Université Côte d'azur, CNRS, Institut de Chimie de Nice UMR7272, 06108 Nice, France.
Université Côte d'azur, CNRS, Institut de Chimie de Nice UMR7272, 06108 Nice, France.
Food Chem. 2017 Apr 15;221:1421-1425. doi: 10.1016/j.foodchem.2016.10.145. Epub 2016 Nov 3.
Based on the most exhaustive database of sweeteners with known sweetness values, a new quantitative structure-activity relationship model for sweetness prediction has been set up. Analysis of the physico-chemical properties of sweeteners in the database indicates that the structure of most potent sweeteners combines a hydrophobic scaffold functionalized by a limited number of hydrogen bond sites (less than 4 hydrogen bond donors and 10 acceptors), with a moderate molecular weight ranging from 350 to 450g·mol. Prediction of sweetness, bitterness and toxicity properties of the largest database of natural compounds have been performed. In silico screening reveals that the majority of the predicted natural intense sweeteners comprise saponin or stevioside scaffolds. The model highlights that their sweetness potency is comparable to known natural sweeteners. The identified compounds provide a rational basis to initiate the design and chemosensory analysis of new low-calorie sweeteners.
基于具有已知甜度值的甜味剂最详尽的数据库，建立了一个新的用于甜度预测的定量构效关系模型。对数据库中甜味剂的物理化学性质的分析表明，大多数强效甜味剂的结构都结合了一个疏水支架，该支架由数量有限的氢键位点（少于 4 个氢键供体和 10 个受体）官能化，分子量适中，在 350 到 450g·mol 之间。对最大的天然化合物数据库的甜度、苦味和毒性特性进行了预测。计算机筛选显示，预测的大多数天然高强度甜味剂都包含皂苷或甜菊糖苷支架。该模型强调，它们的甜度与已知的天然甜味剂相当。所鉴定的化合物为设计和化学感官分析新的低热量甜味剂提供了合理的依据。
