Thèse Découverte et Méchanismes Moléculaires d'Enzymes pour la Synthèse d'Oligosaccharides H/F Doctorat.Gouv.Fr

Établissement : Institut National des Sciences Appliquées de Toulouse École doctorale : SEVAB - Sciences Ecologiques, Vétérinaires, Agronomiques et Bioingenieries Laboratoire de recherche : TBI - Toulouse Biotechnology Institute, Bio & Chemical Engineering Direction de la thèse : Gabrielle VERONESE ORCID 000000034232230X Début de la thèse : 2027-09-01 Date limite de candidature : 2026-11-16T23:59:59 Les oligosaccharides sont d'un intérêt capital pour la nutrition et la santé humaine. Cependant, leur synthèse chimique ou leur extraction à partir de la biomasse posent des défis majeurs en termes de rendement, de coût et de durabilité. Les glycoside phosphorylases (GPs) représentent une alternative attractive pour leur synthèse. Ces enzymes catalysent de façon réversible la rupture et la synthèse de liaisons glycosidiques, permettant une production durable d'oligosaccharides sans avoir recours à des donneurs de sucre activés coûteux, contrairement aux glycosyltransferases de type Leloir. Néanmoins, la diversité fonctionnelle et structurale des GPs, pourtant abondantes dans les écosystèmes microbiens, reste sous-explorée, limitant leurs applications biotechnologiques. De plus, de récentes études ont révélé que l'organisation spatiale et les changements conformationnels de ces enzymes, souvent multimériques, voire multi-domaines, guident leur spécificité et leur efficacité.
En combinant différentes approches d'exploration fonctionnelle des génomes et métagenomes, ce projet de thèse vise à identifier de nouvelles GPs à l'activité inédite, d'intérêt pour la nutrition et la santé humaine, et à décrypter, grâce à des études structurales basées sur la CryoEM, les déterminants moléculaires qui gouvernent la catalyse lors de la synthèse d'oligosaccharides.
Oligosaccharides and polysaccharides are essential carbohydrates that play pivotal roles in a wide range of physiological and structural functions across all living organisms. Oligosaccharides, often found as free molecules or conjugated to proteins and lipids, act as key mediators in cell recognition, signaling, and immune modulation. Some are also prebiotics that shape gut microbiota composition. Polysaccharides, on the other hand, provide structural integrity to cell walls, contribute to energy storage, and form protective barriers against environmental stresses. Their structural diversity, arising from variations in monosaccharide composition, glycosidic linkages, and branching patterns, underpins their multifunctional bioactivities.
Due to their stereochemical diversity, and the need for precise regioselective glycosylation, the chemical synthesis of oligosaccharides remains a challenge, and suffers from low yields, tedious protection-deprotection steps, and scalability issues. Alternatively, extracting oligosaccharides from biomass presents hurdles such as heterogeneous feedstock composition, inefficient hydrolysis techniques, and the need for environmentally sustainable separation processes. In this context, enzymatic synthesis offers a solution to balance efficiency, cost, and sustainability (1). In particular, glycoside-phosphorylases (GPs) represent a unique class of enzymes that play a pivotal role in the metabolism and synthesis of carbohydrates (2). Structurally and mechanistically, GPs share similarities with glycoside hydrolases (GHs) and glycosyltransferases (GTs). Thus, they are classified in both GH and GT families of the CAZy classification of carbohydrate-active enzymes (3). These intriguing GPs utilize inorganic phosphate to catalyze the reversible phosphorolysis of glycosidic linkages. This mechanism, which is distinct from that of true GHs which cleave glycosidic bonds through hydrolysis, results in the formation of sugar-1-phosphates and a shortened oligosaccharide chain. The reversibility of this reaction enables GPs to function not only in the degradation of oligo and polysaccharides, but also in their synthesis (in the so-called reverse-phosphorolysis reaction). The biotechnological potential of GPs is vast, as they enable the synthesis of high-value oligosaccharides and glycoconjugates without the need for expensive activated sugar donors, contrary to Leloir-type GTs. GPs can be used either in vitro (4) or in vivo, as recently exemplified for the synthesis of oligosaccharides of health interest (prebiotics and antigenic oligosaccharides used for the prevention, diagnosis and treatment of infectious diseases (5,6)), offering a sustainable and cost-effective alternative to traditional chemical synthesis or enzymatic synthesis using GTs.
