Measurement and control of dissolved carbon dioxide in mammalian cell culture processes using an in situ fiber optic chemical sensor.

At high viable cell concentrations in large-scale mammalian cell culture processes, the accumulation of dissolved carbon dioxide (dCO(2), typically quantified as an equilibrium gas-phase concentration) becomes problematic as a result of low CO(2) removal rates at reduced surface-to-volume ratios. High dCO(2) concentrations have previously been shown to inhibit cell growth and product formation in mammalian cells and to alter the glycosylation pattern of recombinant proteins. Therefore, reliable monitoring and control of dCO(2) are important for successful large-scale operation. Off-line measurements by instruments such as blood gas analyzers (BGA) are constrained by the low frequency of data collection and cannot be used for on-line control. In a preliminary evaluation of the YSI 8500 in situ sensor, a response time (t(90%)) of 6 min, sensitivity of 0.5% CO(2) (3.6 mmHg), and linearity of measurement (R(2) = 0.9997) between the equivalent gas-phase partial pressure of 0-180 mmHg (0% and 25% CO(2)) were established. Measurements were found to be unaffected by culture pH and typical mammalian cell culture concentrations of glucose, glutamine, glutamate, lactate, and ammonium. The sensor withstood repeated sterilization and cleaning cycles. The reliability of this sensor was demonstrated in microcarrier-based Chinese hamster ovary (CHO) cell perfusion cultures at reactor scales of 30, 40, 340, and 2000 L and was successfully implemented in a dCO(2) control strategy using N(2) sparging.

Formation and cell-medium partitioning of autoinhibitory free fatty acids and cyclodextrin's effect in the cultivation of Bordetella pertussis.

Palmitic, palmitoleic and stearic acids were found in the extracted cellular lipids of virulent Bordetella pertussis as unesterified acids in confirmation of earlier taxonomic analyses. The same free fatty acids (FFAs) were found in the spent culture supernatant in concentrations higher than in the uninoculated medium, indicating that they are released into the extracellular medium. These long-chain fatty acids are known to inhibit the growth of B. pertussis at concentrations as low as 1 ppm. Measurement of palmitate cell-medium partitioning demonstrated a strong tendency of FFAs for cellular adsorption. Inhibition kinetics indicated that the cell-bound FFA was responsible for inhibition and that the specific cellular FFA concentrations actually found during growth were similar to those determined to be inhibitory. Autoinhibition by these endogenous FFAs provides an explanation of the low maximum cell concentrations currently attainable in liquid media. Addition of soluble dimethyl-beta-cyclodextrin (MebetaCD) to FFA-inhibited cultures resulted in a rapid reversal of the inhibition. A corresponding shift in the distribution of FFAs from the cells to the extracellular medium demonstrated that MebetaCD sequesters FFAs. Although MebetaCD did not increase final cell concentrations and even had an adverse effect on growth at concentrations above 1 g l-1, it did (at 1 g l-1 extend the initial period of high growth rate leading to shorter cultivation times.

Improved pertussis toxin production by Bordetella pertussis through adjusting the growth medium's ionic composition.

Ionic composition and total ionic concentration of the growth medium were important factors in limiting productivities in aerated reactors used for the production of pertussis toxin and other antigens by Bordetella pertussis. Salt concentration has opposing effects on cell growth of wild-type B. pertussis and specific toxin formation. Sodium ion concentrations below 140 mM correlated with a precipitous decline in specific yields of pertussis toxin, an otherwise growth-associated product. High salt concentrations in the medium resulted in lower final cell concentrations but did not affect initial growth rates. A new medium is proposed that allows a 60 to 70% increase in both cell and toxin yields by replacing the sodium chloride in the 'cyclodextrin liquid' (CL) medium with additional monosodium glutamate which provides both the sodium and the carbon and energy source.