Kinetic modeling of the enzymatic synthesis of galacto-oligosaccharides: Describing galactobiose formation

• Published in: Food and bioproducts processing 2021-05, Vol.127, p.1-13
• Main Authors: Schultz, Guilhermina, Alexander, Ronald, Lima, Fernando V., Giordano, Roberto C., Ribeiro, Marcelo P.A.
• Format: Article
• Language: English
• Subjects: Bioreactors; Chemical synthesis; Confidence intervals; Design optimization; Enzymatic synthesis; Enzymes; Food industry; Food processing industry; Galacto-oligosaccharides; Galactobiose; Galactooligosaccharides; Galactosidase; Inactivation; Kinetic Modeling; Kinetics; Kluyveromyces lactis; Lactose; Mathematical models; Oligosaccharides; Parameters; Prebiotics; Process control; Process controls; State estimation; Statistical analysis; Statistical methods; Substrates; β-Galactosidase
• ISSN: 0960-3085; 1744-3571
• DOI: 10.1016/j.fbp.2021.02.004

•A new kinetic model is proposed to describe the enzymatic synthesis of galacto-oligosaccharides (GOS) from lactose using beta-galactosidase from Kluyveromyces lactis.•The proposed model is able predict galactobiose formation during the reaction.•The model presents only 9 parameters besides the enzyme inactivation constant. Galacto-oligosaccharides (GOS) are highly valued prebiotics in the food industry with many significant health benefits. The enzymatic transgalactosylation of lactose to produce GOS, catalyzed by β-galactosidase, is a consolidated route in industry. Kinetic models of this reaction network presented in the literature usually adopt broad assumptions by considering only the formation of GOS with a glucose residue in the reaction, neglecting the presence of galactobiose (Glb), which is also a Galacto-oligosaccharide that has bifidogenic effects. Proposed models that take into account the presence of Glb and use lactose as a substrate do not give information about this molecule concentration during the reaction. In this paper, a kinetic model is proposed with nine adjustable parameters to characterize the formation of Glb, trisaccharides, tetrasaccharides and enzyme inactivation. A bootstrap method, based on the residuals obtained during the fitting procedure, is employed to calculate parameters’ confidence intervals. This model fits very well to experimental data and addresses many of the existing issues in the literature. As a result, this model may be useful for bioreactor optimization and process design, as well as process control and state estimation studies.