Symmetrical protein cages have evolved to fulfil diverse roles in nature, including compartmentalization and cargo delivery1, and have inspired synthetic biologists to create novel protein assemblies via the precise manipulation of protein–protein interfaces. Despite the impressive array of protein cages produced in the laboratory, the design of inducible assemblies remains challenging2,3. Here we demonstrate an ultra-stable artificial protein cage, the assembly and disassembly of which can be controlled by metal coordination at the protein–protein interfaces. The addition of a gold (i)-triphenylphosphine compound to a cysteine-substituted, 11-mer protein ring triggers supramolecular self-assembly, which generates monodisperse cage structures with masses greater than 2 MDa. The geometry of these structures is based on the Archimedean snub cube and is, to our knowledge, unprecedented. Cryo-electron microscopy confirms that the assemblies are held together by 120 S–Aui–S staples between the protein oligomers, and exist in two chiral forms. The cage shows extreme chemical and thermal stability, yet it readily disassembles upon exposure to reducing agents. As well as gold, mercury(ii) is also found to enable formation of the protein cage. This work establishes an approach for linking protein components into robust, higher-order structures, and expands the design space available for supramolecular assemblies to include previously unexplored geometries.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Data availability

The data that support the findings of this study are available from the corresponding author on reasonable request. The cryo-EM density maps have been deposited in the Electron Microscopy Data Bank under accession codes EMD-6966 (GNP-produced TRAP-cage), EMD-4443 (left-handed TRAP-cage) and EMD-4444 (right-handed TRAP-cage), and the coordinates have been deposited in the Protein Data Bank under accession numbers 6IB3 (left-handed TRAP-cage) and 6IB4 (right-handed TRAP-cage).

Code availability

Custom codes used to compute the optimal arrangement of TRAP rings and to predict paradoxical cages are available from the authors on reasonable request.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Aumiller, W. M., Uchida, M. & Douglas, T. Protein cage assembly across multiple length scales. Chem. Soc. Rev. 47, 3433–3469 (2018).

  2. 2.

    Bale, J. B. et al. Accurate design of megadalton-scale two-component icosahedral protein complexes. Science 353, 389–394 (2016).

  3. 3.

    Yeates, T. O. Geometric principles for designing highly symmetric self-assembling protein nanomaterials. Annu. Rev. Biophys. 46, 23–42 (2017).

  4. 4.

    Padilla, J. E., Colovos, C. & Yeates, T. O. Nanohedra: using symmetry to design self assembling protein cages, layers, crystals, and filaments. Proc. Natl Acad. Sci. USA 98, 2217–2221 (2001).

  5. 5.

    King, N. P. et al. Computational design of self-assembling protein nanomaterials with atomic level accuracy. Science 336, 1171–1174 (2012).

  6. 6.

    Holliday, B. J. & Mirkin, C. A. Strategies for the construction of supramolecular compounds through coordination chemistry. Angew. Chem. Int. Ed. 40, 2022–2043 (2001).

  7. 7.

    Salgado, E. N., Radford, R. J. & Tezcan, F. A. Metal-directed protein self-assembly. Acc. Chem. Res. 43, 661–672 (2010).

  8. 8.

    Brodin, J. D. et al. Metal-directed, chemically tunable assembly of one-, two- and three-dimensional crystalline protein arrays. Nat. Chem. 4, 375–382 (2012).

  9. 9.

    Der, B. S. et al. Metal-mediated affinity and orientation specificity in a computationally designed protein homodimer. J. Am. Chem. Soc. 134, 375–385 (2012).

  10. 10.

    Zhang, W. et al. Self-assembly of glutathione S-transferase into nanowires. Nanoscale 4, 5847–5851 (2012).

  11. 11.

    Huard, D. J. E., Kane, K. M. & Tezcan, F. A. Re-engineering protein interfaces yields copper-inducible ferritin cage assembly. Nat. Chem. Biol. 9, 169–176 (2013).

  12. 12.

    Antson, A. A. et al. The structure of trp RNA-binding attenuation protein. Nature 374, 693–700 (1995).

  13. 13.

