CoChat
Sign Up
Cited by 201Antimicrobial Peptides and ActivitiesAntimicrobial agents and applicationsMicrobial Natural Products and Biosynthesis

The DOI matches its Crossref record; 64 of the 67 references it cites were found in a registry or an index.

Teixobactin kills bacteria by a two-pronged attack on the cell envelope

Open access
Shukla, Rhythm, Lavore, Francesca, Maity, Sourav, Derks, Maik G. N., Jones, Chelsea R., Vermeulen, Bram J. A., Melcrová, Adéla, Morris, Michael A., Becker, Lea Marie, Wang, Xiaoqi, Kumar, Raj, Medeiros-Silva, João, van Beekveld, Roy A. M., Bonvin, Alexandre M. J. J., Lorent, Joseph H., Lelli, Moreno, Nowick, James S., MacGillavry, Harold D., Peoples, Aaron J., Spoering, Amy L., Ling, Losee L., Hughes, Dallas E., Roos, Wouter H., Breukink, Eefjan, Lewis, Kim, Weingarth, Markus·2022·Nature
DOI 10.1038/s41586-022-05019-yPMC PMC9365693
Open this paper in CoChat — free accountOpen PDF

Abstract

Antibiotics that use novel mechanisms are needed to combat antimicrobial resistance 1–3 . Teixobactin 4 represents a new class of antibiotics with a unique chemical scaffold and lack of detectable resistance. Teixobactin targets lipid II, a precursor of peptidoglycan 5 . Here we unravel the mechanism of teixobactin at the atomic level using a combination of solid-state NMR, microscopy, in vivo assays and molecular dynamics simulations. The unique enduracididine C-terminal headgroup of teixobactin specifically binds to the pyrophosphate-sugar moiety of lipid II, whereas the N terminus coordinates the pyrophosphate of another lipid II molecule. This configuration favours the formation of a β-sheet of teixobactins bound to the target, creating a supramolecular fibrillar structure. Specific binding to the conserved pyrophosphate-sugar moiety accounts for the lack of resistance to teixobactin 4 . The supramolecular structure compromises membrane integrity. Atomic force microscopy and molecular dynamics simulations show that the supramolecular structure displaces phospholipids, thinning the membrane. The long hydrophobic tails of lipid II concentrated within the supramolecular structure apparently contribute to membrane disruption. Teixobactin hijacks lipid II to help destroy the membrane. Known membrane-acting antibiotics also damage human cells, producing undesirable side effects. Teixobactin damages only membranes that contain lipid II, which is absent in eukaryotes, elegantly resolving the toxicity problem. The two-pronged action against cell wall synthesis and cytoplasmic membrane produces a highly effective compound targeting the bacterial cell envelope. Structural knowledge of the mechanism of teixobactin will enable the rational design of improved drug candidates.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41586-022-05019-y

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Oligomerization and membrane remodelling. a, Chemical structure of teixobactin. b, 2D NH ssNMR spectrum of lipid II-bound teixobactin in membranes. c, Secondary chemical shifts (SCS23) show β-structuring of the N terminus. Source data are provided. NA, not applicable. d, 2D CC ssNMR spectra of lipid II-bound teixobactin acquired with 50 ms (cyan) and 600 ms (grey) magnetization transfer time show intermolecular Cα–Cα contacts (red) consistent with the formation of antiparallel teixobactin β-sheets. e, Visualization of lipid II clustering (arrows) in 3D (left), using confocal microscopy of GUVs doped with Atto-labelled lipid II and treated with teixobactin. The transmission image of the same GUV revealed membrane perturbations induced by the oligomers (middle). Control GUVs with Atto-labelled lipid II, without teixobactin (transmission) (right) are also shown. Note that as the plane used for the middle image is not exactly in the middle of the GUV, the diameter is smaller than for the entire GUV as seen in the left image. The experiment was performed in biological triplicates. f, Static 31P ssNMR spectra show membrane perturbation induced by teixobactin (cyan line) in contrast to the untreated lipid II-doped DOPC large unilamellar vesicles (black dashed line). g, Confocal microscopy with B. megaterium incubated in the presence of a fluorescent teixobactin analogue26. Intense, elongated teixobactin clusters form fast and get more compact after 45 min. h, Boxplot of the fluorescence intensity of clusters after 1, 15 and 45 min of incubation with teixobactin. n = 380 clusters from biological triplicates. Data are represented as a boxplot in which the middle line is the median, the lower and upper hinges correspond to the first and third quartiles, the upper whisker extends from the hinge to the largest value and the lower whisker extends from the hinge to the smallest value, no further than 1.5× interquartile range. ***P < 0.0001, unpaired, two-tailed Student's t-test. Source data
Fig. 1 Oligomerization and membrane remodelling. a, Chemical structure of teixobactin. b, 2D NH ssNMR spectrum of lipid II-bound teixobactin in membranes. c, Secondary chemical shifts (SCS23) show β-structuring of the N terminus. Source data are provided. NA, not applicable. d, 2D CC ssNMR spectra of lipid II-bound teixobactin acquired with 50 ms (cyan) and 600 ms (grey) magnetization transfer time show intermolecular Cα–Cα contacts (red) consistent with the formation of antiparallel teixobactin β-sheets. e, Visualization of lipid II clustering (arrows) in 3D (left), using confocal microscopy of GUVs doped with Atto-labelled lipid II and treated with teixobactin. The transmission image of the same GUV revealed membrane perturbations induced by the oligomers (middle). Control GUVs with Atto-labelled lipid II, without teixobactin (transmission) (right) are also shown. Note that as the plane used for the middle image is not exactly in the middle of the GUV, the diameter is smaller than for the entire GUV as seen in the left image. The experiment was performed in biological triplicates. f, Static 31P ssNMR spectra show membrane perturbation induced by teixobactin (cyan line) in contrast to the untreated lipid II-doped DOPC large unilamellar vesicles (black dashed line). g, Confocal microscopy with B. megaterium incubated in the presence of a fluorescent teixobactin analogue26. Intense, elongated teixobactin clusters form fast and get more compact after 45 min. h, Boxplot of the fluorescence intensity of clusters after 1, 15 and 45 min of incubation with teixobactin. n = 380 clusters from biological triplicates. Data are represented as a boxplot in which the middle line is the median, the lower and upper hinges correspond to the first and third quartiles, the upper whisker extends from the hinge to the largest value and the lower whisker extends from the hinge to the smallest value, no further than 1.5× interquartile range. ***P < 0.0001, unpaired, two-tailed Student's t-test. Source dataCC BY · PubMed Central
