Date of this Version
This paper provides an approach for optimizing the cell density (Xc) and dilution rate (D) in a chemostat for a Pichia pastoris continuous fermentation for the extracellular production of a recombinant protein, interferon τ (INF-τ). The objective was to maximize the volumetric productivity (Q, mg INF-τ I-1 h-1), which was accomplished using response surface methodology (RSM) to model the response of Q as a function of Xc and D within the ranges 150 ≤ Xc ≤ 450 g cells (wet weight) l-1 and 0.1 μm ≤D ≤ 0.9 μm (μm =0.0678 h-1, the maximum specific growth rate obtained from a fed-batch phase controlled with a methanol sensor). The methanol and medium feed rates that resulted in the desired Xc and D were determined based on the mass balance. From the RSM model, the optimal Xc and D were 328.9 g h-1and 0.0333 h-1 for a maximum Q of 2.73 mg I-1 h-1. The model of specific production rate (q, mgINF- τ g-1 cells h-1) was also established and showed the optimal Xc=287.7 g I-1 and D=0.0361 h-1 for the maximum ρ(predicted to be 8.92•10-3 mg-3 g-1 h-1). The methanol specific consumption rate (v, g methanol g-1 cells h-1) was calculated and shown to be independent of the cell density. The relationship between v and μ (specific growth rate) was the same as that discovered from fed-batch fermentations of the same strain. The approach developed in this study is expected to be applicable to the optimization of continuous fermentations by other microorganisms.