Abstract: S-adenosyl-L-methionine (SAM) from Saccharomyces cerevisiae is usually extracted with a high concentration of perchloric acid or organic solvents. In order to reduce the use of organic solvents and potential operating risks during SAM extraction from S. cerevisiae, an ultrasonic-assisted extraction method using a low concentration of hydrochloric acid as the extraction solvent was proposed. First of all, dry yeast cells were used to compare the extraction efficiencies of various solvents including perchloric acid, ethyl acetate-sulfuric acid and hydrochloric acid at different concentrations in an effort to select the most efficient solvent for the extraction of SAM. Subsequently, the extraction process was optimized using one-factor-a-at-a-time method, Plackett-Burman design, steepest ascent design and response surface methodology. The results demonstrated that 0.01 mol/L hydrochloric acid was the solvent of choice. The effects of various factors on extraction efficiency decreased in the descending order of were in the descending order of solid-to-liquid ratio > ultrasonic power > irradiation time > irradiation interval. The optimal conditions obtained were as follows: ultrasonic power, 314 W; irradiation time, 4.1 min; and solid-to-liquid ratio, 1:27 (g/mL). Under the optimized conditions, the experimental extraction yield of SAM was 66.56 mg/g, 92.1% as compared to traditional perchloric acid extraction. The relative deviation between the experimental and predicted values was 0.98%. The optimized extraction procedure is applicable to the production of SAM.

Key words: S-adenosyl-L-methionine, ultrasonic-assisted extraction, response surface methodology, optimization of extraction process, Saccharomyces cerevisiae