Fig. 2 Teixobactin sequesters lipid II into supramolecular fibrils. a, Snapshots of a timelapse HS-AFM video (Supplementary Video 1) following the assembly of teixobactin–lipid II fibrils. Images were obtained on a supported lipid bilayer containing 1% (mol) lipid II in the presence of 800 nM teixobactin, added after 24 s. Image acquisition rate of 0.5 frames per second. b, Zoomed in view of an HS-AFM image of the fibrillar sheet on the membrane surface, as marked by a white rectangle in a at 624 s. The inset in the lower right corner shows the height profile at the dashed line. c, HS-AFM image of a lipid bilayer deformed by teixobactin–lipid II fibrils below the membrane surface, 50 min after the addition of 800 nM teixobactin. The inset shows the height profile at the dashed line. d, Histogram showing the relative height of teixobactin–lipid II fibrils above (measured directly when fibrils assemble on the membrane surface) and below (measured 50 min after the addition of 800 nM teixobactin) the lipid bilayer surface (n > 50). e, Side view of a 3D-rendered image in c. f, Model of the mode of action of teixobactin. Teixobactin first forms small β-sheets upon binding of lipid II, then elongates into fibrils that eventually associate into lateral fibrillar sheets, obstructing biosynthesis of peptidoglycan and causing membrane defects. Source data
Fig. 2 Teixobactin sequesters lipid II into supramolecular fibrils. a, Snapshots of a timelapse HS-AFM video (Supplementary Video 1) following the assembly of teixobactin–lipid II fibrils. Images were obtained on a supported lipid bilayer containing 1% (mol) lipid II in the presence of 800 nM teixobactin, added after 24 s. Image acquisition rate of 0.5 frames per second. b, Zoomed in view of an HS-AFM image of the fibrillar sheet on the membrane surface, as marked by a white rectangle in a at 624 s. The inset in the lower right corner shows the height profile at the dashed line. c, HS-AFM image of a lipid bilayer deformed by teixobactin–lipid II fibrils below the membrane surface, 50 min after the addition of 800 nM teixobactin. The inset shows the height profile at the dashed line. d, Histogram showing the relative height of teixobactin–lipid II fibrils above (measured directly when fibrils assemble on the membrane surface) and below (measured 50 min after the addition of 800 nM teixobactin) the lipid bilayer surface (n > 50). e, Side view of a 3D-rendered image in c. f, Model of the mode of action of teixobactin. Teixobactin first forms small β-sheets upon binding of lipid II, then elongates into fibrils that eventually associate into lateral fibrillar sheets, obstructing biosynthesis of peptidoglycan and causing membrane defects. Source dataCC BY · PubMed Central
Fig. 3 Enduracididine governs the interface. a, Chemical structure of lipid II. b, 1D 31P ssNMR data in liposomes show strong shifts of the lipid II PPi signals upon addition of teixobactin. 2D 1H31P (HP) ssNMR shows a direct interaction between the PPi group and the backbone amino-protons of the depsi-cycle and the sidechain of End10. MAS, magic angle spinning. c, Superposition of 2D 13C13C (CC) spectra of the complex with NMR-invisible 12C,14N-labelled (red) and NMR-active 13C,15N-lipid II (cyan) show a dominant presence of enduracididine at the interface. The cyan spectrum was acquired with a magnetization transfer time of 300 ms. Interfacial contacts with MurNAc and A1 of the pentapeptide are shown in orange. d, Sum of interfacial ssNMR contacts with the headgroup of lipid II. e, Binding energetics of the interface obtained by isothermal titration calorimetry. Data show the averages of three experiments for each drug. Data represented as mean ± s.e.m. Source data are provided as a Source Data file. f, Illustration of the differential binding modes of natural teixobactin (left) and the synthetic analogue R4L10-teixobactin18. Enduracididine in teixobactin specifically binds to MurNAc, which stabilizes the entire interface. The substitution of End10 by a leucine residue leads to a fuzzy interface with the headgroup of lipid II and reduces the binding affinity. Source data
Fig. 3 Enduracididine governs the interface. a, Chemical structure of lipid II. b, 1D 31P ssNMR data in liposomes show strong shifts of the lipid II PPi signals upon addition of teixobactin. 2D 1H31P (HP) ssNMR shows a direct interaction between the PPi group and the backbone amino-protons of the depsi-cycle and the sidechain of End10. MAS, magic angle spinning. c, Superposition of 2D 13C13C (CC) spectra of the complex with NMR-invisible 12C,14N-labelled (red) and NMR-active 13C,15N-lipid II (cyan) show a dominant presence of enduracididine at the interface. The cyan spectrum was acquired with a magnetization transfer time of 300 ms. Interfacial contacts with MurNAc and A1 of the pentapeptide are shown in orange. d, Sum of interfacial ssNMR contacts with the headgroup of lipid II. e, Binding energetics of the interface obtained by isothermal titration calorimetry. Data show the averages of three experiments for each drug. Data represented as mean ± s.e.m. Source data are provided as a Source Data file. f, Illustration of the differential binding modes of natural teixobactin (left) and the synthetic analogue R4L10-teixobactin18. Enduracididine in teixobactin specifically binds to MurNAc, which stabilizes the entire interface. The substitution of End10 by a leucine residue leads to a fuzzy interface with the headgroup of lipid II and reduces the binding affinity. Source dataCC BY · PubMed Central
