Fermentation kinetics and modeling of non-growing Clostridium thermocellum JW20

Clostridium thermocellum is an anaerobic, thermophilic bacterium that can directly ferment cellulose into ethanol. The use of non-growing C. thermocellum cells may have potential for industrial applications, but their multiproduct-fermentation characteristics have not been well studied. The overall objective of this study was to quantify end-product formation by nongrowing C. thermocellum cells metabolizing the soluble sugar cellobiose and to develop kinetic models to represent/explain the experimental observations.; Batch and fed batch fermentations with varying cellobiose availability were conducted. An initial cellobiose concentration of 16 mM was sufficient to achieve the maximal ethanol concentration (∼14.5 mM) in static batch fermentation with 3.2 OD600 initial cell density, however significant amounts of lactate (21.3 mM) were produced. Stirring favored acetate production (acetate/ethanol ratio increased from 0.29 to 1.46) and reduced lactate production (lactate/ethanol ratio decreased from 2.85 to 0.82). A constant stirred fed-batch process (SCFR between 0.39 and 1.17 mmole cellobiose/g cells/h) was found to maximize ethanol production (∼25 mM adjusted ethanol concentration) while lowering lactate production and prolong the length of time the cells actively produce end-products.; Growing and non-growing C. thermocellum batch cultures were treated with various exogenous ethanol concentrations from 0 to 870 mM. The specific growth rate and the final cell concentrations decreased by ∼69% and ∼42%, respectively under 217 mM ethanol concentration while the cellobiose consumption and acetate production stayed the same. The overall critical ethanol concentration which completely inhibited cellobiose intake was approximately 652 mM.; Batch and constant fed-batch kinetic models were developed based on an approach which extended beyond the simple material balance normally done for yeast fermentation. For the batch model, the rate of change of specific product formation followed first order kinetics, but the inactivation coefficient varied by product. For the constant fed-batch model, the specific cellobiose feeding rate and specific lumped intracellular materials (SLIM) controlled the specific production rates of ethanol, acetate and lactate and the cellular intake rate of cellobiose. Ethanol inhibition kinetics were incorporated into the batch model by adjusting the model parameters accordingly. The overall performance of the models were satisfactory.