| The iron-chelating siderophore vibriobactin of the pathogenic Vibrio cholerae is assembled by a nonribosomal peptide synthetase system, VibE, B, H and F, using 2,3-dihydroxybenzoate and L-threonine as precursors to two (dihydroxyphenyl)methyloxazolinyl groups in amide linkage on a norspermidine scaffold. We have utilized site-specific and domain-deletion mutagenesis to map the heterocyclization, primary and secondary amine acylation activities of the six domain (Cy1-Cy2-A-C1-PCP-C2) VibF subunit.; The oligomeric structure of nonribosomal peptide synthetases (NRPS), fatty acid synthases (FAS), and polyketide synthases (PKS), multimodular enzymatic assembly lines utilized in natural product biosynthesis, has been a topic of interest because higher order oligomeric quaternary structural arrangements allow for alternate paths of acyl intermediate elongation and present unique challenges for the chimeric engineering of hybrid assembly lines. We report regain of catalytic activity upon mutant VibF protein mixing that argues for heterodimer formation, stable for hundreds to thousands of catalytic cycles. Ultracentrifugation data likewise confirm a dimeric structure for VibF and additionally that the C1 domain is largely responsible for VibF dimerization. Comparative rates of vibriobactin production suggest that the mere presence of C1 does not detectably enhance the catalytic rates of neighboring domains, but may properly orient Cy1-Cy2-A relative to PCP-C2.; The presence and identity of glycosyl moieties on multiple classes of natural product antibiotic scaffolds modulate activity and resistance profiles. Novel combinations of sugar/scaffold pairs will result in new antibiotics with enhanced activities. While most of these novel molecules will be achieved by chemical synthesis, facilitative and production cost considerations make investment into developing chemoenzymatic and/or fermentative access worthwhile. One method to develop novel glycosyltransferase activities is to chimerically cross scaffold and sugar binding domains to yield the desired set of substrate specificities. We report a free energy scoring function that can robustly discriminate real from decoy glycosyltransferases in instances where primary structure analysis and protein threading scoring functions routinely fail. With this free energy scoring function, candidate glycosyltransferase domain pairings that will likely result in functional chimeras are identified, and the optimal chimeric branchpoints for a specific pairing are determined. The free energy scoring function correlates well with the available experimental kinetics measurements. |
