Post-PKS modifications in the biosynthesis of the antifungal antibiotic nystatin

The antifungal polyene macrolide nystatin is produced by Streptomyces noursei ATCC 11455. The nystatin biosynthesis gene cluster of Streptomyces noursei has been cloned and sequenced, and a biosynthesis route has been predicted. In the present work, investigation of genes presumably involved in post-PKS modifications of nystatin is described. The aim of this work was to better understand the nystatin biosynthesis and to further use this information for generation of novel nystatin analogues. Two PKS-modifications of the nystatin molecule were targeted in this study: glycosylation with mycosamine at C-19 and oxidation of the exocyclic methyl group at C-16. Two genes putatively involved in mycosamine biosynthesis (NysDIII and NysDII) and one in attachment of mycosamine to the nystatin aglycone (nysDI) have been identified in the nystatin gene cluster. Their functions have been suggested, respectively, as a putative mannose dehydratase, aminotransferase and a glycosyltransferase. The deoxysugar mycosamine is proposed to have an important function for the activity of nystatin. To better understand the biosynthesis and importance of mycosamine and to perform modifications of nystatin via this post-PKS modifying step, the mycosamine biosynthesis was studied. The NysDIII protein was overexpressed in Escherichia coli and purified, and its in vitro mannose 4,6-dehydratase activity was confirmed. To study the function of nysDII and nysDI, the genes were individually deleted from the S. noursei chromosome. Both mutants were shown to produce a mixture of nystatinolide and 10-deoxynystatinolide, albeit at considerably different levels. Complementation experiments unequivocally confirmed the involvement of these two in mycosamine biosynthesis and attachement. Both antifungal and hemolytic activity of the purified nystatinolides were tested, and were found to be strongly reduced compared to nystatin, confirming the importance of the mycosamine moiety for the biological activity of nystatin. A gene for putative P450 monooxyganse NysN has been identified in the nystatin biosynthesis gene cluster. The function of NysN has been predicted to be oxidation of an exocyclic C16 methyl group on the nystatin molecule in order to afford a C16 carboxyl. The latter group has been implicated in selective toxicity of other polyene macrolides, and thus is considered an important target for manipulation. The nysNgene was inactivated in S. noursei by both in-frame deletion and site-specific mutagenesis, and the resulting mutants were shown to produce 16-decarboxy-16-methylnystatin, supporting the suggested biological role of NysN as C16 methyl oxidase. The recombinant NysN protein was also expressed in Escherichia coli, but its C16-methyl oxidase activity in vitro could not be demonstrated. 16-decarboxy-16-methylnystatin was purified from the nysN mutant, and its antifungal activity was identical with nystatin whereby the toxicit was reduced compared to nystatin. In the work of developing new methods for obtaining nystatin analogues, bioconversion of nystatinolide was performed as a means to modify nystatin aglycone. For this purpose a sub-library of 35 different Streptomyces strains isolated from the Trondheims fjord was selected. One strain was shown to be able to add a water molecule (presumed epoxidation) and another strain was able to chlorinate the nystatinolides. An attempt on alternative glycosylation of nystatinolide was performed by using glycosyltransferase hybrids and deoxysugar biosynthesis gene cassettes. However, these experiments did not afford novel nystatin analogues, suggesting strong preference of the NysDI glycosyltyransferase for its natural sugar substrate GDP-mycosamine