There is an urgent need for new antibiotics against Gram-negative pathogens that are resistant to carbapenem and third-generation cephalosporins, against which antibiotics of last resort have lost most of their efficacy. Here we describe a class of synthetic antibiotics inspired by scaffolds derived from natural products. These chimeric antibiotics contain a β-hairpin peptide macrocycle linked to the macrocycle found in the polymyxin and colistin family of natural products. They are bactericidal and have a mechanism of action that involves binding to both lipopolysaccharide and the main component (BamA) of the β-barrel folding complex (BAM) that is required for the folding and insertion of β-barrel proteins into the outer membrane of Gram-negative bacteria. Extensively optimized derivatives show potent activity against multidrug-resistant pathogens, including all of the Gram-negative members of the ESKAPE pathogens1. These derivatives also show favourable drug properties and overcome colistin resistance, both in vitro and in vivo. The lead candidate is currently in preclinical toxicology studies that—if successful—will allow progress into clinical studies that have the potential to address life-threatening infections by the Gram-negative pathogens, and thus to resolve a considerable unmet medical need.
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.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Mass spectrometric data are available at the ProteomeXchange Consortium (http://www.proteomexchange.org/) with dataset identifier PXD010174.
WHO. Global Priority List of Antibiotic-resistant Bacteria to Guide Research, Discovery, and Development of New Antibiotics (World Health Organization, Geneva, 2017).
Boucher, H. W. et al. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin. Infect. Dis. 48, 1–12 (2009).
O’Neill, J. Project Syndicate – A Call to Antimicrobial Arms https://www.project-syndicate.org/commentary/antibiotics-resistance-economic-costs-by-jim-o-neill-2015-02 (2015).
Paterson, D. L. & Harris, P. N. A. Colistin resistance: a major breach in our last line of defence. Lancet Infect. Dis. 16, 132–133 (2016).
Henderson, J. C. et al. The power of asymmetry: architecture and assembly of the Gram-negative outer membrane bilayer. Annu. Rev. Microbiol. 70, 255–278 (2016).
Konovalova, A., Kahne, D. E. & Silhavy, T. J. Outer membrane biogenesis. Annu. Rev. Microbiol. 71, 539–556 (2017).
Srinivas, N. et al. Peptidomimetic antibiotics target outer-membrane biogenesis in Pseudomonas aeruginosa. Science 327, 1010–1013 (2010).
Werneburg, M. et al. Inhibition of lipopolysaccharide transport to the outer membrane in Pseudomonas aeruginosa by peptidomimetic antibiotics. ChemBioChem 13, 1767–1775 (2012).
Noinaj, N., Rollauer, S. E. & Buchanan, S. K. The β-barrel membrane protein insertase machinery from Gram-negative bacteria. Curr. Opin. Struct. Biol. 31, 35–42 (2015).
Storm, D. R., Rosenthal, K. S. & Swanson, P. E. Polymyxin and related peptide antibiotics. Annu. Rev. Biochem. 46, 723–763 (1977).
Mares, J., Kumaran, S., Gobbo, M. & Zerbe, O. Interactions of lipopolysaccharide and polymyxin studied by NMR spectroscopy. J. Biol. Chem. 284, 11498–11506 (2009).
Roberts, K. D. et al. Antimicrobial activity and toxicity of the major lipopeptide components of polymyxin B and colistin: last-line antibiotics against multidrug-resistant Gram-negative bacteria. ACS Infect. Dis. 1, 568–575 (2015).
Baron, S., Hadjadj, L., Rolain, J.-M. & Olaitan, A. O. Molecular mechanisms of polymyxin resistance: knowns and unknowns. Int. J. Antimicrob. Agents 48, 583–591 (2016).
Olaitan, A. O., Morand, S. & Rolain, J.-M. Mechanisms of polymyxin resistance: acquired and intrinsic resistance in bacteria. Front. Microbiol. 5, 643 (2014).
Raetz, C. R., Reynolds, C. M., Trent, M. S. & Bishop, R. E. Lipid A modification systems in Gram-negative bacteria. Annu. Rev. Biochem. 76, 295–329 (2007).