Fig. 4 Teixobactin–membrane interaction. a, High-resolution ssNMR structure of the teixobactin–lipid II complex. b, Zoomed in view of the complex interface. The backbone amino-protons of the depsi-cycle of teixobactin, the End10 sidechain and the N terminus of an adjacent teixobactin coordinate the lipid II PPi group. In addition, End10 interacts with the MurNAc sugar via hydrogen bonds. Blue spheres represent backbone nitrogens; numbers indicate the residue numbers. c, Membrane topology: lipophilic and hydrophilic residues are sharply separated. d, Oligomerization enhances complex stability: molecular dynamics simulations show that inner lipid II molecules (red) are more stably bound than outer molecules (purple). The plot shows the distance between the centre of mass of the PPi group and the centre of mass of amino-protons of the depsi-cycle averaged over the last 200 ns of two molecular dynamics simulations. Source data are provided as a Source Data file. e, Upon lipid II-induced oligomerization, the hydrophobic side of teixobactin faces the membrane surface, which displaces the polar lipid headgroups and concentrates non-lamellar lipid II tails, causing membrane distortions. f,g, Membrane thickness obtained from molecular dynamics simulations, averaged over the last 50 ns. The membrane is thinner at the site of the complex. The red and blue colours show membrane thickness minima and maxima, respectively. h, Bacterial assays with Staphylococcus simulans were used to study membrane depolarization (DiSC; upper panel) and membrane damage (Sytox; lower panel) caused by teixobactin, nisin, plectasin, vancomycin and sans antibiotic. Arrows indicate the addition of antibiotics. Source data are provided as a Source Data file. Source data
Fig. 4 Teixobactin–membrane interaction. a, High-resolution ssNMR structure of the teixobactin–lipid II complex. b, Zoomed in view of the complex interface. The backbone amino-protons of the depsi-cycle of teixobactin, the End10 sidechain and the N terminus of an adjacent teixobactin coordinate the lipid II PPi group. In addition, End10 interacts with the MurNAc sugar via hydrogen bonds. Blue spheres represent backbone nitrogens; numbers indicate the residue numbers. c, Membrane topology: lipophilic and hydrophilic residues are sharply separated. d, Oligomerization enhances complex stability: molecular dynamics simulations show that inner lipid II molecules (red) are more stably bound than outer molecules (purple). The plot shows the distance between the centre of mass of the PPi group and the centre of mass of amino-protons of the depsi-cycle averaged over the last 200 ns of two molecular dynamics simulations. Source data are provided as a Source Data file. e, Upon lipid II-induced oligomerization, the hydrophobic side of teixobactin faces the membrane surface, which displaces the polar lipid headgroups and concentrates non-lamellar lipid II tails, causing membrane distortions. f,g, Membrane thickness obtained from molecular dynamics simulations, averaged over the last 50 ns. The membrane is thinner at the site of the complex. The red and blue colours show membrane thickness minima and maxima, respectively. h, Bacterial assays with Staphylococcus simulans were used to study membrane depolarization (DiSC; upper panel) and membrane damage (Sytox; lower panel) caused by teixobactin, nisin, plectasin, vancomycin and sans antibiotic. Arrows indicate the addition of antibiotics. Source data are provided as a Source Data file. Source dataCC BY · PubMed Central
Extended Data Fig. 1 High-resolution ssNMR studies of the teixobactin - Lipid II complex in membranes. All ssNMR spectra were acquired with uniformly 13C,15N-labelled teixobactin bound to 12C,14N-Lipid II in DOPC liposomes. a) 2D 13C13C PARISxy33 spectrum acquired at 950 MHz and 18 kHz MAS using 50 ms CC magnetization transfer time. All 13C nuclei of teixobactin are assigned. b) 2D NH spectrum acquired at 700 MHz and 60 kHz MAS. All signals are assigned. c) 2D CN spectrum acquired at 700 MHz and 12 kHz MAS using 5 ms N to C cross-polarization time. Some connectivities, including for the End10 sidechain, are shown by dashed lines. Note that our 2D NH data of the 13C,15N-teixobactin – Lipid II complex in membranes match well to a previously published spectrum of the complex in dodecylphosphocholine (DPC) micelles22. Interestingly, compared to micellar assignments, we observe major (>5 13C ppm) local differences for the two residues that most prominently interact with Lipid II, something that could suggest an impact of the membrane environment for complex formation. See Supplementary Table 3 for a comparison of chemical shifts of the teixobactin – Lipid II complex in micelles and in membranes. Note that backbone assignments were conclusively validated with a pair of 1H-detected 3D CαNH/CONH experiments65.
Extended Data Fig. 1 High-resolution ssNMR studies of the teixobactin - Lipid II complex in membranes. All ssNMR spectra were acquired with uniformly 13C,15N-labelled teixobactin bound to 12C,14N-Lipid II in DOPC liposomes. a) 2D 13C13C PARISxy33 spectrum acquired at 950 MHz and 18 kHz MAS using 50 ms CC magnetization transfer time. All 13C nuclei of teixobactin are assigned. b) 2D NH spectrum acquired at 700 MHz and 60 kHz MAS. All signals are assigned. c) 2D CN spectrum acquired at 700 MHz and 12 kHz MAS using 5 ms N to C cross-polarization time. Some connectivities, including for the End10 sidechain, are shown by dashed lines. Note that our 2D NH data of the 13C,15N-teixobactin – Lipid II complex in membranes match well to a previously published spectrum of the complex in dodecylphosphocholine (DPC) micelles22. Interestingly, compared to micellar assignments, we observe major (>5 13C ppm) local differences for the two residues that most prominently interact with Lipid II, something that could suggest an impact of the membrane environment for complex formation. See Supplementary Table 3 for a comparison of chemical shifts of the teixobactin – Lipid II complex in micelles and in membranes. Note that backbone assignments were conclusively validated with a pair of 1H-detected 3D CαNH/CONH experiments65.CC BY · PubMed Central
