De novo design of potent and selective mimics of IL-2 and IL-15

Abstract

We describe a de novo computational approach for designing proteins that recapitulate the binding sites of natural cytokines, but are otherwise unrelated in topology or amino acid sequence. We use this strategy to design mimics of the central immune cytokine interleukin-2 (IL-2) that bind to the IL-2 receptor βγc heterodimer (IL-2Rβγc) but have no binding site for IL-2Rα (also called CD25) or IL-15Rα (also known as CD215). The designs are hyper-stable, bind human and mouse IL-2Rβγc with higher affinity than the natural cytokines, and elicit downstream cell signalling independently of IL-2Rα and IL-15Rα. Crystal structures of the optimized design neoleukin-2/15 (Neo-2/15), both alone and in complex with IL-2Rβγc, are very similar to the designed model. Neo-2/15 has superior therapeutic activity to IL-2 in mouse models of melanoma and colon cancer, with reduced toxicity and undetectable immunogenicity. Our strategy for building hyper-stable de novo mimetics could be applied generally to signalling proteins, enabling the creation of superior therapeutic candidates.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Computational design of de novo cytokine mimics.
Fig. 2: Characterization of Neo-2/15.
Fig. 3: Structure of Neo-2/15 and its ternary complex with mouse IL-2Rβγc.
Fig. 4: In vivo characterization of Neo-2/15.

Data availability

Structures for Neo-2/15 monomer and its ternary complex with mouse IL-2Rβγc have been deposited in the Protein Data Bank with accession numbers 6DG6 and 6DG5, respectively. Diffraction images have been deposited in the SBGrid Data Bank with accession numbers 587 and 588, respectively, and validation reports are included in the Supplementary Information. Other data and materials are available upon request from the corresponding authors.

References

  1. 1.

    Akdis, M. et al. Interleukins, from 1 to 37, and interferon-γ: receptors, functions, and roles in diseases. J. Allergy Clin. Immunol. 127, 701–721 (2011).

    CAS  Article  Google Scholar 

  2. 2.

    Smyth, M. J., Cretney, E., Kershaw, M. H. & Hayakawa, Y. Cytokines in cancer immunity and immunotherapy. Immunol. Rev. 202, 275–293 (2004).

    CAS  Article  Google Scholar 

  3. 3.

    Lotze, M. T. et al. In vivo administration of purified human interleukin 2. II. Half life, immunologic effects, and expansion of peripheral lymphoid cells in vivo with recombinant IL 2. J. Immunol. 135, 2865–2875 (1985).

    CAS  PubMed  Google Scholar 

  4. 4.

    Moraga, I. et al. Synthekines are surrogate cytokine and growth factor agonists that compel signaling through non-natural receptor dimers. eLife 6, e22882 (2017).

    Article  Google Scholar 

  5. 5.

    Vazquez-Lombardi, R. et al. Potent antitumour activity of interleukin-2–Fc fusion proteins requires Fc-mediated depletion of regulatory T-cells. Nat. Commun. 8, 15373 (2017).

    ADS  CAS  Article  Google Scholar 

  6. 6.

    Sockolosky, J. T. et al. Selective targeting of engineered T cells using orthogonal IL-2 cytokine–receptor complexes. Science 359, 1037–1042 (2018).

    ADS  CAS  Article  Google Scholar 

  7. 7.

    Kureshi, R., Bahri, M. & Spangler, J. B. Reprogramming immune proteins as therapeutics using molecular engineering. Curr. Opin. Chem. Eng. 19, 27–34 (2018).

    Article  Google Scholar 

  8. 8.

    Levin, A. M. et al. Exploiting a natural conformational switch to engineer an interleukin-2 ‘superkine’. Nature 484, 529–533 (2012).

    ADS  CAS  Article  Google Scholar 

  9. 9.

    Sarkar, C. A. et al. Rational cytokine design for increased lifetime and enhanced potency using pH-activated ‘histidine switching’. Nat. Biotechnol. 20, 908–913 (2002).

    CAS  Article  Google Scholar 

  10. 10.

    Spangler, J. B., Moraga, I., Mendoza, J. L. & Garcia, K. C. Insights into cytokine-receptor interactions from cytokine engineering. Annu. Rev. Immunol. 33, 139–167 (2015).

    CAS  Article  Google Scholar 

  11. 11.

    Charych, D. H. et al. NKTR-214, an engineered cytokine with biased IL2 receptor binding, increased tumor exposure, and marked efficacy in mouse tumor models. Clin. Cancer Res. 22, 680–690 (2016).

    CAS  Article  Google Scholar 

  12. 12.

    Dougan, M. et al. Targeting cytokine therapy to the pancreatic tumor microenvironment using PD-L1-specific VHHs. Cancer Immunol. Res. 6, 389–401 (2018).

    CAS  Article  Google Scholar 

  13. 13.

