Identification of two highly promiscuous thermostable sugar nucleotidylyltransferases for glycorandomization

Glycorandomization is a process that improves the efficacy of glycoconjugates by the addition of a diverse array of sugars to secondary metabolites and antibiotics of pharmaceutical importance. This process, which employs sugar nucleotidylyltransferases (SNTs) and glycosyl transferases (GTs) in tandem, would benefit by the employment of promiscuous enzymes, i.e. those with the ability to utilize diverse noncanonical substrates. As promiscuous GTs are available, here we set out to identify promiscuous SNTs. For this, we began with a detailed family-wide characterization of SNTs. Earlier, we had proposed that SNTs could be classified into two major groups – I and II. They share a common structural framework and utilize a similar catalytic mechanism. Subtle variations in the way two magnesium ions — Mg2þ A and Mg2þ B — are stabilized by metal ion coordination motifs led to their classification into diverse subgroups viz. I-A, I-B, I-C, II-A, and II-B. Based on this classification, here we investigate promiscuity across the entire family of SNTs. We study the utilization of several sugar phosphates and nucleotides by the various subgroups of SNTs to understand substrate specificity and promiscuity in these. We find that promiscuity is prevalent among SNTs; and in particular, in the thermophilic homologs. In principle, promiscuity profiling identified four new SNTs that can be employed for the production of sugar-nucleotide libraries. However, assaying for their ability to simultaneously utilize multiple substrates in a single-pot reaction, we find two thermophilic SNTs- TMGA, an adenylyltransferase from Thermotoga maritima and PHGT, a thymidylyltransferase from Pyrococcus horikoshii that are readily employable for the production of diverse sugar-nucleotides.