Development of a wine by-product-based beverage: study of the potential antidiabetic function in humans and possible implications of phenolic metabolites during postprandial state

Winemaking generates underused vegetable residues such as skins and seeds which retain considerable amounts of bioactive compounds, especially polyphenols. In line with the principles of the circular bio-economy, this study was carried out using wine by-products as a functional food ingredient. For this, we have formulated a wine by-products-based beverage rich in anthocyanidins (58 mg/100 mL), flavonols (11.3 mg/100 mL), flavan-3-ols (8.5 mg/100 mL) and pectin (581 mg/100 mL) to study its hypoglycaemic properties. For this, 10 healthy adults of both sexes (50% female, 24-46 y) were recruited to participate in a three independent days¿ pilot trial. Under fasting conditions, the postprandial glycaemia of volunteers was monitored after three different interventions: (i) water + sugar, (ii) beverage + sugar and (iii) beverage + water. In parallel, blood samples were collected using dried blood spot cards to determinate the polyphenol bioavailability by ultra-high liquid chromatography coupled to mass spectrometry (UPLC-MS/MS). Based on the reduction of the iAUC of glucose observed between (ii) beverage + sugar and (iii) water + sugar interventions, subjects were classified into responders (>10% of iAUC reduction, n=6) and non-responders (n=4). Regarded blood polyphenols, hydroxyphenyl propionic acid, hippuric acid, hydroxybenzoic acid and catechol sulphate were detected and quantified at different time points. The iAUC of hydroxyphenyl propionic acid was statistically higher within responders after the (ii) beverage + water intervention compared with the (iii) beverage + water intervention. On the other hand, the iAUC of glucose and hydroxyphenyl propionic acid was statistically higher in non-responders than responders after (ii) beverage + sugar intervention. In addition, individual analysis of blood glucose and phenolic metabolites showed an unclear kinetic association in both, responders and non-responders indicating that glucose absorption rate could be influenced