    Heddle, J. G., Yokoyama, T., Yamashita, I., Park, S.-Y. & Tame, J. R. H. Rounding up: engineering 12-membered rings from the cyclic 11-mer TRAP. Structure 14, 925–933 (2006).

  14. 14.

    Heddle, J. G. et al. Using the ring-shaped protein TRAP to capture and confine gold nanodots on a surface. Small 3, 1950–1956 (2007).

  15. 15.

    Miranda, F. F. et al. A self-assembled protein nanotube with high aspect ratio. Small 5, 2077–2084 (2009).

  16. 16.

    Malay, A. D. et al. Gold nanoparticle-induced formation of artificial protein capsids. Nano Lett. 12, 2056–2059 (2012).

  17. 17.

    Imamura, M. et al. Probing structural dynamics of an artificial protein cage using high-speed atomic force microscopy. Nano Lett. 15, 1331–1335 (2015).

  18. 18.

    Young, G. et al. Quantitative mass imaging of single biological macromolecules. Science 360, 423–427 (2018).

  19. 19.

    Häkkinen, H. The gold–sulfur interface at the nanoscale. Nat. Chem. 4, 443–455 (2012).

  20. 20.

    Bhabak, K. P., Bhuyan, B. J. & Mugesh, G. Bioinorganic and medicinal chemistry: aspects of gold(i)-protein complexes. Dalton Trans. 40, 2099–2111 (2011).

  21. 21.

    Urig, S. et al. Undressing of phosphine gold(i) complexes as irreversible inhibitors of human disulfide reductases. Angew. Chem. Int. Ed. 45, 1881–1886 (2006).

  22. 22.

    Rulíšek, L. & Vondrásek, J. Coordination geometries of selected transition metal ions (Co2+, Ni2+, Cu2+, Zn2+, Cd2+, and Hg2+) in metalloproteins. J. Inorg. Biochem. 71, 115–127 (1998).

  23. 23.

    Arnold, F. H. & Zhang, J. H. Metal-mediated protein stabilization. Trends Biotechnol. 12, 189–192 (1994).

  24. 24.

    Brodin, J. D., Carr, J. R., Sontz, P. A. & Tezcan, F. A. Exceptionally stable, redox-active supramolecular protein assemblies with emergent properties. Proc. Natl Acad. Sci. USA 111, 2897–2902 (2014).

  25. 25.

    Grünbaum, B. & Johnson, N. W. The faces of a regular-faced polyhedron. J. Lond. Math. Soc. s1-40, 577–586 (1965).

  26. 26.

    Garcia-Seisdedos, H., Empereur-Mot, C., Elad, N. & Levy, E. D. Proteins evolve on the edge of supramolecular self-assembly. Nature 548, 244–247 (2017).

  27. 27.

    Sasaki, E. et al. Structure and assembly of scalable porous protein cages. Nat. Commun. 8, 14663 (2017).

  28. 28.

    Huylebrouck, D. & Redondo, A. The Hendecagonal Stars in the Alhambra. In Proc. Bridges Finland 2016 (eds Torrence, E. et al.) 497–500 (Tessellations, Phoenix, 2016).

  29. 29.

    Caspar, D. L. D. & Klug, A. Physical principles in the construction of regular viruses. Cold Spring Harb. Symp. Quant. Biol. 27, 1–24 (1962).

  30. 30.

    Prasad, B. V. & Schmid, M. F. Principles of virus structural organization. Adv. Exp. Med. Biol. 726, 17–47 (2012).

  31. 31.

    Cole, D., Young, G., Weigel, A., Sebesta, A. & Kukura, P. Label-free single-molecule imaging with numerical-aperture-shaped interferometric scattering microscopy. ACS Photonics 4, 211–216 (2017).

  32. 32.

    Li, X. et al. Electron counting and beam-induced motion correction enable near-atomic-resolution single-particle cryo-EM. Nat. Methods 10, 584–590 (2013).

  33. 33.

    Yasunaga, T. & Wakabayashi, T. Extensible and object-oriented system Eos supplies a new environment for image analysis of electron micrographs of macromolecules. J. Struct. Biol. 116, 155–160 (1996).

