Abstract

Bacterial iron acquisition by the means of enterobactin (ENT) is constrained in mammalian hosts due to ENT-binding proteins such as siderocalin and serum albumin. To evade sequestration by these proteins, ENT can be modified by the C glycosyltransferase IroB, which is located in the iroA locus of Salmonella and certain extraintestinal E. coli strains such as uropathogenic E. coli CFT073. The glycosylation of ENT has been reported to be a bacterial evasion mechanism to restore the iron scavenging ability of ENT in the presence of mammalian ENT-binding proteins by the installation of a steric impediment.
The C glycosyltransferase IroB catalyses the transfer of a glucose moiety to the DHB subunit of ENT under formation of a C-C bond between the anomeric C1 of the glucose moiety and the C5 of the 2,3-DHB subunit of ENT. The formation of mono-, di- and triglycosylated Ent (MGE/DGE/TGE) products where observed in vitro. The formation of a C-C bond is remarkable because of its chemical stability and resilience against enzymatic degradation.
In this M.Sc. thesis, we initially identified the iroB gene product in the iroA harbouring E. coli strain Nissle 1917 on transcriptional and translational level and expressed and purified IroB recombinant. Then, we investigated the mechanism of the C-C bond formation catalysed by IroB in vitro. Based on the hypothesis that deprotonation of the catechol 2 hydroxyl renders the catechol C5 para to the 2-hydroxyl nucleophilic, the C-C bond would then be formed in a general SN2 reaction between the attacking nucleophile and the anomeric carbon of glucose, which is further facilitated by the excellent phosphate leaving group of the UDP-glucose donor. By the means of homology modelling and superposition strategies, we were able to identify the binding sites of the glycosyl donor UDP-glucose and the glycosyl acceptor ENT and to locate residues that could potentially act as base catalysts to increase the phenolate anionic character of the 2,3-DHB subunit during catalysis.
We established an activity assay for IroB, separated products arising from IroB activity by reversed phase chromatography and compared so the activity of wild-type IroB and several variants. Additionally, all variants were characterized biophysically, mainly to confirm that the structural integrity was not impaired by mutations. Ultimately, our results enable us to propose a mechanism for C-glycosylation of IroB that is consistent with other glycosyltransferases found in nature.