Natural products constitute one of the richest sources and inspiration for novel pharmaceutical drugs with unprecedented structures and diverse bioactivities. The enzymes responsible for the biosynthesis of these compounds are encoded for by a group of genes organized in a cluster, called biosynthesis gene clusters (BGCs). Conventionally, the isolation of bioactive compounds is performed through cultivation, chemical extraction, and ultimately, bioactivity-guided fractionation and screening methods. However, this method is laborious, has led to the re-isolation of previously characterized compounds, and prevents exploration of bacteria that cannot be fermented under laboratory conditions. Genome mining has revealed that microorganisms have the capacity to produce more compounds than wha...[ Read more ]

Natural products constitute one of the richest sources and inspiration for novel pharmaceutical drugs with unprecedented structures and diverse bioactivities. The enzymes responsible for the biosynthesis of these compounds are encoded for by a group of genes organized in a cluster, called biosynthesis gene clusters (BGCs). Conventionally, the isolation of bioactive compounds is performed through cultivation, chemical extraction, and ultimately, bioactivity-guided fractionation and screening methods. However, this method is laborious, has led to the re-isolation of previously characterized compounds, and prevents exploration of bacteria that cannot be fermented under laboratory conditions. Genome mining has revealed that microorganisms have the capacity to produce more compounds than what is detected from traditional compound isolation methods. Large scale genome mining analysis of bacterial genomes indicate that there are hundreds of thousands of cryptic BGCs awaiting to be explored, and the challenge lies on which BGCs should be prioritized for compound production. In the present thesis, three global genome mining and networking strategies were developed and employed for BGC prioritization. First, a resistance-gene targeted genome mining approach has led to the selection of eight BGCs containing duplicated, resistant copies of the gyrB gene, that putatively produce antibiotics that target DNA gyrase B. The diversity of thioamidated peptides was also described in an in silico approach which included BGC networking, phylogenetic, and motif sequence analyses. Precursor peptide analysis displayed the vast uncovered chemical space encoded by this family of compounds. Lastly, a structure-guided genome mining approach by global analysis of cytochrome P450-associated cyclodipeptide synthase BGCs was performed. This has led to the discovery of cyctetryptomycins, a novel family of cyclodipeptides with an unprecedented large macrocyclic core and promising neuroprotective bioactivity. These results show the utility of applying different global genome mining approaches coupled with networking analyses as an effective way to prioritize BGCs for novel secondary metabolite isolation.