Insight into substrates binding in β-1,3 and β-1,4-glucan phosphorylases by STD NMR spectroscopy

S10.3 Glycosyl phosphorylases
Location (hall): 
Start/end time: 
Thursday, July 4, 2019 - 18:00 to 18:15

Valeria Gabrielli1, Juan Carlos Muñoz-Garcia1, Ridvan Nepravishta1, Giulia Pergolizzi2, Yaroslav  Khimyak1, Robert A. Field2, Jesus Angulo1

1University Of East Anglia, Norwich, United Kingdom, 2John Innes Centre, Norwich, United Kingdom

Glycoside phosphorylases (EC 2.4.x.x) are enzymes able to carry out the reversible phosphorolysis of glucan polymers. Importantly, their ability to build up short-to-medium length sugar chains from donor and acceptor substrates makes them powerful biological tools for the synthesis of new sugars with functional groups introduced in a selective way. The synthesis of cellodextrin oligomers from GH94 cellodextrin phosphorylase (CDP, EC involves the elongation of short β-(1→4)-glucans (acceptors) through the addition of glucose units donated from α-D-glucose-1-phosphate (donor).[1] Recently, the first X ray crystal structure of CDP bound to cellotetraose has been published, allowing a better understanding of the enzyme molecular recognition.[2] Previous studies have covered acceptors permissiveness and donor specificity as well, probing the efficiency of CDP to carry out the synthesis of cellulose derivatives. In addition, a new family of β-1,3-phosphorylase was recently identified and named GH149, which is able to synthesise β-(1→3)-glucans oligomers.[3] To be able to exploit the interesting catalytic properties of these enzymes in the synthesis of novel glucan derivatives, a profound understanding of the molecular basis of substrate recognition in solution is still needed for a diverse sets of potential substrates.

In this study, STD NMR experiments for the investigation of the molecular recognition of substrates by β-1,3- and β-1,4-phosphorylases were performed, and binding epitope maps of the different ligands were obtained. In the case of CDP, several donors (α-D-glucose-1-phosphate, α-D-galactose-1-phosphate, α-D-mannose-1-phosphate, etc.) as well as acceptors (glucose, cellobiose, laminaribiose and cellotriose) have been investigated. Moreover, we focused on cellotriose to look into ligand orientation and solvent-accessibility in the binding pocket by average DEEP-STD NMR methodology[4] and LOGSY-titration experiments[5], respectively. Finally, we probed the impact of inorganic phosphate in the ligands binding affinity. The obtained experimental data were compared to molecular docking calculations run both with and without inorganic phosphate, in order to reveal its impact on ligands conformation and orientation. The same study was started for β-1,3- phosphorylase, focusing on the donors and acceptors binding epitope investigation.

The collected data show differences in the protein contacts to the α  and β-anomers at the reducing ring, as well as to β-(1→4) and β-(1→3)-glucans and highlights enzyme preferences for binding recognition. Cellobiose and cellotriose β-anomers showed a spread binding epitope all along the ligand, suggesting a more complex binding mode. The ligand orientation, with the non-reducing ring inside the binding pocket, was confirmed through DEEP-STD NMR and the analysis of water exposure. On the other hand, combined analysis with docking calculations unveiled the existence in solution of an unproductive binding mode, where the acceptors enter the binding pocket via the reducing ring. Interestingly, the presence of inorganic phosphate enhances ligand binding affinity and affects the conformation around the inter-glycosidic linkage.

Overall, the combination of STD NMR experiments and molecular modelling calculations has allowed us: first, to obtain insights on enzyme-ligand interactions for small molecules reluctant to co-crystallize, and second, to detect and characterize an additional unproductive binding mode of the enzyme acceptors.

  1. M. Hiraishi et al. / Carbohydrate Research 344 (2009) 2468–2473
  2. O'Neill et al. / Carbohydrate Research 451 (2017) 118-132
  3. Kuhaudomlarp et al./ J. Biol. Chem. (2018) 293(8) 2865–2876
  4. Monaco et al. / Angew. Chem. Int.Ed. (2017), 56, 15289—15293
  5. Geist, L. et al. / J. Med.Chem (2017)., 60, 8708-8715