  34. 34.

    Tang, G. et al. EMAN2: an extensible image processing suite for electron microscopy. J. Struct. Biol. 157, 38–46 (2007).

  35. 35.

    Ludtke, S. J., Baldwin, P. R. & Chiu, W. EMAN: semiautomated software for high-resolution single-particle reconstructions. J. Struct. Biol. 128, 82–97 (1999).

  36. 36.

    Scheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J. Struct. Biol. 180, 519–530 (2012).

  37. 37.

    Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017).

  38. 38.

    Heymann, J. B. Bsoft: image and molecular processing in electron microscopy. J. Struct. Biol. 133, 156–169 (2001).

  39. 39.

    Rohou, A. & Grigorieff, N. CTFFIND4: Fast and accurate defocus estimation from electron micrographs. J. Struct. Biol. 192, 216–221 (2015).

  40. 40.

    Hopcroft, N. H. et al. The interaction of RNA with TRAP: the role of triplet repeats and separating spacer nucleotides. J. Mol. Biol. 338, 43–53 (2004).

  41. 41.

    Kucukelbir, A., Sigworth, F. J. & Tagare, H. D. Quantifying the local resolution of cryo-EM density maps. Nat. Methods 11, 63–65 (2014).

  42. 42.

    Pettersen, E. F. et al. UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612 (2004).

  43. 43.

    Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010).

  44. 44.

    Malay, A. D., Watanabe, M., Heddle, J. G. & Tame, J. R. H. Crystal structure of unliganded TRAP: implications for dynamic allostery. Biochem. J. 434, 427–434 (2011).

  45. 45.

    Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D 66, 213–221 (2010).

  46. 46.

    Wang, Z. et al. An atomic model of brome mosaic virus using direct electron detection and real-space optimization. Nat. Commun. 5, 4808 (2014).

  47. 47.

    Natchiar, S. K., Myasnikov, A. G., Kratzat, H., Hazemann, I. & Klaholz, B. P. Visualization of chemical modifications in the human 80S ribosome structure. Nature 551, 472–477 (2017).

  48. 48.

    Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D 66, 12–21 (2010).

  49. 49.

    Krissinel, E. & Henrick, K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797 (2007).

  50. 50.

    Pelicon, P. et al. A high brightness proton injector for the Tandetron accelerator at Jožef Stefan Institute. Nucl. Instrum. Methods Phys. Res. B 332, 229–233 (2014).

  51. 51.

    Vavpetič, P., Kelemen, M., Jenčič, B. & Pelicon, P. Nuclear microprobe performance in high-current proton beam mode for micro-PIXE. Nucl. Instrum. Methods Phys. Res. B 404, 69–73 (2017).

  52. 52.

    Pallon, J. et al. An off-axis STIM procedure for precise mass determination and imaging. Nucl. Instrum. Methods Phys. Res. B 219–220, 988–993 (2004).

  53. 53.

    Garman, E. F. & Grime, G. W. Elemental analysis of proteins by microPIXE. Prog. Biophys. Mol. Biol. 89, 173–205 (2005).

  54. 54.

    Campbell, J. L., Hopman, T. L., Maxwell, J. A. & Nejedly, Z. The Guelph PIXE software package III: alternative proton database. Nucl. Instrum. Methods Phys. Res. B 170, 193–204 (2000).

  55. 55.

    Kondrat, F. D., Struwe, W. B. & Benesch, J. L. Native mass spectrometry: towards high-throughput structural proteomics. Methods Mol. Biol. 1261, 349–371 (2015).

  56. 56.

    Sobott, F., Hernández, H., McCammon, M. G., Tito, M. A. & Robinson, C. V. A tandem mass spectrometer for improved transmission and analysis of large macromolecular assemblies. Anal. Chem. 74, 1402–1407 (2002).

  57. 57.

    Press, W. H., Teukolsky, S. A., Vetterling, W. T. & Flannery, B. P. Numerical recipes in C: the art of scientific computing 2nd edn (Cambridge Univ. Press, Cambridge, 1992).

  58. 58.

