Optimization of Exocytopolysaccharide Production from Fermented Blueberry Juice by Complex Lactic Acid Bacteria Based on Response Surface Method and Artificial Neural Network

Abstract

To improve the content of extracellular polysaccharide (EPS) in fermented blueberry juice, three strains of lactic acid bacteria with high EPS production were selected in this study, and the fermentation conditions of blueberry juice were optimized by single-factor method and response surface methodology (RSM), and the four most influential factors were screened out, namely, initial pH, inoculum amount, fermentation temperature and fermentation time. Based on this, the optimal fermentation process conditions were obtained by artificial neural network (ANN) and genetic algorithm (GA). The optimized process conditions were 2:1:1 ratio of Lactobacillus plantarum 9sh, Lactobacillus fermentum SR2-6 and Citrobacter cepacia GM11, lactose was 6%, soybean peptide was 0.6%, the initial pH of blueberry juice was 4.5, the inoculum amount was 8%, fermentation temperature was 30 ℃, and fermentation time was 60 h. Under these conditions, the EPS content was 3.537 g/L. This study shows that RSM and ANN can be used to optimize the EPS production process of blueberry juice by lactic acid bacteria fermentation.

Keywords

• exocytopolysaccharide
• artificial neural network (ann)
• response surface method (rsm)
• lactic acid bacteria
• mixed fermentation
• fermented blueberry juice