Elucidation and control of substrate recognition during RiPP biosynthesis

Ribosomally synthesized and posttranslationally modified peptides (RiPPs) are a rapidly growing class of natural products. RiPP precursor peptides can undergo extensive enzymatic tailoring to yield structurally and functionally diverse products. Cyclodehydratases are a type of RiPP modifying enzyme that catalyze phosphorylation of the peptide backbone and subsequent nucleophilic attack by the sidechains of Cys, Ser, or Thr to form azoline heterocycless (or azoles upon oxidation). The catalytic unit of the cyclodehydratase is a YcaO-family protein which is often accompanied by a partner protein from the E1-like superfamily (depending on the of type cyclodehydratase). Although primarily known for azoline formation, recent work suggests that YcaO proteins can use different nucleophiles and partner proteins to generate thioamide, macroamidine, and possibly other peptide posttranslational modifications. In Chapter 1, I comprehensively review the biosynthetic gene clusters (BGCs), natural products, functions, mechanisms, and applications of YcaO proteins and outline future directions for this protein superfamily. In Chapter 2, I report my investigations into the substrate recognition of canonical cyclodehydratases. Recent work suggested that unrelated RiPP modifying enzymes contain structurally similar precursor peptide binding domains. Using profile hidden Markov model comparisons, I discovered related and previously unrecognized peptide binding domains in proteins spanning the majority of known prokaryotic RiPP classes. This conserved domain was designated the RiPP precursor peptide recognition element (RRE). Through binding studies, I verified the RRE's role for three distinct RiPP classes: linear azole-containing peptides, thiopeptides, and lasso peptides. Because numerous RiPP biosynthetic enzymes act on peptide substrates, these findings have powerful predictive value as to which protein(s) drive substrate binding, thereby laying a foundation for further characterization of RiPP biosynthetic pathways and the rational engineering of new peptide binding activities. In Chapter 3, I use knowledge gained from precursor peptide binding studies to engineer a peptide that can be recognized and modified by two biosynthetic enzymes from different pathways. Combining enzymes from multiple pathways is an attractive approach for producing molecules with desired structural features, but this strategy thus far has been hampered by limited substrate tolerance of enzymes from unrelated pathways. Because RiPP biosynthetic enzymes modify their substrates by binding motifs located usually in the N-terminal leader region of precursor peptides, RiPP biosynthetic systems are highly amenable to the engineering of new compounds. I exploit this by designing chimeric leader peptides that can be bound and processed by multiple enzymes from unrelated RiPP pathways. Using this broadly applicable strategy, a cyclodehydratase was combined with enzymes from the sactipeptide and lanthipeptide RiPP classes to create new-to-nature hybrid RiPPs. These hybrids provide insight into biosynthetic timing and enzyme compatibility and establish a general platform for the engineering additional hybrid RiPPs