Groisman, E. A. The pleiotropic two-component regulatory system PhoP–PhoQ. J. Bacteriol. 183, 1835–1842 (2001).
McPhee, J. B., Lewenza, S. & Hancock, R. E. Cationic antimicrobial peptides activate a two-component regulatory system, PmrA–PmrB, that regulates resistance to polymyxin B and cationic antimicrobial peptides in Pseudomonas aeruginosa. Mol. Microbiol. 50, 205–217 (2003).
Okuda, S., Sherman, D. J., Silhavy, T. J., Ruiz, N. & Kahne, D. Lipopolysaccharide transport and assembly at the outer membrane: the PEZ model. Nat. Rev. Microbiol. 14, 337–345 (2016).
Hartmann, J.-B., Zahn, M., Burmann, I. M., Bibow, S. & Hiller, S. Sequence-specific solution NMR assignments of the β-barrel insertase BamA to monitor its conformational ensemble at the atomic level. J. Am. Chem. Soc. 140, 11252–11260 (2018).
Mahoney, T. F., Ricci, D. P. & Silhavy, T. J. Classifying β-barrel assembly substrates by manipulating essential Bam complex members. J. Bacteriol. 198, 1984–1992 (2016).
Lee, J. et al. Characterization of a stalled complex on the β-barrel assembly machine. Proc. Natl Acad. Sci. USA 113, 8717–8722 (2016).
Gunasinghe, S. D. et al. The WD40 protein BamB mediates coupling of BAM complexes into assembly precincts in the bacterial outer membrane. Cell Rep. 23, 2782–2794 (2018).
Mitchell, A. M. & Silhavy, T. J. Envelope stress responses: balancing damage repair and toxicity. Nat. Rev. Microbiol. 17, 417–428 (2019).
Storek, K. M. et al. Monoclonal antibody targeting the β-barrel assembly machine of Escherichia coli is bactericidal. Proc. Natl Acad. Sci. USA 115, 3692–3697 (2018).
Browning, D. F. et al. Mutational and topological analysis of the Escherichia coli BamA protein. PLoS ONE 8, e84512 (2013).
Lee, J. et al. Substrate binding to BamD triggers a conformational change in BamA to control membrane insertion. Proc. Natl Acad. Sci. USA 115, 2359–2364 (2018).
Rigel, N. W., Ricci, D. P. & Silhavy, T. J. Conformation-specific labeling of BamA and suppressor analysis suggest a cyclic mechanism for β-barrel assembly in Escherichia coli. Proc. Natl Acad. Sci. USA 110, 5151–5156 (2013).
Dixon, R. A. & Chopra, I. Polymyxin B and polymyxin B nonapeptide alter cytoplasmic membrane permeability in Escherichia coli. J. Antimicrob. Chemother. 18, 557–563 (1986).
Moffatt, J. H. et al. Colistin resistance in Acinetobacter baumannii is mediated by complete loss of lipopolysaccharide production. Antimicrob. Agents Chemother. 54, 4971–4977 (2010).
Choi, M. et al. MSstats: an R package for statistical analysis of quantitative mass spectrometry-based proteomic experiments. Bioinformatics 30, 2524–2526 (2014).
Carlson, M. UniProt.ws: R interface to UniProt web services. R package version 2.14.0. https://bioconductor.org/packages/release/bioc/html/UniProt.ws.html (2018)
We thank M. Gwerder, the Center for Microscopy and Image Analysis at UZH, and the Functional Genomics Center Zurich for technical support. We thank the following agencies for funding: M. Müller and B.W. were supported by ETH grant ETH-30 17-1 and a grant from the Swiss National Science Foundation (grant number 31003A_160259); J.A.R. was supported by a grant from the Commission for Technology and Innovation (CTI/KTI) (grant number 18146.1 PFLS-LS); S.H. was supported by a grant from the Swiss National Science Foundation via the NRP 72 (grant 407240_167125); A.V. was supported by the Swiss National Science Foundation (SystemsX.ch - IPhD project 51PHP0_163556). Polyphor acknowledges funding from CARB-X and the Wellcome Trust (grant number 202728/Z/16/Z) and financial support from the REPAIR Impact Fund (Novo Holdings).
