Structural basis for nutrient acquisition by dominant members of the human gut microbiota (original) (raw)

Accession codes

Primary accessions

Protein Data Bank

References

  1. Bäckhed, F., Ley, R. E., Sonnenburg, J. L., Peterson, D. A. & Gordon, J. I. Host-bacterial mutualism in the human intestine. Science 307, 1915–1920 (2005)
    Article ADS Google Scholar
  2. Marchesi, J. R. et al. The gut microbiota and host health: a new clinical frontier. Gut 65, 330–339 (2015)
    Article Google Scholar
  3. Koropatkin, N. M., Cameron, E. A. & Martens, E. C. How glycan metabolism shapes the human gut microbiota. Nat. Rev. Microbiol. 10, 323–335 (2012)
    Article CAS Google Scholar
  4. Sonnenburg, E. D. et al. Specificity of polysaccharide use in intestinal Bacteroides species determines diet-induced microbiota alterations. Cell 141, 1241–1252 (2010)
    Article CAS Google Scholar
  5. Rogowski, A. et al. Glycan complexity dictates microbial resource allocation in the large intestine. Nat. Commun. 6, http://dx.doi.org/10.1038/ncomms8481 (2015)
  6. Larsbrink, J. et al. A discrete genetic locus confers xyloglucan metabolism in select human gut Bacteroidetes. Nature 506, 498–502 (2014)
    Article ADS CAS Google Scholar
  7. Cuskin, F. et al. Human gut Bacteroidetes can utilize yeast mannan through a selfish mechanism. Nature 517, 165–169 (2015)
    Article ADS CAS Google Scholar
  8. Martens, E. C., Koropatkin, N. M., Smith, T. J. & Gordon, J. I. Complex glycan catabolism by the human gut microbiota: the Bacteroidetes Sus-like paradigm. J. Biol. Chem. 284, 24673–24677 (2009)
    Article CAS Google Scholar
  9. D’Elia, J. N. & Salyers, A. A. Contribution of a neopullulanase, a pullulanase, and an α-glucosidase to growth of Bacteroides thetaiotaomicron on starch. J. Bacteriol. 178, 7173–7179 (1996)
    Article Google Scholar
  10. Reeves, A. R., Wang, G. R. & Salyers, A. A. Characterization of four outer membrane proteins that play a role in utilization of starch by Bacteroides thetaiotaomicron. J. Bacteriol. 179, 643–649 (1997)
    Article CAS Google Scholar
  11. Koropatkin, N. M., Martens, E. C., Gordon, J. I. & Smith, T. J. Starch catabolism by a prominent human gut symbiont is directed by the recognition of amylose helices. Structure 16, 1105–1115 (2008)
    Article CAS Google Scholar
  12. Shipman, J. A., Berleman, J. E. & Salyers, A. A. Characterization of four outer membrane proteins involved in binding starch to the cell surface of Bacteroides thetaiotaomicron. J. Bacteriol. 182, 5365–5372 (2000)
    Article CAS Google Scholar
  13. Noinaj, N., Guillier, M., Barnard, T. J. & Buchanan, S. K. TonB-dependent transporters: regulation, structure, and function. Annu. Rev. Microbiol. 64, 43–60 (2010)
    Article CAS Google Scholar
  14. Martens, E. C. et al. Recognition and degradation of plant cell wall polysaccharides by two human gut symbionts. PLoS Biol. 9, e1001221 (2011)
    Article CAS Google Scholar
  15. Martens, E. C., Chiang, H. C. & Gordon, J. I. Mucosal glycan foraging enhances fitness and transmission of a saccharolytic human gut bacterial symbiont. Cell Host Microbe 4, 447–457 (2008)
    Article CAS Google Scholar
  16. Bolam, D. N. & Koropatkin, N. M. Glycan recognition by the Bacteroidetes Sus-like systems. Curr. Opin. Struct. Biol. 22, 563–569 (2012)
    Article CAS Google Scholar
  17. Krissinel, E. & Henrick, K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797 (2007)
    Article CAS Google Scholar
  18. Sonnenburg, J. L. et al. Glycan foraging in vivo by an intestine-adapted bacterial symbiont. Science 307, 1955–1959 (2005)
    Article ADS CAS Google Scholar
  19. Kumar, S. & Engelberg-Kulka, H. Quorum sensing peptides mediating interspecies bacterial cell death as a novel class of antimicrobial agents. Curr. Opin. Microbiol. 21, 22–27 (2014)
    Article CAS Google Scholar
  20. Mohammad, M. M., Howard, K. R. & Movileanu, L. Redesign of a plugged β-barrel membrane protein. J. Biol. Chem. 286, 8000–8013 (2011)
    Article CAS Google Scholar
  21. Koropatkin, N., Martens, E. C., Gordon, J. I. & Smith, T. J. Structure of a SusD homologue, BT1043, involved in mucin O-glycan utilization in a prominent human gut symbiont. Biochemistry 48, 1532–1542 (2009)
    Article CAS Google Scholar
  22. Shultis, D. D., Purdy, M. D., Banchs, C. N. & Wiener, M. C. Outer membrane active transport: structure of the BtuB:TonB complex. Science 312, 1396–1399 (2006)
    Article ADS CAS Google Scholar
  23. Pawelek, P. D. et al. Structure of TonB in complex with FhuA, E. coli outer membrane receptor. Science 312, 1399–1402 (2006)
    Article ADS CAS Google Scholar
