Molecular basis for short-chain thioester hydrolysis by acyl hydrolase domains in trans-acyltransferase polyketide synthases

Abstract

Polyketide synthases (PKSs) are multi-domain enzymatic assembly lines that biosynthesise a wide selection of bioactive natural products from simple building blocks. In contrast to their cis-acyltransferase (AT) counterparts, trans-AT PKSs rely on stand-alone AT domains to load extender units onto acyl carrier protein (ACP) domains embedded in the core PKS machinery. Trans-AT PKS gene clusters also encode acyl hydrolase (AH) domains, which are predicted to share the overall fold of AT domains, but hydrolyse aberrant acyl chains from ACP domains, thus ensuring efficient polyketide biosynthesis. How such domains specifically target short acyl chains, in particular acetyl groups, tethered as thioesters to the substrate-shuttling ACP domains, with hydrolytic rather than acyl transfer activity, has remained unclear. To answer these questions, we solved the first structure of an AH domain and performed structure-guided activity assays on active site variants. Our results offer key insights into chain length control and selection against coenzyme A-tethered substrates, and clarify how the interaction interface between AH and ACP domains contributes to recognition of cognate and non-cognate ACP domains. Combining our experimental findings with molecular dynamics simulations allowed for the production of a data-driven model of an AH:ACP domain complex. Our results advance the currently incomplete understanding of polyketide biosynthesis by trans-AT PKSs, and provide foundations for future bioengineering efforts.