Development of glycosyltransferase inhibitors for the glycoengineering of therapeutic antibodies

Student thesis: Doctoral ThesisDoctor of Philosophy


Immunoglobulin E (IgE)-based antibodies are currently under investigation as promising novel cancer therapeutics. Structure-function studies are complicated by the structural complexity of the IgE glycoforms. The overall goal of this project was to develop glycosyltransferase inhibitors with suitable properties for the application in IgE cell culture and the generation of defined IgE glycoforms. This would provide a novel method for IgE glycoengineering, establishing a basis for structure-function relationship studies of IgE and optimizing its biological and therapeutic properties.

Many existing glycosyltransferase inhibitors are charged molecules with limited cell permeability. The Wagner research group has previously developed a novel class of uncharged galactosyltransferase inhibitors with modest activity in cells. These inhibitors are based on a uridine scaffold with an additional substituent in position 5, which is critical for inhibitor activity. Starting from this existing inhibitor class, two strategies were pursued to improve potency and cellular activity.

In the first series, alternative substituents in position 5 were investigated. Target molecules 1 – 4 were tested for inhibitory activity against two recombinant galactosyltransferases (β-1,4-galactosyltransferase 1and β-1,4-galactosyltransferase 7) in vitro and, in the form of pro-drugs 5 – 8, for IgE glycoengineering in cell cultures (Chapter 2). Based on these results, a novel class of conformationally restricted analogues with a macrocyclic linkage between the 5- substituent at the uracil base and the 5-position of the ribose was designed. For the generation of target molecules, four different synthetic routes, following three general strategies, were explored, resulting in the delivery of 12 inhibitor candidates, 29 – 41 (Chapter 3). The activity of the conformationally restricted compounds was evaluated against galactosyltransferases in vitro and in cell assays (Chapter 4). Molecular modelling was employed to rationalise the experimental outcomes (Chapter 2 and 4). Additionally, the novel inhibitors were investigated against three bacterial sugar-UDP-dependent glycosyltransferases (LgtC, SetA and LtpM) (Chapter 2 and 4).

Findings revealed that the conformationally restricted macrocycle 31 exhibited the most potent activity. 31 displayed 99 % inhibition of β4GalT1 at 2 mM and 97 % inhibition of β4GalT7 at 1 mM. In a ligand binding assay based on Differential Scanning Fluorimetry (DSF), 31 demonstrated favourable selectivity for LgtC over SetA and LtpM. Subsequent inhibition assays revealed an IC50 of 193 ± 35 μM for LgtC. These suggested 31 could potentially serve as a broad range inhibitor of GalTs (β4GalT, β4GalT7 and LgtC) over glucosyltransferases (SetA and LtpM). Molecular dynamics (MD) simulations revealed intermolecular interactions between 31 and key residues in β4GalT1, β4GalT7, and LgtC that can explain the observed activities against these enzymes. Conversely, steric clashes were observed between 31 and both SetA and LtpM. These differences may provide an explanation for the experimentally observed target preference of 31. Compared to its ring-open analogue 1, the macrocyclic nucleoside 31 appears to exhibit increased inhibitory activity. This suggests that conformational restriction is a successful strategy for improving inhibitory activity. 31 also demonstrated a remarkable 26.0 – 31.0 % cell permeability. Overall, the aim of achieving inhibitors with optimised inhibitory activity, selectivity and cell penetration was successfully accomplished via conformational restriction.

Initial findings from IgE glycoengineering revealed a reduction of four different terminal residues on IgE glycans in the presence of the ring-open nucleoside 5, as analysed by lectin blots, while analogue 6 appeared to slightly increase the levels of these residues. Furthermore, there was no noticeable decrease in the prevalence of galactose or sialic acid in the tentative glycan assignment. These unexpected, and seemingly paradoxical, effects may at least in part be influenced by the relatively modest potency of these inhibitors. In contrast, the novel, conformationally restricted nucleoside inhibitors are promising candidates for IgE glycoengineering due to their enhanced inhibitory activity and promising cell permeability. Although some cytotoxicity was observed at 100 μM for compound 31, it still stands as a promising candidate with the potential for further structural refinement to increase its activity, thereby reducing the amount required for cellular applications. More generally, the successful evaluation of the conformational restriction strategy has opened a new pathway for designing additional analogues suitable for cellular applications.

Date of Award1 Apr 2024
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorSophia Karagiannis (Supervisor) & Gerd Wagner (Supervisor)

Cite this