; Rapid virtual screening; Umami peptide.