Despite this promise for advancing green chemistry and biocatalysis, there are several bottlenecks to the extensive use of GPs for glycoside synthesis. Firstly, GPs have been identified so far in only eleven of the 195 GH and 140 GT families described to date. They represent less than 2 % of the characterized CAZymes, and only a handful have been identified from uncultured bacteria, which however dominate in all microbial ecosystems. The substrate specificity of the known GPs is restricted to a few donor substrates for glycoside synthesis (Glc1P, GlcNAc1P, Man1P and Gal1P) while other sugar-1-phosphates do exist in nature, such as Fuc1P and GalNAc1P (7). Therefore, the discovery of GPs specific for lactose and human milk oligo-saccharides (HMOs, which contain a variety of sugar moieties, including Fuc and GalNAc) would thus unlock technological limitations in the dairy and infant milk industries. Secondly, the activity-based discovery of new GPs from environmental DNA is still limited by the available screening workflows. A pioneering study proved that new GPs could be discovered using functional metagenomics, without any a priori on sequence homology with known GPs (8). Nevertheless, using this approach, hit rates are so low that very large libraries (hundreds of thousands clones) must be screened to get hits. A micro-plate assay has been developed for GP discovery (9), but throughput is too low and this automated strategy is too expensive in terms of consumables/substrates/reactants to be generalized to extensive GP screening campaigns against multiple substrates of interest. Finally, structurally, GPs often form homooligomers and some of them, that we recently identified using sequence-based metagenomics (unpublished data), are multimodular, harbouring additional GH and/or GT catalytic domains, of which the functional role has not been investigated yet. As we recently revealed by solving the first CryoEM structure of a dimeric GP, the assembly, conformational changes and dynamics of the different protomers certainly play a central role in catalysis (10,11).
During this thesis, we will further explore the diversity of GPs in genomes and metagenomes using two complementary strategies: sequence-based and activity-based approaches. We will analyze the sequence diversity of CAZy families containing only GHs active on lactose and HMO motifs, aiming to discover the first GPs specific to these substrates. We will also investigate already known GP families to uncover new specificities. To achieve this, we will employ a sequence clustering strategy and analyze the conservation of catalytic site residues, a method that has previously enabled the identification of novel GP activities (12). Additionally, we will target original sequences identified in an unpublished prior study performed in collaboration with the CAZy group, encoding bimodular and trimodular GPs.
In parallel, to discover new GP activities without sequence-based assumptions, we will develop a novel droplet microfluidics-based screening strategy. This ultra-high-throughput approach, which is orders of magnitude faster and less expensive (due to miniaturization to pL reaction volumes) than traditional screening techniques, will be used to screen metagenomic libraries derived from mammalian gut microbiomes, which are rich in GPs.
Based on the results of these complementary studies, we will establish a shortlist of the ten most original GPs, which we will produce recombinantly in E. coli. The multi-modular GPs will be rationally truncated to elucidate the functional role of each catalytic domain. In-depth functional and structural characterization of the most interesting and best-produced targets will be performed. X-ray crystallography and CryoEM will be used to investigate, at the atomic level, the relationships between quaternary structure, conformational mobility and dynamics to unravel the structural basis for catalytic efficiency, substrate specificity, and chain length of synthesis products.

- In-depth analysis of sequence diversity of glycoside-phosphorylases in genomes and metagenomes
- Development of droplet-microfluidics screening workflows for the discovery and engineering of glycoside-phosphorylases
- Discovery of novel enzymes of biotechnological interest, specific for lactose and/or for the synthesis of prebiotic fiber and human milk oligosaccharides
- Biochemical and structural characterization of original glycoside-phosphorylases and in-depth analysis of their mode of action.
Bioinformatics (sequence diversity analysis, AI-based structural modeling), droplet-microfluidics and functional metagenomics, recombinant production and purification of enzymes, rational engineering of enzymes, enzymatic assays, HPLC (HPAEC-PAD), mass spectrometry and NMR (for reaction product identification), X-ray crystallography, Cryo-EM, Small angle X-ray scattering