    Tzeng, A., Kwan, B. H., Opel, C. F., Navaratna, T. & Dane Wittrup, K. Antigen specificity can be irrelevant to immunocytokine efficacy and biodistribution. Proc. Natl Acad. Sci. USA 112, 3320–3325 (2015).

    ADS  CAS  Article  Google Scholar 

  14. 14.

    Zhu, E. F. et al. Synergistic innate and adaptive immune response to combination immunotherapy with anti-tumor antigen antibodies and extended serum half-life IL-2. Cancer Cell 27, 489–501 (2015).

    CAS  Article  Google Scholar 

  15. 15.

    Kim, D. E., Gu, H. & Baker, D. The sequences of small proteins are not extensively optimized for rapid folding by natural selection. Proc. Natl Acad. Sci. USA 95, 4982–4986 (1998).

    ADS  CAS  Article  Google Scholar 

  16. 16.

    Taverna, D. M. & Goldstein, R. A. Why are proteins marginally stable? Proteins 46, 105–109 (2002).

    CAS  Article  Google Scholar 

  17. 17.

    Foit, L. et al. Optimizing protein stability in vivo. Mol. Cell 36, 861–871 (2009).

    CAS  Article  Google Scholar 

  18. 18.

    Marshall, S. A., Lazar, G. A., Chirino, A. J. & Desjarlais, J. R. Rational design and engineering of therapeutic proteins. Drug Discov. Today 8, 212–221 (2003).

    CAS  Article  Google Scholar 

  19. 19.

    De Groot, A. S. & Scott, D. W. Immunogenicity of protein therapeutics. Trends Immunol. 28, 482–490 (2007).

    Article  Google Scholar 

  20. 20.

    Peyvandi, F. et al. A randomized trial of factor VIII and neutralizing antibodies in hemophilia A. N. Engl. J. Med. 374, 2054–2064 (2016).

    CAS  Article  Google Scholar 

  21. 21.

    Antonelli, G., Currenti, M., Turriziani, O. & Dianzani, F. Neutralizing antibodies to interferon-α: relative frequency in patients treated with different interferon preparations. J. Infect. Dis. 163, 882–885 (1991).

    CAS  Article  Google Scholar 

  22. 22.

    Eckardt, K.-U. & Casadevall, N. Pure red-cell aplasia due to anti-erythropoietin antibodies. Nephrol. Dial. Transplant. 18, 865–869 (2003).

    Article  Google Scholar 

  23. 23.

    Prümmer, O. Treatment-induced antibodies to interleukin-2. Biotherapy 10, 15–24 (1997).

    Article  Google Scholar 

  24. 24.

    Fineberg, S. E. et al. Immunological responses to exogenous insulin. Endocr. Rev. 28, 625–652 (2007).

    CAS  Article  Google Scholar 

  25. 25.

    Boyman, O. & Sprent, J. The role of interleukin-2 during homeostasis and activation of the immune system. Nat. Rev. Immunol. 12, 180–190 (2012).

    CAS  Article  Google Scholar 

  26. 26.

    Blattman, J. N. et al. Therapeutic use of IL-2 to enhance antiviral T-cell responses in vivo. Nat. Med. 9, 540–547 (2003).

    CAS  Article  Google Scholar 

  27. 27.

    Siegel, J. P. & Puri, R. K. Interleukin-2 toxicity. J. Clin. Oncol. 9, 694–704 (1991).

    CAS  Article  Google Scholar 

  28. 28.

    Mott, H. R. et al. The solution structure of the F42A mutant of human interleukin 2. J. Mol. Biol. 247, 979–994 (1995).

    CAS  Article  Google Scholar 

  29. 29.

    Carmenate, T. et al. Human IL-2 mutein with higher antitumor efficacy than wild type IL-2. J. Immunol. 190, 6230–6238 (2013).

    CAS  Article  Google Scholar 

  30. 30.

    Tagaya, Y., Bamford, R. N., DeFilippis, A. P. & Waldmann, T. A. IL-15: a pleiotropic cytokine with diverse receptor/signaling pathways whose expression is controlled at multiple levels. Immunity 4, 329–336 (1996).

    CAS  Article  Google Scholar 

  31. 31.

    Ozaki, K. & Leonard, W. J. Cytokine and cytokine receptor pleiotropy and redundancy. J. Biol. Chem. 277, 29355–29358 (2002).

    CAS  Article  Google Scholar 

  32. 32.

    Lin, J. X. et al. The role of shared receptor motifs and common Stat proteins in the generation of cytokine pleiotropy and redundancy by IL-2, IL-4, IL-7, IL-13, and IL-15. Immunity 2, 331–339 (1995).

    CAS  Article  Google Scholar 

  33. 33.

    Ma, A., Boone, D. L. & Lodolce, J. P. The pleiotropic functions of interleukin 15: not so interleukin 2-like after all. J. Exp. Med. 191, 753–756 (2000).

    CAS  Article  Google Scholar 

  34. 34.

