The temperature-limited fed-batch technique for control of Escherichia coli cultures
2006 (English)Doctoral thesis, comprehensive summary (Other scientific)
The objective of this study was to investigate the physiology and productivity in Escherichia colicultures with emphasis on the temperature-limited fed-batch (TLFB) culture. The TLFB techniquecontrols the oxygen consumption rate of the growing culture by a gradually declining temperaturefrom 37-35 °C down to 20-18 °C. The temperature regulated the DOT around a set-point (30 % airsat.), and all nutrients were in excess. Glucose was fed into the culture continuously, however, highacetate formation was avoided by keeping the glucose at a low, yet excessive, concentration. Thebiomass productivity was approximately the same in TLFB as in glucose-limited fed-batch (GLFB)cultures, since the specific growth rate and the oxygen consumption rate are limited by the oxygentransfer capacity of the reactor in both techniques.High concentrations of endotoxins were found in the medium of E. coli fed-batch cultures at lowspecific growth rates (below 0.1 h-1) and severe glucose limitation. In this thesis the TLFB techniquewas found to suppress the endotoxin release even at low specific growth rates. The repressed release of endotoxins in TLFB cultures was due to the high availability of glucose and not to the low growthrate or the lower temperature. The conclusion was drawn from comparing with the GLFB technique performed at 20 °C, which resulted in high endotoxin release.Extensive release of endotoxin, accompanied with high concentrations of soluble proteins was foundin a TLFB culture exposed to a higher energy dissipation rate, 16 kW m-3, instead of 2 kW m-3, due toa higher stirrer speed (1000 instead of 500 rpm). The hypothesis that this is a result of mechanicalstress is discussed in context with the common view that cells like E. coli, which are smaller than the Kolmogoroff’s microscale of turbulence, should not be influenced by the turbulence.TLFB cultured cells exhibited more stable cytoplasmic membranes when treated with osmotic shockas compared to the GLFB cultured cells. The concentrations of DNA and soluble proteins in the periplasmic extracts from the TLFB cultured cells were lower than from GLFB cultured cells. Inaddition, the specific productivity of periplasmic β-lactamase was higher in the TLFB cultures,suggesting that this technique could be an alternative for protein production. The solubility of apartially aggregated recombinant protein increased in the TLFB compared to the GLFB cultures.However some time after induction, in spite of the gradually declining temperature, the solublefraction decreased.For obtaining better understanding of the performance of the process and for identifying criticalparameters, a mathematical model was developed based on the growth, energy and overflowmetabolism at non-limiting nutrient conditions. The temperature-dependent maximum specific glucoseand oxygen uptake rates were determined in pH-auxostat cultures for temperatures ranging from 18 to37 °C. A dynamic simulation model of the TLFB technique was developed and the results were compared to experimental data. The simulation program was also used for sensitivity analysis of some physiological and process parameters to study the impact on biomass concentration and temperatureprofiles. An effect on the biomass concentration profile but not on the temperature profile wasobserved when changing the oxygen transfer coefficient. If the maximum specific glucose uptake ratewas altered, or if the glucose concentration was permitted to assume other values, the temperatureprofile but not the biomass concentration profile was influenced. Cell death affected both the biomassconcentration profile and the temperature profile.
Place, publisher, year, edition, pages
Stockholm: Bioteknologi , 2006.
Endotoxin, Escherichia coli, fed-batch technique, glucose-limitation, mathematical model
IdentifiersURN: urn:nbn:se:kth:diva-4154ISBN: 91-7178-473-XOAI: oai:DiVA.org:kth-4154DiVA: diva2:10943
2006-11-10, FB54, AlbaNova Universitetscentrum, Roslagstullsbacken 21, Stockholm, 10:00
Villadsen, John, Professor
QC 201009222006-10-232006-10-232010-09-22Bibliographically approved
List of papers