High-resolution crystal structures and STD NMR mapping of human ABO(H) blood group glycosyltransferases in complex with trisaccharide reaction products suggest a molecular basis for product release.

The human ABO(H) blood group A- and B-synthesizing glycosyltransferases GTA and GTB have been structurally characterized to high resolution in complex with their respective trisaccharide antigen products. These findings are particularly timely and relevant given the dearth of glycosyltransferase structures collected in complex with their saccharide reaction products. GTA and GTB utilize the same acceptor substrates, oligosaccharides terminating with α-l-Fucp-(1→2)-ß-d-Galp-OR (where R is a glycolipid or glycoprotein), but use distinct UDP donor sugars, UDP-N-acetylgalactosamine and UDP-galactose, to generate the blood group A (α-l-Fucp-(1→2)[α-d-GalNAcp-(1→3)]-ß-d-Galp-OR) and blood group B (α-l-Fucp-(1→2)[α-d-Galp-(1→3)]-ß-d-Galp-OR) determinant structures, respectively. Structures of GTA and GTB in complex with their respective trisaccharide products reveal a conflict between the transferred sugar monosaccharide and the ß-phosphate of the UDP donor. Mapping of the binding epitopes by saturation transfer difference NMR measurements yielded data consistent with the X-ray structural results. Taken together these data suggest a mechanism of product release where monosaccharide transfer to the H-antigen acceptor induces active site disorder and ejection of the UDP leaving group prior to trisaccharide egress.

Thermodynamic signature of substrates and substrate analogs binding to human blood group B galactosyltransferase from isothermal titration calorimetry experiments.

It has been observed earlier that human blood group B galactosyltransferase (GTB) hydrolyzes its donor substrate UDP-Galactose (UDP-Gal) in the absence of acceptor substrate, and that this reaction is promoted by the presence of an acceptor substrate analog, α-L-Fuc-(1,2)-ß-D-3-deoxy-Gal-O-octyl (3DD). This acceleration of enzymatic hydrolysis of UDP-Gal was traced back to an increased affinity of GTB toward the donor substrate in the presence of 3DD. Herein, we present new thermodynamic data from isothermal titration calorimetry (ITC) on the binding of donor and acceptor substrates and analogs to GTB. ITC data are supplemented by surface plasmon resonance and STD-NMR titration experiments. These new data validate mutual allosteric control of binding of donor and acceptor substrates to GTB. It is of note that ITC experiments reveal significant differences in enthalpic and entropic contributions to binding of the natural donor substrate UDP-Gal, when compared with its analog UDP-Glucose (UDP-Glc). This may reflect different degrees of ordering of an internal loop (amino acids 176-194) and the C-terminus (amino acids 346-354), which close the binding pocket on binding of UDP-Gal or UDP-Glc. As both ligands have rather similar dissociation constants KD and almost identical modes of binding this finding is unexpected. Another surprising finding is that an acceptor analog, α-L-Fuc-(1,2)-ß-D-3-amino-3-deoxy-Gal-O-octyl (3AD) as well as the constituent monosaccharide ß-D-3-amino-3-deoxy-Gal-O-octyl (3AM) effectively inhibit enzymatic hydrolysis of UDP-Gal. This is unexpected, too, because in analogy to the effects of 3DD one would have predicted acceleration of enzymatic hydrolysis of UDP-Gal. It is difficult to explain these observations based on structural data alone. Therefore, our results highlight that there is an urgent need of experimental studies into the dynamic properties of GTB.

A new concept for glycosyltransferase inhibitors: nonionic mimics of the nucleotide donor of the human blood group B galactosyltransferase.

Glycosyltransferases play an important role in the formation of oligosaccharides and glycoconjugates. To find suitable and selective inhibitors for this class of enzymes is still challenging. Here, we describe a novel concept that allows the design of inhibitors based on the structure of the donor substrate binding pocket. As a first step we describe the design, synthesis and analysis of inhibitors of the human blood group B galactosyltransferase (GTB). This enzyme served as a model system to study the concept, which can be used for easy access of glycosyltransferase inhibitors in general. In silico docking of bicyclic heteroaromatic ligands to GTB and experimental verification of binding affinities by saturation transfer difference NMR (STD NMR) spectroscopy gave 9-N-pentityl uric acid derivatives as non-ionic mimics of UDP. Two derivatives were synthesized and showed inhibitory activity for GTB as determined by competitive STD NMR experiments and by a radiolabeled enzyme assay.

Binding of an acceptor substrate analog enhances the enzymatic activity of human blood group B galactosyltransferase.

The hydrolysis of the donor substrate uridine diphosphate galactose (UDP-Gal) by human blood group B galactosyltransferase (GTB) has been followed by nuclear magnetic resonance in the presence and in the absence of an acceptor substrate analog. It is observed that the presence of the acceptor substrate analog promotes hydrolysis of UDP-Gal. Subsequent analysis of the kinetics of the enzymatic hydrolysis suggests that this effect is due to an increased affinity of GTB for UDP-Gal in the presence of the acceptor analog. Isothermal titration calorimetry experiments substantiate this conclusion. As hydrolysis may be understood as a glycosyl transfer reaction where water serves as universal acceptor, we suggest that in general the binding of acceptor substrates to retaining glycosyltransferases modulates the rate of glycosyl transfer. In fact, this may point to a general mechanism used by retaining glycosyltransferases to discriminate acceptor substrates under physiological conditions.

NMR-based exploration of the acceptor binding site of human blood group B galactosyltransferase with molecular fragments.

A substantial body of work has been devoted to the design and synthesis of glycosyltransferase inhibitors. A major obstacle has always been the demanding chemistry. Therefore, only few potent and selective inhibitors are known to date. Glycosyltransferases possess two distinct binding sites, one for the donor substrate, and one for the acceptor substrate. In many cases binding to the donor site is well defined but data for acceptor binding is sparse. In particular, acceptor binding sites are often shallow, and in many cases the dimensions of the binding pocket are not well defined. One approach to glycosyltransferase inhibitors is to chemically link donor site and acceptor site ligands to generate high affinity binders. Here, we describe a novel approach to identify acceptor site ligands from a fragment library. We have chosen human blood group B galactosyltransferase (GTB) as a biologically important model target. The approach utilizes a combination of STD NMR, spin-lock filtered NMR experiments and surface plasmon resonance measurements. Following this route we have identified molecular fragments from a fragment library that bind to the acceptor site of GTB with affinities of the order of a natural acceptor substrate. Unlike natural substrates these fragments allow for straightforward chemical modifications and, therefore will serve as scaffolds for potent GTB inhibitors. In general, the approach described is applicable to any glycosyltransferase and may assist in the development of novel glycosyltransferase inhibitors.

A nonionic inhibitor with high specificity for the UDP-Gal donor binding site of human blood group B galactosyltransferase: design, synthesis, and characterization.

9-(5-O-α-D-galactopyranosyl)-D-arabinityl-1,3,7-trihydropurine-2,6,8-trione (1) was designed and synthesized as a nonionic inhibitor for the donor binding site of human blood group B galactosyltransferase (GTB). Enzymatic characterization showed 1 to be extremely specific, as the highly homologous human N-acetylgalactosaminyltransferase (GTA) is not inhibited. The binding epitope of 1 demonstrates a high involvement of the arabinityl linker, whereas the galactose residue is only making contact to the protein via its C-2 site, which is very important for the discrimination between galactose and N-acetylgalactosamine, the substrate transferred by GTA. The approach can generate highly specific glycosyltransferase inhibitors.