Extended Data Fig. 2 Dynamics of teixobactin in the complex and pyrene excimer fluorescence confirm Lipid II oligomerization upon binding of teixobactin. a) Site-resolved 15N T1rho dynamics of Lipid II-bound teixobactin in membranes acquired at 60 kHz (MAS). The overall dynamics of the molecule is very slow, in agreement with the immobilization of teixobactin in large oligomeric complexes. The error bars indicate the standard error of the fit. Data represented as mean +/− SD. Source data are provided. Monomeric pyrene fluorescence emission maxima occur at 378, 398, and 417 nm. Pyrene molecules exhibit a unique fluorescence at 490 nm only when two pyrene molecules come within 10-20 Å of each other, forming an excited dimer state known as the excimer state. Source data are provided. b) Fluorescence spectra of pyrene-labelled Lipid II in LUVs for increasing concentrations of teixobactin. It can be readily seen that with the increase in teixobactin concentration there is a relative decrease in the monomer signal followed by a respective increase in the excimer signal. Each spectrum is acquired immediately after the addition of teixobactin and averaged over 50 s, demonstrating that the oligomerization event is almost instantaneous. c) The insert shows the excimer region of the plot. d) Quantification of the excimer over monomer ratio as a function of the teixobactin concentration. e) Fluorescence microscopy shows oligomerization and membrane perturbations. A 3D render of the z-stack of DOPC GUVs doped with Atto-labelled Lipid II and treated with teixobactin shows Lipid II oligomerization. f) A 2D plane of the same GUV shown in e) revealed membrane perturbations induced by the oligomers. ii) Transmission image of the GUV. g) Control GUV: i) fluorescent ii) Transmission image; h) i) Confocal microscopy images of GUVs taken from biological triplicates ii) The transmission images. All scale bars represent 5 μm. Source data
Extended Data Fig. 2 Dynamics of teixobactin in the complex and pyrene excimer fluorescence confirm Lipid II oligomerization upon binding of teixobactin. a) Site-resolved 15N T1rho dynamics of Lipid II-bound teixobactin in membranes acquired at 60 kHz (MAS). The overall dynamics of the molecule is very slow, in agreement with the immobilization of teixobactin in large oligomeric complexes. The error bars indicate the standard error of the fit. Data represented as mean +/− SD. Source data are provided. Monomeric pyrene fluorescence emission maxima occur at 378, 398, and 417 nm. Pyrene molecules exhibit a unique fluorescence at 490 nm only when two pyrene molecules come within 10-20 Å of each other, forming an excited dimer state known as the excimer state. Source data are provided. b) Fluorescence spectra of pyrene-labelled Lipid II in LUVs for increasing concentrations of teixobactin. It can be readily seen that with the increase in teixobactin concentration there is a relative decrease in the monomer signal followed by a respective increase in the excimer signal. Each spectrum is acquired immediately after the addition of teixobactin and averaged over 50 s, demonstrating that the oligomerization event is almost instantaneous. c) The insert shows the excimer region of the plot. d) Quantification of the excimer over monomer ratio as a function of the teixobactin concentration. e) Fluorescence microscopy shows oligomerization and membrane perturbations. A 3D render of the z-stack of DOPC GUVs doped with Atto-labelled Lipid II and treated with teixobactin shows Lipid II oligomerization. f) A 2D plane of the same GUV shown in e) revealed membrane perturbations induced by the oligomers. ii) Transmission image of the GUV. g) Control GUV: i) fluorescent ii) Transmission image; h) i) Confocal microscopy images of GUVs taken from biological triplicates ii) The transmission images. All scale bars represent 5 μm. Source dataCC BY · PubMed Central
Extended Data Fig. 3 Intermolecular teixobactin – teixobactin interactions seen by ssNMR. Solid-state NMR 13C-13C PARISxy spectra of the 13C,15N-teixobactin – Lipid II complex in membranes show several cross-peaks (in red) that cannot be intramolecular contacts because this would require magnetization transfer over more than 10 Å. Rather, these cross-peaks correspond to intermolecular contacts between different teixobactin molecules and agree with the formation of antiparallel β-sheets. Spectra were acquired at 950 MHz and 18 kHz MAS using 50 (purple) and 600 (cyan) ms CC magnetization transfer. a) and b) show zooms into spectral Cα-Cα and Cα-Cβ regions, while c) shows long-range interactions with the aromatic sidechain of Phe1.
Extended Data Fig. 3 Intermolecular teixobactin – teixobactin interactions seen by ssNMR. Solid-state NMR 13C-13C PARISxy spectra of the 13C,15N-teixobactin – Lipid II complex in membranes show several cross-peaks (in red) that cannot be intramolecular contacts because this would require magnetization transfer over more than 10 Å. Rather, these cross-peaks correspond to intermolecular contacts between different teixobactin molecules and agree with the formation of antiparallel β-sheets. Spectra were acquired at 950 MHz and 18 kHz MAS using 50 (purple) and 600 (cyan) ms CC magnetization transfer. a) and b) show zooms into spectral Cα-Cα and Cα-Cβ regions, while c) shows long-range interactions with the aromatic sidechain of Phe1.CC BY · PubMed Central
Extended Data Fig. 4 Only an antiparallel arrangement agrees with ssNMR data. ssNMR data (Fig. 1d and Extended Data Fig. 3) provide clear evidence that teixobactin molecules oligomerize upon Lipid II binding and form β-sheets. We confirm oligomerization by several microscopy setups in vitro and in vivo. ssNMR data support that teixobactin molecules feature antiparallel arrangement in oligomers. Unambiguous NMR contacts are shown as continuous blue lines; ambiguous contacts are shown as dashed green lines. For ambiguous contacts, only the contact with the shorter average distance is shown (see Supplementary Tables 5–7). a) Parallel arrangement. This can be ruled out because most intermolecular teixobactin – teixobactin ssNMR contacts would be >>10 Å and the interfacial contact between F1 of teixobactin and MurNAc of Lipid II would be >>20 Å, while the maximum threshold for ssNMR 13C-13C distance measurements is ~8 Å. Moreover, the depsi-cycles of adjacent teixobactin molecules would clash. b) Shifted parallel arrangement. Steric clashes are avoided, but intermolecular contacts between teixobactin molecules are heavily violated, and contacts between the N-terminus of teixobactin (F1) and Lipid II (MurNAc) remain heavily violated. In addition, the formation of teixobactin – teixobactin hydrogen bonds is compromised. c) Antiparallel arrangement. All Intermolecular and interfacial ssNMR restraints are fulfilled. There are no steric clashes and intermolecular hydrogen bonds can form. An antiparallel arrangement also agrees with the X-ray structure of a teixobactin-sulfate complex in solution66. Contacts consistent with antiparallel β-sheets are listed in Supplementary Table T4. Note that some contacts are ambiguous: Contacts with S7α could also be with I11Cα. Note that contacts with I11α would also (even more than with S7α) only be feasible in antiparallel sheets. Likewise, interfacial contacts between MurNAc and F1MeN could also be with Q4γ. Here also, the distance Q4γ - MurNAc – would be >>10 Å in parallel β-sheets.