    Procko, E. et al. A computationally designed inhibitor of an Epstein-Barr viral Bcl-2 protein induces apoptosis in infected cells. Cell 157, 1644–1656 (2014).

    CAS  Article  Google Scholar 

  35. 35.

    Chevalier, A. et al. Massively parallel de novo protein design for targeted therapeutics. Nature 550, 74–79 (2017).

    ADS  CAS  PubMed  PubMed Central  Google Scholar 

  36. 36.

    Jacobs, T. M. et al. Design of structurally distinct proteins using strategies inspired by evolution. Science 352, 687–690 (2016).

    ADS  CAS  Article  Google Scholar 

  37. 37.

    Correia, B. E. et al. Proof of principle for epitope-focused vaccine design. Nature 507, 201–206 (2014).

    ADS  CAS  Article  Google Scholar 

  38. 38.

    Boyken, S. E. et al. De novo design of protein homo-oligomers with modular hydrogen-bond network-mediated specificity. Science 352, 680–687 (2016).

    ADS  CAS  Article  Google Scholar 

  39. 39.

    Ring, A. M. et al. Mechanistic and structural insight into the functional dichotomy between IL-2 and IL-15. Nat. Immunol. 13, 1187–1195 (2012).

    CAS  Article  Google Scholar 

  40. 40.

    Fleishman, S. J. et al. RosettaScripts: a scripting language interface to the Rosetta macromolecular modeling suite. PLoS ONE 6, e20161 (2011).

    ADS  CAS  Article  Google Scholar 

  41. 41.

    Leaver-Fay, A. et al. Rosetta3: an object-oriented software suite for the simulation and design of macromolecules. Methods Enzymol. 487, 545–574 (2011).

    CAS  Article  Google Scholar 

  42. 42.

    Chaudhury, S., Lyskov, S. & Gray, J. J. PyRosetta: a script-based interface for implementing molecular modeling algorithms using Rosetta. Bioinformatics 26, 689–691 (2010).

    CAS  Article  Google Scholar 

  43. 43.

    Wang, X., Rickert, M. & Garcia, K. C. Structure of the quaternary complex of interleukin-2 with its α, β, and γc receptors. Science 310, 1159–1163 (2005).

    ADS  CAS  Article  Google Scholar 

  44. 44.

    Robinson, T. O. & Schluns, K. S. The potential and promise of IL-15 in immuno-oncogenic therapies. Immunol. Lett. 190, 159–168 (2017).

    CAS  Article  Google Scholar 

  45. 45.

    Bouchaud, G. et al. The exon-3-encoded domain of IL-15Rα contributes to IL-15 high-affinity binding and is crucial for the IL-15 antagonistic effect of soluble IL-15Rα. J. Mol. Biol. 382, 1–12 (2008).

    CAS  Article  Google Scholar 

  46. 46.

    Cao, X. Regulatory T cells and immune tolerance to tumors. Immunol. Res. 46, 79–93 (2009).

    Article  Google Scholar 

  47. 47.

    Fontenot, J. D., Rasmussen, J. P., Gavin, M. A. & Rudensky, A. Y. A function for interleukin 2 in Foxp3-expressing regulatory T cells. Nat. Immunol. 6, 1142–1151 (2005).

    CAS  Article  Google Scholar 

  48. 48.

    Chen, X. et al. Combination therapy of an IL-15 superagonist complex, ALT-803, and a tumor targeting monoclonal antibody promotes direct antitumor activity and protective vaccinal effect in a syngenic mouse melanoma model. J. Immunother. Cancer 3, 347 (2015).

    CAS  Article  Google Scholar 

  49. 49.

    Dougan, M. & Dranoff, G. Immune therapy for cancer. Annu. Rev. Immunol. 27, 83–117 (2009).

    CAS  Article  Google Scholar 

  50. 50.

    Roberts, M. J., Bentley, M. D. & Harris, J. M. Chemistry for peptide and protein PEGylation. Adv. Drug Deliv. Rev. 64, 116–127 (2012).

    Article  Google Scholar 

  51. 51.

    Fleishman, S. J. et al. Computational design of proteins targeting the conserved stem region of influenza hemagglutinin. Science 332, 816–821 (2011).

    ADS  CAS  Article  Google Scholar 

  52. 52.

    Benatuil, L., Perez, J. M., Belk, J. & Hsieh, C.-M. An improved yeast transformation method for the generation of very large human antibody libraries. Protein Eng. Des. Sel. 23, 155–159 (2010).

    CAS  Article  Google Scholar 

  53. 53.

    Chang, H. C. et al. A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments. Proc. Natl Acad. Sci. USA 91, 11408–11412 (1994).

    ADS  CAS  Article  Google Scholar 

  54. 54.

    Kabsch, W. XDS. Acta Crystallogr. D 66, 125–132 (2010).

  55. 55.

    Evans, P. Scaling and assessment of data quality. Acta Crystallogr. D 62, 72–82 (2006). 

  56. 56.

