Lipopolysaccharide (LPS) resides in the outer membrane of Gram-negative bacteria where it is responsible for barrier function1,2. LPS can cause death as a result of septic shock, and its lipid A core is the target of polymyxin antibiotics3,4. Despite the clinical importance of polymyxins and the emergence of multidrug resistant strains5, our understanding of the bacterial factors that regulate LPS biogenesis is incomplete. Here we characterize the inner membrane protein PbgA and report that its depletion attenuates the virulence of Escherichia coli by reducing levels of LPS and outer membrane integrity. In contrast to previous claims that PbgA functions as a cardiolipin transporter6,7,8,9, our structural analyses and physiological studies identify a lipid A-binding motif along the periplasmic leaflet of the inner membrane. Synthetic PbgA-derived peptides selectively bind to LPS in vitro and inhibit the growth of diverse Gram-negative bacteria, including polymyxin-resistant strains. Proteomic, genetic and pharmacological experiments uncover a model in which direct periplasmic sensing of LPS by PbgA coordinates the biosynthesis of lipid A by regulating the stability of LpxC, a key cytoplasmic biosynthetic enzyme10,11,12. In summary, we find that PbgA has an unexpected but essential role in the regulation of LPS biogenesis, presents a new structural basis for the selective recognition of lipids, and provides opportunities for future antibiotic discovery.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Structural data are deposited in the Protein Data Bank (PDB) under accession number 6XLP. All mass spectrometry RAW files were uploaded to the MassIVE data repository, accessible by the identifier MSV000083754, and can be downloaded from ftp://MSV000083754@massive.ucsd.edu. DNA sequencing data were deposited at NCBI under BioProject PRJNA541088, BioSample SAMN11572257, experiment SRX5788703, run SRR9010525. The E. coli CFT073 reference genome was deposited at NCBI under BioProject PRJNA624646, BioSample SAMN14575425, accession CP051263. Source data are provided with this paper.
Whitfield, C. & Trent, M. S. Biosynthesis and export of bacterial lipopolysaccharides. Annu. Rev. Biochem. 83, 99–128 (2014).
Shrivastava, R. & Chng, S. S. Lipid trafficking across the Gram-negative cell envelope. J. Biol. Chem. 294, 14175–14184 (2019).
Parrillo, J. E. Pathogenetic mechanisms of septic shock. N. Engl. J. Med. 328, 1471–1477 (1993).
Pristovsek, P. & Kidric, J. Solution structure of polymyxins B and E and effect of binding to lipopolysaccharide: an NMR and molecular modeling study. J. Med. Chem. 42, 4604–4613 (1999).
Poirel, L., Jayol, A. & Nordmann, P. Polymyxins: antibacterial activity, susceptibility testing, and resistance mechanisms encoded by plasmids or chromosomes. Clin. Microbiol. Rev. 30, 557–596 (2017).
Dalebroux, Z. D. et al. Delivery of cardiolipins to the Salmonella outer membrane is necessary for survival within host tissues and virulence. Cell Host Microbe 17, 441–451 (2015).
Fan, J., Petersen, E. M., Hinds, T. R., Zheng, N. & Miller, S. I. Structure of an inner membrane protein required for PhoPQ-regulated increases in outer membrane cardiolipin. MBio 11, e03277-19 (2020).
Dong, H. et al. Structural insights into cardiolipin transfer from the Inner membrane to the outer membrane by PbgA in Gram-negative bacteria. Sci. Rep. 6, 30815 (2016).
Rossi, R. M., Yum, L., Agaisse, H. & Payne, S. M. Cardiolipin synthesis and outer membrane localization are required for Shigella flexneri virulence. MBio 8, e01199-17 (2017).
Guest, R. L., Samé Guerra, D., Wissler, M., Grimm, J. & Silhavy, T. J. YejM Modulates activity of the YciM/FtsH protease complex to prevent lethal accumulation of lipopolysaccharide. MBio 11, e00598-20 (2020).