Extended Data Fig. 4 Only an antiparallel arrangement agrees with ssNMR data. ssNMR data (Fig. 1d and Extended Data Fig. 3) provide clear evidence that teixobactin molecules oligomerize upon Lipid II binding and form β-sheets. We confirm oligomerization by several microscopy setups in vitro and in vivo. ssNMR data support that teixobactin molecules feature antiparallel arrangement in oligomers. Unambiguous NMR contacts are shown as continuous blue lines; ambiguous contacts are shown as dashed green lines. For ambiguous contacts, only the contact with the shorter average distance is shown (see Supplementary Tables 5–7). a) Parallel arrangement. This can be ruled out because most intermolecular teixobactin – teixobactin ssNMR contacts would be >>10 Å and the interfacial contact between F1 of teixobactin and MurNAc of Lipid II would be >>20 Å, while the maximum threshold for ssNMR 13C-13C distance measurements is ~8 Å. Moreover, the depsi-cycles of adjacent teixobactin molecules would clash. b) Shifted parallel arrangement. Steric clashes are avoided, but intermolecular contacts between teixobactin molecules are heavily violated, and contacts between the N-terminus of teixobactin (F1) and Lipid II (MurNAc) remain heavily violated. In addition, the formation of teixobactin – teixobactin hydrogen bonds is compromised. c) Antiparallel arrangement. All Intermolecular and interfacial ssNMR restraints are fulfilled. There are no steric clashes and intermolecular hydrogen bonds can form. An antiparallel arrangement also agrees with the X-ray structure of a teixobactin-sulfate complex in solution66. Contacts consistent with antiparallel β-sheets are listed in Supplementary Table T4. Note that some contacts are ambiguous: Contacts with S7α could also be with I11Cα. Note that contacts with I11α would also (even more than with S7α) only be feasible in antiparallel sheets. Likewise, interfacial contacts between MurNAc and F1MeN could also be with Q4γ. Here also, the distance Q4γ - MurNAc – would be >>10 Å in parallel β-sheets.CC BY · PubMed Central
Extended Data Fig. 5 Growth of teixobactin-Lipid II fibrils and membrane deformation captured by HS-AFM. a) Snapshots of time-lapse High-Speed Atomic Force Microscopy (HS-AFM) images following the assembly of teixobactin-Lipid II fibrils on a supported DOPC-lipid bilayer in the presence of 1% Lipid II and 1 µM teixobactin. Image acquisition rate 2 frames/second. b) Snapshots of time-lapse HS-AFM images following the assembly of fibrils on a supported lipid bilayer in the presence of 4% Lipid II and 1 µM teixobactin. Fibril growth on the membrane (0-91 s), absorption of the fibrils (91-350 s), and membrane deformation (1070-1468 s) can be observed. In the last two frames both fibrillar sheets on top (yellow colour) and below (dark brown/black colour) the membrane can be observed. Image acquisition rate 2 frames/second. c) Snapshot images from the neighbouring regions (of the area panel b) that were not previously imaged. This shows that membrane deformation occurred over the entire surface independent of the imaging. d) Fibrillar sheet below the membrane, observed 50 min after the addition of 800 nM teixobactin on a membrane containing 1% Lipid II. The dark brown structures spanning from the top of the image to the bottom are the fibrils below the membrane (yellow color). The black area at the bottom right is a hole in the membrane where the mica surface can be observed. e) Zoom-in of white rectangle in panel d) including height profile of the fibrillar sheet below the membrane. The red arrows point at the tops of a few adjacent fibrils. f) Histogram of peak-to-peak distance between fibrils in sheets on top of the membrane and below the membrane surface revealing a stable ~8 nm peak-to-peak distance after the fibrils have descended into the surface. Source data
Extended Data Fig. 5 Growth of teixobactin-Lipid II fibrils and membrane deformation captured by HS-AFM. a) Snapshots of time-lapse High-Speed Atomic Force Microscopy (HS-AFM) images following the assembly of teixobactin-Lipid II fibrils on a supported DOPC-lipid bilayer in the presence of 1% Lipid II and 1 µM teixobactin. Image acquisition rate 2 frames/second. b) Snapshots of time-lapse HS-AFM images following the assembly of fibrils on a supported lipid bilayer in the presence of 4% Lipid II and 1 µM teixobactin. Fibril growth on the membrane (0-91 s), absorption of the fibrils (91-350 s), and membrane deformation (1070-1468 s) can be observed. In the last two frames both fibrillar sheets on top (yellow colour) and below (dark brown/black colour) the membrane can be observed. Image acquisition rate 2 frames/second. c) Snapshot images from the neighbouring regions (of the area panel b) that were not previously imaged. This shows that membrane deformation occurred over the entire surface independent of the imaging. d) Fibrillar sheet below the membrane, observed 50 min after the addition of 800 nM teixobactin on a membrane containing 1% Lipid II. The dark brown structures spanning from the top of the image to the bottom are the fibrils below the membrane (yellow color). The black area at the bottom right is a hole in the membrane where the mica surface can be observed. e) Zoom-in of white rectangle in panel d) including height profile of the fibrillar sheet below the membrane. The red arrows point at the tops of a few adjacent fibrils. f) Histogram of peak-to-peak distance between fibrils in sheets on top of the membrane and below the membrane surface revealing a stable ~8 nm peak-to-peak distance after the fibrils have descended into the surface. Source dataCC BY · PubMed Central