    Simms, G. A., Padmos, J. D. & Zhang, P. Structural and electronic properties of protein/thiolate-protected gold nanocluster with “staple” motif: A XAS, L-DOS, and XPS study. J. Chem. Phys. 131, 214703 (2009).

  59. 59.

    Zhang, P. & Sham, T. K. X-ray studies of the structure and electronic behavior of alkanethiolate-capped gold nanoparticles: the interplay of size and surface effects. Phys. Rev. Lett. 90, 245502 (2003).

  60. 60.

    Movasaghi, Z., Rehman, S. & Rehman, I. U. Raman spectroscopy of biological tissues. Appl. Spectrosc. Rev. 42, 493–541 (2007).

Download references


We thank M. Michalak and R. Pałka for initial TRAP protein production, P. Afonine for assistance with PHENIX real space refinement, M. Kobiałka and O. Woźnicka for TEM imaging, A. Telk and D. Dudek-Adamska for electrothermal atomic absorption spectroscopy experiments and evaluation (with financial support no. POIG.02.02.00-12-023/08), and A. Naskalska for additional experiments. This work was performed in part under the International Cooperative Research Program of the Institute for Protein Research, Osaka University CEMCR-17-05. A.D.M. and J.G.H. were funded by RIKEN Initiative Research Funding awarded to J.G.H. A.D.M. was supported by a Kakenhi Grant-In-Aid for Challenging Exploratory Research (JSPS), no. 2556023. J.G.H., A.B., I.S., A.K. and K.M. were funded by the National Science Centre (NCN, Poland) grant no. 2016/20/W/NZ1/00095 (Symfonia-4). Raman spectroscopy experiments were performed using equipment purchased as part of a project co-funded by the Malopolska Regional Operational Program Measure 5.1 Krakow Metropolitan Area as an important hub of the European Research Area for 2007-2013, project no. MRPO.05.01.00-12-013/15. J.L.P.B. acknowledges support from the EPSRC EP/J01835X/1. T.P.W. was supported under grant no. Homing/2016-2/20. A.F., M.S.K. and P.K. were supported by a ERC starting investigator grant (337757). K.I. was supported by the Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research) from the Japan Agency for Medical Research and Development (AMED). Work at JSI was supported by Slovenian research agency grants nos P1-0112, I0-0005, J7-9398 and EU projects no. 227012 “SPIRIT” and no. 824096 “RADIATE”.

Reviewer information

Nature thanks Jeroen Cornelissen, Todd Yeates and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Author notes

    • Ali D. Malay

    Present address: Biomacromolecules Research Team, Center for Sustainable Resource Science, RIKEN, Saitama, Japan

    • Georg K. A. Hochberg

    Present address: Department of Ecology and Evolution, University of Chicago, Chicago, IL, USA


  1. Heddle Initiative Research Unit, RIKEN, Saitama, Japan

    • Ali D. Malay
    •  & Jonathan G. Heddle
  2. Laboratory of Protein Synthesis and Expression, Institute for Protein Research, Osaka University, Osaka, Japan

    • Naoyuki Miyazaki
    •  & Kenji Iwasaki
  3. Bionanoscience and Biochemistry Laboratory, Malopolska Centre of Biotechnology, Jagiellonian University, Kraków, Poland

    • Artur Biela
    • , Soumyananda Chakraborti
    • , Karolina Majsterkiewicz
    • , Izabela Stupka
    • , Agnieszka Kowalczyk
    •  & Jonathan G. Heddle
  4. Department of Cell Biology and Imaging, Institute of Zoology and Biomedical Research, Jagiellonian University, Kraków, Poland

    • Artur Biela
  5. Postgraduate School of Molecular Medicine, Warsaw, Poland

    • Karolina Majsterkiewicz
    •  & Izabela Stupka
  6. David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, Ontario, Canada

    • Craig S. Kaplan
  7. Faculty of Mathematics and Computer Science, Jagiellonian University, Kraków, Poland

    • Agnieszka Kowalczyk
  8. Department of Mathematical Sciences, Durham University, Durham, UK

    • Bernard M. A. G. Piette
  9. Department of Chemistry, Physical & Theoretical Chemistry Laboratory, University of Oxford, Oxford, UK