Fivenson, E. M. & Bernhardt, T. G. An essential membrane protein modulates the proteolysis of LpxC to control lipopolysaccharide synthesis in Escherichia coli. MBio 11, e00939-20 (2020).
Nguyen, D., Kelly, K., Qiu, N. & Misra, R. YejM controls LpxC levels by regulating protease activity of the FtsH/YciM complex of Escherichia coli. J. Bacteriol.JB. 00303-20 (2020).
Sorensen, P. G. et al. Regulation of UDP-3-O-[R-3-hydroxymyristoyl]-N-acetylglucosamine deacetylase in Escherichia coli. The second enzymatic step of lipid a biosynthesis. J. Biol. Chem. 271, 25898–25905 (1996).
Gabale, U., Palomino, P. A. P., Kim, H., Chen, W. & Ressl, S. New functional identity of the essential inner membrane protein YejM: the cardiolipin translocator is also a metalloenzyme. Preprint at https://www.biorxiv.org/content/10.1101/2020.02.13.947838v1 (2020).
Hirvas, L., Nurminen, M., Helander, I. M., Vuorio, R. & Vaara, M. The lipid A biosynthesis deficiency of the Escherichia coli antibiotic-supersensitive mutant LH530 is suppressed by a novel locus, ORF195. Microbiology 143, 73–81 (1997).
Nurminen, M., Hirvas, L. & Vaara, M. The outer membrane of lipid A-deficient Escherichia coli mutant LH530 has reduced levels of OmpF and leaks periplasmic enzymes. Microbiology 143, 1533–1537 (1997).
Cian, M. B., Giordano, N. P., Masilamani, R., Minor, K. E. & Dalebroux, Z. D. Salmonella enterica Serovar Typhimurium uses PbgA/YejM to regulate lipopolysaccharide assembly during bacteremia. Infect. Immun. 88, e00758-19 (2019).
De Lay, N. R. & Cronan, J. E. Genetic interaction between the Escherichia coli AcpT phosphopantetheinyl transferase and the YejM inner membrane protein. Genetics 178, 1327–1337 (2008).
Jia, W. et al. Lipid trafficking controls endotoxin acylation in outer membranes of Escherichia coli. J. Biol. Chem. 279, 44966–44975 (2004).
Qiu, N. & Misra, R. Overcoming iron deficiency of an Escherichia coli tonB mutant by increasing outer membrane permeability. J. Bacteriol. 201, e00340-19 (2019).
Fronzes, R. et al. Structure of a type IV secretion system core complex. Science 323, 266–268 (2009).
Lu, D. et al. Structure-based mechanism of lipoteichoic acid synthesis by Staphylococcus aureus LtaS. Proc. Natl Acad. Sci. USA 106, 1584–1589 (2009).
Anandan, A. et al. Structure of a lipid A phosphoethanolamine transferase suggests how conformational changes govern substrate binding. Proc. Natl Acad. Sci. USA 114, 2218–2223 (2017).
Ogura, T. et al. Balanced biosynthesis of major membrane components through regulated degradation of the committed enzyme of lipid A biosynthesis by the AAA protease FtsH (HflB) in Escherichia coli. Mol. Microbiol. 31, 833–844 (1999).
Yoshimura, M., Oshima, T. & Ogasawara, N. Involvement of the YneS/YgiH and PlsX proteins in phospholipid biosynthesis in both Bacillus subtilis and Escherichia coli. BMC Microbiol. 7, 69 (2007).
Klein, G., Kobylak, N., Lindner, B., Stupak, A. & Raina, S. Assembly of lipopolysaccharide in Escherichia coli requires the essential LapB heat shock protein. J. Biol. Chem. 289, 14829–14853 (2014).
Mahalakshmi, S., Sunayana, M. R., SaiSree, L. & Reddy, M. yciM is an essential gene required for regulation of lipopolysaccharide synthesis in Escherichia coli. Mol. Microbiol. 91, 145–157 (2014).
