Fed-batch production of recombinant beta-galactosidase using the universal stress promoters uspA and uspB in high cell density cultivations.

A high-level production system using the universal stress promoters uspA and uspB in a fed-batch cultivation based on minimal medium was designed. In development it was shown that a standard industrial fed-batch protocol could not be used for this purpose since it failed to induce the levels of product as compared to the basal level. Instead, a batch protocol followed by a low constant feed of glucose was shown to give full induction. The levels of the product protein, beta-galactosidase, corresponded to approximately 25% of the total protein. Higher levels were found using the uspA than uspB vectors where uspA showed considerably higher basal level. The data indicate that the sigma(70) regulated promoter, uspA, although affected by the alarmone guanosine tetraphosphate, ppGpp, worked partly in a similar manner to constitutive promoters. An industrial high cell density fed-batch cultivation on the basis of the suggested fed-batch protocol and the uspA promoter gave a final beta-galatosidase concentration of 7 g/L and a final cell concentration of 65 g/L. The heterogeneity in production of the individual cell was measured by fluorescence microscopy. The data show that there is a process time independent heterogeneity in production, which is suggested to be caused by heterogeneity in the substrate uptake rate of the individual cell.

Limiting factors in Escherichia coli fed-batch production of recombinant proteins.

Fed-batch production of recombinant beta-galactosidase in E. coli was studied with respect to the specific growth rate at induction. The cultivations were designed to induce protein production by IPTG at a glucose feed rate corresponding to high mu = 0.5 h(-1)) or low (mu = 0.1 h(-1)) specific growth rate. Protein production rate was approximately 100% higher at the higher specific growth rate, resulting in the accumulation of beta-galactosidase up to 30% of the total cell protein. Transcription analysis showed that beta-galactosidase-specific messenger RNA was immediately formed after induction (<5 min), but the amount was the same in both cases and was thus not the initial limiting factor. The content of ribosomes, as represented by rRNA, rapidly decreased with specific growth rate from a relative level of 100%, at the high specific growth rate, to 20% at the low specific growth rate. At high specific growth rate, ribosomes were additionally degraded upon induction due to the high production level. Translation therefore seemed to be the initial limiting factor of the protein synthesis capacity. The alarmone guanosine tetraphosphate increased at both high and low feed level inductions, indicating an induction-forced starvation of charged tRNA and/or glucose. The altered physiological status was also detected by the formation of acetic acid. However, the higher production rate resulted in high-level accumulation of acetic acid, which was absent at low feed rate production. Acetic acid production is thus coupled to the high product formation rate and is proposed to be due either to a precursor drain of Krebs cycle intermediates and a time lag before induction of the glyoxalate shunt, or to single amino acid overflow, since the model product is relatively poor in glycin and alanin. In conclusion, it is proposed that production at high specific growth rate becomes precursor-limited, while production at low specific growth rate is carbon- and/or energy-limited.