Edited by Janet L. Smith, University of Michigan, Ann Arbor, MI, and accepted by Editorial Board Member Michael A. Marletta September 16, 2016 (received for review May 27, 2016)
Here we investigate the structural basis for cyclization activity in hybrid polyketide synthase-nonribosomal peptide synthetases. This first structure of a cyclization (Cy) domain reveals an unexpected location for the enzyme active site, providing a fresh perspective on past mutational studies. Our structures also depict two 20-Å-long channels that create routes for the two tethered substrates to simultaneously reach the buried active site, affording substrate condensation and cyclization. Along with the Cy domain, these structures contain a covalently attached docking domain, providing insight into how protein modules work together to achieve uni-directionality in the biosynthesis of natural products.
Epothilones are thiazole-containing natural products with anticancer activity that are biosynthesized by polyketide synthase (PKS)-nonribosomal peptide synthetase (NRPS) enzymes EpoA–F. A cyclization domain of EpoB (Cy) assembles the thiazole functionality from an acetyl group and l-cysteine via condensation, cyclization, and dehydration. The PKS carrier protein of EpoA contributes the acetyl moiety, guided by a docking domain, whereas an NRPS EpoB carrier protein contributes l-cysteine. To visualize the structure of a cyclization domain with an accompanying docking domain, we solved a 2.03-Å resolution structure of this bidomain EpoB unit, comprising residues M1-Q497 (62 kDa) of the 160-kDa EpoB protein. We find that the N-terminal docking domain is connected to the V-shaped Cy domain by a 20-residue linker but otherwise makes no contacts to Cy. Molecular dynamic simulations and additional crystal structures reveal a high degree of flexibility for this docking domain, emphasizing the modular nature of the components of PKS-NRPS hybrid systems. These structures further reveal two 20-Å-long channels that run from distant sites on the Cy domain to the active site at the core of the enzyme, allowing two carrier proteins to dock with Cy and deliver their substrates simultaneously. Through mutagenesis and activity assays, catalytic residues N335 and D449 have been identified. Surprisingly, these residues do not map to the location of the conserved HHxxxDG motif in the structurally homologous NRPS condensation (C) domain. Thus, although both C and Cy domains have the same basic fold, their active sites appear distinct.
↵1Present address: Department of Chemistry, University of Massachusetts, Boston, MA 02125.
↵3Present address: Department of Chemistry, Bryn Mawr College, Bryn Mawr, PA 19010.
↵4Present address: School of Chemistry and Biochemistry and the Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, GA 30332.
Author contributions: D.P.D., Y.K., A.K.C., K.T., and C.L.D. designed research; D.P.D., Y.K., A.K.C., and K.T. performed research; W.L.K. and C.T.W. contributed new reagents/analytic tools; D.P.D., Y.K., A.K.C., K.T., and C.L.D. analyzed data; and D.P.D. and C.L.D. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission. J.L.S. is a Guest Editor invited by the Editorial Board.
Data deposition: The atomic coordinates and structure factors have been deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 5T81 and 5T7Z).
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1608615113/-/DCSupplemental.