Nicolaes, V. et al. Insights into the function of YciM, a heat shock membrane protein required to maintain envelope integrity in Escherichia coli. J. Bacteriol. 196, 300–309 (2014).
Ho, H. et al. Structural basis for dual-mode inhibition of the ABC transporter MsbA. Nature 557, 196–201 (2018).
Wu, T. et al. Identification of a protein complex that assembles lipopolysaccharide in the outer membrane of Escherichia coli. Proc. Natl Acad. Sci. USA 103, 11754–11759 (2006).
Lemmon, M. A. Membrane recognition by phospholipid-binding domains. Nat. Rev. Mol. Cell Biol. 9, 99–111 (2008).
Park, B. S. et al. The structural basis of lipopolysaccharide recognition by the TLR4–MD-2 complex. Nature 458, 1191–1195 (2009).
Li, Y., Orlando, B. J. & Liao, M. Structural basis of lipopolysaccharide extraction by the LptB2FGC complex. Nature 567, 486–490 (2019).
Ferguson, A. D., Hofmann, E., Coulton, J. W., Diederichs, K. & Welte, W. Siderophore-mediated iron transport: crystal structure of FhuA with bound lipopolysaccharide. Science 282, 2215–2220 (1998).
Arunmanee, W. et al. Gram-negative trimeric porins have specific LPS binding sites that are essential for porin biogenesis. Proc. Natl Acad. Sci. USA 113, E5034–E5043 (2016).
Datsenko, K. A. & Wanner, B. L. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl Acad. Sci. USA 97, 6640–6645 (2000).
Chan, W. et al. A recombineering based approach for high-throughput conditional knockout targeting vector construction. Nucleic Acids Res. 35, e64 (2007).
Lawrence, M. et al. Software for computing and annotating genomic ranges. PLOS Comput. Biol. 9, e1003118 (2013).
Diederich, L., Rasmussen, L. J. & Messer, W. New cloning vectors for integration in the lambda attachment site attB of the Escherichia coli chromosome. Plasmid 28, 14–24 (1992).
Baba, T. et al. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2, 2006.0008 (2006).
Miller, J. H. Experiments in Molecular Genetics (Cold Spring Harbor Laboratory, 1972).
Mohammad, M. M., Howard, K. R. & Movileanu, L. Redesign of a plugged beta-barrel membrane protein. J. Biol. Chem. 286, 8000–8013 (2011).
Kitagawa, M. et al. Complete set of ORF clones of Escherichia coli ASKA library (a complete set of E. coli K-12 ORF archive): unique resources for biological research. DNA Res. 12, 291–299 (2005).
Alexander, M. K. et al. Disrupting Gram-negative bacterial outer membrane biosynthesis through inhibition of the lipopolysaccharide transporter MsbA. Antimicrob. Agents Chemother. 62, e01142-18 (2018).
Karimova, G., Pidoux, J., Ullmann, A. & Ladant, D. A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc. Natl Acad. Sci. USA 95, 5752–5756 (1998).
Ladant, D. & Ullmann, A. Bordatella pertussis adenylate cyclase: a toxin with multiple talents. Trends Microbiol. 7, 172–176 (1999).
Bligh, E. G. & Dyer, W. J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917 (1959).
Caffrey, M. & Cherezov, V. Crystallizing membrane proteins using lipidic mesophases. Nat. Protoc. 4, 706–731 (2009).
Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326 (1997).
Strong, M. et al. Toward the structural genomics of complexes: crystal structure of a PE/PPE protein complex from Mycobacterium tuberculosis. Proc. Natl Acad. Sci. USA 103, 8060–8065 (2006).
McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007).
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010).
Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D 66, 213–221 (2010).
Winn, M. D., Isupov, M. N. & Murshudov, G. N. Use of TLS parameters to model anisotropic displacements in macromolecular refinement. Acta Crystallogr. D 57, 122–133 (2001).