  24. Jordan, L. D. et al. Energy-dependent motion of TonB in the Gram-negative bacterial inner membrane. Proc. Natl Acad. Sci. USA 110, 11553–11558 (2013)
    Article ADS CAS Google Scholar
  25. Celia, H. et al. Structural insight into the role of the Ton complex in energy transduction. Nature 538, 60–65 (2016)
    Article ADS CAS Google Scholar
  26. Van Duyne, G. D., Standaert, R. F., Karplus, P. A., Schreiber, S. L. & Clardy, J. Atomic structures of the human immunophilin FKBP-12 complexes with FK506 and rapamycin. J. Mol. Biol. 229, 105–124 (1993)
    Article CAS Google Scholar
  27. Kabsch, W. XDS. Acta Crystallogr. D 66, 125–132 (2010)
    Article CAS Google Scholar
  28. Evans, P. R. & Murshudov, G. N. How good are my data and what is the resolution? Acta Crystallogr. D 69, 1204–1214 (2013)
    Article CAS Google Scholar
  29. Evans, P. Scaling and assessment of data quality. Acta Crystallogr. D 62, 72–82 (2006)
    Article Google Scholar
  30. Pape, T. & Schneider, T. R. HKL2MAP: a graphical user interface for phasing with SHELX programs. J. Appl. Crystallogr. 37, 843–844 (2004)
    Article CAS Google Scholar
  31. Sheldrick, G. M. Experimental phasing with SHELXC/D/E: combining chain tracing with density modification. Acta Crystallogr. D 66, 479–485 (2010)
    Article CAS Google Scholar
  32. Cowtan, K. The Buccaneer software for automated model building. 1. Tracing protein chains. Acta Crystallogr. D 62, 1002–1011 (2006)
    Article Google Scholar
  33. Vagin, A. & Teplyakov, A. MOLREP: an automated program for molecular replacement. J. Appl. Crystallogr. 30, 1022–1025 (1997)
    Article CAS Google Scholar
  34. Vagin, A. A. et al. REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for its use. Acta Crystallogr. D 60, 2184–2195 (2004)
    Article Google Scholar
  35. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132 (2004)
    Article Google Scholar
  36. Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D 66, 12–21 (2010)
    Article CAS Google Scholar
  37. McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007)
    Article CAS Google Scholar
  38. Buchanan, S. K. et al. Crystal structure of the outer membrane active transporter FepA from Escherichia coli. Nat. Struct. Biol. 6, 56–63 (1999)
    Article CAS Google Scholar
  39. Adams, P. D. et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr. D 58, 1948–1954 (2002)
    Article Google Scholar
  40. Terwilliger, T. C. Maximum-likelihood density modification. Acta Crystallogr. D 56, 965–972 (2000)
    Article CAS Google Scholar
  41. The PyMOL Molecular Graphics System, Version 1.7.4 Schrödinger, LLC
  42. Schwarz, G., Danelon, C. & Winterhalter, M. On translocation through a membrane channel via an internal binding site: kinetics and voltage dependence. Biophys. J. 84, 2990–2998 (2003)
    Article ADS CAS Google Scholar
  43. Montal, M. & Mueller, P. Formation of bimolecular membranes from lipid monolayers and a study of their electrical properties. Proc. Natl Acad. Sci. USA 69, 3561–3566 (1972)
    Article ADS CAS Google Scholar
  44. Pronk, S. et al. GROMACS 4.5: a high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 29, 845–854 (2013)
    Article CAS Google Scholar
  45. Klauda, J. B. et al. Update of the CHARMM all-atom additive force field for lipids: validation on six lipid types. J. Phys. Chem. B 114, 7830–7843 (2010)
    Article CAS Google Scholar
  46. Best, R. B. et al. Optimization of the additive CHARMM all-atom protein force field targeting improved sampling of the backbone φ, ψ and side-chain χ1 and χ2 dihedral angles. J. Chem. Theory Comput. 8, 3257–3273 (2012)
    Article CAS Google Scholar
  47. Pothula, K. R., Solano, C. J. & Kleinekathöfer, U. Simulations of outer membrane channels and their permeability. Biochim. Biophys. Acta 1858 (7 Pt B), 1760–1771 (2016)
    Article CAS Google Scholar
  48. Hess, B., Bekker, H., Berendsen, H. J. C. & Johannes, G. E. M. F. LINCS: a linear constraint solver for molecular simulations. J. Comput. Chem. 18, 1463–1472 (1997)
    Article CAS Google Scholar
  49. Eswar, N. et al. Comparative protein structure modeling using Modeller. Curr. Protoc. Bioinformatics Chapter 5, Unit 6 (2006)
    PubMed Google Scholar
  50. Jo, S., Kim, T., Iyer, V. G. & Im, W. CHARMM-GUI: a web-based graphical user interface for CHARMM. J. Comput. Chem. 29, 1859–1865 (2008)
    Article CAS Google Scholar
  51. Coyne, M. J. et al. Phylum-wide general protein O-glycosylation system of the Bacteroidetes. Mol. Microbiol. 88, 772–783 (2013)
    Article CAS Google Scholar
  52. Holm, L. & Rosenström, P. Dali server: conservation mapping in 3D. Nucleic Acids Res. 38, W545–W549 (2010)
    Article CAS Google Scholar

