| D-xylose is the second most abundant carbohydrate in nature. Its efficient utilization is essential for successful utilization of renewable biomass. However, only a few microorganisms can ferment xylose. The goal of this work is to engineer Saccharomyces cerevisiae to convert xylose into ethanol. The initial metabolic engineering approach has focused on extension of substrate ranges in S. cerevisiae by heterologous expression of XYL1 and XYL2 coding for xylose reductase and xylitol dehydrogenase from Pichia stipitis. These attempts have resulted in low ethanol yield from xylose because of xylitol accumulation. To reduce xylitol accumulation, both the expression levels of XYL1 and XYL2 and the amount of aerations were changed. Higher xylitol dehydrogenase (XDH) activity and aeration shifted byproduct formation from xylitol to xylulose. This result supported previous findings that the third enzyme in the pathway, D-xylulokinase, limits flux of metabolites through the xylose assimilation pathway in recombinant S. cerevisiae . To augment enough D-xylulokinase activity in recombinant S. cerevisiae expressing XYL1 and XYL2, XYL3 coding for D-xylulokinase was isolated from P. stipitis and expressed along with XYL1 and XYL2 . Interestingly, overexpression of XYL3 inhibited cell growth on xylose. The higher xylulokinase activity was in recombinant S. cerevisiae, the more cell growth inhibition was observed. This result suggested that uncontrolled xylulokinase expression in recombinant S. cerevisiae is deleterious to the cell. Therefore, we optimized the expression level of XYL3 with a tunable expression vector that utilized a yeast transposon element (Ty) for integration. The resulting low-level expression of XYL3 improved conversion of xylose into ethanol, but ethanol yield and productivity was not equivalent to that of native P. stipitis. Genome-wide analysis of gene expression revealed that xylose was not recognized as a fermentable carbon sugar by a recombinant S. cerevisiae. This resulted in low ethanol yield because of oxidative utilization of xylose. To force the cells to use xylose in a non-oxidative manner, we isolated a respiration-deficient mutant from the recombinant S. cerevisiae. The resulting petite mutant produced more ethanol from xylose than the parent. Surprisingly, it also accumulated less xylitol from xylose. Metabolic flux analysis in conjunction with fermentation trials were employed to understand the differences between glucose and xylose metabolism in S. cerevisiae. A stoichiometric model for S. cerevisiae that describes the main cellular metabolic reactions involved in growth, glycolysis, pentose metabolism, and respiration was developed. The metabolic phenotypes predicted from calculated flux distributions from the stoichiometric model were in accordance with corresponding experimental data sets. |
