Rapid screening and taste mechanism of novel umami peptides from natural tripeptide database

Abstract

Small peptides, particularly tripeptides, play a crucial role in food umami taste. To dig for more umami tripeptides, the novel tripeptides pharmacophore model was established to rapidly screen umami peptides from natural tripeptide database. Twenty peptides with potential umami characteristics from 8000 tripeptides were further screened by molecular docking. The electronic tongue analysis and sensory evaluation suggested that the 20 tripeptides exhibited umami taste characteristics. The thresholds of the 20 tripeptides spanned from 0.137 to 2.237 mmol/L. Molecular dynamics simulations were used on T1R1 and four tripeptides with high umami intensity to reveal their taste mechanisms. In this study, a new screening strategy was established and 20 new umami tripeptides were identified and validated, providing a theoretical reference for rapid screening of umami tripeptides.

Keywords: Common characteristic pharmacophore; Molecular docking; Molecular dynamics simulation; Rapid virtual screening; Umami tripeptides.

Fig. 1

Construction and Evaluation of common feature pharmacophore models. A) Ten common feature pharmacophore models (stacked molecules are Asp-Glu-Lys) generated by 14 umami tripeptides. The red, blue and green balls represent ionizable positive, ionizable negative and represent acceptor, respectively. B) Resulting parameters of the 10 common feature pharmacophore models built with the umami tripeptides as the training set. C) ROC curves of 10 pharmacophore models. The horizontal coordinates in the graph represent the false positive rate; the vertical coordinates represent the true positive rate; and the area under the curve was used to represent the final statistical results. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Evaluation results of Ramachandran plot for two models. A) and B) were Ramachandran plot analysis of the optimized T1R1/T1R3-1EWK model and T1R1/T1R3-5X2M model after molecular dynamic simulation, respectively.

Fig. 3

Results of Electronic tongue analysis for the synthesis of 20 tripeptides. B) was a local zoom-in result for the tripeptide in the unnamed part of A). C) and D) were local zoom-in result for the blue and orange ellipse in B), respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Molecular docking probed the interaction mechanism between umami tripeptides and umami receptors. A) Umami receptor T1R1 and 69 umami tripeptides interaction active site statistics. B) A Number of non-bonded interactions and hydrogen bonds resulting from the docking of 69 umami tripeptides with the structural domain of T1R1 venus flytrap domain.

Fig. 5

MD proved the interaction mechanism between umami tripeptides and umami receptors. A) and C) were RMSD and RMSF analysis of the protein backbone after molecular dynamics simulation of four umami tripeptides complexed with T1R1, respectively. B) The changes of Rg during molecular dynamics simulations of four umami tripeptides complexed with T1R1.

Fig. 6

Disassembly analysis of binding energy of four complex systems. (A)-(D) were DHA-T1R1, DGE-T1R1, ENG-T1R1 and EDN-T1R1, respectively.