Ashkenazy, H. et al. ConSurf 2016: an improved methodology to estimate and visualize evolutionary conservation in macromolecules. Nucleic Acids Res. 44 (W1), W344–W350 (2016).
Holm, L. & Laakso, L. M. Dali server update. Nucleic Acids Res. 44 (W1), W351–W355 (2016).
The PyMOL Molecular Graphics System, version 2.0 Schrödinger, LLS.
Liu, W., Ishchenko, A. & Cherezov, V. Preparation of microcrystals in lipidic cubic phase for serial femtosecond crystallography. Nat. Protoc. 9, 2123–2134 (2014).
Barty, A. et al. Cheetah: software for high-throughput reduction and analysis of serial femtosecond X-ray diffraction data. J. Appl. Crystallogr. 47, 1118–1131 (2014).
White, T. A. et al. CrystFEL: a software suite for snapshot serial crystallography. J. Appl. Cryst. 45, 335–341 (2012).
White, T. A. et al. Recent developments in CrystFEL. J. Appl. Crystallogr. 49, 680–689 (2016).
Sastry, G. M., Adzhigirey, M., Day, T., Annabhimoju, R. & Sherman, W. Protein and ligand preparation: parameters, protocols, and influence on virtual screening enrichments. J. Comput. Aided Mol. Des. 27, 221–234 (2013).
Schrödinger Release 2017-3: Schrödinger Suite 2017-3 Protein Preparation Wizard (New York, 2017).
Schrödinger Release 2017-3 (New York, 2017).
Shivakumar, D. et al. Prediction of absolute solvation free energies using molecular dynamics free energy perturbation and the OPLS force field. J. Chem. Theory Comput. 6, 1509–1519 (2010).
Guo, Z. et al. Probing the α-helical structural stability of stapled p53 peptides: molecular dynamics simulations and analysis. Chem. Biol. Drug Des. 75, 348–359 (2010).
Bowers, K. J. et al. Scalable algorithms for molecular dynamics simulations on commodity clusters. In SC '06: Proc. 2006 ACM/IEEE Conference on Supercomputing, 43–43) (Tampa, FL, 2006).
Hankins, J. V., Madsen, J. A., Needham, B. D., Brodbelt, J. S. & Trent, M. S. The outer membrane of Gram-negative bacteria: lipid A isolation and characterization. Methods Mol. Biol. 966, 239–258 (2013).
Choi, H. et al. SAINT: probabilistic scoring of affinity purification-mass spectrometry data. Nat. Methods 8, 70–73 (2011).
Berman, H. M. et al. The Protein Data Bank. Nucleic Acids Res. 28, 235–242 (2000).
Ma, G., Zhu, Y., Yu, Z., Ahmad, A. & Zhang, H. High resolution crystal structure of the catalytic domain of MCR-1. Sci. Rep. 6, 39540 (2016).
Mi, W. et al. Structural basis of MsbA-mediated lipopolysaccharide transport. Nature 549, 233–237 (2017).
Noland, C. L. et al. Structural insights into lipoprotein N-acylation by Escherichia coli apolipoprotein N-acyltransferase. Proc. Natl Acad. Sci. USA 114, E6044–E6053 (2017).
Vilar, S., Cozza, G. & Moro, S. Medicinal chemistry and the molecular operating environment (MOE): application of QSAR and molecular docking to drug discovery. Curr. Top. Med. Chem. 8, 1555–1572 (2008).
Owens, T. W. et al. Structural basis of unidirectional export of lipopolysaccharide to the cell surface. Nature 567, 550–553 (2019).
Ohto, U., Fukase, K., Miyake, K. & Shimizu, T. Structural basis of species-specific endotoxin sensing by innate immune receptor TLR4/MD-2. Proc. Natl Acad. Sci. USA 109, 7421–7426 (2012).
