Optimization of the fermentation medium for Paecilomyces tenuipes N45 using statistical approach

Abstract

In the present study, a sequential statistical approach was applied to optimize the medium in submerged cultivation of Paecilomyces tenuipes N45. Desirability value was used as response to simultaneously enhance the yields of mycelium, adenosine, polysaccharide and cordyceps acid. Based on single-factor optimization strategy, the suitable carbon sources, nitrogen sources and inorganic salts were obtained. Then key medium components were identified by Plackett-Burman design (PBD) and further optimized by Box-Behnken design (BBD). Finally, response surface methodology (RSM) and artificial neural network – genetic algorithm (ANN-GA) were used to model and optimize the experimental results obtained from BBD. The optimum components of nutrient medium comprised (g/L): glucose 40, beef extract 10, soy peptone 10, KH2PO4 0.688, MgSO4·7H2O 1, NaCl 0.500, VB1 0.201, VB12 0.130. In a word, a mean value of desirability valuesDv = 0.493 was obtained, which was 20.540% higher than the value achieved by the basal medium. The biomass, the production of adenosine, the polysaccharide and the cordyceps acid yields were enhanced by 8.200, 3.580, 23.170 and 31.510% respectively. Key words: Artificial neural network, optimization, Paecilomyces tenuipes, response surface methodology, genetic algorithm.