    • Georg K. A. Hochberg
    • , Di Wu
    • , Adam Fineberg
    • , Manish S. Kushwah
    • , Philipp Kukura
    •  & Justin L. P. Benesch
  10. Institute of Nuclear Physics, Polish Academy of Sciences, Kraków, Poland

    • Tomasz P. Wrobel
  11. Jožef Stefan Institute, Ljubljana, Slovenia

    • Mitja Kelemen
    • , Primož Vavpetič
    •  & Primož Pelicon
  12. Jožef Stefan International Postgraduate School, Ljubljana, Slovenia

    • Mitja Kelemen
  13. Life Science Center for Survival Dynamics, Tsukuba Advanced Research Alliance (TARA), University of Tsukuba, Tsukuba, Japan

    • Kenji Iwasaki


  1. Search for Ali D. Malay in:

  2. Search for Naoyuki Miyazaki in:

  3. Search for Artur Biela in:

  4. Search for Soumyananda Chakraborti in:

  5. Search for Karolina Majsterkiewicz in:

  6. Search for Izabela Stupka in:

  7. Search for Craig S. Kaplan in:

  8. Search for Agnieszka Kowalczyk in:

  9. Search for Bernard M. A. G. Piette in:

  10. Search for Georg K. A. Hochberg in:

  11. Search for Di Wu in:

  12. Search for Tomasz P. Wrobel in:

  13. Search for Adam Fineberg in:

  14. Search for Manish S. Kushwah in:

  15. Search for Mitja Kelemen in:

  16. Search for Primož Vavpetič in:

  17. Search for Primož Pelicon in:

  18. Search for Philipp Kukura in:

  19. Search for Justin L. P. Benesch in:

  20. Search for Kenji Iwasaki in:

  21. Search for Jonathan G. Heddle in:


A.D.M. designed experiments, produced the protein cage, wrote the manuscript and carried out other experiments not listed. J.G.H. designed experiments, carried out TEM and wrote the manuscript. C.S.K. produced mathematical models of the TRAP-cage. A.K. and B.M.A.G.P. wrote the algorithm to predict paradoxical cage structures. A.B. and A.D.M. built the atomic models. N.M. and K.I. carried out cryo-electron microscopy experiments. G.K.A.H., D.W. and J.L.P.B. carried out mass spectrometry experiments. T.P.W. carried out Raman and X-ray photoelectron spectroscopy experiments. A.F., M.S.K., A.B. and P.K. designed, carried out and analysed the single-molecule mass photometry experiments. P.P., P.V. and M.K. executed micro-proton-induced X-ray emission analysis. S.C. and K.M. produced the protein cage, carried out stability tests and transmission electron microscopy sample preparation. I.S. carried out protein production and purification, dynamic light scattering measurements and preparation of samples for TEM. S.C. and A.B. carried out TEM. All authors contributed to writing the manuscript.

Competing interests

A.D.M. and J.G.H. are named as inventors on a patent application related to the use of gold to cross-link cysteine residues aimed at protein assembly construction.

Corresponding author

Correspondence to Jonathan G. Heddle.

Extended data figures and tables

  1. Extended Data Fig. 1 Size determination and optimization of conditions for TRAP-cage formation.

    a, Average size of TRAP-cage determined by dynamic light scattering, based on three separate preparations of purified TRAP-cage, each measured in triplicate. PDI, polydispersity index. The mean diameter is based on volume distribution. bf, Effect of pH on cage formation. Reactions containing 0.8 mg ml−1 TRAP(K35C/R64S) were incubated with Au-TPPMS at the indicated pH values for 3 days, spun down on a desktop centrifuge and subjected to blue native PAGE. The ratios are TRAP(K35C/R64S) monomer:Au(i) molar ratios. Formation of white precipitate was detected in the reactions for which the ratios are underlined, and correlates with a decrease in band intensity. gj, Time course of TRAP-cage formation at pH 7, 8 and 9, visualized by blue native PAGE. Each reaction contains Au-TPPMS and 0.8 mg ml−1 TRAP(K35C/R64S) at a 1:1 molar ratio. Total incubation times are indicated above each gel. Gels bj were repeated once, giving similar results. k, Left, the product of reaction containing 8.15 mg TRAP(K35C/R64S) and Au-TPPMS under standard cage-formation conditions was subjected to SEC using a Superose 6 Increase 10/300 GL column and fractions were collected as indicated. Right, native PAGE of fractions 1–13 demonstrates that they contain almost exclusively cage structures. The inset table demonstrates high recovery yields of TRAP-cage based on A280 measurements of initial and purified samples. SEC is representative of two independent experiments, giving similar results. Positions of molecular-weight marker bands are indicated to the left of gels and arrowheads indicate the position of bands corresponding to TRAP-cage. For gel source data, see Supplementary Fig. 1. Source Data

