Directed Evolution of a ß-Galactosidase into a ß-Transgalactosidase

The classic chemical synthesis of oligosaccharides and glycoconjugates is challenging since regio- and stereochemical control of linkage formation requires multiple protection and deprotection steps. Enzymatic synthesis represents an attractive alternative. Retaining glycosidases hydrolyze glycosidic linkages using a double- displacement mechanism in which a covalent glycosylenzyme intermediate is formed, which is subsequently hydrolyzed in a general acid/base-catalyzed manner. However, under kinetically controlled conditions these enzymes can be used for glycosidic bond formation due to their transglycosylation activity, where the glycoside is transferred onto an acceptor saccharide instead of a water molecule. Despite a number of applications the utility of glycosidases in carbohydrate synthesis has been limited by the challenge of finding reaction conditions that will drive the reaction in the desired direction without allowing significant enzymatic degradation of the reaction product. To overcome these limitations engineering of glycosidases offers great promise. The application of directed evolution will bring forward new important amino acid positions in and around the active site as well as unpredictable, more distant positions connected to the transglycosylation/hydrolysis ratio. The ultimate goal is to engineer a true transglycosidase without significant hydrolytic activity. The objective of the proposed research project is to use a combination of semi-rational design and directed evolution techniques for the conversion of a ß-galactosidase into a ß-transgalactosidase and to identify a mutant enzyme that can efficiently transfer galactose onto lactose or N-acetyl-glucosamine; this latter acceptor was chosen since galactosyl transfer will yield lacto-N-biose, a building block of human milk oligosaccharides. The target protein of this proposal is the ß-galactosidase from Lactobacillus reuteri, a member of the glycoside hydrolase family 2, whose crystal structure is currently elucidated. Screening of mutant glycosidases for improved transfer activity is extremely challenging, because no obvious change in fluorescence or absorbance is associated with bond formation. Therefore a modification of a high- throughput screening developed by the group of Stephen G. Withers for glycosyltransferases will be used, where E. coli cells are sorted by a fluorescence-activated cell sorter (FACS) according to the fluorescence derived from fluorescent transglycosylation products trapped in the cells. Several of the most promising enzyme variants will be characterized in detail including the identification of the transglycosylation products by HPLC and NMR. The crystal structure of the best enzyme variants will be elucidated by our cooperation partner. We expect to gain new insight in the structure-function relationship of ß- galactosidases concerning the balance between hydrolysis and transglycosylation, likely helping to engineer other glycosidases into transglycosidases, especially enzymes of the glycoside hydrolase family 2. The engineered enzymes could be part of a toolbox for the synthesis of tailor-made oligosaccharides with potential applications as food additives or therapeutics.

ß-Galactosidases catalyse both the hydrolysis and the transfer of sugar residues onto other sugars such as lactose or cellobiose (transglycosylation). These properties make these enzymes interesting for biotechnological applications, such as the production of galacto- oligosaccharides (GOS) from lactose. As so-called prebiotics, indigestible food ingredients, GOS have a positive influence on the intestinal microflora. GOS are essential components of human mother's milk, as such they are of great interest for infant milk products, which are among the most commonly industrially produced prebiotics.We investigated the transgalactosylation potential of ß-galactosidases from Streptococcus thermophilus and Bifidobacterium breve. It was shown that all three are suitable for the production of GOS due to their high transgalactosylation activity. A GOS yield of up to 50% of total sugar was achieved.On the basis of variants of the Halothermotrix orenii ß-glucosidase, the mechanism of transglycosylation activity was investigated more closely. For this purpose, the amino acid sequence of the enzyme was changed at single positions by genetic engineering methods. First, five positions (N222F, N294T, F417S, F417Y and Y296F) were changed. Of all variants, F417S and F417Y showed the best GOS yields, which were increased by one third compared to the wild type (WT). Subsequently, the influence of changes in position F417 was examined in detail. The amino acid at this position was replaced by all other possible amino acids and the resulting changes in activity were investigated. One of these variants, F417T, showed a 3- fold higher preference for lactose than for cellobiose, whereas the WT showed a 15-fold higher selectivity for cellobiose versus lactose - a ß-glucosidase was turned into a ß-galactosidase. By other variations the spectrum of the produced GOS could be changed. In summary, the exchange of a single amino acid could greatly alter enzyme activity. These findings can lead to engineering of tailor-made enzymes for the production of prebiotics with an optimal composition.