Extended Data Fig. 6 HS-AFM control experiments. Control experiments show that fibrils solely form in the presence of both teixobactin and Lipid II, and that membrane height at positions without fibrils remains unaltered. a) Snapshots of HS-AFM images of supported lipid bilayer with Lipid II (1 %) in the absence of teixobactin. Image acquisition rate 0.5 frames/second. b) Histogram representing membrane heights before and after (30 min) the addition of teixobactin. c) HS-AFM images of a DOPC-membrane without Lipid II in the absence of teixobactin (0 min) and 50 min after the addition of different concentrations of teixobactin. Source data
Extended Data Fig. 6 HS-AFM control experiments. Control experiments show that fibrils solely form in the presence of both teixobactin and Lipid II, and that membrane height at positions without fibrils remains unaltered. a) Snapshots of HS-AFM images of supported lipid bilayer with Lipid II (1 %) in the absence of teixobactin. Image acquisition rate 0.5 frames/second. b) Histogram representing membrane heights before and after (30 min) the addition of teixobactin. c) HS-AFM images of a DOPC-membrane without Lipid II in the absence of teixobactin (0 min) and 50 min after the addition of different concentrations of teixobactin. Source dataCC BY · PubMed Central
Extended Data Fig. 7 ssNMR analysis of the teixobactin – Lipid II interface and characterisation of the pentapeptide of Lipid II in the complex. a) 2D CC PARISxy33 spectra of 13C,15N-teixobactin and 13C,15N-Lipid II in DOPC liposomes acquired at 950 MHz using 150 (red) and 300 ms (blue) magnetization transfer. Interfacial contacts are labelled in green. Lipid II pentapeptide signals are marked with an asterisk, e.g., A1*αβ. b) Zoom into the spectrum acquired with 300 ms magnetization transfer. Only MurNAc shows a contact with Ala9 of teixobactin depsi-cycle, while GlcNAc is distal from the interface. c) Experiments that rely on scalar couplings can be used to detect highly mobile residues. The spectrum is based on a scalar 2D 13C13C TOBSY35 ssNMR experiment with the complex formed by 13C,15N-teixobactin and 13C,15N-Lipid II in liposomes acquired at 950 MHz, 8 kHz MAS, and 300 K temperature. Teixobactin and the Lipid II sugars are not visible in the spectrum (cyan). Conversely, the last three residues (Lys3-Ala4-Ala5) of the Lipid II pentapeptide give scalar signals indicating high mobility and no involvement in complex formation. While signals of E2 (red) of Lipid II are absent for the complex formed by natural teixobactin, these signals were previously18 observed on R4L10-teixobactin (grey). Likewise, the K3αγ signal of the third pentapeptide residue was only visible for R4L10-teixobactin. This is due to the tight binding of End10 of natural teixobactin to MurNAc, resulting in the rigidification of the first two residues of the pentapeptide. Conversely, pentapeptide residues A1 and E2 give well-defined dipolar signals for natural teixobactin. Residue K3 is visible in both dipolar and scalar spectra, implying intermediate mobility. d) A representative ITC thermogram showing a strong binding of natural teixobactin to Lipid II containing DOPC LUVs. All experiments were performed in triplicates. All the parameter values are shown as mean +/− SD.
Extended Data Fig. 7 ssNMR analysis of the teixobactin – Lipid II interface and characterisation of the pentapeptide of Lipid II in the complex. a) 2D CC PARISxy33 spectra of 13C,15N-teixobactin and 13C,15N-Lipid II in DOPC liposomes acquired at 950 MHz using 150 (red) and 300 ms (blue) magnetization transfer. Interfacial contacts are labelled in green. Lipid II pentapeptide signals are marked with an asterisk, e.g., A1*αβ. b) Zoom into the spectrum acquired with 300 ms magnetization transfer. Only MurNAc shows a contact with Ala9 of teixobactin depsi-cycle, while GlcNAc is distal from the interface. c) Experiments that rely on scalar couplings can be used to detect highly mobile residues. The spectrum is based on a scalar 2D 13C13C TOBSY35 ssNMR experiment with the complex formed by 13C,15N-teixobactin and 13C,15N-Lipid II in liposomes acquired at 950 MHz, 8 kHz MAS, and 300 K temperature. Teixobactin and the Lipid II sugars are not visible in the spectrum (cyan). Conversely, the last three residues (Lys3-Ala4-Ala5) of the Lipid II pentapeptide give scalar signals indicating high mobility and no involvement in complex formation. While signals of E2 (red) of Lipid II are absent for the complex formed by natural teixobactin, these signals were previously18 observed on R4L10-teixobactin (grey). Likewise, the K3αγ signal of the third pentapeptide residue was only visible for R4L10-teixobactin. This is due to the tight binding of End10 of natural teixobactin to MurNAc, resulting in the rigidification of the first two residues of the pentapeptide. Conversely, pentapeptide residues A1 and E2 give well-defined dipolar signals for natural teixobactin. Residue K3 is visible in both dipolar and scalar spectra, implying intermediate mobility. d) A representative ITC thermogram showing a strong binding of natural teixobactin to Lipid II containing DOPC LUVs. All experiments were performed in triplicates. All the parameter values are shown as mean +/− SD.CC BY · PubMed Central