    Evans, P. R. & Murshudov, G. N. How good are my data and what is the resolution? Acta Crystallogr. D 69, 1204–1214 (2013).

  57. 57.

    Winn, M. D. et al. Overview of the CCP4 suite and current developments. Acta Crystallogr. D 67, 235–242 (2011).

  58. 58.

    McCoy, A. J. et al. Phaser crystallographic software. J Appl. Crystallogr. 40, 658–674 (2007).

  59. 59.

    Terwilliger, T. C. et al. Iterative model building, structure refinement and density modification with the PHENIX AutoBuild wizard. Acta Crystallogr. D 64, 61–69 (2008).

  60. 60.

    Emsley, P. et al. Features and development of Coot. Acta Crystallogr. D 66, 486–501 (2010).

  61. 61.

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

  62. 62.

    D’Arcy, A. et al. Microseed matrix screening for optimization in protein crystallization: what have we learned? Acta Crystallogr. F 70, 1117–1126 (2014).

  63. 63.

    Bruhn, J. F. et al. Crystal structure of the Marburg virus VP35 oligomerization domain. J. Virol. 3, e01085-16 (2017).

    Google Scholar 

  64. 64.

    Smart, O. S. et al. Exploiting structure similarity in refinement: automated NCS and target-structure restraints in BUSTER. Acta Crystallogr. D 68, 368–380 (2012).

    CAS  Article  Google Scholar 

  65. 65.

    The PyMOL Molecular Graphics System v.2.1.0 (Schrodinger, LLC., 2010).

  66. 66.

    Morin, A. et al. Collaboration gets the most out of software. eLife 2, e01456 (2013).

    Article  Google Scholar 

  67. 67.

    Yodoi, J. et al. TCGF (IL 2)-receptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like cell line (YT cells). J. Immunol. 134, 1623–1630 (1985).

    CAS  PubMed  Google Scholar 

  68. 68.

    Kuziel, W. A., Ju, G., Grdina, T. A. & Greene, W. C. Unexpected effects of the IL-2 receptor alpha subunit on high affinity IL-2 receptor assembly and function detected with a mutant IL-2 analog. J. Immunol. 150, 3357–3365 (1993).

    CAS  PubMed  Google Scholar 

  69. 69.

    Hondowicz, B. D. et al. Interleukin-2-dependent allergen-specific tissue-resident memory cells drive asthma. Immunity 44, 155–166 (2016).

    CAS  Article  Google Scholar 

  70. 70.

    Liu, L. et al. Inclusion of Strep-Tag II in design of antigen receptors for T-cell immunotherapy. Nat. Biotechnol. 34, 430–434 (2016).

    CAS  Article  Google Scholar 

  71. 71.

    Silva, D.-A., Stewart, L., Lam, K.-H., Jin, R. & Baker, D. Structures and disulfide cross-linking of de novo designed therapeutic mini-proteins. FEBS J. 285, 1783–1785 (2018).

    CAS  Article  Google Scholar 

  72. 72.

    Stumpp, M. T., Kaspar Binz, H. & Amstutz, P. DARPins: A new generation of protein therapeutics. Drug Discov. Today 13, 695–701 (2008).

    CAS  Article  Google Scholar 

  73. 73.

    Marcos, E. & Silva, D.-A. Essentials of de novo protein design: methods and applications. WIREs Comput. Mol. Sci. 8, e1374 (2018).

    Article  Google Scholar 

  74. 74.

    Berger, S. et al. Computationally designed high specificity inhibitors delineate the roles of BCL2 family proteins in cancer. eLife 5, e20352 (2016).

    Article  Google Scholar 

  75. 75.

    Abraham, M. J. et al. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2, 19–25 (2015).

    ADS  Article  Google Scholar 

  76. 76.

    Markidis, S. & Laure, E. Solving software challenges for Exascale. In International Conference on Exascale Applications and Software (eds Markidis, S. & Laure, E.) (Springer, 2015).

  77. 77.

    Lindorff-Larsen, K. et al. Improved side-chain torsion potentials for the Amber ff99SB protein force field. Proteins 78, 1950–1958 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  78. 78.

    Leszczynski, J. & Shukla, M. K. Practical Aspects of Computational Chemistry: Methods, Concepts and Applications (Springer, Dordrecht, 2009).

    Google Scholar 

  79. 79.

    Berendsen, H. J. C., Postma, J. P. M., van Gunsteren, W. F., DiNola, A. & Haak, J. R. Molecular dynamics with coupling to an external bath. J. Chem. Phys. 81, 3684–3690 (1984).

    ADS  CAS  Article  Google Scholar 

  80. 80.

    Parrinello, M. & Rahman, A. Polymorphic transitions in single crystals: A new molecular dynamics method. J. Appl. Phys. 52, 7182–7190 (1981).

    ADS  CAS  Article  Google Scholar 

  81. 81.

    Essmann, U. et al. A smooth particle mesh Ewald method. J. Chem. Phys. 103, 8577–8593 (1995).

