Bacterial genome mining for novel natural product discovery

Bioinformatics analyses have identified gene clusters encoding cryptic polyketide biosynthetic pathways, not associated with the production of known metabolites, in several actinomycete genome sequences. Discovery of the metabolic products of such cryptic gene clusters promises to unearth a hitherto untapped wealth of novel bioactive compounds. However, a major obstacle to the discovery of novel natural products by genome mining is that many cryptic pathways are expressed poorly or not at all under normal laboratory conditions. The discovery of the stambomycins, a new family of 51-membered macrolides with promising anti-cancer activity, as the products of a novel type I modular polyketide synthase (PKS) system identified in the partial genome sequence of Streptomyces ambofaciens will be described. Activation of expression of the pathway by genetic manipulation of a putative pathway-specific regulatory gene was key to this discovery. The structures of these novel macrolides suggests that their biosynthesis involves novel features, including in trans hydroxylation during polyketide chain assembly to provide the hydroxyl group required for offloading of the fully-assembled polyketide chain from the PKS via macrocyclisation. Experiments aimed at probing this unusual transformation will be described.

Identification of coelimycin P, a pigmented metabolic product of a type I modular PKS system encoded by a cryptic gene cluster within the Streptomyces coelicolor genome, by exploiting a genetic engineering strategy aimed at maximising metabolic flux through the pathway will also be described. The structure of the pigment, coupled with bioinformatics analyses of the enzymes encoded by the cryptic gene cluster, suggests an unusual biosynthetic pathway with numerous novel features. Incorporation experiments with labelled precursors and biochemical experiments with purified recombinant enzymes provide experimental support for the proposed biosynthetic pathway.