Professor Daniel Ungar 

Research

Overview

The diversity of eukaryotic glycan structures is produced by a large number of Golgi localized glycosylation enzymes. The precise localization of these enzymes to different Golgi cisternae is essential for accurate glycan synthesis. We would like to understand how the organization of the glycosylation machinery governs glycan biosynthesis.

• More of our research can be found on the Ungar Lab website

Vesicle tethering at the Golgi and glycosylation

The conserved oligomeric Golgi (COG) tethering complex is responsible for coordinating vesicle targeting to sort glycosylation enzymes to their respective cisternae. We have previously mapped protein-protein interactions between the eight mammalian COG subunits and members of other trafficking protein families. More recent projects have investigated the (mis)localization of enzymes within the Golgi upon COG-mutations, and the resulting changes in glycosylation. We are also generating mutations in COG subunits that alter COG’s interaction with selected binding partners with the aim to obstruct a subset of the intra-Golgi vesicles. The characterization of these mutants is done with the support of biologics manufacturers interested in new ways for glyco-engineering mammalian cells.

Glycan profiling and modelling glycan biosynthesis

To understand the consequences of altered enzyme sorting within the Golgi we are developing analytical tools to profile the glycans made by the Golgi. The filter-aided N-glycan separation method was originally developed for N-glycans, and then expanded for O-glycans as well. A further addition to our toolkit is a stochastic computational model of N-glycan processing in the Golgi. By using Bayesian fitting between modelled and experimentally determined glycan profiles, we can generate predictions about the changes of Golgi organization upon COG mutations. The profiling and modelling tools are now employed in projects with strong industrial links to investigate glycosylated biologics and the (bio)synthesis of human milk oligosaccharides.

Investigating glycan functions during cellular differentiation

To test the concept of “forward glycomics”, we started using a mesenchymal stem cell model that can differentiate into osteoblasts. Changing glycosylation using COG-defects could indeed alter differentiation. Interestingly, we found enhanced osteogenic differentiation caused by inhibiting complex N-glycan formation, something that our modelling suggests could be due to a shift in glycan branching during osteogenesis.

Figure 1. Glycans influence osteogenic differentiation of MSCs.

Three week differentiation of MSCs ± the glycan processing inhibitor kifunensine (A) or in the presence of the sugars lactose and sucrose (B). Mineralization was stained by von Kossa staining (brown specs – A) or alizarin red staining (red colour – B).