    ADS  CAS  Article  Google Scholar 

  82. 82.

    Páll, S. & Hess, B. A flexible algorithm for calculating pair interactions on SIMD architectures. Comput. Phys. Commun. 184, 2641–2650 (2013).

    ADS  Article  Google Scholar 

  83. 83.

    Perez, F. & Granger, B. E. IPython: a system for interactive scientific computing. Comput. Sci. Eng. 9, 21–29 (2007).

    CAS  Article  Google Scholar 

  84. 84.

    Oliphant, T. E. Python for scientific computing. Comput. Sci. Eng. 9, 10–20 (2007).

    CAS  Article  Google Scholar 

  85. 85.

    Oliphant, T. E. Guide to NumPy 2nd edn (CreateSpace, 2015).

  86. 86.

    Hunter, J. D. Matplotlib: a 2D graphics environment. Comput. Sci. Eng. 9, 90–95 (2007).

    Article  Google Scholar 

  87. 87.

    Garreta, R. & Moncecchi, G. Learning scikit-learn: Machine Learning in Python. (Packt, Birmingham, 2013).

    Google Scholar 

  88. 88.

    Behnel, S. et al. Cython: the best of both worlds. Comput. Sci. Eng. 13, 31–39 (2011).

    Article  Google Scholar 

  89. 89.

    McKinney, W. Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython. (O’Reilly, Sebastopol, 2017).

    Google Scholar 

  90. 90.

    Minami, S., Sawada, K. & Chikenji, G. MICAN: a protein structure alignment algorithm that can handle multiple-chains, inverse alignments, Cα only models, alternative alignments, and non-sequential alignments. BMC Bioinformatics 14, 24 (2013).

    CAS  Article  Google Scholar 

  91. 91.

    Crooks, G. E., Hon, G., Chandonia, J.-M. & Brenner, S. E. WebLogo: a sequence logo generator. Genome Res. 14, 1188–1190 (2004).

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank B. Nordstrom, J. Nordstrom, P. Barrier and J. Barrier for the IPD Fund (Budget Number: 68-0341); CONACyT SNI (Mexico), CONACyT postdoctoral fellowship (Mexico) and IPD translational research program to D.-A.S.; NIH MSTP grant T32 GM007266 to S.Y.; JDRF (2-SRA-2016-236-Q-R) to U.Y.U.; la Caixa Fellowship (la Caixa Banking Foundation, Barcelona, Spain) to A.Q.-R.; FCT Portugal Ph.D. studentship to C.L.-A.; European Research Council (ERC StG, grant agreement 676832), FCT Investigator (IF/00624/2015), and the Royal Society (UF110046 and URF\R\180019) to G.J.L.B.; Marie Curie International Outgoing Fellowship (FP7-PEOPLE-2011-IOF 298976) to E.M.; Natural Sciences and Engineering Research Council of Canada Postdoctoral Fellowship to C.D.W.; Washington Research Foundation to B.D.W.; NIH grant R35GM122543 to F.P.-A.; Mentored Clinical Scientist Development Award 1K08DK114563-01, and the American Gastroenterological Association Research Scholars Award to M.D.; NIH-RO1-AI51321, NIH-RO1-AI51321, Mathers Foundation, Younger Endowed Chair, and Howard Hughes Medical Institute to K.C.G.; and Howard Hughes Medical Institute and Michelson Medical Research Foundation to D.B. See Supplementary Information for extended acknowledgements.

Reviewer information

Nature thanks Y. Jones, W. Schief and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Affiliations

Authors

Contributions

D.-A.S., S.Y., U.Y.U., J.B.S., M.P., G.J.L.B., M.D., K.C.G. and D.B. designed the research; D.-A.S. developed the method for designing de novo protein mimics, and designed and invented the IL-2/IL-15 mimics; S.Y., D.-A.S. and U.Y.U. characterized and optimized the IL-2/IL-15 mimics; J.B.S. performed BLI binding characterization, in vitro cell signalling, and recombinant IL-receptor production; K.M.J. performed crystallography experiments; A.Q.-R. characterized binding and stability, directed evolution and protein expression; L.R.A., T.B. and S.J.C. performed in vivo naive mouse T cell response, melanoma cancer model and immunogenicity experiments; C.L.-A. performed in vivo colorectal cancer experiments; M.R. performed ex vivo cell signalling and in vivo airway inflammation experiments; I.L. performed in vivo CAR-T cell experiments; C.D.W. designed and characterized single-cysteine mutations; E.M. and J.C. assisted in developing the computational design methods; B.D.W. designed and characterized disulfide-stapled variants; F.P.-A. performed and analysed molecular dynamics simulations; L.C. performed optimization and production of recombinant protein; L.S. supervised and coordinated collaborations; S.R.R. supervised in vivo CAR-T cell experiments, M.P. supervised research for ex vivo cell signalling and in vivo tissue residency; G.J.L.B. supervised research for the in vivo colorectal cancer model; M.D. coordinated research for in vivo naive mouse T cell response, melanoma cancer model and immunogenicity; D.-A.S., S.Y., U.Y.U., J.B.S., M.D., K.C.G. and D.B. wrote the manuscript; D.-A.S., K.C.G. and D.B. supervised and coordinated the overall research.

