Ingénierie à façon de glycoenzymes par des outils de criblage virtuel innovants pour le design de vaccins entériques – GLUCODESIGN

Scientific context: Shigella flexneri are enteroinvasive Gram negative bacteria responsible for endemic shigellosis, or bacillary dysentery. To fight this major burden, especially in children under 5 years of age, vaccines ensuring multi-serotype coverage are in demand. New glycovaccines based on synthetic oligosaccharide mimics of the bacterial lipopolysaccharides of concern, the major targets for protection, were proposed by P1 as a promising alternative to live-attenuated candidate vaccines. Their development needs a better understanding of the molecular bases of the role that bacterial polysaccharides play in pathogenicity and long-term host's protection. All these aspects strongly stimulate the development of synthetic serotype-specific oligosaccharide haptens. The chemical synthesis of these complex motifs was undertaken, but the necessary introduction of serotype-specific a-D-branched glucopyranosyl residues was thought to be a limitation towards multivalency. Biocatalysts open the way to novel chemical diversity and hold great potential to circumvent the boundaries defined by chemistry. However, natural enzymes do not always display the specificity required for efficient chemo-enzymatic pathways. Engineering of the enzyme specificity was demonstrated feasible. Yet poorly exploited, the fast-growing development of protein engineering technologies allows the search and the construction of better performing enzymes with new specificities. However, directed molecular evolution techniques generate large size of variants to be screened and are rapidly overflowed by the size of the combinatorial space to be explored. There is a clear need to develop effective methods enabling reduction of the sequence space to be explored and fasten the enzyme engineering process. Description of project methodology and challenge: The motivation of this project is the computer-aided design of appropriate enzymatic glycosylation tools to allow an optimal combination of the chemical and enzymatic steps involved in the synthesis of complex fragments of S. flexneri O-antigens. This multidisciplinary project combines competencies in glycochemistry, enzyme engineering and protein design, as fulfilled by the three involved partners: UCB/Institut Pasteur (P1), LISBP (P2), RIA-LAAS/CNRS (P3). Chemo-enzymatic pathways designed for the low cost enzymatic 1,2-cis-glucosylation of different disaccharides made of L-rhamnose and/or N-acetyl-D-glucosamine, which answer the need for serotype-specificity are to be investigated. The challenge is to engineer new biocatalysts matching this demand. The selected enzymes, glucansucrases, are transglucosidases of glycoside-hydrolase family 13 and 70 that naturally synthesize glucans from sucrose. Although featuring broad and adaptable acceptor substrate specificity, they are not optimized to glucosylate the selected non natural disaccharide acceptors. To reach our goal, improved specificities will be created. An innovative molecular modelling approach based on robotic motion planning adapted to molecular simulation will be introduced to develop new computational tool enabling virtual screening of combined mutations localized in the enzyme acceptor-sites. The interest is to combine the efficacy of a geometric treatment of the main molecular constraints with the performance of a motion planning based conformational search. This approach should allow structure-function prediction with a rapidity far surpassing classical modeling approaches. This innovative strategy will serve to guide the generation of intelligent libraries of glucansucrase mutants. High-throughput screening techniques will be used to isolate mutants with the desired specificity. Information exchange between partners will permit cross-feeding of experimental and in silico approaches with a prediction-validation logic to give new insight in the structure-activity relationship of glucansucrases and improve the virtual screening accuracy. Finally, the validation step will consist in demonstrating the efficient chemical conversion of the products of enzymatic glucosylation into synthetic intermediates to targeted S. flexneri oligosaccharides. Expected results: Focused on S. flexneri, a biological target of importance for human health, our project will demonstrate the feasibility of new innovative chemo-enzymatic pathways in glycochemistry. New computational tools for virtual screening with broad applicability will be developed. New insights in glucansucrases structure-activity and new glucosylation tools will be disclosed. Standing at the border between chemistry, biology, and computational sciences, this project will open the way to a number of future developments (extrapolation to other enzymes and biological targets…). Important applications in the fields of biotechnology and health are anticipated.