Mechanistic insights into lipid A modification by the phosphoethanolamine transferase MCR-1

Project Summary Phosphoethanolamine (PEA) transferases are bacterial inner membrane proteins that catalyze the addition of PEA from phosphatidylethanolamine (PE) donor substrate to the 1- or 4’-phosphate groups of lipid A, as a result conferring resistance to a class of last-resort, positively charged antibiotic drugs known as polymyxins. Polymyxins treat multidrug-resistant Gram-negative bacterial infections by electrostatically interacting with the anionic lipid A anchor of lipopolysaccharide (LPS) molecules on the outer membrane (OM), in turn disrupting the membrane and leading to cell death. Hence, by cleaving and adding cationic PEA from PE to lipid A, PEA transferases reduce the overall negative charge on the OM and interfere with the ability of polymyxins to interact with lipid A. While most of these enzymes – including the lipid A PEA transferase (EptA) – are chromosomally encoded, the recent discovery of the first variant of mobilized colistin resistance (MCR-1) on plasmids in China entails that, for the first time, bacteria are capable of using horizontal transfer to confer plasmid-mediated polymyxin resistance between different strains, thus constituting a significant additional public health risk. Indeed, the mcr-1 gene has already been reported worldwide in livestock, food, and humans, underscoring the need to better understand the molecular mechanism by which PEA transferases generally and MCR-1 specifically modify lipid A to confer polymyxin resistance. The recently reported crystal structure of EptA from Neisseria meningitidis (NmEptA) has shown that PEA transferases exist as monomers comprised of distinct C-terminal soluble and N- terminal transmembrane (TM) domains with five TM helices. The presence of a small cavity in the NmEptA crystal structure harboring residues conserved among PEA transferases, and the absence of an obvious additional binding cavity, suggests that these enzymes may bind PE and lipid A sequentially, whereby a major conformational change between the soluble and TM domains enables the larger lipid A to bind in the same cavity as PE. However, the lack of a full-length, substrate (either PE, lipid A, or both)-bound structure in a native-like lipidic environment precludes a full understanding of this molecular mechanism. I hypothesize that such a structure would be able to validate the proposed mechanism-of-action and reveal the extent to which residues in the putative substrate binding cavity or elsewhere play a role in donor (PE) and acceptor (lipid A) substrate binding as well as coordination of the PEA intermediate. To address these questions, I will utilize structural and functional analyses of full-length MCR-1 in lipid-filled nanodiscs by single-particle cryo-electron microscopy (cryo-EM) in addition to well-established lipid A modification and polymyxin-resistance assays in whole cells. Two aims are proposed: (1) to determine the structure of substrate-bound MCR-1; and (2) to investigate the specific interactions of MCR-1 with PE and lipid A. Together, these data will provide molecular-level insight into the mechanism that MCR-1 uses to bind its substrates and coordinate and transfer PEA to modify lipid A, which in turn will be critical for drug design efforts to inhibit MCR-1 function and treat polymyxin resistance.