One target of our research is a family of enzymes called non-ribosomal peptide synthetases (NRPSs). The NRPS enzymes catalyze the formation of small peptides. These peptides are important as some exhibit antibiotic or anti-cancer activities; other NRPS clusters are responsible for the synthesis of peptide siderophores, compounds that bind to iron with high affinity. The NRPSs are modular enzymes that, like fatty acid synthases and polyketide synthases, contain multiple catalytic enzyme activities joined in a single multi-domain protein that can be thousands of residues in length. The undecapeptide cyclosporin, for example, is synthesized by the 11-module NRPS cyclosporin synthetase CssA; the full-length protein is 15281 residues. Each module is responsible for the incorporation of a single residue of the final peptide. In the schematic below, the yellow domains represent adenylation domains that activate the substrate amino acid as a thioester on the end of the pantetheine cofactor on the red peptidyl carrier protein (PCP) domains. The blue domains represent condensation domains that catalyze the formation of peptide bonds between residues bound to adjacent PCP domains.
The NRPS adenylation domains are part of a larger family of Adenylate-Forming enzymes that we are studying structurally and functionally. The members of this family can be divided into three subfamilies.
Interestingly, all three subfamilies catalyze two-step chemical reactions. In the first step, the carboxylate reacts with ATP to form an acyl-adenylate. In the second step, the thioester is formed or, in the case of luciferase, the oxidation step is catalyzed. To better understand the function of the adenylate-forming enzymes, we have determined the structures of two members of the family: acetyl-CoA synthetase, and 4-chlorobenzoyl-CoA synthetase. These two enzymes were crystallized in different conformations. The structures of these proteins, as well as other adenylate-forming enzymes from the labs of Mohamed Marahiel, Peter Brick, Masashi Miyano, and Milton Stubbs, have provided structural evidence for the Domain Alternation hypothesis. In this view, the adenylate-forming enzymes adopt two dramatically different conformations to catalyze the two half-reactions.
The figure below shows the orientation of chlorobenzoyl-CoA Synthetase bound to chlorobenzoate (on the left) and acetyl-CoA synthetase bound to 5'-propyladenosine and CoA (on the right). The smaller C-terminal domain adopts two different conformations that result from a 140 degree rotation around the hinge at Asp402 (CBAL) or Asp517 (Acs).
We have examined this domain alternation hypothesis through a series of mutagenesis studies of both Acs and CBAL, as well as with the NRPS adenylation domain of the E. coli protein EntE. These studies have shown that mutations on one face of the mobile C-terminal domain specifically affect the adenylation half-reaction while mutations on the opposite face affect the thioester-forming half-reaction.
The implications of this domain movement are particularly important for the NRPS enzymes. If this movement occurs within the context of a large multi-domain protein such as that shown above, we are curious how the enzyme is able to accommodate this and avoid clashes of all neighboring domains.
To understand the interaction of the NRPS adenylation domain with the carrier protein(1,2), we have determined the structure of the EntE adenylation domain of E. coli in a complex with the EntB partner carrier protein. This structure was determined through the use of a mechanism-based inhibitor that trapped the interaction. We have used a similar strategy to crystallize a natural two-domain Adenylation-PCP protein, the PA1221 protein from P. aeruginosa.