Download references

Acknowledgements

We would like to thank J. Gray for B. theta outer membrane protein identification, R. Lewis for critical reading of the manuscript and S. Buchanan for the pET9 expression vector. We thank O. Davies for help with SEC–MALS analysis. We are also indebted to the staff at beamlines I24, I04 and I03 of the Diamond Light Source UK for beam time (proposal mx9948) and assistance with data collection. A.J.G. acknowledges support from the Barbour Foundation. S.B.P. was funded by EU FP7-PEOPLE-2013-ITN Translocation network Nr. 607694. The research of K.R.P., U.K., M.W. and B.v.d.B. has received support from the Innovative Medicines Initiatives Joint Undertaking under Grant Agreement No. 115525, resources which are composed of financial contributions from the European Union’s seventh framework programme (FP7/2007-2013) and European Federation of Pharmaceutical Industries and Associations companies in-kind contribution.

Author information

Author notes

  1. Amy J. Glenwright and Karunakar R. Pothula: These authors contributed equally to this work.

Authors and Affiliations

  1. Institute for Cell and Molecular Biosciences, The Medical School, Newcastle University, Newcastle upon Tyne, NE2 4HH, UK
    Amy J. Glenwright, Arnaud Baslé, Susan J. Firbank, Hongjun Zheng, David N. Bolam & Bert van den Berg
  2. Department of Physics & Earth Sciences, Jacobs University Bremen, Bremen, 28759, Germany
    Karunakar R. Pothula & Ulrich Kleinekathöfer
  3. Department of Life Sciences & Chemistry, Jacobs University Bremen, Bremen, 28759, Germany
    Satya P. Bhamidimarri & Mathias Winterhalter
  4. Physical and Theoretical Chemistry Laboratory, University of Oxford, South Parks Road, Oxford, OX1 3QZ, UK
    Dror S. Chorev & Carol V. Robinson

