Structure-odor relationship in pyrazines and derivatives: A physicochemical study using 3D-QSPR, HQSPR, Monte Carlo, molecular docking, ADME-Tox and molecular dynamics

Abstract

In this study, using both 2D-QSPR and 3D-QSPR approaches, to understand the structure-odor relationship of 78 1–4-pyrazine odorant molecules and to use this knowledge for the design of new food and flavor products. According to our results, the developed models have good predictability such as the HQSPR/BC model with QLOO2=0.832, R2 = 0.916, CoMSIA/SEH model with Q2 = 0.624, Rcv2 = 0.590, Rncv2 = 0.932, Rbs2 = 0.963, and Topomer CoMFA model withRtraining2 = 0.899, Rtest2 = 0.916. The Monte Carlo method was used in the creation of a Quantitative Structure-Property Relationship (QSPR) model. The molecular structure is represented using optimized Simplified Molecular Input Line Entry System (SMILES) and molecular descriptors. The performance of the model is evaluated using the Correlation Ideality Index (IIC) and the Correlation Contradiction Index (CCI). The best model, designated as TF2, boasts excellent statistical properties with a training R-squared value of 0.957 and a test R-squared value of 0.834. The model was then used to determine promoter activity levels, which formed the basis for the design of 36 new odorant molecules. Molecular docking and pharmacokinetic properties were used to explain the mode of binding between the proposed compounds and the active site of the Porcine Odorant Binding Protein complexed with pyrazine (2-isobutyl-3-methoxypyrazine). Molecular dynamic simulation was used to assess and justify the stability of the ligand in the active site of the receptor. The results of this study provide a basis for the discovery of new compounds with lower olfactory thresholds and diverse pharmacological properties.

Keywords

• Pyrazine
• Odor
• Monte-Carlo
• Molecular Docking
• MD simulation
• 3D/2D-QSAR