Bioinformatics guided synthesis of natural product cyclic peptides

Natural products are a fantastic source of bioactive molecules but are generally challenging to discover due to the rediscovery of known molecules and silent (i.e. transcriptionally inactive) biosynthetic gene clusters. A recently developed method to overcome this issue is the synthetic-bioinformatic natural product (syn-BNP) approach. In this approach, bioinformatics programs are used to predict natural product structures from cryptic biosynthetic gene clusters followed by chemical synthesis to access these otherwise challenging to acquire molecules. This enables access to natural products, or closely structurally related derivatives, from silent biosynthetic gene clusters or strains that are currently not culturable for biological activation. While this approach has been employed by a handful of laboratories, the bioinformatics pipeline can be challenging to scale due to the need to interface with multiple programs. Presented here is the development of a bioinformatics pipeline B-LinESS (Biosynthesis Linker: from Enzymes to Structures and Synthons), which enables more scalable identification of predicted structures and synthons for natural products based on the shared presence of certain genes of interest. Additionally, we describe methods to chemically synthesize the predicted cyclic peptide natural products. The chemical synthesis allows scalable access to molecules that can then be tested in a variety of bioactivity assays.