Corresponding authors

Correspondence to Daniel-Adriano Silva or K. Christopher Garcia or David Baker.

Ethics declarations

Competing interests

D.-A.S., S.Y., U.Y.U., A.Q.-R., C.D.W. and D.B. are co-founders and stockholders of Neoleukin Therapeutics, a company that aims to develop the inventions described in this manuscript. D.-A.S., S.Y., U.Y.U., J.B.S., A.Q.-R., K.C.G. and D.B. are co-inventors on a US provisional patent application (no. 62/689769), which incorporates discoveries described in this manuscript.

Additional information

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

Extended data figures and tables

Extended Data Fig. 1 Therapeutic effect of Neo-2/15 on colon cancer.

a, BALB/C mice were inoculated with CT26 tumours. Starting on day 9 and ending on day 14, mice were treated daily with intraperitoneal injection of mouse IL-2 or Neo-2/15 at the specified concentrations (n = 4 per group), or were left untreated (n = 6 per group). Top, tumour growth curves show data only for surviving mice and stop if the number of mice in a group fell below 50% of the initial number. Bottom, survival curves. Mice were euthanized when weight loss exceeded 10% of initial weight or when tumour size reached 1,300 mm3. The experiments were performed twice with similar results. bd, Bar plots comparing the T cell populations in BALB/C mice (n = 3 per group) that were inoculated with CT26 tumours and treated, starting from day 6, by daily intraperitoneal injection of 10 μg Neo-2/15, 10 μg mouse IL-2 or no treatment (no tx). On day 14 the percentage of Treg cells (CD4+CD45+FoxP3+, top) and CD8:Treg cell ratio (CD45+CD3+CD8+ cells:Treg cells; bottom) were assessed in tumours (b), neighbouring inguinal lymph node (LN) (c), and spleen (d). Data are mean ± s.d., except in growth curves, where data are mean ± s.e.m. Results were analysed by one-way ANOVA (95% confidence interval), except for survival curves that were assessed using the Mantel–Cox test (95% confidence interval). Experiments were performed twice with similar results.

Extended Data Fig. 2 Therapeutic effect of Neo-2/15 on melanoma.

Survival curves (top) and tumour growth curves (bottom) for C57BL/6 mice that were inoculated with B16 tumours (as in Fig. 4a) and treated with low (1 μg per mouse per day) or high (10 μg per mouse per day) doses of Neo-2/15. a, Starting on day 1, mice (n = 5 per group) were treated daily with intraperitoneal injection of single agent Neo-2/15 at 1 μg per mouse or equimolar mouse IL-2 (left), or the same treatments in combination with a twice-weekly treatment with TA99 (started on day 5) (right). Mice were euthanized when tumour size reached 2,000 mm3. Tumour growth curves show data only for surviving mice and stop if the number of mice in a group fell below 50% of the initial number. The experiments were performed twice with similar results. b, Similar to a, but starting on day 4. Mice (n = 5 per group) were treated daily with intraperitoneal injection of single agent Neo-2/15 at 10 μg per mouse or equimolar mouse IL-2 (left), or the same treatments in combination with a twice-weekly treatment with TA99 (started on day 4) (right). Mice were euthanized when tumour size reached 1,000 mm3. The therapeutic effect of Neo-2/15 is dose-dependent (higher doses have a stronger effect) and is potentiated in the presence of the antibody TA99. Tumour growth curves show data only for surviving mice and stop if the number of mice in a group fell below 50% of the initial number. The experiments were performed twice with similar results. c, C57BL/6 mice were immunized with 500 μg KO Neo-2/15 in complete Freund’s adjuvant and boosted on days 7 and 15 with 500 μg KO Neo-2/15 in incomplete Freund’s adjuvant. Reactivity against KO Neo-2/15 and native Neo-2/15, as well as cross-reactivity with mouse IL-2 were determined by incubation of serum (diluted 1:1,000 in PBS) with plate-bound KO Neo-2/15, Neo-2/15 or mouse IL-2 as indicated. Serum binding was detected using an anti-mouse secondary antibody conjugated to HRP followed by incubation with TMB. Data are reported as optical density at 450 nm. Top, naive mouse serum; bottom, immunized mouse serum. The experiments were performed once. In all the growth curves, data are mean ± s.e.m. Results were analysed by one-way ANOVA (95% confidence interval), except for survival curves that were assessed using the Mantel–Cox test (95% confidence interval).

