Contributed by Wayne A. Hendrickson, October 15, 2018 (sent for review July 20, 2018; reviewed by Yifan Cheng and Michael C. Wiener)
Membrane proteins function naturally as imbedded in the lipid bilayers of cell membranes, but isolation into homogeneous and soluble preparations is needed for many biochemical studies. Detergents, which are used traditionally to extract and purify membrane proteins from cells, also remove most protein-associated lipid molecules as they disrupt the membranes. We have devised a detergent-free system to prepare native cell-membrane nanoparticles for biochemical analysis. In application to the membrane transporter AcrB, we demonstrate that these detergent-free nanoparticles are suitable for cryo-EM imaging at high resolution and that the natural lipid-bilayer structure so preserved is important for the functional integrity of AcrB. This nanoparticle system should be broadly applicable in membrane-protein research.
Membrane proteins function in native cell membranes, but extraction into isolated particles is needed for many biochemical and structural analyses. Commonly used detergent-extraction methods destroy naturally associated lipid bilayers. Here, we devised a detergent-free method for preparing cell-membrane nanoparticles to study the multidrug exporter AcrB, by cryo-EM at 3.2-Å resolution. We discovered a remarkably well-organized lipid-bilayer structure associated with transmembrane domains of the AcrB trimer. This bilayer patch comprises 24 lipid molecules; inner leaflet chains are packed in a hexagonal array, whereas the outer leaflet has highly irregular but ordered packing. Protein side chains interact with both leaflets and participate in the hexagonal pattern. We suggest that the lipid bilayer supports and harmonizes peristaltic motions through AcrB trimers. In AcrB D407A, a putative proton-relay mutant, lipid bilayer buttresses protein interactions lost in crystal structures after detergent-solubilization. Our detergent-free system preserves lipid–protein interactions for visualization and should be broadly applicable.
↵1W.Q. and Z.F. contributed equally to this work.
Author contributions: W.Q. and Y.G. designed research; W.Q., Z.F., G.G.X., and Y.G. performed research; Y.G. supervised all of the work; J.F. supervised the EM experiments; R.A.G. set up and maintained EM facilities; Y.Z. gave advice on chemical synthesis; W.Q., G.G.X., Y.Z., and Y.G. contributed new reagents/analytic tools; W.Q., Z.F., W.A.H., and Y.G. analyzed data; W.A.H. gave advice on structure analysis; W.Q., Z.F., G.G.X., R.A.G., Y.Z., J.F., W.A.H., and Y.G. wrote the paper.
Reviewers: Y.C., University of California, San Francisco; and M.C.W., University of Virginia.
The authors declare no conflict of interest.
Data deposition: Three-dimensional density maps and atomic models have been deposited in the Electron Microscopy Data Bank, www.ebi.ac.uk/pdbe/emdb [EMDB entry nos. EMD-7074 (www.ebi.ac.uk/pdbe/entry/emdb/EMD-7074; wild-type AcrB and lipid bilayer) and EMD-7609 (www.ebi.ac.uk/pdbe/entry/emdb/EMD-7609; AcrB D407A mutant and lipid bilayer)], and the Protein Data Bank, www.wwpdb.org [PDB ID codes 6BAJ (wild-type AcrB and lipid bilayer) and 6CSX (AcrB D407A mutant and lipid bilayer)].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1812526115/-/DCSupplemental.
Published under the PNAS license.