Extended Data Fig. 8 The membrane topology of the complex. a) and b) show mobility-edited (‘T2-filtered’) 1H(1H)13C ssNMR experiments37 of 13C,15N-teixobactin bound to Lipid II. Here, magnetization from mobile water and lipid molecules is transferred to the rigid teixobactin protons via 1H-1H mixing and eventually transferred to the 13C nuclei of teixobactin via a short cross polarisation step (200 μs). A 2D implementation of this experiment demonstrates that the sidechains of Ile2, Ile5, and Ile6 partition into the membrane whereas the End10 sidechain remains in the water phase. Spectra were measured at 700 MHz, 16.5 kHz MAS, and 300 K sample temperature. a) Control: 1D 1H(1H)13C ssNMR spectrum (10240 scans) using a T2-filter of 2.5 ms without transfer to teixobactin (0 ms 1H-1H mixing). All signals relate to lipids, demonstrating the effectiveness of the T2-filter. b) 2D 1H(1H)13C ssNMR spectrum using 2.5 ms T2 filter (8192 scans) and 5 ms 1H-1H mixing. c) 13C cross-polarization spectrum (200 μs contact time) of Lipid II-bound teixobactin. d) Normalized relative signal intensities of the correlations with water (blue) and lipid (brown) protons for several residues. For each residue, peak areas were integrated using Topspin 4.06. Signals that were used for the integration are annotated. e) Membrane topology visualized with the ssNMR structure. Hydrophilic (in blue) and hydrophobic (brown) residues are sharply separated above and below the β-sheet, respectively. Note that the N-terminal Phe1 at the rim of the β-sheet has higher water accessibility, potentially because of fraying and mobility effects. f) Normalized contacts to water and lipid-tails, back-calculated over the MD simulation based on our ssNMR structure. g) Membrane topology using a different structure of the calculated ssNMR ensemble. As shown in Extended Data Fig. 9a, the despi-cycle is plastic and can point up towards the water phase, bringing End10 well-above the membrane-plane. Source data
Extended Data Fig. 8 The membrane topology of the complex. a) and b) show mobility-edited (‘T2-filtered’) 1H(1H)13C ssNMR experiments37 of 13C,15N-teixobactin bound to Lipid II. Here, magnetization from mobile water and lipid molecules is transferred to the rigid teixobactin protons via 1H-1H mixing and eventually transferred to the 13C nuclei of teixobactin via a short cross polarisation step (200 μs). A 2D implementation of this experiment demonstrates that the sidechains of Ile2, Ile5, and Ile6 partition into the membrane whereas the End10 sidechain remains in the water phase. Spectra were measured at 700 MHz, 16.5 kHz MAS, and 300 K sample temperature. a) Control: 1D 1H(1H)13C ssNMR spectrum (10240 scans) using a T2-filter of 2.5 ms without transfer to teixobactin (0 ms 1H-1H mixing). All signals relate to lipids, demonstrating the effectiveness of the T2-filter. b) 2D 1H(1H)13C ssNMR spectrum using 2.5 ms T2 filter (8192 scans) and 5 ms 1H-1H mixing. c) 13C cross-polarization spectrum (200 μs contact time) of Lipid II-bound teixobactin. d) Normalized relative signal intensities of the correlations with water (blue) and lipid (brown) protons for several residues. For each residue, peak areas were integrated using Topspin 4.06. Signals that were used for the integration are annotated. e) Membrane topology visualized with the ssNMR structure. Hydrophilic (in blue) and hydrophobic (brown) residues are sharply separated above and below the β-sheet, respectively. Note that the N-terminal Phe1 at the rim of the β-sheet has higher water accessibility, potentially because of fraying and mobility effects. f) Normalized contacts to water and lipid-tails, back-calculated over the MD simulation based on our ssNMR structure. g) Membrane topology using a different structure of the calculated ssNMR ensemble. As shown in Extended Data Fig. 9a, the despi-cycle is plastic and can point up towards the water phase, bringing End10 well-above the membrane-plane. Source dataCC BY · PubMed Central
Extended Data Fig. 9 Structures of the drug in the complex superimpose well. a) Superposition of 25 structures (lines and cartoon representations) of four teixobactin bound to four Lipid II molecules (not shown). The average backbone RMSD is 2.3 +/− 0.6 Å (from the average structure). We calculated the structure of four teixobactin and four (4 x 4) Lipid II (quartet) instead of a 2 x 2 unit (dimer) to obtain a better description of the intermolecular interactions. In a dimer, half of the backbone hydrogen bonds cannot form, and the repetitive interactions between sidechains along the fibril axis are truncated. The description is improved in a quartet, in which intermolecular interactions are complete for the inner dimer, which is the reason why the inner teixobactin molecules are better defined than the outer ones. b) The network of hydrophilic residues on the water-exposed side creates opportunities for hydrogen bonding between the sidechains of D-Gln4. Moreover, MD simulations suggest that the hydroxyl-group of Ser7 favourably interacts with the Lipid II PPi group. c) Green lines show hydrogen bonds and favourable electrostatics between End10 and MurNAcC6OH. The unambiguous, short ssNMR contacts of End10Cζ with MurNAcC5 and MurNAcC6 are highlighted with thick continuous lines. Distances in the NMR structure: Ala9Cβ – MurNAcC1 = 4.0 Å. Ala9Cβ – MurNAcC3 = 6.4 Å. End10Cζ – MurNAcC6 = 4.1 Å. End10Cζ – MurNAcC5 = 5.2 Å. End10Cζ – MurNAcC4 = 5.1 Å. End10Cζ – MurNAcC3 = 6.4 Å. End10Cζ – MurNAcC1 = 6.6 Å. End10Cδ – MurNAcC5 = 5.7 Å. End10Cγ – MurNAcC5 = 5.9 Å. End10Cγ – MurNAcC3 = 7.0 Å. End10Cγ – MurNAcC1 = 6.6 Å. End10Cβ – MurNAcC1 = 5.6 Å. Phe1MeN – MurNAcC3 = 6.0 Å. Phe1MeN – MurNAcC5 = 7.2 Å. End10Nδ – MurNAcC6OH = 2.6 Å (hydrogen bond/electrostatic interaction). End10Nε – MurNAcC6OH = 2.2 Å (hydrogen bond/electrostatic interaction). End10Nζ – MurNAcC6OH = 3.2 Å (hydrogen bond/electrostatic interaction).