We thank our Genentech colleagues for their support, in particular, A. Song, I. Kekessie, J. Toms, E. Hecht, C. Peng, A. Liu, P. Smith, A. Estevez, C. Ciferri, H. Ho, E. Castellanos, A. K. Katakam, I. Zilberleyb, M. Reichelt, M.-W. Tan, J. Kiefer, Y. Franke, C. Koth, E. Brown and S. Hymowitz. We thank D. Cawley for antibody generation and Smartox Biotechnology for peptide synthesis. Use of the Linac Coherent Light Source (LCLS) and the Stanford Synchrotron Radiation Lightsource (SSRL), SLAC National Accelerator Laboratory, are supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. The SSRL Structural Molecular Biology Program is supported by the DOE Office of Biological and Environmental Research, and by the National Institutes of Health, National Institute of General Medical Sciences (P41GM103393). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of NIGMS or NIH. C.G. appreciates support from the SLAC National Accelerator Laboratory and Stanford University through a Panofsky Fellowship. All reagents are available under a materials transfer agreement with Genentech.
All authors, except D.P.D., M.S.H. and C.G., are or were employees of Genentech/Roche.
Peer review information Nature thanks Bert van den Berg, Russell Bishop, Changjiang Dong and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
a, CFUs recovered from UPEC and UPEC ΔpbgA in neutropenic mouse tissues after intravenous injection of BALB/C mice 0.5 and 24 h after injection (n = 5 per group). Data are mean ± s.d. with dashed line indicating lower boundary of detection. b, Rifampicin sensitivity assay with conditional E. coli K-12 ΔpbgA::pBADpbgA strain. Data are mean ± s.d. for at each rifampicin concentration for n = 3 of each strain. c, Rifampicin sensitivity assay with E. coli K-12 and ΔclsABC strains. Data are mean ± s.d. for each rifampicin concentration for n = 3 of each strain. d, Quantification of lipid A and cardiolipin measured by MALDI–TOF and Qtrap liquid chromatography–tandem mass spectrometry (LC–MS/MS), respectively, normalized to total protein amounts in whole cells (left and middle) or outer membrane vesicles (right). AUC, area under the curve. Data are mean ± s.d. for each strain for n = 3 replicates. e, MALDI–TOF mass spectrometry analyses detected no cardiolipin in the ΔclsABC strain (orange) compared to the E. coli K-12 strain (black) when analysed under matched conditions. Representative results are shown. Source data
a, E. coli and S. typhimurium PbgA were purified in the mild detergent dodecylmaltoside and analysed by SEC-MALS. b, Thermostability of purified E. coli PbgA was analysed by differential scanning calorimetry with or without 0.1 mg ml−1 lipid supplementation. c, Left, from PbgA crystalized in space group C2, using data to 2.0 Å, an Fo − Fc map calculated shows bilobal extra electron density along the periplasmic membrane leaflet before the inclusion of LPS into models, 3σ contour. Inset, close-up view of an Fo − Fc map calculated before the inclusion of LPS into the model, rendered at 8σ (yellow) and 2σ (blue), respectively. Final refined coordinates of lipid A are shown for reference. Right, from PbgA crystalized in space group P31, using data to 4.6 Å, an Fo − Fc map calculated before the inclusion of LPS into the model, contoured at 3σ. d, Representative non-protein densities observed surrounding the TMD of PbgA that were assigned as putative phosphatidylethanolamine or monoolein lipids; inset shows Fo − Fc maps calculated before the inclusion of phosphatidylethanolamine or monoolein into the model, 2σ contour (phosphatidylethanolamine, orange; monoolein, blue). e, Schematic illustration of the inter-domain surface area contacts within PbgA. f, Close-up view highlighting the interaction of the Arg215 side chain with a conserved acidic residue, Asp192 on TM5, which appears to stabilize the IFD–TMD interface.
