Efeito do glicerol como co-substrato na bioconversão de xilose em xilitol pela levedura Candida guilliermondii: modelagem e simulação

Abstract

Sucrose is a specialized additive in a wide variety of foods as it offers the best palatability, making it more attractive for consumption. However, in excess, this sugar can cause harm to health. In this context, xylitol presents itself as an alternative, due to its sweetening potential similar to that of sucrose, but with properties that are more advantageous to human well-being. Mostly, xylitol is produced through chemical processes, but that require several costly operations and that reduce the final yield of the process, thus, the biotechnological route can be an alternative for the industrial production of this sweetener. In this sense, yeasts are the most studied for this bioconversion, and among them are the Candida genus. The production of xylitol takes place through the reduction of xylose to xylitol through the enzyme Xylose Reductase (XR) in the presence of NADPH as a cofactor. During a bioconversion, there is the formation of glycerol as a by-product of this metabolism, which is undesirable due to the deviation of the main carbon source for its formation. Due to the high commercial value of the product, it is important to apply tools to help predict the production of this sweetener. Thus, the present work aims to evaluate mathematical models to predict the behavior of Candida guilliermondii yeast using glycerol as a co-substrate for the production of xylitol, using the MATLAB® program for simulation and optimization, considering the data available in the literature. According to the results, the models used are based to predict the behavior of the concentration profiles obtained for each condition tested, with R2 ranging from 0.9574 to 0.9833 for cell growth, 0.9264 to 0.9904 for glycerol, 0.9974 to 0.9995 for xylose and 0.9919 to 0.9983 for xylitol and with no expected variation between the 6 conditions analyzed, indicating good quality in the adjustments made and adequacy of the models to the observed data.