PubAg

Main content area

Effect of Fermentation and Drying on Procyanidins, Antiradical Activity and Reducing Properties of Cocoa Beans

Author:
Di Mattia, Carla, Martuscelli, Maria, Sacchetti, Giampiero, Scheirlinck, Ilse, Beheydt, Bram, Mastrocola, Dino, Pittia, Paola
Source:
Food and bioprocess technology 2013 v.6 no.12 pp. 3420-3432
ISSN:
1935-5130
Subject:
air drying, antioxidant activity, cocoa beans, data collection, decolorization, fermentation, high performance liquid chromatography, least squares, models, polymers, procyanidins, solar drying
Abstract:
This work was aimed to determine the effect of fermentation and drying on the content and profile of procyanidins (from monomers P1 to polymers P10) as well as on the antiradical and scavenging properties of cocoa beans. To this purpose, three experiments were carried out: a traditional fermentation process followed by air drying and two pilot-scale fermentation processes by either natural microbiota or starter followed by sun drying. Procyanidins were evaluated by HPLC analysis, while the total polyphenol index (TPI), the antiradical activity as well as the reducing power were determined by means of the reaction with the Folin–Ciocalteu reagent, the decolorization assays of the ABTS radical (TEAC) and the Ferric Reducing Antioxidant Power (FRAP) methods, respectively. Both the traditional and pilot-scale processes resulted to affect the profile and content of the procyanidins fractions as well as the antiradical and reducing power functionality. Drying caused a severe reduction of compounds and thus resulted to be the critical step for the loss of procyanidins and monomers in particular. The indices of functionality generally showed a decreasing trend as a consequence of processing, and their evolution was similar to that observed in procyanidins content. To study the relationship between the individual procyanidins and the antioxidant activity expressed as TEAC, FRAP and TPI, the data set were processed by modified partial least squares regression. The obtained models presented a good predictive ability. Normalised regression coefficients showed that the relative contribution of each single class of compounds to total antioxidant activity resulted as follows: P1 > P2 > P3 > P4 > P6 > P8 > P5 > P7 > P9 >> P10.
Agid:
611057