Authors

  1. Amy J. Glenwright
    You can also search for this author inPubMed Google Scholar
  2. Karunakar R. Pothula
    You can also search for this author inPubMed Google Scholar
  3. Satya P. Bhamidimarri
    You can also search for this author inPubMed Google Scholar
  4. Dror S. Chorev
    You can also search for this author inPubMed Google Scholar
  5. Arnaud Baslé
    You can also search for this author inPubMed Google Scholar
  6. Susan J. Firbank
    You can also search for this author inPubMed Google Scholar
  7. Hongjun Zheng
    You can also search for this author inPubMed Google Scholar
  8. Carol V. Robinson
    You can also search for this author inPubMed Google Scholar
  9. Mathias Winterhalter
    You can also search for this author inPubMed Google Scholar
  10. Ulrich Kleinekathöfer
    You can also search for this author inPubMed Google Scholar
  11. David N. Bolam
    You can also search for this author inPubMed Google Scholar
  12. Bert van den Berg
    You can also search for this author inPubMed Google Scholar

Contributions

A.J.G. expressed, purified and crystallized proteins and carried out western blot analysis. K.R.P. and U.K. performed molecular dynamics simulations. S.P.B. and M.W. performed and supervised electrophysiology experiments, respectively. D.S.C. and C.V.R. carried out mass-spectrometry analyses. A.B. collected crystallography data and managed the Newcastle Structural Laboratory. H.Z. expressed and purified BT1762 and performed ITC measurements. S.J.F. and A.B. determined the structure of BT1762. D.N.B. designed research and assisted in writing the paper. B.v.d.B. crystallized proteins and solved crystal structures, designed research and wrote the paper.

Corresponding author

Correspondence toBert van den Berg.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Reviewer Information

Nature thanks B. Henrissat, S. White and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Extended data figures and tables

Extended Data Figure 1 High abundance of SusCD complexes in B. theta outer membrane.

a, SDS–PAGE gel of total outer membranes from E. coli (lanes 1, 2, 5, 6) and B. theta (lanes 3, 4, 7, 8). Rich medium, even lanes; minimal medium, odd lanes. Each lane contains approximately 10 μg protein. Samples 5–8 were boiled. ‘P’ denotes E. coli trimeric porins OmpF/C, which migrate at their monomeric molecular weights (around 35 kDa) only after boiling. Note the relative lack of small-molecule outer membrane diffusion channels (around 30–50 kDa) in B. theta (lanes 7, 8) and the low levels of large outer membrane proteins including TBDTs (70–120 kDa) in E. coli (lanes 5, 6). Purified BT2261–64 complex is shown in lanes 9 (non-boiled) and 10 (boiled). b, Representative ion-exchange chromatogram from three separate experiments of B. theta total outer membrane proteins separated on Resource-Q (6 ml; pH 7.5) after extraction in LDAO (Methods). Peaks A and B were further purified by gel filtration. c, SDS–PAGE gel of purified SusCD complexes from peaks A and B. Numbered gel bands were excised and subjected to identification by mass spectrometry. d, SDS–PAGE gel of purified BT1762–BT1763 complex before (asterisk) and after boiling. The SusCD complexes are highly stable and remain intact in 2% SDS.

Extended Data Figure 2 X-ray crystal structures of the small lipoproteins BT2261 and BT2262.

a, Stereo cartoon of BT2261 within the BT2261–64 complex with rainbow colouring (blue; N terminus). BT2261 is O-glycosylated at Ser117, consistent with the presence of the Bacteroidetes glycosylation motif D-(S/T)-(A/L/V/I/M/T)51. Ser117 is shown as a stick model. _F_o − F_c density within 20 Å of Ser117 is shown as a green mesh contoured at 3.0_σ. Three to four sugar moieties can be observed bound to Ser117. b, Stereo cartoon of soluble BT2262 with rainbow colouring. The protein consists of an N-terminal Ig-like domain and a C-terminal eight-stranded β-barrel. The functions of BT2261 and BT2262 are not clear, but both contain a small C-terminal eight-stranded β-barrel that displays structural similarity to lipid binding domains as judged by DALI52. For BT2262, only one copy with a poorly ordered C-terminal domain is visible in the triclinic structure. Analogous to BT2263, the N-terminal segments of BT2261/BT2262 that lead to the lipid anchors on the N-terminal cysteine residues are visible in the electron density; they are closely associated with SusC and do not appear to be flexible. Structures were determined using data obtained from a single crystal in each case.