Extended Data Fig. 3 Single disulfide-stapled variants of Neo-2/15 with higher thermal stability.

a, b, Structural models of disulfide-stabilized variants of Neo-2/15 (grey) are shown superposed on the ternary crystal structure of Neo-2/15 (red) with mutated residues highlighted in magenta and the disulfide bond shown in gold. Two strategies were used to generate the disulfide stapled variants. a, Top, internal placement of the disulfide linking residues 38 and 75. Bottom, experimental CD spectra of the design at 25 °C, 95 °C and then cooled back to 25 °C, showing complete recovery of ellipticity spectrum (full reversibility) upon cooling. b, Top, for the terminal disulfide variant, three residues were added to each terminus in order to allow the disulfide to be formed without distorting the Neo-2/15 structure. Bottom, experimental CD spectra of the design at 25 °C, 95 °C and then cooled back to 25 °C, showing complete recovery of ellipticity spectrum (full reversibility) upon cooling. c, Thermal melting of each disulfide variant in a and b between 25 °C and 95 °C (heating rate ≈ 2 °C min−1) was monitored using circular dichroism at 222 nm. Each of the disulfide-stapled variants shows improved stability relative to native Neo-2/15. d, Binding strength of each disulfide variant was measured by biolayer interferometry, showing that the introduction of disulfide bonds does not disrupt binding. Furthermore, both disulfide variants exhibit improved binding of IL-2Rβγc (Kd ≈ 1.3 ± 0.49 nM and 1.8 ± 0.26 nM for the internal and external disulfide staples, respectively), compared to Neo-2/15 (Kd ≈ 6.9 ± 0.61 nM) under the same experimental conditions. These results are consistent with the expected effect of disulfide-induced stabilization on a de novo protein binding site71. Thermal denaturation experiments were performed 3 times with similar results; binding experiments were performed once.

Extended Data Fig. 4 The stimulatory effect of Neo-2/15 on human CAR-T cells.

a, b, Human primary CD4 (top) or CD8 (bottom) T cells stimulated with CD3/CD28 antibodies (a) or unstimulated (b) were cultured in indicated concentrations of human IL-2 or Neo-2/15. T cell proliferation was measured as fold change over T cells cultured without IL-2 supplement. Experiments were performed 3 times with similar results. Data are mean ± s.d. c, NSG mice inoculated with 0.5 × 106 RAJI tumour cells were treated with 0.8 × 106 anti-CD19 CAR-T cells 7 days post-tumour inoculation. Tumour growth was analysed by bioluminescence imaging. The experiment was performed once.

Extended Data Fig. 5 Immunogenicity of Neo-2/15 in healthy naive mice.

a, Naive C57BL/6 mice were treated daily with Neo-2/15 (n = 10), KO Neo-2/15 (n = 5), mouse IL-2 (n = 5) or left untreated (n = 5). Blood was collected after 28 days and the serum was diluted 1:100 and analysed for IgG against Neo-2/15, mouse IL-2, human IL-2, KO Neo-2/15 and ovalbumin using ELISA. FBS (10%) was used as a negative control. Polyclonal antibody against Neo-2/15 was used as a positive control. All statistical comparisons between sera from treated mice and negative control serum were not significant (two-way ANOVA with a 95% confidence interval). All statistical comparisons between Neo-2/15 and mouse IL-2 treated mice serum were not significant (two-way ANOVA with a 95% confidence interval). The experiments were performed once. b, After 14 days, immune cell populations in the blood of treated mice were quantified by flow cytometry. B cell:T cell ratio (top right) was calculated by dividing the percentage of B220+ cells by the percentage of CD3+ cells. CD8+ cell:CD4+ cell ratio (top left) was calculated by dividing the percentage of CD3+CD8+ cells by the percentage of CD3+CD4+ cells. NK cells (bottom left) were identified by their expression of NK1.1. Results were analysed by one-way ANOVA (95% confidence interval). The experiments were performed once. In all cases, data are mean ± s.d.

Extended Data Fig. 6 Kinetics of STAT5 phosphorylation with Neo-2/15 treatment.

Naive C57BL/6 mice were treated once with 13 µg mouse IL-2 (n = 5) or 10 µg Neo-2/15 (n = 5), or were left untreated (n = 5). Phosphorylation of STAT5 was measured in peripheral blood at the indicated time points by flow cytometry using an anti-pSTAT5 antibody. Mean fluorescence intensity (MFI) is shown at each time point for TCRβ+CD8+ cells (top) and TCRβB220+ cells (bottom). Data are mean ± s.d. Results were analysed by one-way ANOVA (75% confidence interval). The experiments were performed once.

