(346b) Selective Production of Xylitol From Hemicellulosic Sugars - Using a Combined Protein and Metabolic Engineering Approach

Xylitol is a five-carbon sugar alcohol used as a natural sweetener in food, pharmaceutical, and confectionery industries. It has garnered significant interest due to its anti-cariogenic properties as it cannot be metabolized by Streptococci, Lactobacilli and yeasts found in the mouth. In addition, its metabolism is insulin-independent, making it a desirable sugar substitute. Currently, xylitol is produced commercially by catalytic reduction of hemicellulosic hydrolysate. This process is generally unsafe, as it utilizes dangerous Raney-Nickel catalyst and requires the use of high temperatures and pressures, thus fermentative and enzymatic approaches have been developed to make the xylitol production eco-friendly. However, the production cost is high due to the requirement for separation of xylose from its epimer arabinose (the most abundant sugars in hemicellulosic hydrolysate) prior to reduction. To minimize or eliminate the separation steps and lower the cost, fermentative microorganisms or enzymes that could specifically reduce xylose from the hydrolysate sugars to xylitol, leaving arabinose unreacted.

To meet this goal, we engineered a xylose-specific xylose reductase (XR), thereby performing an enzymatic ?kinetic resolution? of substrates. This strategy should prevent the primary contaminant, arabinose, from being reduced to arabinitol, thus allowing for easy purification of xylitol. Using an approach that combines directed evolution and structure-guided mutagenesis, we have engineered a mutant XR called VMQCI. This mutant demonstrates 16.5-fold higher activity toward xylose when compared to arabinose. Preliminary resting cell studies showed that E. coli cultures expressing this mutant produced xylitol to levels comparable to that of the parent enzyme, but significantly lower level of arabinitol.

We have also metabolically engineered an E. coli strain to efficiently transport and utilize pentoses. A culture of this strain expressing VMQCI yields xylitol as the only detectible product when grown in an equimolar mixture of xylose, arabinose, and glucose. Currently, we are in the process of demonstrating the feasibility of this system in bioreactor settings.