Polyene macrolide biosynthesis in streptomycetes and related bacteria: recent advances from genome sequencing and experimental studies

The polyene macrolide group includes important antifungal drugs, to which resistance does not arise readily. Chemical and biological methods have been used in attempts to make polyene antibiotics with fewer toxic side effects. Genome sequencing of producer organisms is contributing to this endeavour, by providing access to new compounds and by enabling yield improvement for polyene analogues obtained by engineered biosynthesis. This recent work is also enhancing bioinformatic methods for deducing the structures of cryptic natural products from their biosynthetic enzymes. The stereostructure of candicidin D has recently been determined by NMR spectroscopy. Genes for the corresponding polyketide synthase have been uncovered in several different genomes. Analysis of this new information strengthens the view that protein sequence motifs can be used to predict double bond geometry in many polyketides. Chemical studies have shown that improved polyenes can be obtained by modifying the mycosamine sugar that is common to most of these compounds. Glycoengineered analogues might be produced by biosynthetic methods, but polyene glycosyltransferases show little tolerance for donors other than GDP-α-D-mycosamine. Genome sequencing has revealed extending glycosyltransferases that add a second sugar to the mycosamine of some polyenes. NppY of Pseudonocardia autotrophica uses UDP-N-acetyl-α-D-glucosamine as donor whereas PegA from Actinoplanes caeruleus uses GDP-α-D-mannose. These two enzymes show 51 % sequence identity and are also closely related to mycosaminyltransferases. These findings will assist attempts to construct glycosyltransferases that transfer alternative UDP- or (d)TDP-linked sugars to polyene macrolactones.