Abstract: Response surface methodology was employed to optimize the aqueous enzymatic extraction of oil and protein from perilla seeds. The extraction yields of protein and oil were investigated as a function of six extraction conditions including enzyme type, solid-to-solvent ratio, pH, temperature, enzyme loading and hydrolysis time. As a result, a mathematic model was established using a Box-Behnken experimental design. The model was statistically significant and allowed accurate prediction of extraction yields of protein and oil under various conditions. Alcalase was identified as the best enzyme for the aqueous enzymatic extraction of oil and protein from perilla seeds and the optimized values of solid-to-solvent ratio, pH, temperature, enzyme loading, and hydrolysis time for extracting oil and protein from perilla seeds were 1:4, 8.94, 61.3 ℃, 1.5% and 4.47 h, respectively. Under the optimized conditions, the extraction yields of oil and protein were e 85.59% and 73.43%, respectively.

Key words: perilla seed, aqueous enzymatic extraction, optimization, response surface analysis, oil extraction yield