Influence of ethanol on xylose metabolism in engineered S. cerevisiae and applicability in continuous ethanol production

Abstract

There is a rapidly growing market for renewable substitutes to fossil-based fuels and chemicals due to increased awareness of global warming and political efforts to reduce emissions. Sekab E-technology AB develops and licenses technologies for bioethanol production from forest residues, converting various sugar types found in cellulose and hemicellulose into ethanol using the yeast Saccharomyces cerevisiae. Beyond their native pathways for utilization of glucose and other hexose monomers, strains used at Sekab have been engineered with a pathway for xylose metabolism. Co-fermentation of glucose and xylose results in a higher yield of ethanol, but productivity is exceedingly reduced if xylose fermentation is allowed to proceed until depletion. Extensive attempts to increase flux through the xylose pathway using metabolic engineering have been reported, but investigations of the impact of process-related parameters on co-fermentation are scarce. At Sekab, recent experiments have indicated that externally added ethanol may have a different effect on the fermentative rate of xylose compared to glucose in co-fermenting cultures on hydrolysate. This thesis aimed towards further investigating how external addition of ethanol influences the rate of fermentation in glucose and xylose co-fermenting batch cultures in defined media. In the present study, the fermentation rate decreased by 22 % during glucose consumption and increased by 25 % during xylose consumption when 20 gL−1 ethanol was added to co-fermenting cultures. Maximum xylose consumption rate appeared independent of both glucose concentration and ethanol addition, although the timing of xylose consumption onset was highly dependent on both. No difference in allocation of carbon between growth and ethanol formation was observed. As such, the effect of ethanol could not be attributed to prolonged glucose availability nor to increased growth during the glucose consuming phase. Co-fermenting batch cultures were also simulated computationally based on a combination of previously suggested kinetic models. A comparison of model-predicted behaviour with experimental data indicated that glucose kinetics cannot describe all aspects of xylose metabolism, in particular when glucose levels are low. Nevertheless, Sekab’s continuous fermentation regime was simulated using the existing models. While further modification of the model is needed for accurate representation of co-fermentation, feed-rate control based on measurements of refractive index (RI) appeared to be an efficient control system for continuous fermentation. The simulation also constituted a proof-of-concept that, when further developed, will provide demonstrative value for costumer-adapted in silico test runs. Overall, results presented in this report highlight the lack of knowledge about glucose and xylose co-fermentation dynamics and the need for further research.