Improved utilization of hemicellulose sugars (specifically D-xylose and L-arabinose) for biosynthesis of value-added products is crucial for fermentative processes to become more economically feasible. In yeast and filamentous fungi, both pentose sugars enter the pentose phosphate pathway (PPP) through a common intermediate, D-xylulose 5-phosphate. D-xylose is converted to D-xylulose 5-phosphate through two redox reactions, while L-arabinose conversion requires four redox reactions prior to entering the PPP. These redox reactions require alternative forms of nicotinamide cofactor, as the reductases (xylose reductase and L-xylulose reductase) typically favor reduced NADPH and the dehydrogenases (L-arabinitol 4-dehydrogenase and xylitol dehydrogenase) favor oxidized NAD+. This results in a cofactor imbalance that potentially prohibits the efficient utilization of these hemicellulose sugars, resulting in the excretion of sugar alcohol by-products formed by the initial xylose reductase.

This work describes the cloning, characterization, and engineering of an L-arabinose pathway comprised mainly of enzymes from the filamentous fungi Neurospora crassa for co-utilization of D-xylose and L-arabinose for improved production of xylitol, a five carbon sugar alcohol with a growing market as a sweetener. In particular, the rational design and directed evolution of L-arabinitol 4-dehydrogenase towards the utilization of NADP+ as a cofactor was accomplished in an attempt to partially relieve the cofactor imbalance involved in conversion of L-arabinose to xylitol, and is currently being investigated for improvement in xylitol production from both pentose sugars in a model organism Escherichia coli. The engineered strain generated for this work includes deletions of the endogenous bacterial pentose sugar pathway genes (xylA and araBAD) as well as a mutant carbon catabolite regulator (crp*) to allow for utilization of pentose substrates with glucose as a co-substrate.

Extending this concept even further, engineering both dehydrogenases for NADP+ utilization could conceivably create closed redox loops between the reductases and the dehydrogenases, resulting in the utilization of a single cofactor pair, NADP+/NADPH. Through rational design, N. crassa xylitol dehydrogenase cofactor specificity was completely reversed to NADP+, and the entire engineered pathway leading from L-arabinose and D-xylose to D-xylulose 5-phosphate prior to introduction into the PPP is being tested for improved fermentative ethanol production from the pentose sugars.