OPTIMIZATION OF FERMENTATION MEDIUM FOR ACETOIN PRODUCTION BY Bacillus subtilis SF4-3 USING STATISTICAL METHODS

To improve the acetoin-producing ability of Bacillus subtilis SF4-3, isolated from “natto,” a Japanese traditional food, the fermentation medium was optimized in shake-flask fermentation by statistically designed methods. Based on results of the single-factor experiment, orthogonal experiment, and Plackett–Burman design, yeast extract, corn steep liquor, and urea were identified as showing significant influence on the acetoin production. Subsequently, the optimum combination of the three factors was investigated by the Box–Behnken design (BBD) of response surface methodology (RSM) in order to further enhance the acetoin production. The maximum acetoin yield of 45.4 g/L was predicted when the concentrations of yeast extract, corn steep liquor, and urea were 8.5 g/L, 14.6 g/L, and 3.8 g/L, respectively. The results were further confirmed in triplicate experiments using the optimized medium (glucose 160 g/L, yeast extract 8.5 g/L, corn steep liquor 14.6 g/L, urea 3.8 g/L, manganese sulfate 0.05 g/L, ferrous sulfate 0.05 g/L), and an acetoin yield of 46.2 g/L was obtained in the validation experiment, which was in agreement with the prediction. After the optimization of medium components, an increase of 36.28% in acetoin production was achieved in comparison to that at the initial medium levels.