  2. Extended Data Fig. 2 Procedure for cryo-EM single-particle reconstruction for TRAP-cage formed with Au-TPPMS.

    a, Representative micrograph of the TRAP-cage. Scale bar, 50 nm. b, Summary of the image processing procedure (see Methods). c, 2D class averages from reference-free 2D classification by RELION 2.0. The selected 2D class-averaged images (5 classes including 578,865 particles) for further image processing are highlighted with red squares. d, The selected five 2D classes (box size: 220 × 220 pixels, 382.8 × 382.8 Å).

  3. Extended Data Fig. 3 Initial density map of TRAP-cage displaying aberrant features and final map quality and resolution for TRAP-cage formed with Au-TPPMS.

    a, b, The initial cryo-EM map, refined to 5.6 Å resolution, showing a distinct lack of chirality at the level of the individual rings. a, Overall map depicted in transparency, showing ring densities resembling radial wheel spokes. b, Magnification of a ring density at low contour level, viewed from the interior of the cage, showing exclusively radial features (densities and gaps). c, For comparison, the atomic model of TRAP(K35C/R64S) (based on PDB: 4V4F), is simulated to a resolution of 5.9 Å, showing that chiral properties (for example, curved propeller-like features, slanted grooves) should be readily visible on the rings at this resolution. d, e, Map quality and resolution of the two final cryo-EM density maps bearing opposite chirality: left-handed and right-handed cages are shown on the left and right, respectively. d, Surface representations coloured according to the distance from the centre of the particle. e, Gold-standard FSC curve for the cryo-EM map of left-handed and right-handed cages with C1, C4, D4, and octahedral (Oct) symmetries from 82,125 and 94,338 particles, respectively. The estimated resolutions at the 0.143 criterion were 3.7 Å for the two octahedral symmetry maps. f, The refined density maps coloured by local resolution in surface view.

  4. Extended Data Fig. 4 Details of the refined TRAP-cage structure.

    a, b, Overall fits of the final TRAP-cage models onto their respective density maps: left-handed (a) and right-handed (b) structures. Cysteine residues are rendered as ball and stick models, whereas gold atoms are shown as spheres. c, Magnified view of the left-handed cage, to show fitting of TRAP(K35C/R64S) structural elements into the density. d, Magnified view of the interior of the left-handed cage, showing flexible loop (residues 22–32) with missing density, consistent with the non-tryptophan-bound TRAP structure44. e, Slightly unequal dihedral angles are formed between neighbouring TRAP(K35C/R64S) rings in the final TRAP-cage model, averaging 135.8°, 135.3° and 137.25° around the two-, three-, and four-fold rotational axes, respectively, with a mean value of 136.2° across the entire cage. It is notable that in the canonical pentagonal icositetrahedron, a constant dihedral angle of 136.3° is formed between any two adjacent faces. f, The equivalence of the two chiral forms of TRAP-cage may explain their roughly equal proportions, and is clear if the cage assemblies are decomposed into two congruent hemispheres of 12 rings each. The hemispheres are themselves achiral, but together two can take either chiral form depending on their relative orientations when assembled. The two chiral forms can be interconverted through the relative rotation of hemispheres by about 24.1°. In the figure, each achiral half of the cage is coloured in red and blue, and the relative rotation can be tracked as the change in position of the highlighted ring in the direction of the arrow.

