Linear Correlation between Online Capacitance and Offline Biomass Measurement up to High Cell Densities in Escherichia coli Fermentations in a Pilot-Scale Pressurized Bioreactor

Abstract

To yield high concentrations of protein expressed by genetically modified Escherichia coli, it is important that the bacterial strains are cultivated to high cell density in industrial bioprocesses. Since the expressed target protein is mostly accumulated inside the E. coli cells, the cellular product formation can be directly correlated to the bacterial biomass concentration. The typical way to determine this concentration is to sample offline. Such manual sampling, however, wastes time and is not efficient for acquiring direct feedback to control a fedbatch fermentation. An E. coli K12-derived strain was cultivated to high cell density in a pressurized stirred bioreactor on a pilot scale, by detecting biomass concentration online using a capacitance probe. This E. coli strain was grown in pure minimal medium using two carbon sources (glucose and glycerol). By applying exponential feeding profiles corresponding to a constant specific growth rate, the E. coli culture grew under carbon-limited conditions to minimize overflow metabolites. A high linearity was found between capacitance and biomass concentration, whereby up to 85 g/L dry cell weight was measured. To validate the viability of the culture, the oxygen transfer rate (OTR) was determined online, yielding maximum values of 0.69 mol/l/h and 0.98 mol/l/h by using glucose and glycerol as carbon sources, respectively. Consequently, online monitoring of biomass using a capacitance probe provides direct and fast information about the viable E. coli biomass generated under aerobic fermentation conditions at elevated headspace pressures.

Real-time estimation of biomass and specific growth rate in physiologically variable recombinant fed-batch processes

“…Different calibrations can be necessary for different process modes, media composition, and physiological states of the culture and also for different strains. Model maintenance or very good generalization (however, this often comes at the cost of estimation accuracy) is required to account for variability of the relation between the probe response and the actual dry cell weight, since the relation can vary, e.g., dependent on the media constituents and other process characteristics [13, 14], especially in process development with constantly changing process environments, culture conditions, strains, etc. it can be misleading to use these hard-type sensors as a sole quantification of biomass in process development.…”

Control of specific carbon dioxide production in a fed-batch culture producing recombinant protein using a soft sensor

“…Gnoth et al, 2007;Clementschitsch and Bayer, 2006;Knabben et al, 2011). Studies carried out with GFP expression in fed-batch cultivation underscore this further (DeLisa et al, 1999).…”

A Soft Sensor for Bioprocess Control Based on Sequential Filtering of Metabolic Heat Signals

“…The capacitance probe has previously been applied in yeast and mammalian cell cultures [28–30] as well as in E. coli recombinant protein production [31]. Capacitance measurement has also previously been combined with bio-calorimetric monitoring of bioreactors [32]. Thus, the capacitance methodology seems as a realistic alternative to compare our data with.…”