Extended Data Fig. 7 Conformational flexibility of Neo-2/15 in molecular dynamics simulations.

a, Molecular dynamics simulations started from the computational model of Neo-2/15 (top) converged into structures similar to the crystal conformation. Apo Neo-2/15 is shown in red thick tubes (chain A from PDB ID: 6GD6) and 45 (randomly selected) molecular dynamics conformations from 5 independent simulations are shown in thin grey tubes. Bottom, the plot shows the r.m.s.d. along 5 independent simulations (average r.m.s.d. = 1.93 Å). b, Similar to a, but for (control) molecular dynamics simulations started from the crystallographic structure of human IL-2. Top, crystal conformation of human IL-2 (chain A from PDB ID: 2B5I) is shown in blue thick tubes and 45 (randomly selected) conformations from 5 independent molecular dynamics simulations are shown in thin grey tubes (average r.m.s.d. = 2.02 Å). c, Top, similar to a and b, but showing molecular dynamics structures for simulations started from the computational model of Neo-2/15 bound to human IL-2Rβγc. The plot shows the r.m.s.d. along 5 independent molecular dynamics simulations (average r.m.s.d. to apo Neo-2/15 (model) = 1.28 Å). The lower structure shows the nearest conformation (to the apo Neo-2/15 computational model) that was sampled on each of the 5 independent simulations (structures from the first 50 ns of molecular dynamics simulations were not considered). Bottom, a 2D scatter plot (and the underlying density plot, in which yellow, blue, green and purple represent decreasing densities) comparing the r.m.s.d. (after discarding the first 50 ns of each simulation) for apo Neo-2/15 (computational model) versus the r.m.s.d. for the holo crystal structure of Neo-2/15 (in complex with the mouse receptor). The conformations sampled by Neo-2/15 when in complex with human IL-2Rβγc are more similar to the apo Neo-2/15 structure (computational model) than to the Neo-2/15 conformation observed in complex with mouse IL-2Rβγc. d, As in c, but for molecular dynamics simulations started from the computational model of apo Neo-2/15 in complex with the crystallographic structure of mouse IL-2Rβγc. The model of apo Neo-2/15 was generated by aligning (using TMalign) the ternary computational model of Neo-2/15 with human IL-2Rβγc (from c) into our crystallographic structure of mouse IL-2Rβγc (PDB ID: 6GD5) (average r.m.s.d. to holo Neo-2/15 (mouse) = 1.43 Å). Bottom, 2D scatter plot (and the underlying density plot, in which yellow, blue, green and purple represent decreasing densities) comparing the r.m.s.d. (after discarding the first 50 ns of molecular dynamics simulation) for apo Neo-2/15 (computational model) versus the r.m.s.d. for the holo crystal structure of Neo-2/15 (in complex with the mouse receptor). Unlike in c, the conformations sampled by Neo-2/15 when in complex with mouse IL-2Rβγc are more similar to the Neo-2/15 conformation observed in the crystallographic structure of the ternary complex of Neo-2/15 with mouse IL-2Rβγc (Fig. 3). For clarity, all the r.m.s.d. plots were filtered (running average filter, 5 frames = 100 ps), and points in the 2D scatter plots were subsampled every 25 conformations (that is, every 500 ps); however, the density plot corresponds to all the analysed conformations (that is, the last 40 ns of 5 molecular dynamics simulations that were analysed, and conformations were recorded each 20 ps).

Extended Data Fig. 8 Overall sequence conservation in binding residues for each of the four common helices, combining information from the three different de novo-designed IL-2 mimics.

Sequence logos were generated using combined data from binding experiments (using the heterodimeric mouse IL-2Rβγc, see Methods) from 3 independent SSM mutagenesis libraries for G2_neo2_40_1F_seq27, G2_neo2_40_1F_seq29 and G2_neo2_40_1F_seq36 (Supplementary Figs. 8–10). All of these proteins are functional high-affinity mimetics of mouse and human IL-2 (see Supplementary Figs. 6–11), some having topologies that differ from that of Neo-2/15, but all containing the four Helices H1, H3, H2′ and H4. The logos show the combined information for each helix independently. Below each logo, a line graph shows the probability score (higher means more conserved) for each amino acid in the Neo-2/15 sequence. The red line highlights positions where the Neo-2/15 amino acid has a probability score ≥30% (that is, these amino acids contribute more generally to receptor binding as they are globally enriched in the binding populations across all of the de novo IL-2 mimics tested). The topology of each helix in Neo-2/15 is shown left of each logo. The sequences of the Neo-2/15 helices and those of the corresponding helices (structurally aligned) in human IL-2 and IL-15 are shown below the graphs, highlighting the distinctiveness of the Neo-2/15 helices and binding interfaces.

Extended Data Table 1 Characterization of several de novo designed mimics of IL-2/IL-15
Extended Data Table 2 Crystallographic data table for monomeric Neo-2/15 and the quaternary complex of Neo-2/15 with mouse IL-2Rβγc

Supplementary information

Supplementary Information

This file contains Supplementary Figures 1-15, Supplementary Tables 1-9, Appendices A-D and Extended Acknowledgments.

Reporting Summary

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Silva, D., Yu, S., Ulge, U.Y. et al. De novo design of potent and selective mimics of IL-2 and IL-15. Nature 565, 186–191 (2019). https://doi.org/10.1038/s41586-018-0830-7

Download citation

Further reading

Comments

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.

Search

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing