Optimization of Metabolic Oligosaccharide Engineering with Ac₄GalNAlk and Ac₄GlcNAlk by an Engineered Pyrophosphorylase

Abstract

Metabolic oligosaccharide engineering (MOE) has fundamentally contributed to our understanding of protein glycosylation. Efficient MOE reagents are activated into nucleotide-sugars by cellular biosynthetic machineries, introduced into glycoproteins and traceable by bioorthogonal chemistry. Despite their widespread use, the metabolic fate of many MOE reagents is only beginning to be mapped. While metabolic interconnectivity can affect probe specificity, poor uptake by biosynthetic salvage pathways may impact probe sensitivity and trigger side reactions. Here, we use metabolic engineering to turn the weak alkyne-tagged MOE reagents Ac 4 GalNAlk and Ac 4 GlcNAlk into efficient chemical tools to probe protein glycosylation. We find that bypassing a metabolic bottleneck with an engineered version of the pyrophosphorylase AGX1 boosts nucleotide-sugar biosynthesis and increases bioorthogonal cell surface labeling by up to two orders of magnitude. A comparison with known azide-tagged MOE reagents reveals major differences in glycoprotein labeling, substantially expanding the toolbox of chemical glycobiology.

Cell-specific Bioorthogonal Tagging of Glycoproteins

“…21,25,26 In contrast, a shorter 5-carbon pent-4-ynoate chain in the substrate GalNAlk can be used as a substrate in low efficiency. 17 This is corroborated by the crystal structures of both GALK2 and AGX1 (Fig. S1).…”

“…13 Many MOE reagents are based on chemically tagged analogues of Nacetylgalactosamine (GalNAc) that are activated by the biosynthetic GalNAc salvage pathway to derivatives of the nucleotide-sugar UDP-GalNAc. [14][15][16][17] Epimerisation at C4 position by the epimerase GALE yields UDP-Nacetylglucosamine (UDP-GlcNAc) analogues that can be used by different glycosyltransferases to chemically tag glycan structures in the secretory pathway and in the cytosol. [16][17][18] In turn, UDP-GalNAc analogues are used by members of the cancer-relevant N-acetylgalactosaminyltransferase (GalNAc-T) family consisting of 20 members (T1…T20).…”

“…[14][15][16][17] Epimerisation at C4 position by the epimerase GALE yields UDP-Nacetylglucosamine (UDP-GlcNAc) analogues that can be used by different glycosyltransferases to chemically tag glycan structures in the secretory pathway and in the cytosol. [16][17][18] In turn, UDP-GalNAc analogues are used by members of the cancer-relevant N-acetylgalactosaminyltransferase (GalNAc-T) family consisting of 20 members (T1…T20). 14,19,20 We have recently used engineered double mutants (DM) of GalNAc-Ts to enable recognition of bulky analogues of UDP-GalNAc that are not used by WT-GalNAc-Ts.…”

“…1a). 14,[16][17][18] First, the 20 members of the GalNAc transferase family (GalNAc-T1…T20) use UDP-GalNAc to form the linkage GalNAc-Ser/Thr and thereby prime cancer-relevant O-GalNAc glycans. 14,19,20 Second, epimerisation at the GalNAc C4 position by the UDP-galactose-4-epimerase (GALE) yields UDP-N-acetylglucosamine (UDP-GlcNAc) that can be incorporated into different glycan subtypes, for instance Asn-linked N-glycans.…”

“…14,19,20 Second, epimerisation at the GalNAc C4 position by the UDP-galactose-4-epimerase (GALE) yields UDP-N-acetylglucosamine (UDP-GlcNAc) that can be incorporated into different glycan subtypes, for instance Asn-linked N-glycans. 17,18,21 Certain chemical 3 modifications at the N-acyl moiety can render GalNAc analogues recalcitrant to these metabolic processes. For instance, analogues of UDP-GalNAc with long alkyne-containing N-acyl substituents are not biosynthesised by the GalNAc salvage pathway and not used as substrates by wild type (WT)-GalNAc-Ts.…”