Metabolic optimization of Escherichia coli for the de novo synthesis of 2'-fucosyl-lactose from glycerol and lactose in a fed-batch process

• Applicants: Professorin Dr. Kathrin Castiglione; Professor Dr. Sebastian Aljoscha Wahl; Professor Dr.-Ing. Dirk Weuster-Botz
• Subject Area: Biological Process Engineering
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 545517165

Human milk oligosaccharides are found in large quantities in breast milk. The trisaccharide 2'-fucosyl-lactose (2′-FL) is one of the most abundant oligosaccharides and 2’-FL is an approved ingredient of infant formula. Chemical production is inefficient, therefore biosynthetic routes are favored - especially whole-cell syntheses with recombinant Escherichia coli. In de novo synthesis, mannose-6-phosphate is converted in four enzymatic steps to GDP-L-fucose, which is then used by an α-1,2-fucosyltransferase for the fucosylation of lactose. Today, metabolically engineered E. coli production strains are available that enable 2'-FL synthesis using glycerol as the energy and carbon source and lactose as co-substrate in a fed-batch process. To gain a better understanding of the synthesis of 2'-FL with recombinant E. coli in the fed-batch process and to use this knowledge for further strain optimization, the limiting steps in the metabolic network of 2'-FL producers will first be identified. For this purpose, short-term analyses (metabolome, fluxome) will be used. Specifically, rapid media transitions of E. coli cells from the fed-batch process will be performed, building the basis for a metabolic control analysis (MCA). The targets for genetic changes identified from MCA will be implemented in the 2′-FL production strain (overexpression of genes of limiting enzymes, expression of genes of deregulated and optimized enzymes or deletion of genes to reduce unwanted carbon fluxes). A hypothesis based on the previous findings is that low α-1,2-fucosyltransferase activity is most likely one of the limiting factors in 2'-FL synthesis. This is due to two properties of the enzymes known so far: low soluble expression in E. coli in combination with low turnover numbers with the substrate lactose. Therefore, we will develop an α-1,2-fucosyltransferase with higher soluble expression. For this purpose, a taxonomy-based screening for new natural variants is combined with protein engineering of selected enzymes. The introduction of stabilizing mutations can be done based on good homology models. Active site modifications to increase the activity with lactose would greatly benefit from a crystal structure that currently does not exist for this class of enzymes. The structural elucidation will be carried out with a variant of a cyanobacterial α-1,2-fucosyltransferase that has already shown spherulitic crystal growth in preliminary work. The resulting improved enzymes will be fed into the MCA-guided metabolic engineering process. The production performance of the engineered E. coli strains will subsequently be evaluated in an optimized fed-batch process to enable particularly efficient production of 2'-FL.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Yves André Muller