A. Luther, P.C., S.S., C.V., T.R., M.S., C.K., M.-A.W., N.D., E.B., S.H., K.L., A.V., R.J., V.R., G.U., A. Lederer, P.Z., A.W., H.H.L., F.B., G.E.D. and D.O. declare competing interests as employees of Polyphor AG who pursue clinical studies.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Peer review information Nature thanks Paul Hergenrother, Lynn Silver and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
a, In vitro killing kinetics of 3 and 8 against representative Gram-negative species, E. coli ATCC 25922, P. aeruginosa ATCC 27853 and A. baumannii ATCC 19606. b, Resistance development of 8 by serial passage against E. coli ATCC 25922, P. aeruginosa ATCC 27853 and A. baumannii ATCC 19606. The y axis indicates the MIC measured directly from the tubes during the serial passages (mg l−1) and the x axis is the number of passages. Antibiotics used as comparisons are indicated (colistin and meropenem). In a, b, the curves show typical examples of n = 2 biologically independent experiments. c, In vitro killing kinetics of 3 and 8 against K. pneumoniae SS3010 and resistance development of 8 by serial passage against K. pneumoniae SSI3010. The y axis indicates the MIC measured directly from the tubes during the serial passages (mg l−1) and the x axis is the number of passages. The asterisks represent the clones taken for whole-genome sequencing. Antibiotics used as comparisons are indicated (colistin and meropenem). The curves show representative examples of n = 2 biologically independent experiments. d, In vivo efficacy of peptide 3 in a mouse model of septicaemia against E. coli 5799. Each point represents the percentage of survival of n = 6 mice. e, f, In vivo efficacy of 3 and 8, in mouse models of peritonitis, against E. coli AF45 (mcr-1, colistin resistant) after a single subcutaneous administration (reduction in CFU counts in peritoneal wash fluid and blood). The mean and s.e.m. of n = 6 mice are shown. g, In vivo efficacy of 3, 7 and 8, in mouse models of peritonitis, against K. pneumoniae SSI3010 after a single subcutaneous administration (reduction in CFU counts in peritoneal wash fluid). The mean and s.e.m. of n = 4 (n = 6 for vehicle) mice are shown. Start of treatment and vehicle were repeated in three experiments. h, In vivo efficacy of 3 and 8, in mouse models of peritonitis, against E. coli mcr-3 SNT R36B6 (colistin resistant) after a single subcutaneous administration (reduction in CFU counts in peritoneal wash fluid). The mean and s.e.m. of n = 4 mice (n = 6 for vehicle) are shown. i–k, In vivo efficacy of 8, in mouse models of thigh infection, against E. coli ATCC BAA2469 (NDM-1 strain), P. aeruginosa PA14 and A. baumannii NCTC 13301 (extensively drug resistant, OXA-23 strain). The total daily dose (TDD) indicated was administered in 2 doses over the course of 24 h (q12h). The mean and s.e.m. of n = 6 mice are shown (n = 4 mice for start of treatment). On each mouse, two technical replicates were doneSource data.
Relative incorporations of 3H label into macromolecules in E. coli from the labelled precursors shown over 20 min, at 37 °C, with increasing concentrations of peptide 3 and peptide 4. The incorporation is relative to a control with no addition of antibiotic (100%). The results show no inhibition of protein, RNA, DNA biosynthesis or cell-wall polysaccharide. In control experiments, the expected effects of known antibiotics (tobramycin, rifampicin, ciprofloxacin or ceftriaxone, on protein, DNA, RNA or cell-wall biosynthesis, respectively, were observed (data not shown)). The red dotted line indicates the MIC of each antibiotic. The results shown are representative of n = 3 biologically independent experimentsSource data.
