Predicting vegetative inoculum performance to maximize phytase production in solid-state fermentation using response surface methodology.

Microbial phytase is used to reduce the environmental loading of phosphorus from animal production facilities. The limiting factors in the use of this enzyme in animal feeds can be overcome by solid-state fermentation (SSF), which is a promising technology for commercial enzyme production with lower production costs. Inoculum quality and the influence of inoculum quality on phytase production are important factors which need in-depth investigation before scaling-up of high-yielding fermentation process. A full factorial experimental design for 240 h with sampling at every 24 h was used to determine the effects of the treatments, inoculum age (plate and liquid culture), media composition and the duration of SSF on the production of fungal biomass and phytase in SSF systems using Aspergillus niger. The optimal treatment combination for maximal phytase production was determined by statistically comparing all treatments at each sampling time. Both 7- and 14-day plate cultures and M1 + medium composition with 72-h-old liquid inoculum treatments resulted in optimal phytase production at 144 h of SSF, which was the shortest duration observed for maximal phytase production. This resulted in maximal phytase production with a mean of 884 +/- 121 U/g substrate, while the maximal phytase production observed at 216 h of SSF (mean phytase activity of 1,008 +/- 121 U/g substrate), with the same treatment combinations, was not statistically significant from that at 144 h of SSF. Phytase production was strongly growth-associated with younger inocula. The significant treatment variables, age of liquid inoculum and the duration of SSF, were used to predict the system response for phytase production using response surface methodology. From the response surface model, the optimal response of the experiment was predicted and the reliability of the prediction was checked with the verification experiment.