Professor Douglas Mitchell is the recipient of the 2015 Pfizer Award in Enzyme Chemistry. The Mitchell laboratory has a long-standing interest in the biosynthetic construction, structure, and mechanism of action of complex molecules. Our interest in this area derives from the fact that most of today’s medicines can trace their origins to natural products. Not only do architecturally complex “small” molecules from nature provide challenging targets for synthetic chemists, sometimes even inspiring the development of new reaction methodologies, the study of the naturally occurring route often uncovers novel biosynthetic enzymology. Indeed, the study of natural products continues to stimulate significant advances at the chemistry-biology-medicine interface.
One focus of our group has been to characterize a class of ribosomal peptide-derived natural products adorned with thiazole and oxazole heterocycles. The biosynthetic gene clusters for such compounds ubiquitously encode a two-domain cyclodehydratase that converts Cys and Ser/Thr residues to thiazolines and (methyl)oxazolines. Subsequently, these heterocycles sometimes undergo further processing by a FMN-dependent dehydrogenase to yield the corresponding azole heterocycle. It has long been known that the heterocycle-forming cyclodehydratase required ATP, but pinpointing the mechanism of catalysis and ATP usage presented significant unresolved challenges. The N-terminal domain (C-protein) of the cyclodehydratase is homologous to known ATP-utilizing enzymes, thus it was believed to overcome the energetically uphill reaction by coupling cyclodehydration with the hydrolysis of ATP. The C-terminal domain (D-protein) belongs to the cryptically named “YcaO” superfamily, of which there were no characterized functions. Thus, the D-protein was thought to play a regulatory or scaffolding role. Our group overturned these functional assignments by studying a heterocycle synthetase where the activities of the C- and D-proteins could be separated. The D-protein (YcaO) was found to be a catalytically competent, ATP-dependent cyclodehydratase in the absence of the C-protein. By bringing the peptide substrate in close proximity, the C-protein potentiated the activity of the D-protein by nearly 1000-fold. Using a series of isotope-labeling experiments, we determined that the D-protein utilizes ATP to directly activate the peptide backbone via a phosphorylated hemiorthoamide intermediate. Thus, the leaving group during cyclodehydration is phosphate, not water. We since have crystallized an uncharacterized YcaO protein from E. coli in collaboration with Prof. Satish Nair’s group in various nucleotide-bound and -free forms. The most conserved residues in the superfamily comprise a novel ATP-binding motif, which gratifying explained the earlier inability to bioinformatically identify this site. Our results strongly support a model where ATP utilization is a universal feature of all YcaOs, which sheds light on the probable enzymatic functions of many thousands of additional uncharacterized proteins. A minor fraction of these are expected to be involved in synthesizing heterocycles on ribosomal peptide scaffolds; the majority is expected to use ATP for novel biochemical reactions that require amide backbone activation.