Extended Data Figure 3 The oligomeric nature of SusCD complexes is not a consequence of crystal packing.

a–c, Cartoon side views of BT2261–64 complexes rotated by 90° for space groups P1 (a), P212121 (b) and SeMet P21 (c). d, Cartoon side view of BT1762–BT1763 (P212121). The protein backbones are coloured on the same scale by their B-factors (blue; 20 Å2, red; 130 Å2). The grey bars indicate the hydrophobic phase of the outer membrane. Structures were determined using data obtained from a single crystal in each case.

Extended Data Figure 4 BT2261–64 and BT1762–BT1763 form oligomeric complexes.

a, Mass spectrum of BT2261–64 shows two prominent masses corresponding to an octamer and a ligand bound octamer in m/z = 12,000–15,000. b, Mass spectrum of BT1762–BT1763 indicates that these two proteins form dimers and tetramers. The numbers in parentheses on the right are the theoretical masses. c, Analytical gel filtration chromatography of BT2261–64 (blue) and BT1762–BT1763 (green). For comparison, samples were run for soluble horse spleen ferritin (440 kDa; red) and for the membrane protein ammonium transporter Mep2 from Candida albicans (160 kDa; black). The following buffer was specifically for this experiment: 10 mM HEPES/100 mM NaCl/0.12% DM pH 7.5. Column: Superdex-200 Increase 10/300 GL. d, SEC–MALS analysis of BT1762–BT1763. Light scattering (LS) and differential refractive index (dRI) are plotted alongside the fitted total protein-detergent complex molecular weight (diamonds), and constituent protein (pluses) and detergent (crosses) molecular weights, across each peak. BT1762–BT1763 eluted as two species of 499 kDa (protein component 319 kDa, corresponding to a SusCD dimer) and 269 kDa (protein component 214 kDa). Chromatograms shown are from single experiments.

Extended Data Figure 5 Unbiased electron density for the bound ligand in BT2261–64.

Stereo views of simulated annealing omit maps using a starting temperature of 1000 K. a, 2_F_o − F_c maps contoured at 1.5_σ; carve, 2. b, _F_o − F_c map contoured at 3.0_σ; carve, 2. Selected residues contacting ligand are shown (yellow; BT2264/SusC, magenta, BT2263/SusD). Density for at least six amino acid side chains is present (denoted by an asterisk in the 2_F_o − _F_c map). c, Interaction table showing hydrogen-bond distances between the putative peptide ligand backbone and residues in BT2263 and BT2264.

Extended Data Figure 6 Molecular dynamics simulations for BT2261–64.

a, Plots of BT2264 (SusC) Cα r.m.s.d. versus simulation time for holo- and apo-complexes. b, c, Plots of BT2263 (SusD) Cα r.m.s.d. versus simulation time for holo and apo simulations, relative to the starting conformation (b) and after SusD superposition (c). d, e, Plots showing the number of hydrogen bonds between SusC and SusD versus simulation time (d) and between holo-SusCD and the modelled peptide (e). Simulations are numbered as follows: sim1–3, apo-BT2263–BT2264 (dimer); sim7–9, apo-BT2261–64 (tetramer); sim13, apo-(BT2261–64) × 2 (octamer); sim4–6, holo-BT2263–BT2264; sim10–12, holo-BT2261–64; sim14, holo-(BT2261–64) × 2. With the exception of those of the octamer (owing to its very large size), the simulations were repeated three times with different initial atomic velocities to allow sampling in order to obtain a measure of the possible spread in results.