  5. Extended Data Fig. 5 Confirmation and quantitation of gold in the TRAP-cage.

    a, Results of five electrothermal atomic absorption spectroscopy measurements of TRAP-cage, each performed in triplicate, showing the measured mass of gold and its translation into the number of gold atoms per TRAP-cage. Measurement 3 was discarded in the calculation of overall averages owing to the large observed error. b, Micro-PIXE measurements of purified TRAP-cage showing the X-ray spectrum. Au L peaks and the S K peak were used to estimate the S:Au molar ratio in the sample, which was calculated to a range of approximately 5.1–6.9 S per Au. The dots represent the measured data, whereas the continuous line represents the fit with the Gupixwin program54. Results are representative of two independent experiments, each giving similar results. c, Au 4f XPS spectra (black line) of the TRAP-cage with the expected spectra for Au0 and Aui shown in blue and orange, respectively. The cumulative fit (red) and residuals (small red squares) are also shown. The shift in binding energy from 84.19 to 84.99 eV (that is, 0.80 eV) matches well with values for Au–S reported previously58,59. The presence of signal in the Au0 binding-energy range can be attributed to a weakly interacting Au coordination bond with the second sulfur in the bridge, because repeated measurements of different purification methods excluded the chance of X-ray degradation or unreacted substrates giving a Au0 signal. di, Raman spectroscopy suggests the absence or low abundance of S–S and S–H bonds in TRAP-cage. d, Raman spectra of TRAP-cage, oxidized and reduced TRAP-rings (TRAP-SS and TRAP-SH respectively), with oxidized and reduced glutathione (GSSG and GSH, respectively), showing the full spectral range. e, f, Enlargements of the S–H vibration (e) and S–S (f) vibration regions. Locations of peaks corresponding to S–S and S–H vibrations are labelled with asterisks. Because the exact positions depend on the molecular species and its conformation, the peak maxima for glutathione are shifted relative to cysteine-based signals (for example, the range for S–S vibrations is 509–540 cm−1)60. Because some peaks corresponding to S–H and S–S bonds were small, their absence from the TRAP-cage spectra was assessed by subtracting relevant spectra (that is, TRAP-SS or TRAP-SH) for TRAP-rings from the TRAP-cage data. g, The resulting spectra shown in the full spectral range. h, i, Enlargements of the S–H (h) and S–S (i) regions. Spectra after normalization were offset for clarity. For XPS and Raman spectra, experiments were independently repeated at least once, each giving similar results. Source Data

  6. Extended Data Fig. 6 Further probing TRAP-cage assembly.

    a–c, Testing the ability of different metal ions to induce protein cage formation. Reactions containing 0.8 mg ml−1 TRAP(K35C/R64S) were incubated under standard cage-formation conditions, except that Au-TPPMS was replaced with the indicated metal ions, then spun down on a desktop centrifuge and subjected to blue native PAGE. TRAP(K35C/R64S) monomer:metal ion molar ratios are indicated above each lane. TRAP(K35C/R64S) was incubated with Cui (a), Znii (b) and Auiii (c). White precipitate was detected in the reactions for which the ratios are underlined, and correlates with a decrease in band intensity. d, Modelling alternative locations for placement of cysteine residues on the surface of the TRAP ring, based on PDB: 4V4F, and shown in orthogonal views: K35C, S33C and D15C, with the location of the substituted residues rendered in red, cyan and purple, respectively. e, Reaction of TRAP(S33C/R64S) with Au-TPPMS produces uniform cage structures, as shown by native PAGE (left) and TEM (right). Scale bar, 100 nm. f, Reaction of TRAP(D15C/R64S) with Au-TPPMS fails to produce higher-order structures, as shown by native PAGE. Gels in ac, e, f and TEM were repeated independently twice, each giving similar results. Positions of molecular-weight markers on gels are indicated and arrowheads indicate position of TRAP-cage produced using Au-TPPMS. For gel source data, see Supplementary Fig. 1. Source Data

  7. Extended Data Fig. 7 Additional tests of TRAP-cage stability.

    TRAP-cage was prepared as for cages in Fig. 1d. Purified TRAP-cage samples were incubated at room temperature overnight in 2 M guanidinium HCl (a), 5% SDS (b), 7 M urea (c), 0.01 and 0.1 mM TCEP (d), 0.1 and 10 mM GSH (e) and 10 mM GSSG (f), and subsequently imaged under TEM. All TEM images shown are representative of data that was repeated once, each giving similar results. Scale bars, 100 nm.