a, b. Membrane permeabilization elicited by 3, 4 or PMB monitored by uptake of SYTOX-Green, and increase in fluorescence intensity. a, Permeabilization with E. coli ATCC 25922. b, Permabilization with PMB-resistant E. coli 926415 strain. No fluorescence increase is seen from cells in the presence of SYTOX-Green without antibiotic. The change in cell density (OD600) in the cuvette over 1 h is shown on the right. For experimental methods, see Supplementary Information. In a, the curves represent the mean of n = 3 biologically independent experiments (except for PMB on the left, and peptide 4 (0.8 μg ml−1), for which n = 2 biologically independent experiments). The error bars are s.d. In b, the curves represent the mean of n = 3 biologically independent experiment (except for PMB (40 μg ml−1), peptide 3 (0.3 μg ml−1) and peptide 4 (0.3 μg ml−1), for which n = 2 biologically independent experiments). The error bars are s.d. c, d, Release from E. coli of β-lactamase (c) and of β-galactosidase (d) in the presence of 3, 4 or PMB, monitored by enzymatic assays. Ciprofloxacin does not cause detectable release of either enzyme from cells, whereas protegrin I caused rapid and complete release of both enzymes (100% value corresponds to enzyme released by sonication of cells). For experimental methods, see Supplementary Information. The curves represent the mean of n = 3 or 4 biologically independent experiment. The error bars show the s.dSource data.
Thin layer chromatography (TLC) and mass spectrometry analyses of lipid A. a, TLC analysis of lipid A extracted from E. coli K12 (1), ATCC 25922 (2) and PMB-resistant strain 926415 (3). b, Matrix-assisted laser desorption/ionization–time of flight (MALDI–TOF) mass spectrometry of the lipid A mixture isolated from each strain. c, Individual lipid A species identified in the lipid A extracted from strains ATCC 25922 and 926415 by TLC-MALDI–TOF mass spectrometry. All lipid A in sample 3 contains phosphoethanolamine and/or l-4-amino-4-deoxyarabinose units. For experimental methods, see Supplementary Information. The results shown are representative of n = 2 biologically independent experiments.
Scanning electron microscopy of E. coli ATCC 25922 cells untreated or grown with 3 or 4, at concentrations that cause a growth inhibition of about 50% (about 0.1 mg l−1). Scale bars, 1 μm. The scanning electron microscopy scans were in completed in duplicate, and one typical result is shown. For experimental methods, see Supplementary Information.
a, Western blots (10% SDS–PAGE gel, blotted to a PVDF membrane) of membrane protein extract from PAL-3- (4 mg l−1), PAL-4- (10 mg l−1) and PAL-7- (4 mg l−1) labelled E. coli ATCC 25922 with chemiluminescence detection of biotinylated macromolecules. For gel source data, see Supplementary Fig. 1. b, Volcano plot showing relative abundance of proteins captured by streptavidin, and quantified by mass spectrometry, from E. coli cells photolabelled with PAL-7 versus control cells treated with 7 (n = 3 biologically independent samples each). Protein abundance changes (expressed in log2) were calculated by linear mixed-effect model and tested for statistical significance using a two-sided t-test. P values obtained were further corrected for multiple comparisons using Benjamini–Hochberg method. Proteins are represented on the basis of the UniProt annotated subcellular location as dots (outer membrane) or crosses (no, or other, location); symbol size is scaled according to statistical significance. Significantly enriched proteins (abundance ratio ≥ 1.5 and adjusted P ≤ 0.05, shown as blue lines) are coloured in green. Outer membrane proteins that were also enriched in PAL-3 and PAL-4 photo-affinity interaction mapping experiments are highlighted. BamA is among the most significantly upregulated proteins, and is the only common outer membrane interaction candidate that was identified by all three photoprobes. A full list of proteins quantified by mass spectrometry in these experiments is supplied as Source DataSource data.