a, Structural superposition of PbgA crystal structures determined in the present study (space group C2 and P31) and both chain A and chain B from PDB code 6V8Q. The overall root mean square deviation for main chain atoms between the most divergent structures is <0.8 Å. b, Molecular dynamics study of PbgA, results (top) and experiments (bottom) are summarized by illustration. Simulations were performed following preparation of the 2.0 Å PbgA crystal structure and its placement into a phosphatidylethanolamine: phosphatidylglycerol mixed membrane bilayer, as described in Methods. Top, superimposed are coordinates from the last frames of the four molecular dynamics simulation runs with the starting (non-relaxed) X-ray model to compare the extent of domain movements. c, Views of the previously proposed cardiolipin-binding site8 are shown on the right. Residues proposed to be involved in cardiolipin binding are shown as orange sticks, but are seen here to form an integral part of the hydrophobic protein core; furthermore, the periplasmic domain of PbgA contains no recognizable sequence or structural homology to previously established lipid binding modules56,70. d, Structure-based alignment of the hydrolase superfamily domains from PbgA (periplasmic domain, green), S. aureus LtaS22 (ECD, blue) and E. coli phosphoethanolamine transferase MCR-171 (periplasmic domain, purple). e, Structure-based alignment of PbgA and EptA isolated periplasmic domains (left) and TMDs (right), respectively.
a, Calculated using data to 2.0 Å, an Fo − Fc map near the α7 helix of the IFD (pink) before inclusion of any ligand into refinement, 2σ contour (green). LPS refines well into this electron density whereas cardiolipin does not (see Extended Data Fig. 2c). Modelling and crystallographic refinement was pursued for cardiolipin, phosphatidylethanolamine, phosphatidylglycerol, monoolein and lauryl maltose neopentyl glycol (LMNG) detergent, but all efforts returned unacceptable refinement outcomes and maps. A 2Fo − Fc map following the inclusion of LPS into the refinement (blue, 0.8σ contour) is shown for reference. b, LPS quantification from proteins purified under matched conditions and subjected to a limulus amebocyte lysate assay. MsbA, the inner membrane LPS transporter from E. coli29,72, was purified from a recombinant E. coli expression host and HEK293 cells (MsbA293) for comparison. Lnt is an inner membrane protein involved that is not known or expected to bind or transport LPS73, and was expressed and purified from E. coli for comparison. Experiments were run in duplicate at three different protein concentrations with similar results, where duplicate experiment with 25 ng ml−1 and 100 ng ml−1 protein are shown. c, MALDI–TOF mass spectrometry detects various lipid A species from purified PbgA, including an arabinose-modified species (black). No lipid A species were detected from Lnt purified and analysed under matched conditions (orange).
Sequence alignment of ten PbgA sequences from Enterobacteriaceae Gram-negative bacteria. Domain boundaries are based on E. coli PbgA structure are indicated, including the lipid A-binding motif (red shade) and pseudo-hydrolase active site residues (orange triangles).
a, All UPEC-ΔpbgA bacteria tested in the rifampicin sensitivity assay were probed by western blot analysis to confirm PbgA–Flag expression. GroEL was assessed as a loading control. Representative blots for n = 3 or more experiments are shown. b, Outer membrane permeability of UPEC ΔpbgA strains with pBADpbgA plasmids expressing wild-type or mutant pbgA assessed by rifampicin sensitivity, where MTR-AVA is the M212A/T213V/R216A PbgA triple mutant. Data are representative and presented as mean ± s.d. for n = 3 or more independent cultures. Note, see Extended Data Fig. 2f for a view of the salt-bridge interaction between R215 (IFD) and a conserved TMD acidic residue, D192.