  8. Extended Data Fig. 8 Procedure for cryo-EM single-particle reconstruction, map quality and resolution for TRAP-cage formed with GNP.

    a, A representative micrograph of the TRAP-cage formed using GNP. Scale bar, 50 nm. b, Images of selected 2D classes from reference-free 2D classification by EMAN 2.1, used for automated particle picking with Gautomatch (box size: 220 × 220 pixels, 400.4 × 400.4 Å). c, Summary of the image processing procedure (see Methods). d, 2D class averages from reference-free 2D classification by RELION 2.0. The selected 2D class averages (22 images) for further image processing are highlighted with red squares. It is notable that in some cases (for example, row 1 panel 9) structures inside the TRAP-cage are visible, probably reflecting stochastic capture of TRAP rings as cargo in the cage interior. e, Initial structure used for the high-resolution analysis. The surface representations are coloured according to the distance from the centre of the particle. f, Gold-standard FSC curve for the cryo-EM map from 58,157 particles. The calculated spatial frequency at the 0.143 criterion was 5.6 Å. g, The refined density map coloured by local resolution in surface and slice views.

  9. Extended Data Fig. 9 Formation and stability of TRAP-cage formed with GNPs (TRAP-cage(GNP)).

    TRAP-cages were formed under GNP-cage formation conditions (see Methods). a, TEM image of purified TRAP-cage(GNP), with a magnified view on the right. b, Thermal stability of TRAP-cage(GNP) upon incubation at the indicated temperatures and times. The TEM image shows that structural integrity is maintained after incubation at 95 °C for 180 min. c, Stability of TRAP-cage(GNP) as a function of pH. d, Stability of TRAP-cage(GNP) upon the addition of different chaotropic agents: SDS, urea and guanindine HCl. e, f, Reducing agents trigger disassembly of TRAP-cage(GNP), as shown by native PAGE (e) and TEM (f). Assay conditions are indicated above each lane on the gel. Positions of molecular weight markers are indicated and arrowheads show the position of TRAP-cage. For TEM, scale bars are 100 nm. All gels and TEM images shown are representative of experiments repeated independently at least once, each giving similar results. For gel source data, see Supplementary Fig. 1.

  10. Extended Data Table 1 Cryo-EM data collection, refinement and validation statistics

Supplementary information

  1. Supplementary Figure

    Uncropped photographs of gels with molecular weight marker indications. Boxed areas indicate the cropped area.

  2. Reporting Summary

  3. Supplementary Video 1

    Data obtained using single molecule mass photometry as described in the methods section. Video 1 shows TRAPCS at time point 30 s after addition of Au-TPPMS. All scale bars are 1 µm. Each of videos 1-9 shows the first portion of the video only (due to size constraints). Contrast is scaled to +/- 2.5 MDa. Experiments were run twice, giving similar results with representative results shown.

  4. Supplementary Video 2

    Video 2 shows TRAPCS at time point 270 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  5. Supplementary Video 3

    Video 3 shows TRAPCS at time point 510 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  6. Supplementary Video 4

    Video 4 shows TRAPCS at time point 750 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  7. Supplementary Video 5

    Video 5 shows TRAPCS at time point 990 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  8. Supplementary Video 6

    Video 6 shows TRAPCS at time point 1110 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  9. Supplementary Video 7

    Video 7 shows TRAPCS at time point 1230 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  10. Supplementary Video 8

    Video 8 shows TRAPCS at time point 1350 s after addition of Au-TPPMS. See legend of Supplementary Video 1 for more information.

  11. Supplementary Video 9

    Video 9 shows control in the absence of Au-TPPMS. See legend of Supplementary Video 1 for more information.

Source data

About this article

Publication history




Issue Date



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.