Extended Data Fig. 9 Structures of the drug in the complex superimpose well. a) Superposition of 25 structures (lines and cartoon representations) of four teixobactin bound to four Lipid II molecules (not shown). The average backbone RMSD is 2.3 +/− 0.6 Å (from the average structure). We calculated the structure of four teixobactin and four (4 x 4) Lipid II (quartet) instead of a 2 x 2 unit (dimer) to obtain a better description of the intermolecular interactions. In a dimer, half of the backbone hydrogen bonds cannot form, and the repetitive interactions between sidechains along the fibril axis are truncated. The description is improved in a quartet, in which intermolecular interactions are complete for the inner dimer, which is the reason why the inner teixobactin molecules are better defined than the outer ones. b) The network of hydrophilic residues on the water-exposed side creates opportunities for hydrogen bonding between the sidechains of D-Gln4. Moreover, MD simulations suggest that the hydroxyl-group of Ser7 favourably interacts with the Lipid II PPi group. c) Green lines show hydrogen bonds and favourable electrostatics between End10 and MurNAcC6OH. The unambiguous, short ssNMR contacts of End10Cζ with MurNAcC5 and MurNAcC6 are highlighted with thick continuous lines. Distances in the NMR structure: Ala9Cβ – MurNAcC1 = 4.0 Å. Ala9Cβ – MurNAcC3 = 6.4 Å. End10Cζ – MurNAcC6 = 4.1 Å. End10Cζ – MurNAcC5 = 5.2 Å. End10Cζ – MurNAcC4 = 5.1 Å. End10Cζ – MurNAcC3 = 6.4 Å. End10Cζ – MurNAcC1 = 6.6 Å. End10Cδ – MurNAcC5 = 5.7 Å. End10Cγ – MurNAcC5 = 5.9 Å. End10Cγ – MurNAcC3 = 7.0 Å. End10Cγ – MurNAcC1 = 6.6 Å. End10Cβ – MurNAcC1 = 5.6 Å. Phe1MeN – MurNAcC3 = 6.0 Å. Phe1MeN – MurNAcC5 = 7.2 Å. End10Nδ – MurNAcC6OH = 2.6 Å (hydrogen bond/electrostatic interaction). End10Nε – MurNAcC6OH = 2.2 Å (hydrogen bond/electrostatic interaction). End10Nζ – MurNAcC6OH = 3.2 Å (hydrogen bond/electrostatic interaction).CC BY · PubMed Central
Extended Data Fig. 10 Oligomeric teixobactin – Lipid II complexes thin the membrane by displacing lipids. a) MD simulation of the teixobactin – Lipid II complex after 248 ns of free evolution. The membrane surface is shown in light orange, and the phosphate-atoms of phospholipids are shown as orange spheres. The complex is not shown for clarity but is localised centrally on the upper membrane leaflet where membrane distortions are most pronounced. Lipids below the complex, for which the tails show major distortions, are shown in stick representation and in different colours. Polar headgroups of distorted lipids avoid the hydrophobic underbelly of the teixobactin β-sheet, which causes lipid tails to orient almost parallel to the membrane plane, resulting in a thinning of the membrane. The high concentration of non-lamellar C55-tails of Lipid II (not shown) likely also contributes to membrane perturbations at the complex site. Permeability assays with Bacillus subtilis: Bacterial assays with Bacillus subtilis were used to study b) membrane depolarization and c) membrane damage caused by teixobactin (in blue), nisin (magenta), plectasin (orange), vancomycin (yellow), and sans antibiotic (gray). Source data are provided as a Source Data file. Source data
Extended Data Fig. 10 Oligomeric teixobactin – Lipid II complexes thin the membrane by displacing lipids. a) MD simulation of the teixobactin – Lipid II complex after 248 ns of free evolution. The membrane surface is shown in light orange, and the phosphate-atoms of phospholipids are shown as orange spheres. The complex is not shown for clarity but is localised centrally on the upper membrane leaflet where membrane distortions are most pronounced. Lipids below the complex, for which the tails show major distortions, are shown in stick representation and in different colours. Polar headgroups of distorted lipids avoid the hydrophobic underbelly of the teixobactin β-sheet, which causes lipid tails to orient almost parallel to the membrane plane, resulting in a thinning of the membrane. The high concentration of non-lamellar C55-tails of Lipid II (not shown) likely also contributes to membrane perturbations at the complex site. Permeability assays with Bacillus subtilis: Bacterial assays with Bacillus subtilis were used to study b) membrane depolarization and c) membrane damage caused by teixobactin (in blue), nisin (magenta), plectasin (orange), vancomycin (yellow), and sans antibiotic (gray). Source data are provided as a Source Data file. Source dataCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 1

  1. O’Neill, J. Antimicrobial resistance: tackling a crisis for the health and wealth of nations. jpiamr http://www.jpiamr.eu/wpcontent/uploads/2014/12/AMR-Review-Paper-Tackling-a-crisis-for-the-health-and-wealth-of-nations_1-2.pdf (2014).

Cited by

Sorted by citations

See all 201

Related

See all

Metadata

Authors
Shukla, R., Lavore, F., Maity, S., Derks, M. G. N., Jones, C. R., Vermeulen, B. J. A., Melcrová, A., Morris, M. A., Becker, L. M., Wang, X., Kumar, R., Medeiros-Silva, J., van Beekveld, R. A. M., Bonvin, A. M. J. J., Lorent, J. H., Lelli, M., Nowick, J. S., MacGillavry, H. D., Peoples, A. J., Spoering, A. L., Ling, L. L., Hughes, D. E., Roos, W. H., Breukink, E., Lewis, K., & Weingarth, M.
Affiliations
  • Utrecht University (NL)
  • University of Groningen (NL)
  • University of California, Irvine (US)
  • Consorzio Interuniversitario Risonanze Magnetiche di Metallo Proteine (IT)
  • NovoBiotic Pharmaceuticals (US)
  • Northeastern University (US)
Corresponding author
Weingarth, Markus
ORCID
Venue
Nature
Publisher
Springer Science and Business Media LLC
Published
2022-08-03
Volume
608
Issue
7922
Pages
390-396
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
DOI
10.1038/s41586-022-05019-y
PMC
PMC9365693
OpenAlex
W4289638129
Cited by
201
References
67
Citations by year
2026: 30 · 2025: 65 · 2024: 53 · 2023: 48 · 2022: 5
FWCI
26.73
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • Utrecht University Repository (Utrecht University) (submittedVersion, other-oa)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
  • Florence Research (University of Florence) (submittedVersion, other-oa)
  • University of Groningen research database (University of Groningen / Centre for Information Technology) (submittedVersion, cc-by)
Funders
  • Dipartimenti di Eccellenza (Dipartimenti di Eccellenza 2018-2022, 2018–2022)
  • European Commission (675728, 101045485, 823830, 2018-2022, HORIZON2020, 871037)
  • Nederlandse Organisatie voor Wetenschappelijk Onderzoek (711.018.001, 680-91-007, 723.014.003, 718.015.001, 184.032.207)
  • Università degli Studi di Firenze
Topics
  • Life Sciences › Immunology and Microbiology › Microbiology › Antimicrobial Peptides and Activities
  • Life Sciences › Immunology and Microbiology › Applied Microbiology and Biotechnology › Antimicrobial agents and applications
  • Life Sciences › Immunology and Microbiology › Applied Microbiology and Biotechnology › Microbial Natural Products and Biosynthesis
Keywords
teixobactin, lipid II, peptidoglycan precursor, antibiotic mechanism of action, bacterial cell envelope, membrane damage, antimicrobial resistance
MeSH terms
Sugars, Alanine, Anti-Bacterial Agents, Bacteria, Cell Membrane, Cell Wall, Humans, Lipids, Microbial Sensitivity Tests, Diphosphates, Pyrrolidines, Protein Structure, Secondary, Microscopy, Atomic Force, Nuclear Magnetic Resonance, Biomolecular, Drug Resistance, Bacterial, Depsipeptides, Microbial Viability, Molecular Dynamics Simulation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review