Development of a biotechnological toolkit for the synthesis of diverse cyclic peptides

Abstract

Cyclic peptides possess desirable characteristics as potential pharmaceutical scaffolds. The cyanobactin family of cyclic peptide natural products boast diverse structures and bioactivity. Exemplars are the patellamides, which have attracted attention due to their ability to reverse the effects of multi-drug resistance in human leukemia cells. In addition to their macrocyclic architecture patellamides contain azol(in)e heterocycles and d-amino acids. This structural complexity makes them challenging targets for chemical synthesis. Understanding their biosynthesis will enable the development of a biotechnological ‘toolkit’ for the synthesis of new pharmaceutical compounds. Patellamides are ribosomally-synthesised and post-translationally modified peptides (RiPPs) and much of their biosynthesis has been elucidated, however there are still elements of their biosynthesis that are not yet fully understood. PatA and PatG contain C-terminal domains of unknown function (DUFs). The crystal structure of PatG-DUF has been solved and subsequent to biochemical and biophysical investigation PatG-DUF was found not to constitute an essential part of the biotechnological ‘toolkit’ and can be excluded from in vitro enzyme-based synthesis of cyanobactin-like cyclic peptides. The cyanobactin heterocyclases are able to introduce heterocycles into a peptide backbone, seemingly irrespective of the neighbouring residues; however a molecular rational governing substrate recognition is unknown. Additionally the mechanism of heterocyclisaton is disputed. Analysis of crystal structures of LynD in complex with cofactor and substrate (solved by Dr Jesko Koehnke) enabled the active site and substrate recognition site to be located. A new mechanism for heterocyclisation has been proposed. Guided by the substrate recognition observed in complex structures a constituently active heterocyclase (AcLynD) has been engineered, which is able to process short, leaderless peptide substrates. Epimerisation in cyanobactin biosynthesis is believed to be spontaneous, but its precise timing is uncertain. NMR analysis of selectively labelled peptide substrates processed by the modifying enzymes, identified epimerisation to be spontaneous on the macrocycle, regardless of whether the neighbouring heterocycles have been oxidised. A one-pot in vitro synthesis of cyanobactins has been developed, and employed to create a number of patellamide D analogues to ascertain structural-activity relationships.