Application of statistically-based experimental designs for the optimization of eicosapentaenoic acid production by the diatom Nitzschia laevis.

Statistically based experimental designs were applied to the optimization of medium components and environmental factors for eicosapentaenoic acid (EPA) production by the diatom Nitzschia laevis in heterotrophic conditions. First, the Plackett-Burman design was used to evaluate the effects of variables including medium components and environmental factors on cell growth and EPA production. Among these variables, NaCl, CaCl(2), PI metal solution, pH, and temperature were identified to have the significant effects (with confidence level > 90 %). Subsequently, the concentrations of NaCl, CaCl(2), PI metal solution as well as the values of pH and temperature were optimized using central composite design. The cell growth and EPA production were found to respectively correlate to NaCl, CaCl(2), pH, and temperature that could be represented by second-order polynomial models. The optimal values of the four parameters were determined by response surface and numerical analyses as 8 g/L NaCl, 0.10 g/L CaCl(2), pH 8.5 and 19.8 degrees C for cell dry weight (DW), and 14 g/L NaCl, 0.10 g/L CaCl(2), pH 8.5 and 18 degrees C for EPA production, respectively. The subsequent verification experiments confirmed the validity of the models. This optimization strategy led to a DW of 9 g/L, an EPA yield of 280 mg/L and an EPA productivity of 28 mg/L/d, respectively, which were considerably higher than those obtained in the previous studies. Copyright 2001 John Wiley & Sons, Inc.