a, Biotinylated LAB peptides were captured and interferometry measurements measured upon presenting peptides to different concentrations of detergent solubilized lipids (LPS, phosphatidylethanolamine, phosphatidylglycerol and cardiolipin). Three independent experiments were performed and data shown are representative. b, CFUs of E. coli ATCC 25922 measured over time with LABv2.1 and polymyxin B. Data are mean ± s.d. for n = 3 independent cultures. c, A red blood cell (RBC) lysis assay evaluated after 18 h in the presence of indicated compounds (Methods). Data are mean ± s.d. (n = 3) for each compound tested. d, A RBC lysis assay comparing LABv2.1 precursors (LABWT, LABWT+, LABv2.0) and LABv2.1 analogues designed, based on the LPS–PbgA crystal structure, to disrupt specific interactions of lipid A (LABv2.1_Dap213Thr, LABv2.1_Dap213Arg, LABv2.1_Dap212-Met213). Data are mean ± s.d. for n = 3 independent assay of each compound at each concentration.
a, Proteins identified by mass spectrometry following co-immunoprecipitation of endogenous PbgA using the anti-PbgA monoclonal antibody 7E7 (n = 3 independent experiments). Hits were classified based on abundance (sum of PSMs) and enrichment in PbgA IPs compared to control purifications (SAINT logOddsScore: anti-PbgA monoclonal antibody 7E7 versus anti-gp120). Identified proteins with a Bayesian FDR <10% are highlighted in red. b, Bacterial two-hybrid system using PbgA-prey and different bait proteins in E. coli cells. Interacting proteins lead to blue colonies on agar plates containing X-gal, whereas non-interacting proteins produce white colonies. A representative agar plate is shown (n = 3) and activity was confirmed in broth cultures. c, Growth of a conditional E. coli K-12 ΔpbgA::pBAD-pbgA after depletion of PbgA in the presence of a IPTG-inducible plasmid expressing wild-type lapB or plsY (Methods) demonstrates that lapB expression does not rescue growth after PbgA depletion. Representative plates are shown and growth assay was repeated three or more times. d, Cell lysates prepared from overnight streaks of E. coli K-12 with pBADpbgA wild-type or mutant plasmids were probed with anti-LpxC, anti-PbgA and anti-GroEL antibodies (Methods), indicating that disturbing the LPS–PbgA interaction interface leads to LpxC stabilization. Representative blots from n = 3 biological replicates are shown. e, Western blot analysis of LpxC after treatment with 1 μM (2× MIC) or 4 μM (8× MIC) of the small molecule MsbA inhibitor G’913, indicating that selective inhibition of MsbA29,44 and LPS transport impacts LpxC levels; GroEL is the loading control and a representative experiment (n = 3 independent experiments) is shown. f, E. coli K-12 ΔlptD::pBADlptD lysates prepared from cells grown in indicated concentration of arabinose were probed with anti-LpxC, anti-LptD and anti-GroEL antibodies (Methods). Representative blots from n = 3 biological replicates are shown. g, Bacterial two-hybrid assays using LapB-bait (pUT18-lapB) and indicated PbgA-mutant prey constructs (pKT25-pbgA) in E. coli DHM1 cells were performed (Methods). Interacting proteins lead to blue colonies, whereas non-interacting proteins produce white colonies. Note that EptATM–PbgAIFD+PD is a chimeric construct in which the TMD of PbgA has been replaced with the TMD region from EptA23. Representative plates from n = 3 culture streaks are shown. h, Growth of conditional PbgA strain (E. coli ΔpbgA::pBADpbgA) in the absence of arabinose inducer complemented with, clockwise from the top of plate, wild-type pbgA (PbgAWT), pbgA encoding only the TMD (PbgATM only), or a negative control (malE) on plasmids. A representative plate (n = 3) is shown. i. Cell lysates of the conditional pbgA strain (E. coli ΔpbgA::pBADpbgA) in the absence of arabinose inducer complemented with wild-type pbgA or pbgA encoding only the TMD were probed with anti-LpxC antibody (Methods). A representative blot for n = 3 independent experiments is shown. j, Plasmids encoding acpT (right side of plate) or acpS (left side of plate) in conditional-pbgA strain grown in the absence of the pBADpbgA inducer arabinose, with 0.1 mM IPTG at 30 °C. A representative growth plate (n = 3) was imaged. k, Cultures with plasmids expressing pbgA, acpT, acpS, or malE (control) were shifted to no arabinose/plus IPTG if necessary to deplete PbgA (Methods). A representative blot from at least n = 3 biological replicates is shown.
