Chemoenzymatic Synthesis of Carbohydrates and Derivatives with Engineered D-Fructose-6-Phosphate Aldolase

Tesi realitzada a l'Institut de Química Avançada de Catalunya (IQAC-CSIC) Biocatalysis is the chemical transformation of organic substrates by enzymes. This is the driving force of some of the oldest transformations known to humans. Nowadays, methods involving biocatalysts are gaining importance in organic synthesis owing to their high efficiency and selectivity, and driven by the need to develop processes that are economical and environmentally sustainable. Recombinant DNA technology has paved the road for mass production of enzymes and recently their major limitation, i.e., narrow substrate spectrum is also being addressed by the development of enzyme engineering. Rational design and directed evolution has been used to tackle a number of limitations associated with the use of wild-type aldolases as catalysts in synthetic organic chemistry. Aldolases, which catalyze carbon-carbon bond formations (one of the most challenging transformations in synthetic organic chemistry) with high stereospecificity have substantial utility as biocatalysts in the synthesis of chiral complex bioactive compounds such as carbohydrates, amino acids and their derivatives. D-fructose 6-phosphate aldolase (FSA) from E. coli is a unique member of this class of enzymes for accepting non-phosphorylated analogues of dihydroxyacetone phosphate (DHAP), and constitutes an exception to the strict donor specificity of aldolases. This dihydroxyacetone-dependent aldolase accepts a wide range of donor analogues of dihydroxyacetone (DHA), becoming an extremely promising catalyst for direct stereoselective aldol reactions. In this thesis an unprecedented expansion in the nucleophile and electrophile substrate scope of FSA is presented with the assistance of enzyme engineering. Strategies for the structure guided redesign by site-directed and site-saturation mutagenesis was used to modify the enzyme to improve its tolerance for a wide range of donor and acceptor substrates. We report the structure-guided rational protein engineering of FSA to accept a unique variety of 1-hydroxy-2-alkanones and related ether components as novel donor substrates. The double-active-site variant FSA L107A/L163A was found to convert outstanding variety of donor structures with good reaction rates, retaining high diastereoselectivy for the D-threo configuration. This newly designed FSA variant opens new avenues towards the synthesis of novel product families that were before inaccessible by biological catalysis. In