a, Left, titration of BamA(POTRA1–5) to peptide Cy3-3, monitored by fluorescence anisotropy. Measurements were performed in triplicates; s.d. around the mean is shown. No interaction is detected. Right, titration of LDAO detergent with Cy3-3. Measurements were performed in triplicates; s.d. around the mean is shown. No interaction is detected in the range up to 6% LDAO. The binding experiments between peptide Cy3-3 and BamA-β were carried out at 0.6% LDAO concentration. b, Left, SDS–PAGE gel of purified OmpX and LamB in LDAO and octyl glucoside (OG) micelles, respectively. Samples in lanes 2 and 4 were boiled, and samples in lanes 1 and 3 were not boiled, before the electrophoresis run. The resulting difference in migration indicates the presence of folded protein in the unboiled samples. For gel source data, see Supplementary Fig. 1. Middle, titration of LamB and OmpX to Cy3-3 peptide as monitored by fluorescence anisotropy. No interaction was observed compared to BamA-β. Error bars are s.d. around the mean from triplicate measurements. Right, titration of BamA-β to Cy3-9 (scrambled) peptide as monitored by fluorescence anisotropy. No interaction was observed compared to Cy3-3. Error bars are s.d. around the mean from triplicate measurements. c, Binding of Cy3-3 to BamA-β (left) and to BamA(POTRA1–5) (right) by microscale thermophoresis. Measurements were performed in triplicate with s.d. around the mean shown. For thermophoresis studies of Cy3-3 with BamA(POTRA1–5), a constant concentration of Cy3-3 (10 nM) was titrated with BamA(POTRA1–5) in 20 mM HEPES buffer with 150 mM NaCl, pH 7.5, at room temperature, from 107.5 mM to about 3.2 nM. No thermophoresis signal was observed (right).
a, Interaction-site mapping of BamAext with peptide 3 (n = 1 experiment). Chemical shift perturbations of amide moieties plotted against the BamA residue number upon addition of 600 μM peptide 3 to 300 μM of BamAext. A threshold of 0.07 ppm is indicated by a red dashed line. Three residues, at which the signal goes into intermediate exchange upon peptide titration, are indicated by red lines and their Δδ(HN) was arbitrarily set to 0.1 ppm for visualization. b, Chemical shift perturbations of amide moieties plotted against the BamA residue number upon addition of 500 μM peptide 8 to 250 μM of BamAext (n = 1 experiment). A threshold of 0.1 ppm is indicated by a red dashed line. Six residues, at which the signal goes into intermediate exchange upon peptide titration, are indicated by red lines and their Δδ(HN) was arbitrarily set to 0.2 ppm for visualization. c, Interaction-site mapping of BamAext with peptide 9 (n = 1 experiment). Chemical shift perturbations of amide moieties plotted against the BamA residue number upon addition of 700 μM peptide 9 to 350 μM of BamAext. No statistical tests were done for data shown in this figure.
The interaction of 8 with the BamA β-barrel domain was characterized using NMR spectroscopy (n = 1 experiment). a, Two-dimensional [15N,1H]TROSY spectra of 250 μM [U-2H,15N]BamAext (red) titrated with 0.5 (yellow), 1 (magenta) or 2 (blue) stoichiometric equivalents of peptide 8. b, Close-up views of selected residues from the titration in a, and from a corresponding titration with BamA-β. c, Representation of the interactions of peptide 8 on the BamA β-barrel structure viewed from the top and from the side of the barrel. Labelled residues have substantial chemical shift perturbations or intensity changes upon peptide binding (crystal structure from PDB 6FSU). d, Overlays of two-dimensional [15N, 1H]TROSY spectra of titrations points of BamAext + scrambled peptide 9. BamAext 350 μM (apo), red; +0.5 equivalent of 9, yellow; +1 equivalent of 9, pink; and +2 equivalent of 9, blue.
This file contains Supplementary Information Sections 1-12, including Supplementary Tables 1-5 and Schemes S1-5. Section 13, Proteins quantified in photoaffinity interaction mapping experiments, is provided as 2 separate Excel files - Source Data for Fig.3 and Source Data for Extended Data Fig.6b.
This file contains the uncropped blots.
About this article
Cite this article
Luther, A., Urfer, M., Zahn, M. et al. Chimeric peptidomimetic antibiotics against Gram-negative bacteria. Nature 576, 452–458 (2019). https://doi.org/10.1038/s41586-019-1665-6
Drug Discovery Today (2021)
A molecular architectural design that promises potent antimicrobial activity against multidrug-resistant pathogens
NPG Asia Materials (2021)
Bioorganic & Medicinal Chemistry Letters (2021)
Annals of the New York Academy of Sciences (2021)
Antimicrobial peptides with symmetric structures against multidrug-resistant bacteria while alleviating antimicrobial resistance
Biochemical Pharmacology (2021)