Extended Data Fig. 9 A previous PbgA crystal structure reported to have cardiolipin bound at the IFD is, instead, more consistent with bound lipid A.
a, At the inner membrane–periplasmic interface that we term the IFD: cardiolipin (named CL2)7 from chain A (left) and chain B (middle) of PDB 6V8Q are shown in stick representation; PbgA is removed for clarity. Similarly, lipid A is shown in stick representation taken from the high-resolution crystal structure presented in this work (right). Molecular clashes calculated using the MOE software74 indicate high-energy atomic distance and poor geometry (green lines) in both chains A and B from PDB 6V8Q. The extent of the intramolecular clash is indicated by the relative size of the green circle. b, An Fo − Fc map calculated using coordinates and structure factors from PDB 6V8Q chain A (left) and chain B (middle) shows a strong negative peak (−3σ, red mesh; −4σ, blue mesh) on the assigned modelled P2 phosphate position of the CL2 ligand. Right, the LPS–PbgA complex determined in this work is superimposed onto chain B of PDB 6V8Q for reference, with no further adjustments. c, An Fo − Fc map calculated using coordinates and structure factors from PDB 6V8Q, with CL2 omitted from the calculation, shows strong positive peaks (4σ and 7.5σ for chain A and B, respectively; green mesh), which, in both cases, appear better described by the LPS–PbgA complex structure determined in this work. Shown (right) is the LPS–PbgA complex superimposed onto chain B of PDB 6V8Q with no further adjustments. d, The same Fo − Fc map calculation as in c, only contoured to 3σ (green mesh). As seen on the right, when superimposed onto chain B of 6V8Q, the proximal 1-phospho-GlcNAc group of lipid A in our LPS–PbgA structure appears especially well accounted for by positive density peaks, and density consistent with a KDO sugar head group is also observed; and similar conclusions are reaching upon inspection of superposition onto chain A of 6V8Q (not shown).
Extended Data Fig. 10 Comparison of LPS coordination in PbgA to known selective and passive LPS-binding proteins.
PbgA (this study), MsbA (PDB code 6BPP), a selective LPS transporter29,72, LptB2FG (PDB code 6MHU), a selective LPS33,75, and TLR4-MD2 (PDB code 3VQ2), a high-affinity LPS receptor32,76, represent the examples of selective LPS-binding proteins with known structures. In these latter cases, the hydrophobic acyl chains of lipid A are increased and the bivalent and polar nature are the lipid A head group is exploited. Furthermore, note that Arg216 of PbgA, shown in stick representation, does not appear essential for binding LPS in vivo (see Fig. 3c). In addition, FhuA (PDB code 2FCP), found with LPS complexed along the outer leaflet region of this outer membrane protein barrel34, and OmpE36 (PDB code 5FVN), which has also revealed numerous LPS contacts along the barrel35, are shown for completeness and comparison. Notably, analogous to MsbA, LptB2FG and TLR4, hydrophobic and aromatic side chains make several contacts in FhuA and OmpE36 with the acyl chains of lipid A (not shown for clarity) and polar and basic side chains coordinate the bivalent lipid A head group. In all cases, the lipid A coordination schemes are distinct from what is observed in the LPS–PbgA complex (also see Fig. 3).
About this article
Cite this article
Clairfeuille, T., Buchholz, K.R., Li, Q. et al. Structure of the essential inner membrane lipopolysaccharide–PbgA complex. Nature 584, 479–483 (2020). https://doi.org/10.1038/s41586-020-2597-x
Chemical Reviews (2020)
Carbohydrate Research (2020)
Scientific Reports (2020)
Trends in Microbiology (2020)