Genome Mining Approaches for the Discovery of Novel Lasso Peptides

Abstract

Lasso peptides are a class of ribosomally-synthesized, post-translationally modified natural products made by a variety of species of bacteria. They are characterized by an extraordinarily stable slipknot-like structure, which, along with their diverse range of known functions, make them strong candidates for peptide therapeutics. Because of their resistance to prote-olytic degradation, they are ideal for use as scaffolds for peptide epitopes. Genome mining has previously been shown to be effective in identifying novel lasso peptides, diversifying the collection of potential therapeutic molecules and expanding the library of scaffolds for engineering. Further exploration of mining approaches allows us to continue this discovery as well as determine whether past methods are sensitive, specific, and efficient in their analysis. We build on previous work by exploring three methods for querying full bacterial genome sequences for putative lasso peptide gene clusters. These methods are based upon looking for homology in three different features of the clusters – leader peptides, precursors, and maturation enzymes. We created a website to display results from genome mining studies and further analysis in a graphically accessible way in order to simplify the identification of promising candidate peptides. In order to validate the results of our genome mining studies and to determine potential practical applications, we have also isolated and characterized two lasso peptides, benonodin-1 and benenodin-2, found in the genome of As- ticcacaulis benevestitus. We have determined an NMR solution structure for benenodin-1, revealing a salt bridge that likely contributes to the thermostability of the peptide.