Extended Data Figure 7 Dynamics of the bound peptide during molecular dynamics simulations.

a–c, Side views (left panels) and top views showing the bound peptide in the BT2263–BT2264 dimers (a), BT2261–64 tetramers (b) and the (BT2261–64) × 2 octamer (c). For clarity, only one final conformation for BT2264 (SusC) is shown together with the starting conformation of the peptide (green) and the final peptide conformations after 500 ns of simulation (red). For orientation purposes, the assigned N termini of the peptides are coloured blue.

Extended Data Figure 8 Molecular dynamics simulation root-mean-square-fluctuation analyses.

a, Cα root-mean-square-fluctuation values for SusC in holo- and apo-complexes with the conformational fluctuations of the hinge loop L7 highlighted separately. b, Cα root-mean-square-fluctuation values for SusD in apo and holo simulations. Simulations are numbered as in Extended Data Fig. 6: sim1–3, apo-BT2263–BT2264 (dimer); sim7–9, apo-BT2261–64 (tetramer); sim13, apo-(BT2261–64) × 2 (octamer); sim4–6, holo-BT2263–BT2264; sim10–12, holo-BT2261–64; sim14, holo-(BT2261–64) × 2.

Extended Data Figure 9 Structure of the BT2261–64 apo-octamer after 500 ns of molecular dynamics simulation (sim13), demonstrating the independent bin opening of the two SusCD hubs.

a, Views from the plane of the membrane rotated by 90°. b, View from the extracellular side. For clarity, the SusC and SusD subunits are shown in different colours (yellow and orange for SusC/BT2264, magenta and red for SusD/BT2263). BT2261 and BT2262 are shown in green and blue, respectively. c, Side view of the opened SusCD monomer highlighting the remaining interactions between SusC (yellow) and SusD (magenta) mediated by the SusC hinge loop L7 and loop L8.

Extended Data Figure 10 ITC analysis of levan binding for recombinant BT1762 SusD-like wild-type and mutants.

a, Titration curves from single experiments. Upper panels show raw injection heats of ligand (levan) into protein, lower panels show the integrated binding heats fit to a single set of sites binding model to determine _K_a for all proteins except reduced wild type (10 mM TCEP), W85A and C298A mutants that display no binding. Levan stock solution was between 0.5–2% (w/v) and protein ranged from 50–60 μM. b, Levan affinity of recombinant BT1762 SusD-like wild type and mutant proteins determined by ITC. _K_a values shown are averages and standard deviations from at least two independent titrations. Residue numbering is that of the mature protein (first residue Cys1).

Supplementary information

Supplementary Information

This file contains X-ray Crystallographic Tables showing data collection and refinement statistics for BT2261-64 (Supplementary Table 1), soluble proteins (Supplementary Table 2) and BT1762-63 (Supplementary Table 3). (PDF 50 kb)

Pedal bin opening of BT2263-64 (dimer)

Side-by-side movies showing 500 ns unbiased MD simulations of BT2263-64 in the presence (left; sim6) and absence (right; sim3) of the bound peptide ligand, shown as green spheres. BT2263 (SusD) is coloured magenta, BT2264 (SusC) yellow. The plug domain of SusC is coloured dark blue. (MOV 15918 kb)

Pedal bin opening of BT2261-64 (tetramer)

Side-by-side movies showing 500 ns unbiased MD simulations of BT2261-64 in the presence (left; sim11) and absence (right; sim8) of the bound peptide ligand, shown as green spheres. BT2263 (SusD) is coloured green, BT2264 (SusC) cyan. The small lipoproteins BT2261 and BT2262 are coloured grey and dark blue, respectively. (MOV 17469 kb)

Pedal bin opening of (BT2261-64)x 2 (octamer)

Side-by-side movies showing 500 ns unbiased MD simulations of 2 x (BT2261-64) in the presence (left; sim14) and absence (right; sim13) of the bound peptide ligand, shown as green spheres. BT2263 (SusD) is coloured green, BT2264 (SusC) cyan. The small lipoproteins BT2261 and BT2262 are coloured grey and dark blue, respectively. (MOV 19802 kb)

PowerPoint slides

Rights and permissions

About this article

Cite this article

Glenwright, A., Pothula, K., Bhamidimarri, S. et al. Structural basis for nutrient acquisition by dominant members of the human gut microbiota.Nature 541, 407–411 (2017). https://doi.org/10.1038/nature20828

Download citation